Pump

Pump, Hydraulic Description
All hydraulic systems require a source of hydraulic power. In most applications, the source of hydraulic
power is a variable delivery axial piston pump. Variable displacement means that pump outlet flow
varies according to system flow demands (as more sources actuator, motors, etc. are operating,
the pump will increase output to maintain maximum pump outlet pressure). Other sources of hydraulic
power are vane or gear pumps (see Motors, Hydraulic Descriptionfor description of vane or gear
rotating pumps/motors), or an accumulator (see Accumulators, Hydraulic - Description). An aerospace
vehicles main hydraulic pumps are usually mounted on the engine and connected to the engine
rotating shaft through a gearbox. Pumps may also be driven by an electric motor, APU, ram air
turbine, or second hydraulic system (using a hydraulic motor and pump combination).
To understand how a variable delivery axial piston pump operates refer to the pump cross sectional
view shown in Figure 1. The figure shows pump components and also how pump outlet pressure is
controlled through a compensator valve and control piston arrangement. The key element in control of
pump outlet flow is control of the swashplate angle, , which in turn controls piston relative
displacement and hence pump flow. Ideally, the pump delivers zero flow when there are no flow
demands and the required flow when required while maintaining system design pressure at all times.
Variable displacement pumps dont obtain this ideal goal, but they do come close when flow demands
are within their design flow range.
Prior to the introduction of variable delivery pumps 50-60 years ago, pumps were fixed delivery which
means they always delivered the same amount of fluid irregardless of system flow needs. Unused
fluid was then ported back to the reservoir though a pressure relief valve. Fixed delivery pumps
wasted a lot of energy through heat. A fixed delivery pump has a fixed swashplate angle and therefore
no compensator valve or control piston (refer to Figure 1). Fixed delivery pumps may be found today
in very specialized applications where the fixed delivery (or flow) is tailored to the application to
minimize wasted energy. The benefit in this case is a lighter, less expensive pump. An example might
be a standalone hydraulic system to power a large cargo door where flow demands are constant while
the door is moving and the pump would only be on during door movement. A pressure relief valve
should also be installed in this application.

Figure 1 Simplified Pump Schematic

Referring again to Figure 1, pump flow rate is determined by the swashplate angle, , which in turn
controls pump piston displacement. Swashplate angle is controlled by the compensator valve and
control piston. The compensator valve sets the no flow outlet pressure of the pump (e.g., 3000 50
psi) and meters flow to the compensator piston. The compensator is fed hydraulic pressure from the
pump outlet and is positioned based on a force balance between compensator chamber pressure
acting on the piston and spring force (plus, to a lesser extent, friction and flow forces). The housing is
fixed and does not rotate. The housing must remain fixed so that inlet/outlet ports, compensator valve,
solenoids, and other equipment can be mounted to the pump. All other parts rotate at the pump
speed.
At the compensator pressure setting (e.g., 3000 50 psi for a 3000 psi system), the swashplate angle,
, is zero. As the pump flow increases, the pump outlet pressure decreases. As the outlet pressure
decreases, the compensator moves towards the closed position and the pressure on the control piston
is reduced. As the control piston pressure is reduced, the control piston spring pushes the control
piston so that the swashplate angle, , increases, resulting in greater piston stoke and increased flow
rate.
Figure 2 shows 3-dimensional view of a piston hydraulic pump. This figure provides a better view of
how a pump is built and how the pistons and swashplate operate. Not shown in the figure is a
compensator valve and control piston, so this is more representative of a fixed delivery pump. The
pistons are attached to the swashplate via a spherical bearing arrangement. The swashplate does not
rotate. As the cylinder rotates the piston is at the lower end of the cylinder during one part of a
revolution (intake) and at the top end of the cylinder during the other of the rotation (high pressure
outlet side of the pump). During 1 revolution of the swashplate, each piston will pull in fluid and push
out high pressure fluid once. For a nine piston pump, this will lead to 9 pressure pulsations per 1
revolution of the swashplate.

Figure 2 Pump Cross Section

The relationship of outlet flow to outlet pressure for a variable delivery pump is shown in Figure 3. This
plot can be used to estimate pump flow for a given outlet pressure. The plot also shows the flow rate
where pump outlet pressure starts to drop off dramatically. It is important to note that Figure 3
represents pump characteristics for a fixed pump rotational speed (RPM) or displacement (in
3
/rev).
Normally these curves are provided for the rated pump speed, but in aircraft engine speeds vary -
hence pump speed varies and maximum flow varies also. Therefore, the curve shown in Figure 3 will
shift down for lower pump speeds and shift up for higher pump speeds up to the maximum flow of the
pump.

Figure 3 Typical Flow vs Outlet Pressure Plot (for a given pump rotational speed)

The response of the pump to a change in outlet flow is on the order of 50 milliseconds.
Figure 4 shows the relationship between pump flow, efficiency and outlet pressure. The drop off in
flow occurs due to hydraulic fluid leakage at higher pump outlet pressures (higher delta pressure
across the piston seals), which is equivalent to volumetric efficiency. As volumetric efficiency drops off,
pump outlet flow drops by the same amount. The other curve in Figure 4 is the overall efficiency curve
(overall efficiency = volumetric efficiency x mechanical efficiency). Pump horsepower increases
linearly with pump speed.

Figure 4 Typical Performance Curves for a Variable Delivery Piston Pump

Pump Design Considerations
The most important characteristics for a hydraulic pump are listed below. These parameters assume a
variable delivery constant pressure pump.
Rated Pressure this is the nominal pressure setting of the pump and must be compatible with the
design operating pressure for the system (e.g., 3000 50 psi for 3000 psi system or 5000 50 psi for
a 5000 nominal psi system).
Rated Speed this is the nominal speed rating of the pump. The gearbox connecting the pump to the
driver (engine, APU, etc) will need to be compatible with the drive unit speed and the rated pump
speed. The minimum RPM of the pump may also need to be considered
Design Displacement this is the flow per revolution of the pump (in
3
/rev) that the pump is capable of
achieving without a significant reduction in outlet pressure.
Flow vs Pressure Curve this is a plot that has flow rate on the y-axis and pump outlet pressure on
the x-axis. This plot shows the drop off in pressure at a given flow conditions and shows where the
knee in the flow curve lies (see Figure 4). This graph is required for a simulation model. Nominally, this
graph is provided at the rated speed of the pump. If available, this graph for various pump speeds
would be helpful - otherwise the flow can be ratioed using the design displacement for different pump
speeds.
Temperature Rating the pump must be rated for the temperature extremes that it will see in
operation, such as engine nacelles. Pump seals are the most critical component when considering
temperature.
Case Drain Pressure this is the nominal pressure that would be in the pump case. All pumps have a
case drain line to provide a flow path to the reservoir for hydraulic fluid that flows by the piston seals
and fluid that flows through the compensator. Without a case drain the pressure would blow out a
case or shaft seal. The case drain pressure needs to be greater than the reservoir pressure (and line
resistance) to assure drainage from the case to the reservoir.
Inlet Line Size Standard pumps will have an inlet port sized by the manufacturer. The connecting
inlet line/hose will need to be compatible with the port size and type.
Outlet Line Size - Standard pumps will have an outlet port sized by the manufacturer. The connecting
outlet line/hose will need to be compatible with the port size and type.
Case Drain Line Size - Standard pumps will have a case drain port sized by the manufacturer. The
connecting line/hose will need to be compatible with the port size and type.
Recommended Inlet Pressure to operate properly a pump must be supplied sufficient hydraulic fluid
at a pressure level sufficient to fill the piston cylinders as the pump rotates. A pump manufacturer will
provide recommended inlet pressures and the reservoir and reservoir to pump hydraulic lines need to
designed/sized to meet this requirement. In some instances, a boost pump may be required to
achieve desired inlet pressures and flows.
Number of Pump Pistons most aircraft piston pumps have 9 pistons. An odd number of pistons have
been shown to have smaller output pressure fluctuations than an even number of pistons and
experience has shown 9 pistons to be an optimum number for performance.
Power Requirements what horsepower is required to drive the pump at its maximum operating
condition. Horsepower is the product of flow and delta pressure across the pump, divided by the
overall efficiency of the pump.
Maximum Fluid Viscosity Pump manufacturers will provide a recommendation for the maximum
recommended fluid viscosity. If fluid viscosity is greater than the recommended value, then the pump
may start to cavitate with the pump inlet at the recommended inlet pressure. Maintaining the fluid
within the necessary viscosity range will affect the pump inlet system design.
Seals Seals must be compatible with the specific type of hydraulic fluid used in the pump.
Specifically, seal material for carbon based fluids and synthetic fluids are different.
Filtration Requirements A pump manufacturer will provide recommended fluid cleanliness
requirements to help ensure reliable pump operation. This filtration requirement needs to be taken into
consideration when designing/sizing the return filters in the hydraulic system.
Weight Is always a concern on aircraft. A oversized pump is not desirable from both a weight and
cost standpoint.
Envelope engine nacelles and APU installations usually have limited space available and the pump
installation must be compatible with available volumes
Shaft Type Shafts are usually splined and the spline characteristics must be defined to ensure a
proper interface to the
Direction of Rotation Pumps can rotate either clockwise or counterclockwise and may be of a
concern in certain installations.
Mounting The bolt flange mounting of the pump must be compatible with the attachment on the
gearbox or other attaching plate.
Relief Valve A relief valve should be installed in the outlet (high pressure) line downstream of the
pump. This relief valve will provide protection against hydraulic shock loads, thermal expansion and
any possible overpressure condition. The relief valve setting should be 5-10% greater than maximum
pump pressure.

Pump Installation Considerations
Considerations for the mounting/installation of pumps include vibration, temperature, alignment of
drive motor to pump, spline matching and torque requirements. In some applications, a means of
quick installation and removal is required. Quick installation devices must uphold rigidity of the pump
installation and maintain alignment of rotating shaft.
Vibration Need to consider vibration from power source such as the engine or APU, vibration
characteristics from the pump, g-loading and possibly flutter. Mounting should be sufficient to
withstand these loads from both a stress and fatigue standpoint. Testing to appropriate levels from
RTCA-DO160 or MIL-STD-810, Method 510 should be conducted.
Temperature Due to the high speed and compression of fluid, pumps operate at high temperatures.
Additionally, the power source for the pump, such as the engine, is at high temperature. Temperature
considerations should include pump seals, fluid temperature, mounting pads, connecting hoses or
tubes, etc.
Pump/Motor Alignment Alignment of pump to drive motor splines needs to held to tight tolerances.
Considerations are tolerance stickups, relative motion between drive motor and pump, possible
angular displacements on installations, spline teeth dimensions, etc. Improper alignment can cause
excessive vibration (leading to premature failure), or failure of the pump shaft seal.
Splines Beyond alignment, spline wear is an important consideration. Usually some lubrication
(grease) is applied to the splines to minimize wear. Selection of lubrication should include
temperature, corrosion inhibiting and reasonable life of the grease before breakdown occurs.
Torque Requirements Both start-up torque and running torque should be considered. Start-up
torque is higher than running torque. Start-up torque accelerates the mass/inertia of the pump and
fluid, leading to temporary high stresses within the mounting hardware and pump. This is more of a
concern on APU and RAT installations. Obviously, the speed of the drive motor must match the
manufacturers recommended speed for the pump (usually accomplished through a gear box).
Axial and Radial Shaft Load Capability Ensures pump shaft and splines are adequately sized for
static and fatigue loads that the pump will see over its operating life.
Case Drain Line A case drain line is installed to drain pump leakage flow back to the reservoir. Case
drain back pressures affect seals and bearings (via load balance across them), balance and loading of
the pump rotating hardware, and piston leakage characteristics. In most aircraft installations, the case
drain line back pressure is equal to the reservoir pressure (which is approximately the pump inlet
pressure). This minimizes leakage on the intake side of the pump. Normally, back pressure from the
case drain line is not an issue, but if back pressures are abnormally high (such as clogged filter), the
effects on the pump should be looked at more closely. Case drain pressure < 150 psi is a rule of
thumb for good pump life. The case drain line must be large enough to cover the maximum case drain
flow at the nominal case drain pressure or even a slightly lower pressure. The case drain line will
normally flow back to return through the return line filters. In some applications, a separate filter is
used on the case drain line.
Inlet Line The inlet line to the pump is designed as part of the pump inlet system. The pump inlet
system consist of reservoir and reservoir pressurization and tubes/hoses from the pressure to the
pump. This system should be designed to ensure fluid is provided to the pump inlet within the inlet
pressure range and viscosity recommended by the manufacturer. In sizing the inlet line, the length of
the tubing, bends, height fluid is pumped, reservoir pressurization, additional components (such as a
heat exchanger) in the line and other factors should be taken into account. Sizing of a pump inlet line
uses basic pipe flow equations and reservoir (or supply) pressurization (see Reservoir, Hydraulic
Description).
Outlet Line The pump outlet line should be sized to system pressure drop requirements and to
minimize affects of pump pressure pulsations. Primary considerations in design of the outlet line are
pump pressure pulsations, accumulators (see Accumulator, Hydraulic - Description), pump system
response and parallel pump installation.
Regarding pressure fluctuations, hydraulic fluid has mass and is compressible. Hence the oil in the
pump downstream tubing behaves like a very stiff spring, with variable stiffness as the downstream
configuration changes. Pulsations are a result of each piston within the pump transferring a discrete
amount of fluid to the system, leading to a pulsed input in the hydraulic system. The flow pulses decay
over time from the damping provided by internal flow friction in the downstream tubing and
components. The pulsation frequencies for a odd numbered piston pump are

Example: For a pump running at 2700 rpm with 9 pistons

For aircraft systems, both pulsation frequencies are usually above the response frequency of the
downstream components, however, in some cases the effects may need to be analyzed. The lower
frequency is usually more dominant with pump noise, but both should be analyzed.

System Interaction
System interaction occurs when the natural frequency of the pump compensator is at or near the
natural frequency of a downstream component (such as a servo valve or actuator). Generally there is
sufficient difference between the natural frequencies so that system interactions do not occur.
Another source of interaction occurs when pumps are connected in a parallel arrangement. This
interaction can be stopped by installation of check valves in the outlet lines of each pump.

Pump Pulsation Damping
Several methods exist to dampen the effect of pressure pulsations from a pump:
1. Change configuration (geometry, parts, characteristics, etc.)
2. Increase volumes in pump outlet line
3. Install accumulator close to the pump. Some research shows that for the accumulator to be effective, it
should be installed with 0.3 meters of the pump, and the supply line between the main line and
accumulator should be between 5 and 10 centimeters in length. Also, the volume of the gas accumulator
should be sufficient so that its resonance (response) frequency is less than the pump pulsation
frequency.
4. Install a hose at the pump outlet, or downstream plumbing.
5. Install a Helmholtz resonator (H-filter) in the pump outlet line. A H-filter consists of two lines in
series, of different volumes, that branch away from the main line. By properly selecting the
lengths and cross-sectioal area of both lines, the H-filter can be tuned to a specific frequency.
6. Install a Quincke tube in the main line. The Quincke tube is a side line with areas based on main
line area and lengths sized for a specific frequency.

Accumulator, Hydraulic - Description
Accumulators store hydraulic energy and then provide this energy back to the system when required.
Accumulators store energy when hydraulic system pressure is greater the accumulator and provide
hydraulic energy when the accumulator pressure is greater than the system pressure. By storing and
providing hydraulic energy, accumulators can perform 5 basic functions for hydraulic systems:
- Supply oil for high transient flow demands when pump cant keep up
- Help reduce pump ripple and pressure transients
- Absorb hydraulic shock waves (due to valve closures or actuators hitting stops)
- Used as a primary power source for small (low demand) systems
- Help system accommodate thermal expansion of the fluid
Almost all aerospace hydraulic systems use accumulators for one of the above reasons. In fact, most
hydraulic systems use an accumulator to dampen pressure transients in the power generation system.
The pressure transients result from pump ripple, opening/closing of valves, actuators bottoming out
and so on. Some practitioners believe accumulators are over utilized and systems can be designed
without an accumulator in the power generation system. However, this has not been standard practice
and if an accumulator is not used, other design considerations should be considered. The selection
and design characteristics of accumulators will vary between the applications.
Hydraulic accumulators store hydraulic fluid under pressure. Pressure is supplied through a bag,
diaphragm or piston by either a spring, or pressured gas (most common). Accumulators are inherently
dynamic devices they function when configuration changes (actuators moving, valves opening, etc.)
are occurring within a hydraulic system. Accumulators respond very fast to configuration changes,
nearly instanteously for gas accumulators. The capability and affect of the accumulator is determined
by the overall volume of the accumulator and preload/precharge of the spring/gas
Gas accumulators take advantage of the fact that the gas (nitrogen) is compressible. A gas
accumulator has a gas precharge that is less than nominal hydraulic system pressure. As hydraulic
fluid enters the accumulator the gas is compressed to the nominal system pressure, which is an
equilibrium position and represents the maximum amount of energy stored by the accumulator. As
system hydraulic pressure drops, the gas will expand pushing hydraulic fluid back into the system.
The gas precharge level is an important parameter for gas accumulators since the precharge and
overall accumulator volume determine the maximum amount of hydraulic energy that will be available
to the system.
There are 4 types of accumulators: bladder, diaphragm bladder, piston (either spring or gas
controlled) and metal bellows. The choice of accumulator to use in a given application depends on
required speed of accumulator response, weight, reliability and cost. Pressurized gas accumulators
will have the faster dynamic response and are reliable. Metal bellows accumulators are very reliable,
but will not respond as fast as a pressurized gas accumulator. Accumulators with seals generally have
the lowest reliability.
Accumulators are either spherical or cylindrical in design. Bag, piston and metal bellows accumulators
are cylindrical. Diaphragm accumulators may be spherical or cylindrical. Accumulators are usually
manufactured into 2 halves which are either welded or threaded together. A fill port is installed at one
end of a gas accumulator and the hydraulic connection fitting (with poppet valve, if required) is
installed at the opposite end. For a spring accumulator, the non pressure side usually has a fitting that
connects to the hydraulic reservoir (for seal leakage and to alleviate back pressure on a piston).
Materials are usually steel, but accumulators may also be made from aluminum or a composite
(filament wound) material.
Bladder Accumulator
A bladder accumulator consists of pressure vessel with an internal elastomeric bladder with
pressurized nitrogen on one side and hydraulic fluid on the other side (system side). Figure 1 shows a
bladder accumulator with the 3 stages of operation, plus an overexpanded bag schematic. The
accumulator is charged with nitrogen through a valve installed in the top. The accumulator will be
precharged to nominal pressure when the pumps are not operating, shown in Figure 1a. When
nominal hydraulic system pressure is applied the bag will be compressed to its fully compressed state
as shown in Figure 1b. When the bag is fully compressed, the nitrogen pressure and the hydraulic
pressure are equal. As system pressure drops the bag expands, forcing fluid from the accumulator
into the system as shown in Figure 1c. As the bag expands, pressure in the bag decreases. The bag
will continue to expand until the bag pressure equals the hydraulic pressure (which will be lower than
nominal system pressure) or the bag fills the entire accumulator volume as shown in Figure 1d (an
undesirable situation). A poppet valve keeps the bag in accumulator from being pulled into the
downstream tubing should the bag overexpand. If the bag was pulled into the downstream tubing, the
accumulator would never recharge and normal flow from the pump would be constricted. The
maximum flow rate of the accumulator is controlled by the opening area (orifice) and the pressure
difference across the opening.
The main advantages of a bladder accumulator are fast acting, no hysteresis, not susceptible to
contamination and consistent behavior under similar conditions. Accumulators are easy to charge with
the right equipment. Because there is no piston mass, the speed of the bladder accumulator is
governed by the gas, which reacts very fast to changes in hydraulic system pressure. Hence bladder
accumulators are the best choice for pressure pulsation damping. Also, the bladder attachment
internal to the accumulator has proven to be very reliable in service. Of course there is always the
potential for bladder failure, which is a failure that would not usually be detectable in service. Also,
temperature differences on the gas will have some affect on performance. The main limitation of
bladder accumulators is the compression ratio (maximum system pressure to precharge pressure)
which is limited to approximately 4 to 1. Hence gas accumulators will be larger than other
accumulators for the same flow requirements. The precharge pressure is typically set to approximately
80% of the minimum desired hydraulic system pressure.

Figure 1 Bladder Accumulator Schematic

Diaphragm Accumulator
A diaphragm accumulator is similar to bag accumulator except an elastomeric diaphragm is used in
lieu of a bag. This would typically reduce the usable volume of the accumulator so the diaphragm
accumulator may not have volume capacity of a bladder accumulator. A schematic of a diaphragm
accumulator is shown in Figure 2. The behavior characteristics of a diaphragm accumulator are similar
to a bag accumulator.

Figure 2 Diaphragm Accumulator Schematic

Diaphragm accumulators behavior will be similar to a bladder accumulator and have the same
advantages and disadvantages. However a diaphragm accumulator may be spherical or cylindrical (or
possibly other shapes) which may be an advantage in some installations. The main difference with
bladder accumulators is an increased maximum compressions ratio (maximum system pressure to
precharge pressure) of approximately 8 to 1.
Piston Accumulator
A gas piston accumulator is shown in Figure 3. A gas piston accumulator has a piston which slides
against the accumulator housing on seals. On one side of the piston is nitrogen and on the other side
is the hydraulic fluid and connection to the system. A fill port allows pressurization of the nitrogen.

Figure 3 Piston Accumulator Schematic

A gas piston accumulator will not respond to transient pressures as fast as a bladder accumulator due
to the mass of the piston (frequency characteristics depend on piston mass and spring characteristics
of the nitrogen). However, a piston accumulator will have better damping due to hydraulic leakage
(viscous damping) and friction between the piston and housing (coulomb friction & seal friction). Piston
accumulators may also be more prone to leakage than other types of accumulators due to the seals.
Piston accumulators will generally provide higher flow rates than gas accumulators for equal
accumulator volumes. This is because piston accumulators can accommodate higher pressure ratios
(maximum system pressure to precharge pressure) than gas accumulators, up to 10 to 1, compared
with bladder accumulator ratios of 4 to 1. The disadvantages of piston accumulators are that they are
more susceptible to fluid contamination, have a lower response time than bladder (unless the piston
accumulator is at a very high pressure) and will have hysteresis from the seal friction. The precharge
for a gas piston accumulator is typically set to around 90% of minimum desired hydraulic system
pressure.
A schematic of a spring piston accumulator is shown in Figure 4.

Figure 4 Spring Controlled Accumulator Schematic
In a spring accumulator, the spring applies a force to a piston which compresses (or pressurizes) the
fluid in the accumulator. As normal system pressure, the spring will be fully compressed. As system
flow demands exceed the pump capacity, the spring will extend pushing the piston which in turn
pushes fluid into the adjoining pipe. Hence the accumulator supplements pump flow.
The maximum response time of the accumulator is set by the natural frequency, which is computed
using
(rad/sec)

Metal Bellows
Figure 5 shows a metal bellows accumulator. Metal bellows accumulators are used where a fast
response time is not critical yet reliability is important. Emergency brake accumulators are a good
application for metal bellows accumulators. A metal bellows accumulator is shown in Figure 5. The
metal bellows accumulator consists of a pressure vessel with a metal bellows assembly separating
fluid and nitrogen. The accumulator is similar to a piston accumulator, except a metal bellows replaces
piston and piston seals. Metal bellows accumulators are very reliable and long life components, and
have a proven service history. Metal bellows accumulators are pre-charged by supplier and then
permanently sealed leading to a maintenance free accumulator. Metal bellows accumulators will be
slow in responding to pressure changes due to increased mass of piston and bellows.

Figure 5 Metal Bellows Accumulator Schematic

Gas Accummulator Precharging
The precharge is the pressure of the gas in the accumulator without hydraulic fluid in the fluid side. A
gas accumulator is precharged with nitrogen gas when there is no hydraulic fluid in the accumulator to
the desired pressure. A rule of thumb for bladder accumulators is to set the precharge pressure to
approximately 80% of the desired minimum hydraulic system pressure. A rule of thumb for gas piston
accumulators is to set the precharge pressure to approximately 90% of the of the desired minimum
hydraulic system pressure.
The gas accumulator pre-charge is a very important variable for ensuring optimal accumulator
performance and maintaining long life of the accumulator. Too high of a precharge pressure and the
fluid volume capacity is reduced. Furthermore, if a bag accumulator charge is too high than the bag
may hit the poppet valve which could damage the bag through repeated hits in service, or cause a
fatigue failure in the poppet valve assembly. For a piston accumulator, the piston may be driven into
the stops repeatedly affecting seals or cause a fatigue failure in the piston stop. Too low of a
precharge pressure and the accumulator may not maintain desired minimum hydraulic system
pressure. Also a low precharge pressure will allow a piston accumulator to repeatedly hit the up
stops leading to premature failure of the accumulator. For a bag accumulator, the bag may be forced
into an unnatural shape (e.g., with folds) leading to bag damage and premature bag failure.
When sizing an accumulator the precharge pressure is an input to the sizing process. However, once
the accumulator is sized the minimum and maximum gas volumes should be computed (under worst
case conditions) and analyzed to ensure piston stops are not hit or that a bag cannot fully collapse or
expand completely in the accumulator.

Accumulator Design Considerations
The most important characteristics for hydraulic accumulators are listed below.
Accumulator Type as described above there are 4 basic types of hydraulic accumulators (bladder,
diaphragm, piston, and metal bellows). Each type has advantages and disadvantages and the
specifications will vary between the 4 types.
Accumulator Volume total volume of the accumulator (both gas and fluid volume)
Nominal Hydraulic System Pressure this is the nominal hydraulic system pressure in the system,
which will usually be the no flow rating of the hydraulic pump
Minimum Hydraulic System Pressure this is the minimum pressure that the accumulator must
maintain in the hydraulic system. This is a design requirement used to size the accumulator.
Precharge Pressure precharge is the pressure of the nitrogen in an accumulator without any
hydraulic fluid in the accumulator. The precharge pressure determines the amount of fluid that an
accumulator can hold at the system pressure and the desired minimum hydraulic system pressure.
Required Flow Rate to maintain minimum hydraulic system pressure, the accumulator must be able
to supply sufficient flow over a determined period of time. The required flow rate is a key requirement
that drives the size of the accumulator. The accumulator volume for hydraulic flow is equal to Q * t
(required flow rate times the time required for this flow). The accumulator must provide this flow when
the gas (or spring) is between the nominal hydraulic system pressure and the minimum desired
hydraulic system pressure. Note that the flow rate provided by the accumulator will be nonlinear
because as the gas expands the pressure drops off nonlinearly. This is a design requirement used in
sizing accumulators.
Output Volume Capability the output volume capacity of the fluid volume the accumulator is capable
of providing between the nominal hydraulic system pressure and the required minimum hydraulic
pressure. This volume must be provided at the required flow rate (see Required Flow Rate). The value
is also called the working volume.
Response Time this is the time for the accumulator to provide the desired fluid volume. The
response time times the output volume capability equals the flow rate of the accumulator. This will be
a function of precharge value and the flow opening (orifice) in the accumulator.
Recharge Time this is the time fully charge an accumulator from a fully drained state (i.e., at
minimum volume, which is the volume at the minimum desired hydraulic system pressure). This
should be evaluated when there is a fast duty cycle requirement. The recharge time will be the amount
of time for fluid to fill the accumulator based on the available flow rate from the pump (minus other
system demands).
High Frequency Cycling Capability only a concern when accumulators are used for damping of
pressure pulsations or very fast pressure transients. In this type of application, the frequency response
capability of the accumulator should be computed to ensure it is compatible with the transient
phenomena.
Fluid Type accumulator seals and elastromeric bladder/diaphragm material must be compatible with
the hydraulic fluid used in the system
Failure Modes the main failure modes for an accumulator will be failure of a bladder or piston seal,
or a pressure vessel burst. The affects of a loss of accumulator performance should be evaluated in
the hydraulic system to ensure no unacceptable affects may occur within the system. For a potential
pressure vessel burst, the installation should be reviewed with respect to surrounding components
and also for drainage of fluid and compartment ventilation.
Operating Temperature Range the behavior of the gas (nitrogen) varies with temperature.
Accumulator performance should be evaluated over the expected temperature range of the nitrogen.
Mounting Position vertical is always preferred with fluid outlet at the bottom. Horizontal installations
will tend to wear a bladder or diaphragm on the down side leading to earlier failures and lower
reliability. For piston accumulators, the seals will also tend to wear unevenly leading to earlier leakage.
If a non-vertical installation is required some evaluation of accumulator life should be accomplished
and the appropriate maintenance inspections (or life limits) put in place.
Mounting Flange Determine method of mounting accumulator is acceptable in your application and
that the mount is capable of withstanding all mounting forces, including crash g loads. Analysis should
use the mass of the accumulator when fully charged with fluid.
Connection Fitting the hydraulic interface fitting must be known so that a mating fitting can be
included in the design of the hydraulic system.

Applications of Accumulators
One of the main applications of hydraulic accumulators is to supply flow for brief periods of time when
a pump cannot keep up. A benefit of using an accumulator in this regard is that it allows the pump size
to be smaller. Usually the accumulator only assists during a worst case duty cycle or after a particular
failure has occurred in the system. This requires having an accumulator of sufficient volume to supply
the flow needs while still maintaining adequate system pressure. The approach to sizing an
accumulator for this application is shown in the sizing section (see Accumulators, Hydraulic Sizing).
Another application of accumulators is to damp pressure spikes from pumps or downstream
configuration changes (such as actuators hitting stops and valves closing). This is most often done in
the power generation portion of a hydraulic system, but accumulators can be put anywhere in the
system for pressure pulsation damping. In this application, as a pressure wave moves up and down
the piping, the energy is partially absorbed by the accumulator each time the wave flows by the
accumulator. Hence the wave damps out much faster than in a system without an accumulator.
Standard practice has shown this to be a proven technique, but no well proven design procedure
exists for both sizing and placement of accumulators for pressure pulsation damping. Hence some
experimentation may be required if an initial design does not achieve the desired results. One of the
reasons for experimentation is that laboratory research has shown pressure waves in pipe to be both
a function of time and location along a pipe. Thus at some locations along a tube there will only be
small changes in pressure magnitude (high peak to low peak) while at other locations the pressure
fluctuations (high peak to low peak) will be much larger.
A secondary function of accumulators is to absorb volume changes in fluid due to temperature
fluctuations. If an accumulator is not used and a rise in temperature increases pressure above system
pressure, then the fluid must flow through a thermal relief valve to the reservoir. This is wasted flow
and hence results in wasted energy thereby decreasing system efficiency. Furthermore, a pressure
relief valve exhibits hysteresis and must flow a sufficient fluid so that pressure drops below the level
where the valve will close (which could be a significant flow amount). With an accumulator, reasonable
volume changes can be accommodated without having flow to the reservoir. Computing accumulator
size to accommodate temperature variations is relatively straightforward.
Of course, an accumulator can be sized and installed to do multiple functions. Therefore, a single
accumulator can perform any or all of the above functions. The size and type of accumulator chosen
will depend on the functions that accumulator is addressing.
Lastly, the loss of pressurized gas in a sealed accumulator (or spring failure in spring accumulators) is
generally a latent failure. This latency may be an issue when conducting a safety analysis on a system
where the accumulator plays an important role (such as emergency gear extend or emergency
braking). In this case, an acceptable functional test procedure will usually need to be implemented at
an appropriate interval.

Check Valve, Hydraulic, Description
The function of a check valve is to prevent flow in one direction and
allow flow in the other direction. Check valves commonly use a poppet
and light spring to control flow as shown in the figure below. If P
1
A
1
>
P
2
A
2
+ spring force + friction, then flow occurs in the direction of the
arrows. If P
1
A
1
< P
2
A
2
+ spring force + friction, then the poppet would be
pushed to the left, against the stop, prohibiting flow in the reverse
direction.

Figure 1 Simplified Check Valve Schematic
The most common method for designing a check valve is illustrated in
the Figure 1. Different manufacturers may utilize other design
approaches. For example, another type of check valve is a ball that
pushes against a spring. Operation is similar to the check valve shown
in Figure 1 except a ball replaces the piston.
Check valves are used in hydraulic systems anytime flow in a selected
direction is not desirable or may create a problem. Check valves are not
used in bidirectional flow lines, such as to and from actuators. Some
examples where check valves are used are
Return lines - to prevent a leak
upstream of the reservoir from draining
the reservoir
Return lines to prevent a pressure
spike from migrating back up a return
line to a component (this is especially
important where actuators have
mechanical locks and a pressure spike
in the return line could cause the lock to
disengage)
Ground Service lines to ensure flow
only flows in the proper direction when
servicing and also to prevent a ground
service line leak from draining the
system
Pressure lines to ensure no reverse
flow through a hydraulic motor
Low Pressure (Feed) Line to a Pump
to prevent fluid in the pump from
flowing back to the reservoir without
going through a filter
Charged Accumulator Lines - check
valves can be used to trap pressure in
a given volume and maintain the
charge pressure for a specified time
interval. An example of this application
would be using a pump to charge an
emergency brake accumulator through
a check valve. In this case, when the
pump is turned off pressure is
maintained downstream of the check
valve and is available for emergency
braking (see Figure 2).

Figure 2 Charged Brake Accumulator using Check Valve

Power Control Units (PCUs) to
prevent backflow out of a PCUs supply
pressure port should the supply
pressure to that PCU fail. This is
usually required to ensure hydraulic
fluid is maintained in the PCU in order
to prevent lost of the actuator stiffness.
This is generally required on PCUs that
drive flight control surfaces where flutter
is of concern. The inlet check valve is
generally used with a return line
pressure compensator that will maintain
a low pressure (50-150 psi) in the PCU
internal porting for a given amount of
time following the removal of the inlet
supply pressure to the PCU.
Parallel Pump Installations to prevent
outlet flow from one pump at a slightly
higher pressure from flowing to the
other pump (see Figure 3). In this type
of pump installation, one of the two
check valves is designed to open at a
slightly higher pressure than the other.
For example, say the left pump is
powered by the aircrafts engine and
the right pump is power by an electric
motor. In this type of application the
engine driven pump generally has a
larger flow capacity over that of the
electrical driven pump. The electrical
driven pump is intended as a backup
for the engine driven pump. In this case
the check valve on the electrical driven
pump could be set to open (crack)
about 50 psi more than the engine
driven pump. This will ensure the
engine driven pump supplies the
system leakage flow during periods of
no hydraulic flow demands and will also
help reduce electrical motor noise
variability.

Figure 3 Use of Check Valves in a Parallel Pump Arrangement

When considering the use of a check valve, the following factors should
be evaluated:
Pressure Rating make sure the valve is rated appropriately for the
system pressure
Regulation Range what is the minimum and maximum octop
required to go into and out of the checked flow position
Pressure Drop Across the Valve this will affect design pressure
available to downstream components and thus, the sizing of those
components
Temperature Rating valve should be rated for fluid temperatures and
applicable environmental temperatures
Valve Materials valve should be sufficient to pass proof and burst
testing, not be susceptible to corrosion and other environmental
considerations, operate properly under temperature extremes
Seals/Clearances affects overall reliability of the valve. Some valves
may not use seals and will maintain tight clearances between piston and
housing to minimize leakage through the valve. The design
characteristics can be affected by environmental conditions and
aging/wear over time.
Leakage what is the leakage through the valve in the checked position
under all environmental conditions?
Failure Modes the dominant failure modes consist of the piston
jamming in either the open or the closed position. Clogging is also a
possibility.
Chattering valve should be evaluated for potential to exhibit chattering
or limit cycle behavior under certain upstream or downstream
conditions. This will be a function of the natural frequency of the servo
and the damping.

Directional Control Valve, Hydraulic - Description
Hydraulic valves function to control pressure, control flow or direct flow in response to external
commands. Directional valves are valves that direct flow in response to external commands.
Directional valves are usually servos (see servos) where the servo is positioned in response to
solenoids, torque motors or mechanical input. Directional valves do not provide flow or pressure
regulation and functional only to direct flow (much like a switch). Sometimes directional valves are
packaged with other components such as orifices or check valves. This has the advantage of
combining several functions into 1 assembly (or 1 part number) to simplify installation. Directional
valves are either open or closed (in 1 position or another). Directional valves do not utilize a spool in a
sleeve design, but quite frequently use this configuration. Also, the spool may be zero lapped or have
underlapping or overlapping designs.
Directional valves are referenced by the number of positions the spool will take (2, 3 or 4 positions are
typical) and the number of hydraulic ports in the valve (2 way, 3 way and 4 way are typical). Examples
are shown below.

Two Way, Two Position (2/2) Valve

In a two way, two position valve, the servo can be in one of two positions and the two ways because
there are 2 fluid ports in the valve (or, if you prefer, the valve housing). Although a spool arrangement
is shown, any type of check valve could be considered a two way, two position valve.

(Two Positions Shown)

Three Way, Two Position (3/2) Valve

In a three way, two position valve, there are three inlet/outlet ports in the valve and the spool can be in
one of two positions. A 3/2 valve would be used to allow fluid flow into or out of actuator or motor.

Four Way, Two Position (4/2) Valve

In a four way, two position valves there are four inlet/outlet ports in the valve and the spool can be located in one of
two positions. For 4/2 valve fluid is always flowing through the valve with system pressure supplied to one of the two
outlet ports at all times. The other port would then be ported to return. 4/2 valves would normally be used in hydraulic
systems in conjunction with an upstream shut valve (or 2/2 valve). In this case a 4/3 valve usually makes more sense.
However, 4/3 valves can be found in power control units (PCUs), where a shutoff valve is installed in the PCU where
a shut valve is not packaged with the 4/2 valve due to other design considerations in the PCU.

Four Way, Three Position (4/3) Valve

In a four way, three position valve, the spool is in one of three positions and there are 4 inlet/outlet
ports in the valve. In the midstroke position there is no flow through the valve. A good application of a
4/3 valve is actuator control, where the actuator control goal is to extend, retract or hold a position. 4/3
valves are used in servovalves, where the spool is controlled by a flapper valve or a jet pipe valve.

When specifying a directional control valve, the following parameters should be evaluated:
Pressure Rating make sure valve is rated for your system pressure
Pressure Drop this is the manufacturers pressure drop at a rated flow through the valve. There may
be a tolerance on the pressure drop which may need to be evaluated
Pressure Drop Variance how does the pressure drop change for non-rated flow conditions?
Flow Control A directional flow may also incorporate flow control. This would normally be
accomplished through port sizing or putting an orifice in the outlet flow. If flow control is part of the
valve, then the method of flow control should be ascertained and the accuracy and tolerance of the
flow for design and off-design conditions should be evaluated.
Temperature Rating valve should be rated for fluid temperatures and applicable environmental
temperatures
Actuation Time time to move from open to closed may be important in some applications
Valve Materials should be sufficient to pass proof and burst testing, not be susceptible to corrosion
and other environmental considerations, and not cause any problems under temperature extremes
Seals/Clearances affects overall reliability of the valve. Some valves may not use seals and will
maintain tight clearances between spool and housing to minimize leakage across the servo pistons.
The design characteristics can be affected by environmental conditions and aging/wear over time.
See Seals - Hydraulic Components for discussion on seals.
Method of Spool Position Control the directional valve can be controlled by a solenoid, a linkage, a
torque motor, a pneumatic source or a hydraulic pressure source. This source needs to be evaluated
under all foreseeable conditions to ensure if will open and close the valve as required. Additional
specifications will likely be required for the control element.
Leakage does the valve have high or significant leakage levels? Leakage is wasted energy.
Chattering a directional valve should be evaluated for potential to exhibit chattering or limit cycle
behavior under certain upstream or downstream conditions. This will be a function of the natural
frequency of the servo as well as the damping and friction levels. See Friction - Hydraulic
Components for further discussion of friction characteristics.
Failure Modes the main failure mode jam in any position from full closed to full open and
contamination.

Flow Characteristics
Flow ratings (pressure drop for a rated flow) for directional valves are normally provided by the
manufacturer. Directional valves may incorporate provide flow control through either outlet port sizing
or an installation of an orifice in the valve outlet. Flow behavior would follow the orifice flow equation
(see Orifice Flow Hydraulic). For the affects of valve lapping on valve flow see servo section.

Component Qualification
See Qualification - Hydraulic Components for discussion on directional valve qualification and required
certification testing.

Pressure and Flow Characteristics for a 4/3 Valve

Reference the 4/3 directional schematic above, note the flow areas from P
s
to P
A
and P
B
to P
r
are
equal (matched and symmetrical valve). Ignoring leakage through the servo piston, the flow rates are
characterized by the orifice flow equations,
(1)
(2)
The flow areas A
1
and A
2
are functions of valve position, x
v
. The load pressure drop and load flow are
given by
(3)
(4)
Since Q
1
= Q
2
, equations (1) and (2) can be combined, such that
(5)
Since return pressure is small compared to system pressure, we can let P
r
= 0, leading to equation (6)
(6)
Equations (3) and (6) can be combined to obtain
(7)
(8)
Equations (7) and (8) relate P
A
and P
B
to supply pressure and load pressure. If the load pressure is
zero, P
A
and P
B
are equal to of the supply pressure.
For the load flow rate, equations (7) and (8) along with equations (1) and (2) can be substituted into
equation (4) to yield an equation for load flow in terms of supply pressure and load pressure. The final
equation is
(9)
This equation assumes P
r
is negligible. Also, note that A
1
= A
2
.
The above relationships for P
L
and Q
L
are important for understanding theoretical pressure flow
characteristics of a four way, three position valve where the servo can be positioned in any position
along the bushing (infinitely variable flow area). These equations are used is the discussion of
hydraulic servovalves (seeServovalve, Hydraulic Description).

Filters, Hydraulic - Description
Filters are necessary in hydraulic systems for filtering out contamination and debris. Contamination
and debris in hydraulic systems come from many sources. Contamination includes metal flakes, glass,
ash, lint, various fibers, rubber, sand, etc. Contamination is very small generally in the 1 100
micron range (1 micron = 1.0E-06 meter). Over time contamination can impact component reliability
and performance of components. The purpose of hydraulic filters is to remove contamination from a
system and keep the fluid cleanliness within design tolerances.
During the formation of the hydraulic oil, the fluid will not be pure and contamination will be present.
Therefore, when put into the system there will be some amount of contamination in the fluid. This type
of contamination includes dust and ash plus small particulates/residues from processing equipment.
Small amounts of water may also be present in the fluid. During storage and shipping, dust, paint,
chips of various materials and possibly paint can be introduced into the fluid.
Once installed in a vehicle, the main sources of contamination is contamination that exists in
components and tubes (from manufacture or introduced during installation) and wear that occurs
during normal operation. During manufacture and assembly a common practice is to flush
components to remove any debris arising during the manufacturing process. On airplanes, fluid lines
can be flushed prior to connection of the hydraulic components. This is also a common practice. The
greatest source of in-service contamination is the hydraulic pump. Pumps operate at high speeds in a
severe environment and contamination (metal filings and seal pieces) from the rotating components
occurs in every pump. This contamination makes its way into the system through the case drain line,
which is why case drain fluid must go through its own filter or through the return filter. Should a pump
lose inlet fluid the pump runs dry and contamination from the pump increases dramatically. Normally
after a failure of this nature, all filters are replaced in a system. Other components with moving parts
will also create contamination, but usually at a rate less than a pump. Another source of contamination
is hydraulic ground power carts. Sometimes carts are not well maintained or the filter on the cart is not
to the level required by the airplane. This is why ground service connections always flow directly to the
filters before entering the system.
The overall fluid cleanliness for a hydraulic system is defined by a Class number. Class numbers run
from 1 to 12. The Class number is determined by the number and size of particulates (contamination)
in the fluid. Class number is determined through a laboratory analysis of a representative fluid sample
from the system. The lower the Class number the cleaner the fluid. Aerospace vehicles are usually in
the 7 to 9 class range. Table 1 shows the maximum amount of particles for a given particle diameter
for Class 6, 7, & 8 systems. As shown in the table, Class 7 system will have 32,000 particles/100 mL
in the 5 to 15 m range, 5,700 particles/100 mL in the 15 to 25 m range, 1012 particles/100 mL in
the 25 to 50 m range, 180 particles/100 mL in the 50 to 100 m range, and 32 particles/100 mL in
> 100 m range. Going to Class 6 divides the number of particulates by a factor of 2. Going to Class 5
halves the allowable particulate count from the Class 6 level (or divides by 4 from the Class 7 levels).
Similarly a Class 8 system is allowed twice as many particulates in each size range as a Class 7.

Particle
Size
Class 6 Class 7 Class 8
5 to 15
m
16,000 particles/100
mL
mL
mL
15 to 25
m
2,850 particles/100
mL
5,700 particles/100
mL
mL
25 to 50
m
506 particles/100 mL 1012 particles/100
mL
2024 particles/100
mL
50 to 100
m
90 particles/100 mL 180 particles/100 mL 360 particles/100 mL
> 100 m 16 particles/100 mL 32 particles/100 mL 64 particles/100 mL

Table 1 Particle Size for Class Rating

A simple filter would consist of fine mesh screen or more appropriately, a number of fine mesh
screens put in series so that the fluid has to flow though many meshes. Other methods to filter
contaminates would be a membrane, woven wire cloth element, synthetic fiber, cellulose, micro-glass
and metal edge element. A mesh filter is shown in Figure 1. In this example, the filter assembles into a
container which threads into a filter manifold (not shown). A filter manifold is a machined housing
containing pressure, return and case drain (if used) filters. Inlet and outlet ports are also included in
the housing for connecting the appropriate pressure and return tubes. The advantage of the filter
manifold is all filters are contained in one location for ease of inspection and maintenance.

Figure 1 Mesh Filter

A filter schematic is shown in Figure 2. This schematic shows a common filter arrangement and the
flow paths through the filter. An additional feature is shown in Figure 2. This feature is a high p pop
up switch or button. During normal operation low p across the filter the delta pressure across the
switch is not sufficient to overcome the spring force and the button remains recessed. If the pressure
drop across the filter increases to a predetermined level high p across the filter the delta
pressure across the switch piston will overcome the spring and push the button up. This is illustrated
in the enlarged portion in Figure 2. In lieu of a pop button, an electrical switch can be used. With an
electrical switch the high p piston will actuate a switch which will provide indication to a ground
service panel and/or flight compartment.

Figure 2 Hydraulic Filter Schematic with p Pop Up Indicator

The main considerations of a filter are filtration level, efficiency rating and flow versus pressure drop
characteristic. Filtration capability is listed as a micron level. So a 5 micron filter will filter out particles
with a diameter or width 5 microns or greater. Particles smaller than 5 microns will be able to flow
through the filter. Filters can be rated as 100% efficient which means that all particles equal to or
greater than the filter rating will be caught in the filter. If the rating is less than 100% than some
particles in the filters range will make it through the filter. Efficiency ratings are sometimes given in
terms of a beta rating. Beta ratings are defined in the following way: For a 5 micron filter, a beta rating
of 200 means that for every 200 particles > 5 microns only 1 particle greater than 5 microns will make
it through the filter ( = 200/1 = 200) and a beta rating of 2 means that that for every 200 particles > 5
microns 100 particles greater than 5 microns will make it through the filter (=200/100=2). Thus,
higher beta ratings imply a more efficient filter. Filters that are 100% efficient (i.e. = ) are referred to
as absolute filters.
Since filters must inherently restrict flow there will be a pressure drop (flow resistance) though the
filter. As a general rule, as filter efficiency goes up (i.e., more and/or finer meshes) pressure drop will
increase. Also, as filters get dirty flow resistance goes up. When evaluating a filter in the system the
pressure drop versus flow rate characteristic should be evaluated. The characteristics should be
evaluated at the most dirty condition (i.e., just prior to filter indication). Hydraulic filters are often
manufactured as a filter manifold assembly. This means that more than one filter may be part of the
filter manifold. Also, the manifold allows easy access to the filter to support periodic filter
replacements. Old filters may be discarded. In some filters, it may be possible to send filters back to
the manufacture for cleaning and recycling.
Filters normally have a indicator as part of the assembly (see Figure 2). The indicator will use a delta
pressure device that senses p across the filter. When the p is sufficiently high the indicator piston
will move to a position where a switch is closed or visual indication shows. The indicator alerts
maintenance personnel that the filter needs replacement. Filter indications are often checked on a pre-
flight walk around by flight crew and maintenance personnel. If a switch is used, the switch can be
connected to crew compartment indication. In some bypass filter designs, an indication is provided
when the bypass valve goes into bypass mode.
A filter may include a filter bypass provision, as shown in Figure 3. A filter bypass allows fluid to flow
around (bypass) the filter when the p across the filter is high (i.e., filter is clogged). A bypass
functions similar to a priority valve where the p overcomes a spring when the p is sufficiently high.
A good practice is put bypass provisions on filters in the return line filter. Generally it is preferable to
have dirty fluid flow to the pump rather than no flow. No flow to the pump will require the pump to be
replaced and will put much more contamination into the system as the pump starts to run dry. Bypass
provisions on pressure lines is generally not required nor preferred. If a bypass provision is provided in
the pressure line filter then when the filter is clogged and goes into bypass contamination will make its
way to all downstream components. This can affect performance of many components in a system
and lead to (expensive) replacement of numerous components in a system. If a filter is used on a case
drain line, a good practice would be to install a filter with a bypass provision. This will prevent
excessive case drain back pressure and premature pump failure.

Figure 3 Hydraulic Filter with Bypass Valve

Another type of filter is shown in Figure 4. This filter is an inline filter and would be representative of a
filter used to filter flow just upstream of a hydraulic component, such as a PCU.

Figure 4 Inline Filter

Filter requirements come from component cleanliness requirements. can be put on main system
filter or a separate filter can be put in the lines flowing to the component or internal to the component
(as is often the case with PCUs). As discussed in System Design, Hydraulic Power Generation, a
table should be built that lists the cleanliness requirements for each component. The worst case
requirement should be used to select an appropriate filter.
Tighter filter requirements will lead to a cleaner system and a lower Class rating. However, there is no
standard relationship between filter capability and Class rating. To determine Class rating for a given
hydraulic system, a fluid sample will need to be taken from the system and sent to a certified lab for
evaluation. Care should be taken to not introduce additional contamination into the fluid when
gathering the sample. Statistical sampling methods are recommended due to variances in vehicle
operation and lab methods. There will likely be a difference in Class rating based on the length of time
the vehicle has been in service.

When selecting a filter, the following factors should be evaluated:
Maximum Capacity Rating capacity rating is the maximum flow that can be accommodated by the
filter at the rated pressure drop
Maximum Pressure Rating pressure rating is the maximum pressure that will be seen by the filter in
service. Return pressure spikes should be evaluated when determining the maximum pressure
requirement. Proof and burst pressures will be based on the maximum pressure requirement.
Flow vs. p Characteristics filter manufacturers can provide a flow vs. p curve for the filter. This
curve should be evaluated to ensure that the filter will not restrict flow up to the maximum p (clogged
filter) indication. If a simulation model for the return system has been constructed this curve should be
used for the filter model (see also Filter, Hydraulic Sizing).
Maximum Temperature filter should be rated for the maximum combined fluid and environmental
temperatures expected in service. This will normally be determined by the seal material used in the
filter.
Minimum Particle Size Rating what is the filter cleanliness capability in microns (e.g., 5 micron
filter)?
Efficiency what is the efficiency rating of the filter, in terms of a rating or whether is absolute (100%
efficient)?
Filter Element (Cartridge) Type will filter be wire mesh, membrane, woven wire cloth element,
synthetic fiber, cellulose, micro-glass or metal edge element? What are specific maintainability issues
associated with available filter types? Is the filter compatible with your fluid? What is the expected life
of the filter? The pros and cons of available choices in your application should be reviewed with the
filter manufacturer.
Reusable Element is the selected filter element reusable after cleaning or is the filter disposable? If
reusable, what is the method for cleaning?
Container (Housing) what is the material used for the housing? Is the material corrosion resistance?
Clogged Filter Indication will a high p (clogged filter) indicator be used? If so, what are the design
characteristics? What type of indication is provided manual pop up button or electrical switch?
Bypass Provision will the filter contain a bypass provision? If so, what are the design features of the
bypass valve? Does the bypass valve provide indication independent of a clogged filter indicator?
Seal Type & Material what type of seal is used in the filter? Is the seal material compatible with the
fluid and operating temperatures of the fluid?
Fittings the inlet and outlet tubing fittings need to be compatible with the fittings on the filter.
Normally, these are threaded fittings.
Mounting the method of mounting or supporting the filter within the system should be robust and not
allow any loads to be imparted to the connecting tubes.

Filter Qualification
See Qualification - Hydraulic Components for discussion on hydraulic filter qualification and required
certification testing. Temperature, proof and burst pressure, vibration, impulse pressure and
endurance testing would be important.
Flow Control Valve, Hydraulic - Description
The two methods of controlling flow rate in a hydraulic circuit are (i) using a fixed orifice and (ii) using a
flow control valve.
For accurate flow control, a device that regulates to a p across an orifice is required referred to as
pressure compensated flow control valve. Figure 1 shows a simplified pressure compensated flow
control valve.

Figure 1 Simplified Schematic of a Inlet Pressure Regulated Flow Control Valve

In this valve, the p across the flow metering orifice is maintained at a constant value producing a
constant flow rate. The general orifice flow equation indicates that holding the orifice area and p
constant, where the fluid properties (bulk modulus and density) are relatively constant, will yield a
constant flow rate through the orifice. To hold p constant, the upstream and downstream pressures
are ported to different sides of a servo (piston). An adjustable spring assists the lower pressure to hold
the valve open at low input pressures. As thep (force balance) across the servo varies the flow
opening by the servo, the metering orifice inlet pressure is regulated. Hence as P
1
increases (or
P
2
decreases), the servo moves to the left and reduces the servo flow area. And as P
1
reduces (or
P
2
increases), the servo moves to the right thereby increasing the flow area. The flow control valve
shown in Figure 1 modulates P
1
to control flow. Calibration of the flow control valve is obtained by
adjusting the metering orifice. The spring preload may also be adjustable.
A second example of the flow control valve is shown in Figure 2. In this valve, the servo modulates P
2
.
However, the overall function of the valve is similar to the valve in Figure 1. Regulating P2 may be an
advantage over regulating P1 if the servo port in Figure 1 could become the controlling orifice (flow
area becomes smaller than the metering orifice). In this case the servo port opening would be
controlling flow.

Figure 2 Simplified Schematic of a Outlet Pressure Regulated Flow Control Valve

Fixed Orifice vs Flow Control Valves
A comparison between the use of orifices and pressure compensated flow control valves is shown in
Figure 3. The orifice flow varies dependent on (p). The amount of variation seen in a practical
application of orifice flow depends on the range of p seen over the operating range of the orifice in
the system. As shown in Figure 3, the flow control valve holds flow constant over a wide range of p
(from min regulation p to maximum p).

Figure 3 Notional Graph Showing Flow Characteristics for an Orifice and Flow Control Valve

For an orifice (see Figure 4), flow is governed by the orifice flow equation (seeOrifice Flow - Hydraulic)
(1)
As can be seen in Equation (1), for a fixed p flow can be controlled by controlling area. In practice,
P
1
and P
2
are never constant and therefore orifices do not provide constant flow rates over all
operating conditions (system pressure, downstream pressures, temperatures, etc.). Nevertheless, in
many applications, fixed orifices can be sized to limit flow under the worst case condition and the
accuracy of a simple orifice is sufficient. As an example, simple orifices are common in landing gear
actuator circuits, where the time to retract or extend the gear can be in the range of 6-10 seconds. In
this case, an orifice can be sized to maintain the 6-10 second requirement under all operating
conditions.
In Figure 3, note the flow rate for a flow control valve is constant over a wide pressure range.
However, as the input pressure range varies so will the output pressure. This will affect the inlet
pressure to a downstream component. So, while the flow to a component (actuator or motor)
downstream of the flow control valve will be constant, the inlet pressure to the component will change.
This will affect the p across the component and hence the power output of the component (Power
= p x Q).

Sharp Edge Orifice

Short Tube Orifice

Figure 4 Sharp Edge and Short Tube Orifices

When considering the use of a flow control valve, the following factors should be evaluated
Pressure Rating ensure valve is rated for your system pressure
Regulated Flow Rating should be in the range desired for the specific application
Regulation Range What is the minimum and maximum p required across the orifice for regulation?
For example, one flow control valve manufacturer has a maximum p of 3000 psi and a minimum p
of 100 psi across the metering orifice to maintain flow regulation within the flow tolerance.
Flow Tolerance Ensure tolerance on flow regulation is sufficient for your application. For example,
some valves will control flow to within 10% of the setting. Tolerances result primarily from friction
forces on the servo. When using flow control valves in parallel systems, such as thrust reversers or
ground spoiler systems, flow tolerances should be sufficiently tight to avoid undesirable asymmetric
operation between the two actuators (this may also be affected by piping differences lengths &
bends - between the two actuators and source manifold).
Pressure Drop Across the Valve in the Regulation Range This will affect design pressure available
to a downstream component and will affect sizing of that component. Keep in mind that while flow is
constant over a wide pressure range, the outlet pressure of the flow control valve can vary
significantly. This will affect available power to the downstream component.
temperatures
Valve Materials valve material(s) should be sufficient to pass proof and burst testing, not be
susceptible to corrosion and other environmental considerations, and operate properly under
temperature extremes
Failure Modes the dominant failure modes in the flow control consist of the servo valve jamming in
any position from full closed to full open and degraded performance due to contamination. It may also
be possible for the adjustable device on the metering orifice to fall out of adjustment leading to lower
or higher flow regulation settings.
Chattering valve should be evaluated for potential to exhibit chattering or limit cycle behavior under
certain upstream or downstream conditions. This will be a function of the natural frequency of the
servo as well as the damping and friction levels. See Friction - Hydraulic Components for further
discussion of friction characteristics.
Hysteresis how does the flow regulation change when approaching a control point from a low
pressure condition or high pressure condition? Hysteresis affects flow control accuracy and
contributes to chattering.

Flow Control Valve Qualification
See Qualification - Hydraulic Components for discussion on control valve qualification and required
Hydraulic Fluid - Properties
Fluids, either liquids or gases, are bodies without shape. Fluids experience great changes in shape under applied
forces unless constrained in some manner. The behavior of fluids is characterized by the fluid properties listed
below.
Fluid Pressure Normal tension on the surface element of a fluid
(psi) (1)
Fluid Density Fluid mass per unit volume. Density is a function of both pressure and temperature.
(lbm/ft
3
) or (lbf- sec
2
/ in
4
)
Viscosity - During the movement of fluid, there is a tangential force that resists movement, called viscosity.

Referring to the above figure, suppose 2 fluid layers are moving at a distance apart of dy, at a relative velocity of
dvx. Shear stress occurs between the fluid layers and is given by
(psi) (2)
is the proportional constant called dynamic viscosity, with units lbf-sec/in
2
called reyn. is also called absolute
viscosity or the coefficient of viscosity. varies significantly with the type of fluid and fluid temperature. also
varies with pressure but the effects are smaller.
Kinematic Viscosity - The ratio of coefficient of dynamic viscosity to fluid density is called the coefficient of
kinematic viscosity
(in
2
/sec) (3)
Density The density, , of a fluid is the mass per unit volume
(lbm/ft
3
) or (lbf - sec
2
/ in
4
) (4)
Density is a function of both pressure and temperature

Bulk Modulus - Expanding the above equation for density in a Taylor Series (for 2 variables)

(5)
Mass density increases as pressure increases and decreases as temperature increases as the sign of
is positive and the sign of is negative.
Assuming constant temperature so that = 0, then equation (5) becomes
(6)
The quantity
(psi) (7)
is the change in pressure divided by a fractional change in volume at constant temperature. The minus sign
indicates a volume decrease with pressure increase. is called the isothermal bulk modulus (or simply bulk
modulus) since it was derived assuming constant fluid temperature. The bulk modulus represents fluid
compressibility and has a significant effect on the performance of hydraulic systems. Effects of pressure on bulk
modulus are large and effects of temperature are usually negligible.

Effective Bulk Modulus - Both entrained air in the fluid or mechanical compliance of tubing/hoses can substantially
lower the bulk modulus. Effects of both are additive.
Effects of Entrained Air:
For liquid-air mixtures, an empirical formula for the effective bulk modulus, , is
(8)

where
isen isentropic bulk modulus of the fluid w/o entrained air
VG0 volume of gas entrained in the liquid at atmospheric pressure
VL0 volume of the liquid at atmospheric pressure
p0 atmospheric pressure
p fluid pressure
k isentropic exponent (normally, k=1.4)
rv air to liquid volume ratio
Using equation (8), the effects of entrained air are shown in Figure 1 below.

Figure 1 Effects of Entrained Air on Bulk Modulus

Effects of Mechanical Compliance:
For cylindrical pipelines, the effective bulk modulus, , can computed using
(9)
p is the bulk modulus of the pipeline (available in Materials or Engineering Handbooks) and w is given by
for (thick walls) (10)
for (thin walls) (11)
where
do outer pipe diameter
di inner pipe diameter
Poissons number (0.3 for steel)
S pipe wall thickness

Fluid Properties Example: For a steel pipe with isen = 200000 psi, 0.5 O.D., S = 0.012, compute the effects of
mechanical compliance on bulk modulus.
p = ratio of normal stress (on all faces of a cube) to a change in volume

thin walled

Using equation (9)

In this example, the effects of thin walled pipe reduce bulk modulus by 25%.

Combined Effects of Air and Mechanical Compliance:
Effects of entrained air and mechanical compliance combine together like springs in series, i.e.,
(12)
where
effective bulk modulus (psi)
l bulk modulus of the hydraulic fluid
MC bulk modulus for mechanical compliance
g bulk modulus of air
Vg volume of air in the fluid
Vtotal total volume of fluid and air

Note the effective bulk modulus will be less than any of the individual bulk modulus (l, MC or Vtg/Vg). For air,
the bulk modulus is computed using

Basic Effects of Fluid Properties
The effects of temperature and pressure on hydraulic system fluid properties and flow characteristics are listed
below.
Density Effects orifice and valve volume flow rates. As density increases, orifice and valve flow rates will
decrease (see orifice flow equations).
- Increasing pressure increases density
- Increasing temperature decreases density
Kinematic Viscosity Effects pipe (tube) volumetric flow rate. As viscosity increases, pipe flow rate will decrease
(see orifice flow equations). Kinematic viscosity increases with increased pressure and decreasing temperature.
- Increasing pressure increases kinematic viscosity
- Increasing temperature decreases kinematic viscosity
Bulk Modulus Effects compressibility of fluid and system response time (see pressure derivative equation). As
bulk modulus decreases, the pressure derivative will decrease leading to slower response times. Compressibility
will affect the performance of actuators, motors and pumps because the stiffness of the fluid is less as bulk
modulus is reduced.
- Increasing pressure increases bulk modulus
- Increasing temperature decreases bulk modulus
- Entrained air and compliance of hoses/tubes/parts decreases bulk modulus

Fluid Properties for Standard Hydraulic Fluids

Mil-Prf-5606
-54
o
C -40
o
C 40
o
C 100
o
C
Viscosity (centistokes) 2500 600 13.2 4.9
Bulk Modulus at 40
o
C and 4000 psi: 200,000 psi (minimum)
Specific Gravitity at 60
o
F Approx 0.88*
Nominal Density: 50 lbm/ft
3

Pour Point: -60
o
C
Flash Point: 82
o
C
Coefficient of Thermal Expansion: 8.6E-04 cm
3
/ (cm
3

o
C)
* Not specified in MIL-PRF-5606

Mil-Prf-87257
-40
o
C 40
o
C 100
o
C
Viscosity (centistokes) 550 max 6.7 min 2.0 min
Bulk Modulus at 40
o
C and 4000 psi: 200,000 psi (minimum)
o
F Approx 0.88*
3

Pour Point: -60
o
C
Flash Point: 160
o
C minimum
3
/ (cm
3

o
C)
* Not specified in MIL-PRF-87257

Mil-Prf-83282
-40
o
C 40
o
C 105
o
C 205
o
C
Viscosity (centistokes) 2200 14 3.45 1.0
Bulk Modulus at 60
o
C and up to 10000 psi: 200,000 psi (minimum)
o
F Approx 0.85*
3

Pour Point: -55
o
C
Flash Point: 205
o
C
3
/ (cm
3

o
C)

AS1241 Type V (Phosphate Ester)
-54
o
C 38
o
C 99
o
C
Viscosity (centistokes) 2600
(max)
9-12.5 3-4
Bulk Modulus at 40
o
C and 4000 psi: 200,000 psi
Density at 25
o
C 0.97 1.02 g/mL
3

Pour Point: -62
o
C
Flash Point: 149
o
C
Coefficient of Thermal Expansion: 1.0E-03 in
3
/ (in
3

o
F)
Motors, Hydraulic - Description
The function of hydraulic motors is to convert hydraulic pressure and flow into rotational mechanical
energy via an output shaft. Motors are used where a rotational output is required and actuators are
used where linear output is required. However, in operation motors are more similar to pumps than
actuators and, in fact; the equations for motors are identical to the equations for pumps.
The output of a motor is a torque and an angular velocity (note: power = torque x angular velocity).
Motors operate exactly opposite of pumps and, in fact, some motors/pumps take on dual roles (will
operate as a pump or a motor depending on the position of controlling valves) in a hydraulic system.
Like pumps, the governing equation for an ideal motor is
(1)
where Q is the flow rate, p is the pressure drop across the motor, T is the output motor torque and
is the motor angular velocity.
The main parameter of interest for hydraulic motors is the displacement (similar to pumps).
Displacement is the amount of fluid that is displaced (flows through) the motor for each revolution of
the shaft, and relates motor torque to the p across motor.
(2)
In equation (2), D is the displacement in units of in
3
/rev (or similar) and p = p
1
p
2
, where p
1
is the
inlet pressure and p
2
is the output pressure. The steady state performance of a pump is completely
defined by equations (1) and (2).
There are 3 main types of hydraulic motors: piston, vane and gear.
A piston motor can be radial or axial. In a radial piston motor, the pistons are oriented radially from the
shaft. An axial piston motor is similar to an axial piston pump and is the most common motor used in
aerospace. A schematic of an axial piston motor is shown in Figure 1. As can be seen in the figure,
the axial piston pump is identical to a piston pump, except that the swashplate angle is now fixed (i.e.,
there is no compensator and control piston). The inlet side of the motor is the high pressure side and
the outlet is low pressure. The pressure difference causes the pump to rotate. Since the swashplate is
fixed, speed of this motor is controlled by either controlling inlet pressure (p across the motor) or the
flow rate. Also, like pumps, hydraulic motors tend to have 9 pistons, or possibly 7 (more pistons
increase displacement and hence increase output torque). Piston motors provide the best sealing for
high input pressures and work best in high torque, low speed applications. They have the best sealing
and will be the most efficient. An axial piston motor with a fixed swashplate is unidirectional (rotate in 1
direction only). To be bi-directional, the swashplate would need to be variable position. Lastly, piston
motors will have a case drain line to allow piston leakage to flow to return.

Figure 1 Axial Piston Hydraulic Motor

A vane motor is shown in Figure 2. Vane motors can be a good choice for a motor in high speed
applications. The vane motor rotates as hydraulic fluid at high pressure flows through the motor to the
outlet, or low pressure side of the pump. More vanes reduce output torque ripple, but also lead to
higher pump friction. The vanes are attached to the drive shaft and fit closely to the housing ring (or
cam ring) to minimize leakage. The vanes are pushed out by hydraulic pressure, centrifugal force or
springs (springs are shown in Figure 2). For vanes which rely on centrifugal force to extend, the vanes
are may be attached to the drive shaft (or rotor) via a slot, which allows the vanes to rotate in the slot
and also move slide in the radial direction. As pressure is applied, the vanes will start to rotate the
rotor (shaft). As speed is increased, the vanes move outward in the slots and contact the wall,
providing a seal on the outer surface. The housing (or cam ring surface) can have a ramp shape to
further reduce pressure at the pump outlet this is possible because the rotor slot allows a vane to
move radially and rotate to adjust to the housing (or cam ring) shape. For vane motors that utilize
spring loaded vanes, the spring that helps to hold the vane against the housing (or cam ring) to ensure
sealing at low p across the pump. This helps the motor develop starting torque faster. At higher
speeds, centrifugal force helps hold the vanes out. The vane pump shown in Figure 2 has two ports,
an inlet and outlet. It is possible to have a four port vane motor, which splits the flow through two
separate paths. A four port vane motor will have twice the torque, but will operate at the speed as a
similar sized two port vane motor. Inlets for a four port motor will be at opposite ends of the motor to
balance bearing loads. For vane motors, a valve can control which port has high pressure and which
port has low pressure, leading to a bi-directional motor. Vane motors are not as efficient as piston
motors but are better suited to high speed operation. Vane motors become less efficient at high inlet
pressures (due to potential for more slippage and leakage past the vanes). Vane motors are also
more inefficient at low speeds. Vane motors can be operated in a reversible (bi-directional) manner
and will be less noisy then other pump types. They will generally be less expensive than piston
motors.

Figure 2 Vane Hydraulic Motor

A gear motor schematic is shown in Figure 3. This gear shows two spur gears rotating in a common
housing. Both gears rotate, although only one gear is connected to the output shaft. Fluid enters
where the gears mesh together as shown in Figure 3. The gears then rotate in the direction of the
arrows, as the greatest pressure drop is around the outside of the housing (if the gears rotated the
other direction then they would be pushing against system pressure). Also, by putting the input port
where the gears mesh together puts the effective area of 2 gear teeth against the resisting pressure
acting on 1 gear tooth. Gear motors work best in high speed applications. The efficiency of gear motor
is lower at low speeds and increases (gets better) at high speeds.

Figure 3 Gear Hydraulic Motor

Gear motors are generally very compact relative to their displacement and are able to operate at high
speed. They are also less expensive than a piston or vane motor. However, they can be noisy and are
the least efficient of the 3 motor types. Gear motors can be operated in a reversible (bi-directional)
manner.
Pump performance varies with speed and load. A typical performance chart for a hydraulic piston
motor is shown in Figure 4. As can be seen in the top chart, the volumetric efficiency is high and
constant over the range of speeds. However, the overall efficiency (which is the product of the
volumetric and mechanical efficiencies) drops off at higher speeds, which implies there is an optimal
speed range for each particular hydraulic motor. The lower chart shows a linear relationship between
flow and speed, where the slope of the line is the displacement. Also shown in the lower chart is the
motor output torque, which tends to drop off at higher speeds.

Figure 4 Typical Piston Hydraulic Motor Performance Graphs

As stated above, motors can be uni-directional or bi-directional, but in aerospace bi-directional motors
are almost always used. An exception would be pump/motor units connected by a power transfer
shaft, which tend to be a piston pump/motor (whether a pump or motor depends on direction of
rotation) see Figure 5. Most motors also tend to be fixed displacement, unlike a pump where the
compensator changes the pump displacement based on pump outlet pressure. In motor design, one
of the key elements to a good motor is minimizing internal leakage past piston and vane seals, under
both low load and low speed operation. Leakage is the flow that passes through the motor without
creating any work output. Another key element is the ability of the pressure/load bearing surface to
withstand the applied loads this is bigger issue with vane pumps.
The overall benefits of hydraulic motors include the ability to control to an infinite speed (as opposed
to discrete increments), ability to withstand stall torque easily and relatively high power to weight
ratios. Motors are self-contained units and simple in operation, which leads to high reliability. The main
issues with hydraulic motors in service are excessive leakage, seal failures and noise.

Figure 5 Piston Pump and Motor Unit (Uni-Directional)

Motor Design Considerations
The most important characteristics for hydraulic motor performance are listed below.
Motor Type as described above there are 3 basic types of hydraulic motors (piston, vane and gear).
Each type has advantages and disadvantages and the specifications will vary between the 3 motor
types.
Displacement fluid volume displaced (flows through the motor) per revolution of the output shaft
Operating Torque output torque of the motor, usually given at the rated speed and rated flow for a
given inlet pressure. A plot of output torque may also be available. Note this is related to operating
power through motor speed.
Operating Power output power of the motor, usually given at the rated speed and rated flow for a
given inlet pressure. A plot of output power may also be available. Note this is related to operating
torque through motor speed.
Operating Pressure expected inlet pressure of the motor during normal operation, which must be
compatible with the design operating pressure for the system and the expected pressure drop
between the pump and the motor. This pressure will be less than the pump outlet pressure.
Maximum Operating Pressure maximum allowable inlet pressure to the motor. This should normally
be equal to system pressure or higher.
Operating Speed this is the speed of the motor at the operating pressure.
Operating Flow flow rate through the pump at the nominal operating pressure and operating torque
Maximum Flow maximum flow rate that is possible through the motor. Maximum flow will occur at
the maximum motor inlet pressure under no load conditions.
Maximum Allowable Viscosity motor manufacturers provide recommended ranges of fluid viscosity.
Motor operating parameters provided by the manufacturer will be valid within the recommended range
of viscosity. At high viscosity levels, motor capability will start to decline.
Maximum Operating Temperature highest temperature the motor is expected to see in service. This
temperature is used to ensure motor materials and seals are compatible with the operating
environment of the motor.
Motor Efficiency higher efficiencies are preferred. Efficiency is the difference between power in and
power out. The difference is caused by frictional losses and seal leakage.

Motor Installation Considerations
The most important parameters for the installation of hydraulic motors are listed below.
Dimensions overall dimensions for the motor is required to determine the installation volume
required
Interfaces interfaces include tubing/hose connections and locations on the motor, as well as
mounting lugs and output shaft location.
Weight weight of the motor, which is usually provided as a dry weight. When filled with fluid, weight
will be higher.
Noise Motors run at high speed and create noise. A specification for maximum noise level should be
considered for motor installations.
Attachments method of attachment of the motor to the airframe affects structural rigidity of the motor
and also noise (vibration) transmission into the airframe. Motors can have high starting torque causing
large loads to be reacted through the attachments on start up. Continuous operational torques will
affect the fatigue and durability of the attachments.
Motor/Shaft Alignment Alignment of motor to shaft needs to held to tight tolerances. Considerations
are tolerance stickups, relative motion between motor and shaft, possible angular displacements on
installations, spline teeth dimensions, etc. Improper alignment can cause excessive vibration (leading
to premature failure), or failure of the motor seals.
torque is higher than running torque. Start-up torque accelerates the mass/inertia of the motor and
load, leading to temporary high stresses within the mounting hardware and motor. This is more of a
concern on installations with high load inertias. Obviously, the speed of the motor must match the
manufacturers recommended speed for actuation device (such as a ball screw actuator). A gear
arrangement may be used, if required.
Axial and Radial Shaft Load Capability Ensures motor shaft and splines are adequately sized for
static and fatigue loads that the motor will see over its operating life.
Inlet/Outlet Line Inlet and outlet lines/fittings must be matched to the motor
Direction of Rotation Motors can be uni-directional or bi-directional. Aerospace applications tend to
be bi-directional pumps. Obviously, the control valve position must be matched to the appropriate
direction of rotation.

Speed Control of Hydraulic Motors
Usually, the goal of a hydraulic motor is to produce a constant rotational speed at a sufficient torque to
move the load. The two methods to control speed are use of a hydraulic pump (either fixed delivery or
variable delivery, but usually variable delivery) or through a valve. Valve control may be through a flow
control valve, pressure regulating valve or a servovalve. As discussed in the mathematical equation
section for motors (see Motor, Hydraulic Equations), rotational speed of the motor will vary with load
and flow varies with speed. Also, torque is a function of p across the motor.
Consider a fixed delivery pump. A fixed delivery pump will produce a constant flow. Assuming a
pressure relief valve does not open, a constant flow rate from the pump will flow through the motor but
the inlet pressure will vary with load. Therefore, a fixed delivery pump will produce a constant speed
but the torque will vary with load.
A variable delivery pump will vary flow to produce a constant pump outlet pressure. At the motor, this
will maintain a constant motor inlet pressure, so that the p is constant across the pump and hence
the torque is constant across the pump. Therefore, a variable delivery pump, in the absence of any
valve control, will produce a constant torque but a speed that varies with load.
A flow control valve regulates flow rate to a constant value. This is accomplished by controlling the p
across an orifice. Therefore, a flow control valve will result in a constant flow through the motor with a
varying motor inlet pressure. In effect, use of flow control valve produces a constant motor speed with
a torque that varies with load, similar to a fixed delivery pump. This assumes system pump flow and
pressure to the valve are constant.
A pressure regulating valve regulates valve outlet pressure by controlling flow. This effect will produce
a constant motor torque with a speed that varies with load, similar to a variable delivery pump. This
assumes system pump flow and pressure to the valve are constant.
In the case of servovalve control, a digital control system will sense rate somewhere in the system
(such as flap rate) and modulate the servovalve to maintain a desired rate. This is a speed control,
feedback control system. Parameters within the control loop, such as valve constant and loop gain,
are sized to give optimal performance at the design conditions. Ideally, the speed control loop is a
linear control loop without nonlinearities. However, in practice, hydraulic systems are inherently
nonlinear which may cause some stability and performance issues if the system is not designed
properly. In particular, high stiction and start-up forces in the system may lead to jerky starts as the
control loop attempts to find equilibrium (in terms of speed). Another factor in use of servovalves to
control a motor is that 4 way spool valve will regulate both supply pressure and return pressure to the
motor (so inlet and outlet port pressures will be affected by a servo valve). The pressures will vary
according to the relations

where P
S
= P
1
+ P
2
and Load Pressure = P
L
= P
1
- P
2
.
For other types of control, the downstream pressure, P
2
, will be the return pressure.
Lastly, if hydraulic motor speed is too high - which is often the case a gear box can be installed on
the motor output shaft to reduce speed. For large gear reductions, a planetary gear train can be used.

Applications of Hydraulic Motors
The most common application of hydraulic motors in aerospace vehicles are operation of wing trailing
edge flaps and leading edge slats. In these applications, a hydraulic motor drives the flap or slat via a
torque tube that runs along the trailing edge or leading edge. Gearboxes (90 degree, bevel and offset
gear arrangements) connect the torque tube along the trailing edge or leading edge. Typically a
ballscrew actuator with a gearbox extends and retracts the flap or slat. There are usually at least 2
ballscrew actuators per flap or slat panel. Other applications for hydraulic motors are folding wing
control, cargo doors and ramps and landing gear. Motors are usually high speed with low torque that
are geared down to provide a lower speed and higher torque.

Motor Qualification
See Qualification - Hydraulic Components for discussion on hydraulic motor qualification and required

Orifice Flow, Hydraulic - Equations
An orifice is a sudden flow restriction of short length (zero length for sharp edge orifice). Orifices are
treated as either a sharp edge orifice or a short tube orifice. Orifices are primarily used to control flow
or create a pressure differential (drop). Orifices may be fixed or variable (valve). Many types of valves
and flow devices can essentially be viewed as orifices. In valves, there can be numerous flow
passages, but usually somewhere in the flow passage is a restriction that controls flow, which is why a
valve often behaves like an orifice.
Flow characterization for orifices is done mathematically. However, by understanding the nature of the
mathematical equations the behavior of fluid flow in orifices (as well as pipes and servos) can be
understood and intuition developed. Also, the equations can be used to model component and system
behavior for enhanced analysis and understanding. The orifice flow equation is a key equation for
hydraulic systems.
Like pipe flow, fluid flow in orifices can be either laminar or turbulent (see Figure 1). In laminar flow,
each fluid particle follows a well defined trajectory, with velocity only in the direction of flow. In
turbulent flow (most common in hydraulic systems due to small line diameters and small orifices) each
particle flows in the general direction (velocity) of the flow, but is subjected to fluctuating cross current
velocities. Equations for computing orifice flow are different for laminar and turbulent flow.

Figure 1 Flow Streamlines for Laminar and Turbulent Orifice Flow

Determination of laminar or turbulent is determined using the Reynolds number, Re
(1)
where
hydraulic diameter
A flow section area
S flow section perimeter
v flow velocity
dynamic viscosity
kinematic viscosity
For low values of Re, flow is laminar. For high values of Re, flow is turbulent.
From Bernoullis equation, the total energy loss is the energy converted to heat by friction of particles
against the wall and each other
(2)
Assuming, away from the orifice, that v
1
= v
2
and A
1
= A
2
, the flow becomes a product of the area and
speed
(3)
where is a dimensionless loss coefficient, representing the energy loss associated with the pressure
drop. For hydraulic systems, this equation is normally written as
(4)
where
d
is the discharge coefficient and represents the energy loss in the fluid. Equation (4) is the
orifice flow equation. The discharge coefficient is the key element to estimate for laminar and turbulent
flow regimes. Inspection of the equation (4) indicates that the flow rate varies proportionally with area
if the p is held constant, and that the flow rate varies with the square root of p if the flow area is
held constant. Figure 2 shows notional charts of the flow behavior.

Figure 2 Flow Rate Behavior for a Orifice

Turbulent Orifice Flow
For a sharp edge orifice, with turbulent flow and with orifice flow area, A
o
<< A (pipe flow area), the
theoretical
d
is
(5)
For short tube orifices of length L, pipe diameter d, and orifice diameter d
o
,

Figure 3 graphically shows the variation in
d
for the above two equations.

Figure 3 Turbulent Flow Discharge Coefficient for Short Tube Orifice

Laminar Orifice Flow
Equation (4) can be used in the turbulent-laminar (transition) region and the laminar flow region using
(6)
where
(Sharp Edged Orifice)
(Rounded Off Orifice)
is called the laminar flow coefficient and depends on orifice geometry.
d,turb
comes from equation (5).
Re
crit
is the critical (breakpoint) Reynolds number found during empirical testing for the type of orifice
(see Figure 4). These equations are theoretical, but have been validated by experiment.
Equation (6) is valid up to the critical Reynolds number (see Figure 4). Experimental data for
d
as a
function of Re
crit
for various orifices are shown in the figure below.

Figure 4 Laminar Flow Discharge Coefficient for Orifices

The sloped lines can be used for
d
in the laminar flow region. At the breakpoint, turbulent flow occurs
and
d
= 0.611 should be used.

Pipe Flow, Hydraulic - Description
Flow in a pipe occurs as a result of pressure drop and always travels from a high pressure location
(usually a pump or charged accumulator) to a lower pressure location (such as occurs when a valve
opens). When there is no flow in a circuit, then the pressure is equal throughout the circuit. Note a
hydraulic circuit consists of tubing and components connected together to perform some specific
function. These hydraulic circuits generally have a pressure source which route fluid to a specific
function, such as the landing gear. A pipe cross section is shown in Figure 1.

Figure 1 Pipe Cross Section

Generally speaking, fluid flow behaves much like flow in an electrical circuit with a power supply,
ground path and wires/resistances. In this analogy, pressure drop is equivalent to voltage drop, fluid
flow is equivalent to current, and pipe geometry and flow viscosity combine to form the resistance.
Flow is divided up between branches depending upon the comparative resistance between each
branch branches (pipes) with low resistance get more flow and branches (pipes) with high
resistances get less flow. Pipes that are closed at one end get zero flow. Flow is proportional to
pressure (or voltage) drop through the pipe.
For design purposes we can use simplified equations to analyze performance and to size pipes. The
equation to use depends on whether the flow is laminar or turbulent. Laminar or turbulent flow is
determined using Reynolds number, which is a dimensionless number that is a function of pipe
geometry, fluid properties, and fluid velocity. Analysis of piping networks can also be accomplished
using 1
st
order pipe models in a simulation model.
In reality though, the specific behavior of flow of hydraulic fluid in pipes is very complex. The
complexity is a result of (i) non-steady fluid properties, (ii) the fact that hydraulic fluid has mass, is
compressible and stores energy, and (iii) valves/actuators are changing the piping configuration during
operation. This end result is pressure waves of varying frequencies and magnitudes within the pipes,
which cause unforeseen behavior in hydraulic systems. Determination of these pressure waves and
their behaviors is very difficult, usually requiring detailed analysis using some form of Navier Stokes or
exhaustive testing. Analysis of pressure wave effects in hydraulic systems is tedious and time
consuming. A common approach in industry is to first use simplified equations to size pipes. Secondly,
pipe network models using 1
st
order pipe equations are constructed for analyzing flow sharing
between hydraulic circuits and to validate sufficient pressure is available at each actuator/motor/servo
component. Lastly, high frequency (pressure wave) effects can then be evaluated for components or
subsystems that may be susceptible to pressure waves. There are some simple checks that can be
utilized and also more complex models for pressure wave analysis.
The design of hydraulic piping systems is a tradeoff between keeping pressure drop/flow velocities low
while minimizing weight and cost. Low fluid velocities lead to reduced pressure drop and thus higher
performance at the downstream component. Additionally, low fluid velocities reduce the magnitude of
pressure waves and water hammer. However, low fluid velocity requires larger diameter pipes and
therefore higher weight and cost. In aerospace, the goal is to minimize pipe diameters as much as
possible while still ensuring adequate performance of the component(s) and that pressure waves can
be tolerated by the system without leading to premature failures in the piping and connectors.
Hydraulic pipes (tubing) are usually made from stainless steel, aluminum or titanium. For high
pressure tubing, stainless steel is strong, handles installation pre-stresses well (connections never
seem to match up perfectly) and is easy welded. Steel is inexpensive and more resistant to damage
than Titanium. The downside of steel is weight due to its approximate density of 0.28 lb/in
3
.
Consequently, many manufacturers use titanium, which has an approximate density of 0.16 lb/in
3
, on
high pressure lines. Titanium is strong and weighs much less than steel for comparable stress levels.
Titanium is expensive. The other downside of titanium, which has not shown to be a critical problem, is
that titanium is not as receptive to welding as steel and requires better welding controls to obtain a
good weld. Also, titanium is not as forgiving when it comes to installation pre-stresses, unless wall
thicknesses are increased (which increases costs). Aluminum tubing is low cost, light weight with an
approximate density of 0.10 lb/in
3
, and relatively easy to form, but does not have the strength
properties for high pressure lines (unless wall thicknesses are large). Aluminum tubing is typically
used in low pressure return lines. One disadvantage of aluminum tubing is that it is easier to damage
when compared to steel and titanium tubes. Aluminum tubing material properties are affected at
higher temperatures, generally above 200F; hence, they cannot be used in a fire zone or other high
temperature region of the hydraulic system. Due to these characteristic some aircraft use Titanium
tubing for all return tubes.
Hydraulic tube installations require robust connectors (fittings) and tubing supports that allow for some
motion of the hydraulic tubes. Motion occurs in hydraulic lines due to aircraft bending and flexure as
well as hydraulic pressure spikes. All hydraulic line installations utilize pipe bends and flexible clamps
(such as cushioned p-clamps) to accommodate for movement of the hydraulic lines. If pipes are rigidly
attached, then premature tubing failure will result. In cases, where movement of the hydraulic line is
significant or the component is moving around a pivot point, hydraulic hoses are used. Connectors for
hydraulic tubing are either permanent swaged fittings or standard AN/MS type fittings. Swaged fittings
require repair equipment and procedures, but are often used today because of their lifetime reliability
and light weight. Several manufacturers supply swaged fittings and swaging equipment. AN fittings
require tubes to be flared to 37 degrees on the ends, which does add to the cost of the tubing (swaged
fittings do not require flaring). MS fittings also incorporate a flareless design where the sleeve bites
into the tube. AN/MS fittings need to be selected based on their pressure rating and are readily
available as a standard part.

Pipe Flow, Hydraulic - Equations
For pipe analysis and sizing purposes, piping is usually treated using a 1
st
order lumped parameter
equations. Lumped parameter analysis only provides information at the inlet and outlet of a pipe. If
desired, pipes can be broken down into smaller increments, but this is usually not practical or
necessary.
Figure 1 shows a pipe cross section with relevant input and output parameters. Pipe parameters are
inside diameter and length. Subscripts 1 and 2 refer to the pipe inlet and outlet, respectively.

Figure 1 Pipe Cross Section

In the figure,
P Pressure (psi)
p = P
2
P
1
pressure drop (psi)
Q flow rate (in
3
/sec)
l pipe length (in)
D inside pipe diameter (in)
fluid density (lb
f
-sec
2
/in
4
)
absolute viscosity (lb
f
-sec/in
2
)
kinematic viscosity (in
2
/sec)

For analysis and sizing, one of the following 2 methods is used:
1. Friction Factor Method
2. Test Data:
The test data method for a pipe (or any component) is preferred whenever it is available.

Friction Factor Method
When using the friction factor method, the first step is to compute the Reynolds number. For pipes, the
Reynolds number is computed using
(1)
where
Re Reynolds number [dimensionless]
d internal diameter of the pipe [in]
v flow velocity [in/sec]
kinematic viscosity [in
2
/sec]
For pipe flow, the flow regime based on Reynolds number is listed below:
Laminar: Re < 2300
Transition: 2300 < Re < 4000/FONT>
Turbulent: Re > 4000
The equations are different for each flow regime.
For laminar flow, the pressure drop is proportional to dynamic pressure via
(2)
Where
p pressure drop through pipe (p
1
p
2
)
l length
d internal flow diameter
fluid density
V fluid velocity
The volumetric flow equation is
(3)
where Q is the volumetric flow rate, A is the cross sectional area of the pipe and v is the velocity, and
the Reynolds number equation is
(4)
where v is the flow velocity, d
h
is the hydraulic diameter and is the kinematic viscosity. Substituting
(3) and (4) into (2) yields
(5)
In terms of Q
(6)
Equation (6) relates the pressure drop through a pipe to the flow rate for laminar flow. The relationship
between the pressure drop, p, and flow rate, Q, is governed by fluid properties and flow geometry.
The equation for turbulent flow is
(7)
This equation is derived using p = k (1/2 V
2
) and using a curve fit for turbulent flow from friction
factor graphs. Solving (7) for Q yields
(8)
Equation (8) relates the pressure drop through a pipe to the flow rate for turbulent flow.

Test Data Method
This method can be used when test data for pipes or any other component is available. Manufacturer
data is often available for valves and this method is equally applicable to them.

For incompressible flow, pressure drop can be computed using friction factor and associated K
pressure drop factors. The standard equation is
(9)
where is the dynamic pressure and K is the pressure drop factor given by
(10)
In equation (10), f is the friction factor (f = 64/Re for laminar flow and f = 0.332/Re
.25
for turbulent flow),
L is the length of the pipe, D is the diameter and K
t
accounts for bends, exits, etc. Values for K
t
are in
the SAE AIR1168/1 and some fluids texts. Equation (10) allows the pressure drop for numerous
components to be combined. For example, pressure drop through a pipe, check valve and a pipe
connected in series could be combined into a single equation using equation (10).
When test data flow vs. pressure drop is available for a given pipe or piping configuration, the data
is usually plotted on log-log axis and a relationship of the form
(11)
can be used. In this equation
density ratio
K pressure drop constant
Q volumetric flow rate
n exponent (for air flow, n is usually close to 2)
Comparing this equation to the general orifice or servo flow equation,
(12)
leads to
(13)
where
0
is the density of the fluid when the test data was obtained and is the density of the fluid for
the condition of interest.
For equations of the form,
(14)
a is the y-intercept and n is the slope. Thus
(15)
and using this value of n,
(16)

An example using check valve illustrates use of the above equations. A valve manufacturer provides
the following flow data for a check valve:

For the size 4 check valve, two points are read from the graph
Q1 = 1 gpm (3.85 in3/sec), (p)
1
= 5.8 psi
Q2 = 2 gpm (7.8 in3/sec), (p)
2
=20 psi
Using equation (15),

Using Q
1
and (p)
1
condition in equation (11)

Therefore,

which is the governing flow equation for the check valve.

Pipe Temperature Losses

Temperature losses through a pipe adhere to the following relationship
(17)

where
T
1
inlet temperature
T
2
outlet temperature
T
amb
ambient temperature outside of pipe
U
overall
overall heat transfer coefficient
Q volumetric flow rate
A heat transfer area
c
p
specific heat coefficient at constant pressure
fluid density

Equation (17) computes pipe outlet temperature based on inlet temperature, ambient temperature,
fluid flow rate and pipe material. U
overall
consists of forced convection heat transfer from the fluid to the
pipe wall, conduction through the pipe wall and free (still air) convection away from the pipe to the
ambient. There are also conduction paths from the pipe to structure though pipe supports. The area is
the heat transfer area (pipe circumference x length)

Power Control Unit, Hydraulic - Description
Hydraulic power control units (typically referred to as PCUs) are used to
position primary (or secondary) flight control surfaces. PCUs are a
single assembly (i.e., single part number) that combines individual
hydraulic components into an assembly. PCUs perform two critical
functions: (1) they position the flight control surface in response to
mechanical or electrical commands, and (2) they represent a principle
structural element capable of withstanding flight loads and providing
protection against flutter (unsteady, aerodynamic loads). The
requirements of these two functions make PCU design a challenging
task.
At the heart of PCU is a servovalve and an actuator (servo actuator).
The servovalve can be controlled by a flapper nozzle, jet pipe, solenoid,
torque motor or mechanical linkage. Some PCUs contain only a servo
actuator. However, PCUs often contain other components to meet
performance and failure mode performance criteria. Other components
typically found in PCUs are shutoff valves, pressure relief valves, input
filter, check valves, and compensator.Another component that is often
part of a PCU is one or more servos whose control is based on some
performance criteria. These servos can be positioned electronically or
by hydraulic pressure (where loss of hydraulic pressure allows a spring
to position the servo).
A simple PCU is shown in Figure 1.

Figure 1 Simple PCU Diagram

This PCU shows the pressure input going through a filter to a shut off
valve. If the shut off valve is open (as shown), then pressure is applied
to a solenoid operated servovalve. When closed, the SOV ports
pressure to return. The servovalve controls actuator position. (Typically,
actuator position is measured by a LVDT on the actuator piston and fed
back to a controller that provides closed loop position feedback see
position control system.) The pressure relief valves will bleed off
pressure if the pressure in the actuator chamber exceeds a certain level.
Pressure can increase above acceptable limits due to external loads or
thermal expansion of the fluid.
As shown in Figure 1, PCUs are built from other hydraulic components,
which are packaged into an assembly. An understanding of PCU
behavior and characteristics can be ascertained by understanding each
component in the system and how it interacts with other components.
An analysis or simulation model can be built using models for the
individual components to better understand behavior and performance
characteristics (see Power Control Unit, Hydraulic Equations).
A more complex PCU is shown in Figure 2.

Figure 2 Flight Surface PCU

In Figure 2, the hydraulic fluid flows through a filter and check valve
before going to the servovalve. The servovalve is positioned
mechanically via a linkage from the cockpit. The servovalve positions
the actuator. Connecting the input link to the actuator piston rod
provides feedback and helps stabilize the PCU. Under normal operation,
flow to and from an actuator chamber is through the servovalve. If
actuator chamber pressure exceeds a maximum allowable level, then
fluid is drained through the pressure relief valves at the bottom of the
schematic. These settings will be roughly 20% above nominal system
pressure. The return line pressure relief valve (top right portion of Figure
2) provides a constant backpressure to the actuator and servo. This
helps minimize variance from system back pressure to improve PCU
performance. In addition, should the main system hydraulics fail, the
return line relief valve will help maintain sufficient back pressure to push
fluid through the anti-cavitation check valves (overcome spring force) as
the actuator moves in response to airloads. The relief valve setting will
be above system reservoir pressure setting, generally in the 100-200 psi
range. The compensator acts like an accumulator to help maintain a
constant backpressure on the system under small disturbances. The
main function of the compensator, however, is to store sufficient fluid to
prevent cavitation after a main hydraulic system failure. The
compensator must therefore be sized to have sufficient volume so that
there is enough fluid to maintain fluid in the actuator chambers after
considering worst case leakage losses over the length of the longest
possible flight.
A third example is shown in Figure 3. This PCU is similar to the PCU
shown in Figure 2, except it is now electrically actuated and has a mode
control valve or mode control servo that is also electrically actuated. The
control valve is spring loaded to the center position but has two
solenoids to push the servo to the left or right. The three modes would
be normal (center position), damped bypass (left position) and blocked
(right position). These are 3 common modes used in flight surface PCUs
(a 4
th
mode would be undamped bypass). In normal operation, the PCU
operates similarly to the PCU in Figure 2, except that the servovalve is
electronically controlled instead of via a mechanical linkage. In bypass
mode, solenoid 1 would be energized which would push the mode servo
to the left. This would connect both chambers of the actuator through a
damped orifice which would limit the flow rate between the chambers.
The effect of the damped orifice is to significantly dampen the actuator
so that movement is allowed but at a rate such that the actuator is
insensitive to flutter. In blocked mode, hydraulic fluid would be trapped
in the actuator chamber which would hold (lock) the actuator in the
current position. The pressure relief valves, compensator and anti-
cavitation check valves would function the same as described for the
PCU in Figure 2. In this PCU example, both solenoid1 and solenoid 2
would be controlled by a digital controller. They would be two position
solenoids (energized or unenergized). The solenoid on the servovalve
would be an infinitely positioned solenoid where position is proportional
to applied current. As mentioned above, but not shown in the Figure 3
PCU example is a bypass mode. A PCU that has a bypass mode
connects both actuator chambers through a fluid path without an orifice
if the orifice is removed in left side of the mode valve this would
indicate a bypass mode. Bypass mode allows hydraulic fluid to flow
freely between the actuator chambers. Bypass mode is normally seen
on flight surfaces with 2 or more PCUs connected to the surface. An
actuator will be put into bypass mode when the actuator controller
senses a failure, which allows the other actuator(s) to control the surface
with minimal resistance from the failed actuator. Damped bypass mode
(bypass with a orifice) is used when it is acceptable for the surface to
float at a controlled rate and be insensitive to flutter forces. Later
examples show PCUs with bypass and damped bypass modes.

Figure 3 Flight Surface PCU with Electric Mode Control

Mode valves can be controlled electrically or hydraulically. Manual (or
linkage) control of mode valves is not used in PCUs. The fourth example
of a PCU shown in Figure 4 illustrates a PCU with a hydraulically
actuated mode valve. In this example, hydraulic pressure will put the
mode valve in normal position and loss of hydraulic pressure will allow
the spring to push the mode valve to the right, which would be damped
bypass mode. So, this PCU has two modes: normal and damped
bypass. Note the PCU in Figure 3 has 3 possible mode settings. The
configuration in Figure 4 could be used where there is a single PCU
installed on a surface. In the PCU shown, when hydraulic power is lost,
the PCU automatically switches to damped bypass mode to provide
surface damping and flutter protection. The pressure relief valves, anti-
cavitation check valves and compensator function as before. In some
PCUs where hydraulic pressure is used to position the mode valve
(servo), an electrically controlled, two position servo will be used to
control hydraulic pressure to the valve. This is shown in Figure 5, where
the two position, three way control valve has been added.

Figure 4 Flight Surface PCU with Hydraulic Mode Control

Figure 5 Flight Surface PCU with Electrical / Hydraulic Mode Control

The above examples of PCUs show a natural evolution of a PCU which
are variations of the same basic scheme. The PCUs shown in Figures 1
through 5 are more representative of commercial aircraft PCUs,
although there still is a wide variation in PCU designs between
manufacturers. A different type of PCU is shown in Figure 6, which is
more representative of a military application.
The PCU in Figure 6 shows a triplex redundant electrical, dual
redundant hydraulic PCU. Not shown is the dual redundant electronics
that control the servo and contain the fault detection logic. LVDTs (linear
variable differential transducers) are shown notionally on the dual
tandem servo and the actuator ram. In normal operation, pressure from
P1 and P2 are ported though the dual tandem servo to the actuator.
Position control is maintained using the dual tandem servo. The two
bypass valves are in the position shown under normal operation. The
two mode valves are powered open (valves close without power).
Response to failure scenarios is as follows:
- Electronic failure associated with left
side: Solenoid current is removed
from left mode valve, which removes
hydraulic pressure from the left bypass
valve, and the left bypass valve goes
into bypass mode (the right bypass
valve remains in the position shown).
The left side actuator chambers are
then in bypass mode.
- Electronic failure associated with right
side: Solenoid current is removed
from right mode valve, which removes
hydraulic pressure from the right
bypass valve and the right damped
bypass valve. goes into damped
bypass mode. The right bypass valve
goes into bypass mode, so the
actuator right side actuator chambers
are in bypass mode.
- Loss of P1: Left side bypass valve
loses its pilot pressure and goes into
bypass mode. The left side actuator
chambers are then in bypass mode.
- Loss of P2: Right side damped
bypass valve loses its pilot pressure
and goes into damped bypass mode.
Right side bypass valve loses its pilot
pressure but stays in the position
shown due to pilot pressure from P1.
The right side actuator chambers are
then in bypass mode.
- Loss of P1 and P2 or Loss of both
electronic channels: Left side bypass
valve goes to bypass mode, right side
damped bypass valve goes into
damped bypass mode, and right side
bypass valve stays in the position
shown (blocked position) due to spring.
The actuator is then in damped mode.
Note that loss of one hydraulic source
and the opposite electronic channel
has the same effect.

Figure 6 Flight Surface PCU with Dual Redundant Command Inputs

The above examples show several different types of PCU designs. The
examples are by no means a thorough coverage of the possible options
and ways to mechanize a PCU. The examples are provided to assist
you in gaining an understanding of some basic configurations of PCU
operation.

Power Control Unit Qualification
See Qualification - Hydraulic Components for discussion on PCU
qualification and required qualification testing.
In addition to environmental qualification, PCUs will undergo significant
functional testing to validate all operating modes and that responses to
failures are as required in the component specification. This functional
testing can be done under environmental conditions, such as at
temperature or at altitude.

FMEAs & Safety Analysis
Due to the numerous components installed in a PCU assembly an
analysis of failure modes and effects is an important analysis to
complete. The failure modes and effects analysis (FMEA) should
examine all single potential failures associated with each component in
the PCU. The effects of the failures on PCU performance and operation
should be established. Failures to consider should include part jams,
part runaways, excessive leakage, burst failures, loss of hydraulic
pressure, loss of electrical connection, misleading sensor inputs, loss of
sensor inputs, failed check valves, relief valves stuck closed or open,
clogged filters, high friction levels, etc.
The failure effects within the PCU will then need to be examined at the
airplane level and will also be used in fault trees.
When creating the FMEA, latent failures (undetectable failures) should
be identified. Latent failures need to be analyzed separately to ensure
they dont create a hazardous condition or put the actuator in a position
where an additional failure would cause a hazardous/catastrophic
condition on the aircraft. An example of a latent failure would be a
pressure relief valve failed to the closed position (i.e., wont relieve
pressure when required).

Fault Detection
Fault detection and response can be done in the controller electronics
(using appropriate sensors and logic) or though mechanical means. An
example of a mechanical response to a failure is shown in PCU example
of Figure 6. When hydraulic pressure is lost, a bypass valve
automatically slides to the bypass position. The controller may or may
not know this has occurred. However, the controller could be notified by
either monitoring system pressure or through a switch located in the
bypass valve.
Electronic fault detection will be more comprehensive and more
sophisticated. Examples of items that the electronic controller will
monitor include
- Open of shorted coils in torque motors
and LVDTs
- Jam in a spool valve
- Jam of the actuator
- Hardover or uncommanded motion
failures
- Broken wires
- Mechanical disconnect of an LVDT
- Failure to switch modes
Electronically failures are checked on power up BIT (PBIT) or through
continuous BIT (CBIT), where BIT is built in test. To minimize nuisance
failures, most detection routines utilize a threshold (must be outside of a
range before flagging a fault) and persistence (must remain outside
range for a given time period). Threshold and persistence are used
primarily due to circuit noise and performance variation over
temperature and loads. Multiple controller channels (separate circuit
cards and power supplies) are used for control and monitoring.
Monitoring (fault detection and isolation) may be done within a control
channel or within a separate channel. The controller and monitoring
logic as well as the method for distributing decision making within the
processors is called redundancy management.
The benefit of using multiple channels for redundancy management of a
PCU is illustrated in Figure 7.

Figure 7 Triplex Redundant Controller Channel Comparison Scheme

Figure 7 shows 3 compare monitors in the electronic controller. Each
compare monitor looks at the output command from two channels. If the
channel outputs agree, then the compare flag is set to, say, 1. If the
channels disagree, then the compare flag is set to, say, 0. A truth table
(or other means for voting logic) exists in the voter. When the voter
detects a failed channel, the votor will remove power from the failed
channel. With triplex controller voting logic as shown in Figure 7,
operation can continue with a failed channel. However, when operating
with 2 channels any miscompare between the 2 remaining channels will
shut the system down. For a quadraplex (4 channel) system, the voting
is similar but more complicated with the additional channel. Using a
quadraplex controller, operation can continue with 2 failed channels.
Note that all failures that are not detected either mechanically or
electronically are considered latent failures. Latent failures are generally
detected until either a specific functional test is run by maintenance
personnel or during a teardown inspection of the unit. Items checked
through PBIT are latent for one power down/power up cycle.

Functional Test Procedures
Functional test procedures will be done before, after and, in some
cases, during environmental and operational tests. The Functional Test
Procedures will contain any critical performance criteria that the actuator
must meet such as maximum friction levels, velocity, rate, loads, etc.
In addition functional testing will need to validate all modes of operation,
mode switching, operation under failure, etc as defined in the PCU
specification. For complex PCUs, which contain many features and
modes, functional testing is a significant test activity.

Pressure Regulating Valve, Hydraulic - Description
Pressure regulation uses a servo to control flow area and hence pressure drop through the valve to
control outlet pressure. A simplified schematic is shown in Figure 1.

Figure 1 Simplified Schematic of a Pressure Regulation Valve

In the pressure regulating valve shown in Figure 1, the outlet pressure, P
2
, is ported to the servo. The
position of the servo and hence outlet flow area is based on a force balance between P
c
and the
spring force plus P
rtn
. The spring preload needs to be adjustable (adjustment device is not shown in
Figure 1) to properly calibrate outlet pressure. Flow rate through the valve will change with changing
inlet pressure conditions, depending on the flow area through the servo. With no applied pressure, the
spring will push the valve completely open. The relationship between inlet pressure, outlet pressure
and flow rate is governed by the orifice flow equation. Manufacturers of pressure regulating valves can
provide charts of inlet pressure, outlet pressure and flow rate for a specific part number valve. These
charts can be used to assess performance of a particular pressure regulator over the desired range of
operating conditions. The flow control valve shown in Figure 1 only works when there is a flow
demand on the system. With no flow, the regulator may close off the flow area trapping hydraulic fluid
in the line downstream of the pressure regulator unless a stop on the spool is provided (if there is no
stop a relief valve downstream of the pressure regulator may be necessary in some applications).
One application of a pressure regulating valve shown in Figure 1 would be control of a motor, where
the torque is a function of the p across the motor (inlet motor pressure is controlled by a pressure
regulator and downstream motor pressure is constant at the return pressure). A pressure regulator
would ensure that motor p is sufficient to produce a necessary torque. However, this would lead to
changes in flow rate with changes in inlet pressure. Another application is in a landing gear system
where it is desirable to maintain a high pressure level and some variation in landing gear actuator rate
is tolerable. Many landing gear installations are such that the landing gear actuator loses mechanical
advantage as the gear is retracted. If there is a small margin in the hydraulic pressure required to pull
the gear in the wheel well, the gear can stall out prior to full retraction. A pressure regulating valve will
reduce flow to maintain pressure. Generally speaking, pressure regulators of the type shown in Figure
1 are used sparingly in aerospace applications. This is because a properly sized hydraulic pump will
regulate to design system pressure with minimal variation (seePumps, Hydraulic Description for
pump flow characteristics) and actuation components (actuators and valves) are designed for system
pressure. In effect eliminating a pressure regulator is not absolutely required saves cost and weight.
The pressure regulator shown in Figure 1 provides pressure regulation when there is flow through the
valve. Another type of pressure regulator bleeds off pressure to return to ensure the downstream
pressure is constant under all flow conditions (see Figure 2). In this valve, as the pressure, P
c
,
increases the spool is pushed to left against the spring and the flow opening increases. Note that this
valve, when regulating, wastes energy since flow directed back to the reservoir does no work.

Figure 2 Bleed Off Pressure Regulator

The type of pressure regulating valve shown in Figure 2 is common in brake control systems, where
the goal is to maintain a commanded pressure at the brake under no flow conditions. The brake valve
essentially acts as a pressure regulator where a proportional servo (or direct mechanical linkage) is
used to control spool position and hence downstream pressure.
Pressure regulating valves can be devised in a multitude of ways. Another example of a pressure
reducing valve is shown Figure 3 below.

Figure 3 Pressure Regulator

This valve closes down the inlet opening as outlet pressure increases, which is similar in operation to
the valve in Figure 1. The spring preload needs to be adjustable (adjustment device not shown in
Figure 3) to properly calibrate outlet pressure. Also, if outlet pressure gets too high then the piston
chamber pressure (force) would be able to sufficiently compress the spring so that outlet pressure
would be ported to return. This acts as a load limiting device.

When considering the use of a pressure regulating valve, the following factors should be evaluated
Regulated Pressure Rating should be in the range you desire
Regulation Range what is the minimum and maximum inlet pressure required for regulation. Also,
manufacturer performance charts should be evaluated for output pressure and flow rate variation over
the range of input pressure conditions.
Regulation Tolerance Need to ensure tolerance on pressure regulation is sufficient for you
application. Tolerances result primarily from friction forces on the servo.
temperatures
Leakage leakage rate of the valve should also be evaluated. Leakage becomes wasted energy and
reduced efficiency, however, some leakage will ensure fluid temperatures are warm during cold soak
conditions.
Failure Modes the main failure modes in the flow control are servo valve jamming in any position
from full closed to full open and contamination. It may also be possible for the adjustable device on the
metering orifice to fall out of adjustment leading to lower or higher flow regulation settings.
Chattering valve should be evaluated for potential to exhibit chattering or limit cycle behavior under
servo as well as damping and friction levels. See Friction - Hydraulic Components for further
discussion of friction characteristics.

Pressure Regulating Valve Qualification
Pressure Relief Valve, Hydraulic - Description
A relief valve functions to bleed pressure off to return (by creating an open flow path to return) when a
prescribed pressure level is reached. Relief valves are simple devices (similar to check valves) where
a piston or poppet is held closed by a spring. Figure 1 shows a schematic of a pressure relief valve.
Also shown in Figure 1 is a variable delivery pump and the flow path to the system components
(valves, actuators, etc.). The spring is set such that the poppet will open, providing a flow path to
return, when pressure reaches a prescribed level. Every hydraulic systems requires at least one
pressure relief valve to protect the system against thermal expansion and excessive pressure
transients. The relief valve setting provides burst protection for all components in a system.

Figure 1 Relief Valve Schematic

The cracking pressure for the relief valve is

where F
0
is the spring preload, F
f
is the friction and A
s
is the relief valve poppet area. Below the
cracking pressure the valve has no effect on the system. Once the relief valve is open, system
pressure will need to drop below the cracking pressure to close (re-seat) the valve. This is a hysteresis
characteristic and is primarily due to the friction in the poppet.
When selecting a relief valve, ensure the relief valve is able to flow maximum pump flow under a failed
pump compensator condition. Otherwise, line pressure will build up to a level for potential rupture in a
tube or component will occur. Rupture will lead to leakage in undesirable areas. In addition, when a
pump compensator has failed (pump becomes a fixed delivery pump) and full pump flow is going
through the relief valve, significant heat is generated by the system that is undergoing flow. Therefore,
a thermal analysis of system temperatures and analysis of critical aircraft zones should be performed
to ensure no components or structure would be heated beyond acceptable limits. This analysis will
include all components installed in the vehicle that could be affected by overheating in a hydraulic
system.

When selecting a pressure relief valve, the following parameters should be considered
Relief Setting should be in the range you desire. Note that the relief valve setting will be above
normal system pressure by approximately 5-10%.
Relief Pressure Range what is the minimum and maximum range required for the valve to open
Reseat Pressure - what is the re-seat pressure (the pressure where the valve will close, once open)
Flow Rate what is the flow rate when the valve is fully open? Is the opening sufficient to keep up
with maximum pump flow, should a pump compensator fail?
temperatures

Seals/Clearances since relief valves dont see many operating cycles seals are not as critical as in
modulating valves. However, the method of sealing pressure should be examined. For relief valves
with ball or conical poppets, damage may be possible when re-seating against sharp edges (see
Figure 2). Wear around the poppet seats will lead to leakage. For relief valves with elastomeric seals,
see Seals - Hydraulic Components for discussion on hydraulic seals.

Figure 2 Poppet Sealing Areas for Potential Wear

Failure Modes the main failure mode is a jam of the poppet in the closed position, which would be a
latent failure. It may also be possible for the adjustable device on the valve spring to fall out of
adjustment leading to lower or higher regulation settings.
Hysteresis what is the difference between the relief pressure setting and reseat? This difference
should be large enough to prevent any chattering of the valve. Chattering may occur in a situation
where the valve opens and system pressure drops quickly due to flow through the relief valve and the
valve re-seats. Then the pressure builds up and the valve re-opens, repeating the cycle at a very high
frequency.

Pressure Relief Valve Qualification
certification testing. Note that endurance and fatigue testing will be much less for a relief valve than
other valves due to the low cycles expected to see in service.
Priority Valve, Hydraulic - Description
The function of a priority valve is to allow hydraulic fluid flow to certain functions within a hydraulic
system when the pressure is greater than or equal to a specified level. In effect, the priority valve
gives priority to certain components over less critical components. For example, in a 3000 psi
hydraulic system, a 2000 psi priority valve may be installed such that components downstream of the
priority valve are only supplied hydraulic pressure when the pressure in the system is above 2000 psi.
Below 2000 psi, the priority valve will be closed.
A typical application of a priority valve is to preserve hydraulic pressure for primary flight controls. In
this case, when pressure drops below a specified level, one or more priority valves will close leaving
flow available to primary flight control hydraulic actuators. For example, in the F-14 hydraulic system,
a priority valve isolates landing gear, brakes, nosewheel steering, ram air door servo actuator,
overwing fairing and other components when the engine driven pump pressure drops below 2400 psi.
When the isolation valve shuts at 2400 psi, the engine driven pumps provide hydraulic pressure to
pitch servo, roll servo, yaw servo, rudders, speedbrakes and hook lift. In other applications, a
hydraulic system is powered by a main system pump and an auxiliary pump. Under normal main
system pump, all hydraulic services can be powered. However, should the main pump fail, the
auxiliary pump takes over. The auxiliary pump may be powered through a power transfer unit or an
electric motor. Auxiliary pumps are normally sized smaller than a main system pump since under
normal operation there is a weight penalty when the system is not being used. If the auxiliary pump
has a lower output pressure, than a priority valve can be used to isolate non-critical components at the
lower aux pump pressure capability.
Priority valves have been designed in a number of ways. Two of the most common types are shown
here. The simplest priority valve is similar in design and operation to a check valves (see Check
Valves, Hydraulic Description) or relief valves (see Pressure Relief Valves, Hydraulic Description).
Referring to Figure 1, flow through the valve only occurs when inlet pressure, P
1
is sufficient to
overcome downstream pressure, valve friction and spring force. By setting the spring force sufficiently
high so that high pressure is required to overcome the spring force, the valve can act like a priority
valve (reference figure below). Some priority valves allow free flow in the reverse flow direction (the
priority valve shown in Figure 1 does not allow reverse flow).

Figure 1 Simple Priority Valve Schematic

A 2
nd
arrangement for a priority valve is shown in Figure 2. In this valve, flow goes through a single
port and flows out through port A and port B. As the pressure at the inlet drops, the spring will push
the piston to left. This will start to reduce the flow area through port B and eventually when the
pressure is low enough, the flow area through port B will be zero and all available flow will go through
port A. Hence at the outlet of this valve will be 2 sets of functions. The lower priority functions will
receive flow through port B while the high priority functions will receive flow through port A. Also, note
that flow rates through both port A and B can be tailored based on port sizes in the housing and by
controlling critical flow related dimensions in the piston.

Figure 2 Priority Valve Schematic

A third type of priority has a single inlet and two outlets primary and secondary. To control flow
between the two outlets, a servo (spool) is monitors downstream pressure in both lines. When the
pressure in the primary outlet drops a force imbalance causes the spool to move reducing the flow
area to the secondary outlet. When pressure returns to nominal level, the spool moves back to the
neutral position and normal flow occurs in both outlets. In this way, the valve acts a pressure
compensated flow control valve on the primary outlet with excess flow going to secondary services or
possibly back to return. This type of valve is more common in industrial hydraulic systems and is not
used much in aerospace. There may be some military applications (cargo doors, ramps, etc.) where
this type of valve may be useful.

When selecting a priority valve, the following parameters should be considered
Pressure Setting this is the pressure at which the priority valve opens and allows flow
Valve Pressure Drop what is the pressure drop across the valve in normal operation (when pressure
is sufficient to hold valve open). This will affect pressure available to downstream components and
must be taken into account when sizing downstream components.
Reseat Pressure - what is the re-seat pressure (the pressure where the valve will close, once open)?
Flow Rate what is the rated flow rate when the valve is open? Ensure the valve does not become a
flow limiting device. If there are 2 outlets, what are the flow rate characteristics as a function of poppet
position (or inlet pressure)? Are the flow characteristics acceptable from a system point of view?
Reverse Flow Capability Is reverse flow capability a requirement? Is so, what is the pressure drop
vs. reverse flow characteristics?
Leakage What is the valve leakage rate when pressure is below priority pressure setting? What is
the affect of leakage on downstream components? Can this leakage be tolerated?
temperatures
Seals/Clearances although there may not be seals on a piston or poppet, there will be seals on the
housing. The seals should be compatible with type of fluid and the environment where the valve
operates. For priority valves with ball or conical poppets, damage may be possible when re-seating
against sharp edges (see Figure 3). Wear around the poppet seats will lead to leakage. For relief
valves with elastomeric seals, see Seals - Hydraulic Components for discussion on hydraulic seals.

Figure 3 Poppet Sealing Areas for Potential Wear

Failure Modes the main failure mode is a jam of the poppet in the closed position or open position. A
failure in the open position would be a latent failure. It may also be possible for the adjustable device
on the valve spring to fall out of adjustment leading to lower or higher regulation settings.
Hysteresis what is the difference between the relief pressure setting and reseat? This difference
should be large enough to prevent any chattering of the valve. Chattering may occur in a situation
where the valve shuts and system pressure increases quickly due to reduction in flow through the
valve and the valve then re-opens. Then the pressure reduces and the valve closes, repeating the
cycle at a very high frequency. This could happen if the priority valve setting and the degraded
pressure level are close to each other.

Priority Valve Qualification
Pump, Hydraulic Description
All hydraulic systems require a source of hydraulic power. In most applications, the source of hydraulic
power is a variable delivery axial piston pump. Variable displacement means that pump outlet flow
varies according to system flow demands (as more sources actuator, motors, etc. are operating,
the pump will increase output to maintain maximum pump outlet pressure). Other sources of hydraulic
power are vane or gear pumps (see Motors, Hydraulic Descriptionfor description of vane or gear
rotating pumps/motors), or an accumulator (see Accumulators, Hydraulic - Description). An aerospace
vehicles main hydraulic pumps are usually mounted on the engine and connected to the engine
rotating shaft through a gearbox. Pumps may also be driven by an electric motor, APU, ram air
turbine, or second hydraulic system (using a hydraulic motor and pump combination).
To understand how a variable delivery axial piston pump operates refer to the pump cross sectional
view shown in Figure 1. The figure shows pump components and also how pump outlet pressure is
controlled through a compensator valve and control piston arrangement. The key element in control of
pump outlet flow is control of the swashplate angle, , which in turn controls piston relative
displacement and hence pump flow. Ideally, the pump delivers zero flow when there are no flow
demands and the required flow when required while maintaining system design pressure at all times.
Variable displacement pumps dont obtain this ideal goal, but they do come close when flow demands
are within their design flow range.
Prior to the introduction of variable delivery pumps 50-60 years ago, pumps were fixed delivery which
means they always delivered the same amount of fluid irregardless of system flow needs. Unused
fluid was then ported back to the reservoir though a pressure relief valve. Fixed delivery pumps
wasted a lot of energy through heat. A fixed delivery pump has a fixed swashplate angle and therefore
no compensator valve or control piston (refer to Figure 1). Fixed delivery pumps may be found today
in very specialized applications where the fixed delivery (or flow) is tailored to the application to
minimize wasted energy. The benefit in this case is a lighter, less expensive pump. An example might
be a standalone hydraulic system to power a large cargo door where flow demands are constant while
the door is moving and the pump would only be on during door movement. A pressure relief valve
should also be installed in this application.

Figure 1 Simplified Pump Schematic

Referring again to Figure 1, pump flow rate is determined by the swashplate angle, , which in turn
controls pump piston displacement. Swashplate angle is controlled by the compensator valve and
control piston. The compensator valve sets the no flow outlet pressure of the pump (e.g., 3000 50
psi) and meters flow to the compensator piston. The compensator is fed hydraulic pressure from the
pump outlet and is positioned based on a force balance between compensator chamber pressure
acting on the piston and spring force (plus, to a lesser extent, friction and flow forces). The housing is
fixed and does not rotate. The housing must remain fixed so that inlet/outlet ports, compensator valve,
solenoids, and other equipment can be mounted to the pump. All other parts rotate at the pump
speed.
At the compensator pressure setting (e.g., 3000 50 psi for a 3000 psi system), the swashplate angle,
, is zero. As the pump flow increases, the pump outlet pressure decreases. As the outlet pressure
decreases, the compensator moves towards the closed position and the pressure on the control piston
is reduced. As the control piston pressure is reduced, the control piston spring pushes the control
piston so that the swashplate angle, , increases, resulting in greater piston stoke and increased flow
rate.
Figure 2 shows 3-dimensional view of a piston hydraulic pump. This figure provides a better view of
how a pump is built and how the pistons and swashplate operate. Not shown in the figure is a
compensator valve and control piston, so this is more representative of a fixed delivery pump. The
pistons are attached to the swashplate via a spherical bearing arrangement. The swashplate does not
rotate. As the cylinder rotates the piston is at the lower end of the cylinder during one part of a
revolution (intake) and at the top end of the cylinder during the other of the rotation (high pressure
outlet side of the pump). During 1 revolution of the swashplate, each piston will pull in fluid and push
out high pressure fluid once. For a nine piston pump, this will lead to 9 pressure pulsations per 1
revolution of the swashplate.

Figure 2 Pump Cross Section

The relationship of outlet flow to outlet pressure for a variable delivery pump is shown in Figure 3. This
plot can be used to estimate pump flow for a given outlet pressure. The plot also shows the flow rate
where pump outlet pressure starts to drop off dramatically. It is important to note that Figure 3
represents pump characteristics for a fixed pump rotational speed (RPM) or displacement (in
3
/rev).
Normally these curves are provided for the rated pump speed, but in aircraft engine speeds vary -
hence pump speed varies and maximum flow varies also. Therefore, the curve shown in Figure 3 will
shift down for lower pump speeds and shift up for higher pump speeds up to the maximum flow of the
pump.

Figure 3 Typical Flow vs Outlet Pressure Plot (for a given pump rotational speed)

The response of the pump to a change in outlet flow is on the order of 50 milliseconds.
Figure 4 shows the relationship between pump flow, efficiency and outlet pressure. The drop off in
flow occurs due to hydraulic fluid leakage at higher pump outlet pressures (higher delta pressure
across the piston seals), which is equivalent to volumetric efficiency. As volumetric efficiency drops off,
pump outlet flow drops by the same amount. The other curve in Figure 4 is the overall efficiency curve
(overall efficiency = volumetric efficiency x mechanical efficiency). Pump horsepower increases
linearly with pump speed.

Figure 4 Typical Performance Curves for a Variable Delivery Piston Pump

Pump Design Considerations
The most important characteristics for a hydraulic pump are listed below. These parameters assume a
variable delivery constant pressure pump.
Rated Pressure this is the nominal pressure setting of the pump and must be compatible with the
design operating pressure for the system (e.g., 3000 50 psi for 3000 psi system or 5000 50 psi for
a 5000 nominal psi system).
Rated Speed this is the nominal speed rating of the pump. The gearbox connecting the pump to the
driver (engine, APU, etc) will need to be compatible with the drive unit speed and the rated pump
speed. The minimum RPM of the pump may also need to be considered
Design Displacement this is the flow per revolution of the pump (in
3
/rev) that the pump is capable of
achieving without a significant reduction in outlet pressure.
Flow vs Pressure Curve this is a plot that has flow rate on the y-axis and pump outlet pressure on
the x-axis. This plot shows the drop off in pressure at a given flow conditions and shows where the
knee in the flow curve lies (see Figure 4). This graph is required for a simulation model. Nominally, this
graph is provided at the rated speed of the pump. If available, this graph for various pump speeds
would be helpful - otherwise the flow can be ratioed using the design displacement for different pump
speeds.
Temperature Rating the pump must be rated for the temperature extremes that it will see in
operation, such as engine nacelles. Pump seals are the most critical component when considering
temperature.
Case Drain Pressure this is the nominal pressure that would be in the pump case. All pumps have a
case drain line to provide a flow path to the reservoir for hydraulic fluid that flows by the piston seals
and fluid that flows through the compensator. Without a case drain the pressure would blow out a
case or shaft seal. The case drain pressure needs to be greater than the reservoir pressure (and line
resistance) to assure drainage from the case to the reservoir.
Inlet Line Size Standard pumps will have an inlet port sized by the manufacturer. The connecting
inlet line/hose will need to be compatible with the port size and type.
Outlet Line Size - Standard pumps will have an outlet port sized by the manufacturer. The connecting
outlet line/hose will need to be compatible with the port size and type.
Case Drain Line Size - Standard pumps will have a case drain port sized by the manufacturer. The
connecting line/hose will need to be compatible with the port size and type.
Recommended Inlet Pressure to operate properly a pump must be supplied sufficient hydraulic fluid
at a pressure level sufficient to fill the piston cylinders as the pump rotates. A pump manufacturer will
provide recommended inlet pressures and the reservoir and reservoir to pump hydraulic lines need to
designed/sized to meet this requirement. In some instances, a boost pump may be required to
achieve desired inlet pressures and flows.
Number of Pump Pistons most aircraft piston pumps have 9 pistons. An odd number of pistons have
been shown to have smaller output pressure fluctuations than an even number of pistons and
experience has shown 9 pistons to be an optimum number for performance.
Power Requirements what horsepower is required to drive the pump at its maximum operating
condition. Horsepower is the product of flow and delta pressure across the pump, divided by the
overall efficiency of the pump.
Maximum Fluid Viscosity Pump manufacturers will provide a recommendation for the maximum
recommended fluid viscosity. If fluid viscosity is greater than the recommended value, then the pump
may start to cavitate with the pump inlet at the recommended inlet pressure. Maintaining the fluid
within the necessary viscosity range will affect the pump inlet system design.
Seals Seals must be compatible with the specific type of hydraulic fluid used in the pump.
Specifically, seal material for carbon based fluids and synthetic fluids are different.
Filtration Requirements A pump manufacturer will provide recommended fluid cleanliness
requirements to help ensure reliable pump operation. This filtration requirement needs to be taken into
consideration when designing/sizing the return filters in the hydraulic system.
Weight Is always a concern on aircraft. A oversized pump is not desirable from both a weight and
cost standpoint.
Envelope engine nacelles and APU installations usually have limited space available and the pump
installation must be compatible with available volumes
Shaft Type Shafts are usually splined and the spline characteristics must be defined to ensure a
proper interface to the
Direction of Rotation Pumps can rotate either clockwise or counterclockwise and may be of a
concern in certain installations.
Mounting The bolt flange mounting of the pump must be compatible with the attachment on the
gearbox or other attaching plate.
Relief Valve A relief valve should be installed in the outlet (high pressure) line downstream of the
pump. This relief valve will provide protection against hydraulic shock loads, thermal expansion and
any possible overpressure condition. The relief valve setting should be 5-10% greater than maximum
pump pressure.

Pump Installation Considerations
Considerations for the mounting/installation of pumps include vibration, temperature, alignment of
drive motor to pump, spline matching and torque requirements. In some applications, a means of
quick installation and removal is required. Quick installation devices must uphold rigidity of the pump
installation and maintain alignment of rotating shaft.
Vibration Need to consider vibration from power source such as the engine or APU, vibration
characteristics from the pump, g-loading and possibly flutter. Mounting should be sufficient to
withstand these loads from both a stress and fatigue standpoint. Testing to appropriate levels from
RTCA-DO160 or MIL-STD-810, Method 510 should be conducted.
Temperature Due to the high speed and compression of fluid, pumps operate at high temperatures.
Additionally, the power source for the pump, such as the engine, is at high temperature. Temperature
considerations should include pump seals, fluid temperature, mounting pads, connecting hoses or
tubes, etc.
Pump/Motor Alignment Alignment of pump to drive motor splines needs to held to tight tolerances.
Considerations are tolerance stickups, relative motion between drive motor and pump, possible
angular displacements on installations, spline teeth dimensions, etc. Improper alignment can cause
excessive vibration (leading to premature failure), or failure of the pump shaft seal.
torque is higher than running torque. Start-up torque accelerates the mass/inertia of the pump and
fluid, leading to temporary high stresses within the mounting hardware and pump. This is more of a
concern on APU and RAT installations. Obviously, the speed of the drive motor must match the
manufacturers recommended speed for the pump (usually accomplished through a gear box).
Axial and Radial Shaft Load Capability Ensures pump shaft and splines are adequately sized for
static and fatigue loads that the pump will see over its operating life.
Case Drain Line A case drain line is installed to drain pump leakage flow back to the reservoir. Case
drain back pressures affect seals and bearings (via load balance across them), balance and loading of
the pump rotating hardware, and piston leakage characteristics. In most aircraft installations, the case
drain line back pressure is equal to the reservoir pressure (which is approximately the pump inlet
pressure). This minimizes leakage on the intake side of the pump. Normally, back pressure from the
case drain line is not an issue, but if back pressures are abnormally high (such as clogged filter), the
effects on the pump should be looked at more closely. Case drain pressure < 150 psi is a rule of
thumb for good pump life. The case drain line must be large enough to cover the maximum case drain
flow at the nominal case drain pressure or even a slightly lower pressure. The case drain line will
normally flow back to return through the return line filters. In some applications, a separate filter is
used on the case drain line.
Inlet Line The inlet line to the pump is designed as part of the pump inlet system. The pump inlet
system consist of reservoir and reservoir pressurization and tubes/hoses from the pressure to the
pump. This system should be designed to ensure fluid is provided to the pump inlet within the inlet
pressure range and viscosity recommended by the manufacturer. In sizing the inlet line, the length of
the tubing, bends, height fluid is pumped, reservoir pressurization, additional components (such as a
heat exchanger) in the line and other factors should be taken into account. Sizing of a pump inlet line
uses basic pipe flow equations and reservoir (or supply) pressurization (see Reservoir, Hydraulic
Description).
Outlet Line The pump outlet line should be sized to system pressure drop requirements and to
minimize affects of pump pressure pulsations. Primary considerations in design of the outlet line are
pump pressure pulsations, accumulators (see Accumulator, Hydraulic - Description), pump system
response and parallel pump installation.
Regarding pressure fluctuations, hydraulic fluid has mass and is compressible. Hence the oil in the
pump downstream tubing behaves like a very stiff spring, with variable stiffness as the downstream
configuration changes. Pulsations are a result of each piston within the pump transferring a discrete
amount of fluid to the system, leading to a pulsed input in the hydraulic system. The flow pulses decay
over time from the damping provided by internal flow friction in the downstream tubing and
components. The pulsation frequencies for a odd numbered piston pump are

Example: For a pump running at 2700 rpm with 9 pistons

For aircraft systems, both pulsation frequencies are usually above the response frequency of the
downstream components, however, in some cases the effects may need to be analyzed. The lower
frequency is usually more dominant with pump noise, but both should be analyzed.

System Interaction
System interaction occurs when the natural frequency of the pump compensator is at or near the
natural frequency of a downstream component (such as a servo valve or actuator). Generally there is
sufficient difference between the natural frequencies so that system interactions do not occur.
Another source of interaction occurs when pumps are connected in a parallel arrangement. This
interaction can be stopped by installation of check valves in the outlet lines of each pump.

Pump Pulsation Damping
Several methods exist to dampen the effect of pressure pulsations from a pump:
1. Change configuration (geometry, parts, characteristics, etc.)
2. Increase volumes in pump outlet line
3. Install accumulator close to the pump. Some research shows that for the accumulator to be effective, it
should be installed with 0.3 meters of the pump, and the supply line between the main line and
accumulator should be between 5 and 10 centimeters in length. Also, the volume of the gas accumulator
should be sufficient so that its resonance (response) frequency is less than the pump pulsation
frequency.
4. Install a hose at the pump outlet, or downstream plumbing.
5. Install a Helmholtz resonator (H-filter) in the pump outlet line. A H-filter consists of two lines in
series, of different volumes, that branch away from the main line. By properly selecting the
lengths and cross-sectioal area of both lines, the H-filter can be tuned to a specific frequency.
6. Install a Quincke tube in the main line. The Quincke tube is a side line with areas based on main
line area and lengths sized for a specific frequency.

Pump Qualification
See Qualification - Hydraulic Components for discussion on hydraulic pump qualification and required
Reservoir, Hydraulic - Description
All hydraulic systems have a reservoir. A reservoir is similar to an accumulator, except that the fluid
pressure is constant over all fluid levels. A reservoir performs several functions. First and foremost,
the reservoir holds fluid not required by the system under any given operating condition and accounts
for fluid capacity needs over time in the system. Fluid volume needs will vary during different
operational scenarios, such as gear extension. Secondly, the reservoir provides for thermal expansion
of the fluid over the operational temperature range of the system. Thirdly, the reservoir provides fluid
to the inlet side of the hydraulic pump. Reservoir pressurization levels are a critical aspect of reservoir
installations.
Reservoirs consist of a container or volume, fluid inlet port, fluid outlet port, fill/drain port, and a means
to pressurize the fluid in the volume. Reservoirs used in aerospace are pressurized using pressurized
gas, through a piston which has high pressure hydraulic pressure on one side (commonly referred to a
as a bootstrap reservoir), or through a mechanical piston and spring. The terms separated and non-
separated are used with reservoirs. Separated means that the reservoir fluid is separated from the
pressure source (spring, high pressure fluid or gas) though a separation device (usually a piston).
Non-separated refers to a gas pressurized reservoir where there is no separation between the fluid
and the gas.
Figure 1 shows a pressurized gas reservoir. For a gas reservoir a low pressure gas source is required.
Engine bleed air, fed through a pressure regulator, is often used for pressurized gas reservoirs. A
fixed pressure source, such as a charged nitrogen bottle, is another possible configuration. A bleed air
configuration is shown in Figure 1. The pressure regulator setting will be in the 50-200 psi range.

Figure 1 Pressurized Gas Reservoir Schematic

In the pressurized gas reservoir, P
Air
= P
Fluid
. The air does not mix with the fluid due to the density
differences. However, air can escape from the fluid into the air providing some system removal of air.
This characteristic does not replace system bleeding.
For a gas reservoir, the amount of gas should be sufficient to provide uninterrupted flow of fluid to the
pump under all operating scenarios and vehicle attitudes. For example, a high nose up (pitch) attitude
should not cause fluid to block the air inlet port. In addition, there should be enough gas to absorb
high surge pressure that might occur with high return flow to the reservoir. Surge pressure should be
less than the pressure relief valve setting and, of course, be much less than the proof pressure rating
of the reservoir. The minimum recommended gas volume is 10% of the total reservoir volume. The
inlet line should be located at a place so that inlet flow does create a vortex within the reservoir or
cause foaming of the fluid. The fluid spin caused through vortex action will push fluid to the outside
and can starve the reservoir outlet by opening the outlet up to air. Baffles or diverters can be used in
the reservoir (as part of the reservoir design) to prevent vortexes or foaming. Lastly, if the air
pressurization source is lost there should be a means to allow atmospheric air to enter the gas
chamber of the reservoir.
A bootstrap reservoir is shown in Figure 2. A bootstrap uses a differential area piston where high
pressure hydraulic pressure from the pump outlet is applied to the small area of the piston. This
produces a low pressure on the reservoir side of the piston. A major advantage of bootstrap reservoirs
is that reservoir pressurization is maintained during aggressive flight maneuvers, including negative g
flight. Additional hydraulic plumbing and some components are required for bootstrap reservoir
implementation (see examples in System Design, Hydraulic Power Generation). Also, note the check
valve in the high pressure line. The purpose of this check valve is to maintain reservoir pressure after
the pump has shut down so that the pump inlet is maintained when the engine driven pump is not
rotating. Accumulators may also be used in this circuit to assist in maintaining pump inlet pressure.
The accumulator will be between the check valve and the reservoir.

Figure 2 Bootstrap Reservoir

When the reservoir is at equilibrium P
1
A
1
= P
2
A
2
. Since A
1
>> A
2
, P
1
<< P
2
. The differential piston
areas are set by the pump nominal pump outlet pressure and the required level of reservoir fluid
pressure. Reservoir fluid pressure levels are discussed in Reservoir, Hydraulic Sizing.
Figure 3 shows a mechanical piston reservoir. In this reservoir a spring with an appropriate preload
and a low spring rate pushes on a piston and provides a fairly constant reservoir pressure.

Figure 3 Basic Reservoir Schematic (Piston Spring Type)

Metal bellows is another (relatively new) type of reservoir. A metal bellows reservoir would be similar
to a metal bellows accumulator (see Accumulator, Hydraulic Description). Operation of a metal
bellows accumulator would be similar to a piston/spring reservoir type.

Reservoir Features
The three basic types of reservoirs used in aerospace are shown in Figures 1 through 3. Additional
features may be contained in reservoirs as shown in Figures 4 and 5.
Figure 4 shows a gas reservoir configuration with an emergency reserve capability. The tallest pipe
would be used for normal operation and would therefore feed the normal system pump (usually the
engine driven pump). The shorter standpipe would feed some type of auxiliary pump, such as a
backup electric motor driven pump. The volume difference between the two standpipe inlets is the
emergency reserve capacity for this reservoir. The auxiliary pump would be used for emergency
services such as emergency braking and landing gear. For bootstrap reservoir systems, any
emergency reserve would be contained in separate reservoir. System plumbing must provide proper
isolation between normal operation system functions and auxiliary functions through the use of check
valves, shuttle valves or shut off valves.

Figure 4 Single Reservoir With Emergency Reserve Capacity

When computing emergency reserve capacity, fluid returning to the reservoir should not be assumed.
Therefore, the capacity should be what is required for emergency operation (e.g., to lower the landing
gear and/or run emergency brakes). Military specifications recommend the minimum volume for
emergency reserve be equal to 125% of the calculated emergency reserve needs. Emergency
reserves can also be contained in a separate reservoir. For bootstrap reservoir systems any
emergency reserve will need to be a separate reservoir since the reservoir would provide fluid to each
system and piston movement would not pressurize beyond the to the normal system outlet standpipe.
Some hydraulic systems contain a reservoir with 2 compartments which feed 2 different pumps. This
arrangement is only possible with a gas pressurized reservoir. An example is shown in Figure 5. The
two compartments are connected during normal operation at normal fluid levels. However, should a
leak develop in the main system, fluid will only drain from the main portion of the reservoir. Note that
the return flow is prioritized to the backup compartment and only flows to the main system volume
when the level is sufficiently high. Fluid would be retained in the backup system compartment should a
leak develop in the main system. This arrangement is similar to the emergency reserve reservoir
shown in Figure 4 but this design assumes fluid flow back to the reservoir during backup system
operation. Therefore, full functionality is maintained for the backup functions in the hydraulic system.
For this type of arrangement the hydraulic system plumbing must incorporate check valves and/or
shuttle valves to isolate non-critical functions from the backup pump. For a 2 compartment reservoir,
protection is provided against certain failures in the system primarily loss of fluid in non-critical
hydraulic lines and components. It also protects against a main system pump failure. Normal (or
slightly less than normal if backup pump capability is below that of the normal pump) operation of the
critical functions is maintained in this arrangement. In contrast a reservoir with emergency reserve
only supplies enough for a one time application of critical functions.

Figure 5 Single Reservoir Feeding Normal and Backup Hydraulic Systems

A convenient means for filling and draining hydraulic reservoirs should be part of any reservoir design
and installation. Typically, ground service panels are installed. Ground service panels provide a fluid
connection to the reservoir so that the reservoir can be filled. A reservoir drain can also be installed
into the ground service panel. Ground service panels also provide a means to connect hydraulic
power cart pressure and return lines into the system. This function allows functional and operational
checks of the system while the vehicle is on the ground. The pressure line from the cart will flow into
the normal system pressure line through a tee fitting. The return line from the cart will connect into the
normal system return line through a tee fitting. A shut off valve or check valve can be installed in the
ground service line to prevent normal flow from flowing into the ground service lines. Installation of a
check valve will prevent system loss should a leak develop in the ground service line. However, when
using a valve it is important to not let fluid remain trapped in a ground service line as thermal
expansion will lead to a pipe failure. Ground service panels are for ease of maintenance and are
typically found on most commercial aircraft. A ground service panel (or equivalent) is required for each
independent hydraulic system on the vehicle. In lieu of a panel, tubes with appropriate fittings (such as
quick disconnect) can be located wherever convenient on the vehicle, such as maintenance bays or
wheel wells. If a ground service panel is not used, the functions described above (filling, draining,
external pressure line, external return line) are required for maintaining a system.
Another important feature for a reservoir is fluid level indication. Indications should be provided when
the reservoir fluid level is low, when the reservoir is overfilled and when fluid level is ok. Ideally, fluid
level indication would be continuous throughout the range of fluid in a reservoir. This could be
accomplished through a sight gauge located directly on the reservoir. The drawback of gauges located
directly on the reservoir is that the reservoir may not be easily accessible in the vehicle. Also, on pre-
flight checks flight crews dont normally like to access maintenance bays. Electronic means through a
continuous sensor and indicator are preferred. The indication could then be provided in the cockpit
and the ground service panel. Another indication method is to install switches on the reservoir at the
low level point, fill point and overfill point. The switches can then be used to power indication lights on
a ground service panel. Switches are a less expensive alternative and are appropriate for hydraulic
systems whose criticality is less severe (i.e., not required for continued safe flight and landing). The
drawback of switches is the potential for a latent failure in a switch or indicator light. Since fluid volume
varies with temperature, electronic sensors can provide fluid level indications that are compensated
for fluid temperature. For flight safety critical hydraulic systems, the indication system should be fail
safe such that no single failure in the indication system will lead to an erroneous indication.
In addition the above, reservoirs should provide over pressure protection and also provide a means to
relieve pressure on the reservoir for servicing. Over pressure protection is normally done using a
pressure relief valve for fluid or air depending on type of reservoir. A manual relief valve is used to
release reservoir pressure for servicing. In some applications, a bleed port for bleeding air from the
system may be contained in the reservoir. A bleed valve would only be required in the reservoir could
collect trapped air and/or is located in the system such that entrained air would migrate to the
reservoir.

Reservoir Construction
Reservoirs are normally made from aluminum. High strength materials are not required since
operating pressures are low. Non boot strap reservoirs are typically cylindrical in shape, but may also
resemble a box with rounded corners. Bootstrap reservoirs will consist of two separate cylinders
connected together along a flat surface (see Figure 3). Materials will be corrosion resistance.
Externally, there will be an inlet fitting and outlet fitting plus connections for fill and drain ports. The
inlet fitting should be below the minimum fluid level. Fittings are normally internally threaded bosses.
Permanent welded or swage connections are not used on reservoir connections. No weight or
preloading should be carried by connecting tubes. In some cases, a standpipe would protrude a small
amount above the bottom of the reservoir so that debris would be trapped in the bottom and not pulled
into the system. The outlet should be located below the minimum required fluid level for obvious
reasons.
Baffles and diverters are also common to prevent fluid from sloshing around and also to stop any
swirling action from flow coming into the reservoir. Baffles can be installed in the up/down direction or
side to side.
The indication system will also be part of the reservoir. Indication systems can be sight glass, switches
located at the fill level and full level, or a capacitance based sensor which provides an analog signal
proportional to fluid volume.

Reservoir Capacity
Reservoirs are sized to have sufficient capacity to provide flow under all operating conditions, account
for expansion and contraction of fluid at different temperature and system leakage over time (between
maintenance intervals). See Reservoir, Hydraulic Sizing for details on computing capacity
requirements.

Reservoir Fluid Pressure
The pressure setting for a reservoir is determined by pump inlet pressure requirement and the
pressure loss characteristics between the reservoir and the pump. Longer tube runs, tube bends, tube
diameter, difference in elevation between pump and reservoir, fitting/connector losses and shut off
valve losses all contribute to pressure loss between the reservoir and the pump. The best designs
minimize these losses as much as possible. On large vehicles where there is large separation
between reservoir and pump and/or large change in elevation, boost pumps are installed in the pump
inlet line. The boost pump could be driven from high pressure hydraulic fluid or through an electric
motor. Alternatively, inlet impellers powered by the pump shaft can be part of the pump to boost inlet
pressure and accelerate fluid into the pump. More information on determining the reservoir
pressurization level can be found in Reservoir, Hydraulic Sizing.
Another aspect of reservoir pressurization is the ability of the reservoir to maintain pump inlet pressure
during negative g flight. Negative g flight is usually only a transient phenomenon so the system need
only be designed for short durations of negative g. For highly maneuverable vehicles where longer
durations of negative g flight are required, a bootstrap reservoir should be used.

Installation Considerations
Reservoirs are considered items of mass and must be mounted to meet g-loading and
crashworthiness requirements. No loads should be imparted to connecting tubes.
Reservoirs should be located as close as possible to the pump(s). The primary installation design
goals are to minimize the length of the pump inlet line and fittings/components in the line to minimize
pressure losses. Elevation differences between the reservoir outlet and the pump inlet should be kept
as small as possible. However, reservoirs must not be located in any rotor noncontainment zone. Shut
off valves should be installed in pump inlet line to shut off flow when required and provide for engine
fire protection. Shut off valves should also not be located in a rotor noncontainment zone.
Check valves are normally in the pressure line to the pump so that fluid would be held in the pump
inlet line during reservoir pressurization transients. These transients could occur during engine start or
at other conditions when reservoir pressure is momentarily lost. Check valves also maintain fluid in the
line to minimize fluid flow delays that could result from an empty tube. Negative g flight is another
reason to install a check valve in the pump inlet line.
Reservoirs also remove heat from the system. The amount of heat removed depends on the reservoir
size and the location in the aircraft. If the reservoir does not transfer sufficient heat to the environment,
then a heat exchanger is required (see Thermodynamics, Hydraulic Systems).

As a summary of the above, when selecting a reservoir, the following factors should be evaluated:
Capacity Reservoir capacity should meet design requirements. Capacity determination is discussed
in Reservoir, Hydraulic Sizing.
Type of Reservoir Separated or non-separated? Will reservoir be pressurized through mechanical
means (spring), pressurized gas or hydraulically pressurized (bootstrap)?
Pressurization System Aspects The choice of reservoir type has a big affect on power generation
system design and possible interface to other systems. For example, a pressurized gas reservoir will
require interfacing to the engine bleed air system (see Figure 1).
Emergency Reserve Capacity Does the reservoir require an emergency reserve capacity? If so,
what is this capacity over and above the normal capacity?
Fill and Drain Features How is the reservoir filled with fluid? How is the reservoir drained for
maintenance? How is the reservoir pressure bled off?
Fluid Level Indication Fluid level indication method should be compatible with the vehicle indication
system. This system may need to interface with crew compartment indication systems.
Pressure Rating Ensure the reservoir is rated appropriately for the system pressure
Pressure Relief Valve Is an overpressure relief valve incorporated into the reservoir or is this
installed in the pressure line?
Temperature Rating Reservoir should be rated for maximum rated fluid temperatures in combination
with applicable environmental temperatures
Materials Reservoir materials must be non-susceptible to corrosion from hydraulic fluid, any fluid
contamination and any possible moisture that could be in the system. In addition, reservoir strength
should be sufficient to pass proof and burst testing and vibration.
Seals Seals should be appropriate for the hydraulic fluid used. Different fluids may require different
seal materials.
Mounting Method of mounting the reservoir to vehicle structure should be robust. Reservoirs are
considered items of mass and the attachments are subject to substantial g loading design
requirements.
Connections What interfaces are provided (type of fitting) for connections with the hydraulic tubing?
Operational Characteristics The reservoir needs to provide sufficient fluid to the pump without
cavitation. In addition there should be no froth or foam characteristics in the fluid exiting the reservoir.
To accomplish this requirement, reservoir design should include baffles, diverters, sufficient inlet and
outlet flow tube size (diameter), sufficient level of pressurization. These aspects in the design should
be review carefully and operational tests completed to prove out operating characteristics under all
foreseeable operating conditions.

Reservoir Qualification
See Qualification - Hydraulic Components for discussion on reservoir qualification and required
certification testing. For a reservoir, many of the categories listed will not apply due to low pressure
and simplicity of the component. At a minimum, temperature, proof and burst pressure, vibration, and
endurance should be run. Some specifications may require an immersion test to verify materials are
not subject to corrosion or breakdown in service.
Operational tests should also be run on the pressurization system and reservoir operation at minimum
and maximum fill levels. Reservoir capacity should be verified by test. Relief valve operation and
indication systems should also be verified by test.
Seals Hydraulic Components
Seals can be static seals that provide sealing between non-moving parts, or dynamic seals such as on an
actuator piston or switch plunger. Historically, seal and groove geometry and overall seal requirements have been
specified in MIL-G-5514. However, this document is no longer used for new military designs but is still available
as a reference. MIL-G-5514 has been replaced by a SAE document
Today, the best source of data on types and seals and their applications are seal manufacturers. Seal material
(specifically material type, hardness and friction characteristics) has a big impact on seal performance over time.
O-rings are the most common type of seal and have been used in both static and dynamic applications. However,
in dynamic operations o-ring seals can roll, tear and may have reliability issues. Other types of seal configurations
include T-seals, H-seals (Hat Seals), L-seals, wedge or triangular type seals, plus others. A T-seal cross-section
is shown in Figure 1.

Figure 1 T-Seal Cross Sectional View (Dynamic Seal)

Associated with most seal installations are backup rings. Backup rings provide support for the seal and do not
usually provide any sealing. The also function to keep the surfaces clean so that small debris (hydraulic
contamination) does not damage the seal.
When evaluating a seal for a particular application, the main design variables are fluid compatibility, operating
pressures, expected impulse pressures (levels and cycles), design life goals, external environment, desired
friction levels and leakage. Service history data for a particular seal configuration in a similar environment can
assist evaluations. Seal manufacturer data and experience is also very helpful. Selected seal arrangements
should be tested in a thorough qualification test program, which includes endurance testing in an environmental
chamber.
In applications where seal reliability is critical, dual seals can be used in selected locations. The dual seals can be
different designs, for example, a T-seal with a backup wedge seal. Seals should be qualified individually to
establish the robustness of both seal designs in the application. The disadvantage of dual seals is increased
friction (heat generation), increased assembly time and cost.

Servo, Hydraulic - Description
A servo is shown in Figure 1 and consists of a spool (two lands connected by a rod) and an outer
sleeve (sometimes called a bushing) with flow ports drilled in the sleeve. The position of the spool
determines the flow areas and hence controls the amount of flow through the valve. The spool can be
positioned in a number of ways. Some of the more common methods are two position solenoid
controlled (open/closed), proportional solenoid controlled (position is directly proportional to applied
current), mechanical lever controlled and a servovalve arrangement (see section on Servovalves,
Hydraulic - Description). Some form of servo type valve is used in almost all of the hydraulic paths to
an actuation component.

Figure 1 Basic Servo Showing Flow Geometry and Parameters

Flow through servovalve openings is characterized by the orifice flow equation using a relationship
between valve spool position, x
v
, and flow cross-sectional area, A
v
. As can be seen in Figure 1, the
flow cross sectional area is a function of spool position. Turbulent flow is usually assumed since in
most cases the pressure drop across the servo is sufficiently large. The flow rate in a servo is normally
controlled by the outlet flow area and hence the outlet flow area is smaller than the inlet flow area.
Servos are either zero lapped, over lapped or under lapped as shown in Figure 2. In zero lapped
servos, the width of the land and the width of the flow port are equal. Thus there is only one position
for zero flow. This configuration generally results in the tightest control and is commonly selected for
high precision servo valves. For over lapped servos, the land width is greater than the flow port width
resulting in a deadband in the flow area vs. pressure curve. The means a minimum amount of spool
movement (equal to the overlap dimension) must occur before flow will occur. The main advantage of
an overlapped servo is the servo is more tolerant to noise factors in the control and also to
manufacturing tolerances. This occurs at the price of accuracy. For under lapped servos, the land
width is smaller than the port width, which results in servo flow at all positions of the spool. The
benefits of a under lapped valve is faster response time with a cost of high leakage flows.

Figure 2 Servo Lapping

Servo Flow Characteristics
Using the turbulent orifice equation, the flow expression for flow through servo flow ports is
(1)
where is the area of the valve orifice (servo port). The flow area depends on port geometry,
which varies with manufacturer, valve type, and spool position. Inspection of the equation (1) indicates
that the flow rate varies proportionally with area if the p is held constant, and that the flow rate varies
with the square root of p if the flow area is held constant. Figure 3 shows notional charts of the flow
behavior for a servo which are similar to orifice flow graphs.

Figure 3 Flow Rate Behavior for a Servo

The effects of a lapping can be seen in Figure 4. Figure 4 assumes a 4 way servo and illustrates ideal
flow curves. In the figure, control flow (flow through the valve port) is plotted against valve position
(where p is assumed constant) for under lapped, zero lapped and overlapped valve. For a zero
lapped valve, the curve goes through the origin. For an overlapped valve, the flow is zero until the
valve spool has moved sufficiently to allow flow. For the under lapped valve, there is flow through both
directions of the servo yielding a zero flow to the load at the null position. However, as the under
lapped valve moves off of null, flow to the load will change quickly (higher gain) and change less
rapidly once the spool has moved to the point where flow through one side goes to zero.

Figure 4 Effects of Valve Lapping on Flow

For a two position (open/close) solenoid servo, the flow area is either A
min
(which is usually zero) and
A
max
. Changing between A
min
and A
max
takes less than 100 milliseconds in most applications (see
Figure 5 below).

Figure 5 Flow Rate Behavior for a Two Position Servo

For a mechanically actuated servo valve the flow area will be a function of the mechanical linkage,
Flow Area = f(Input, Geometry). Figure 6 shows an example of a mechanically actuated servo.

Figure 6 Mechanically Positioned Servo

Lateral Spool Forces
If an unequal pressure distribution occurs in the small clearance between the spool outer diameter
and the inner diameter of the bushing, a lateral force will be applied to the spool. This force can
sometimes result in a spool experiencing a hydraulic lock condition. Here the lateral force is so great
that the spool cannot be moved as long as hydraulic pressure is applied. Minor machining tolerances
on the spool outer diameter can result in slightly varying leakage flow across the spool periphery
setting up the unequal pressure distribution on the spool periphery. This type of lateral force is
commonly minimized by machining grooves around the spool outer diameter since the grooves allow
the pressure to equalize around the periphery of the spool. The grooves are perpendicular to the bore
the spool rides in (see Figure 7).

Figure 7 Servo Showing Spool Grooves

When selecting a servo, the following factors should be considered
Pressure Rating make sure servo is rated for your system pressure
Pressure Drop Across the Servo in the Flow Range this will affect design pressure available to a
downstream component and will affect sizing of that component. In some case, an orifice or flow
restriction is included in a servo to regulate flow or pressure drop downstream of the servo.
Temperature Rating servo should be rated for fluid temperatures and applicable environmental
temperatures
Spring A spring is often used in servos to return the servo to the non-powered position. Spring
forces should be examined to ensure they are sufficient to return the servo to the non-powered
position in all operating conditions. When powering, the solenoid or mechanical linkage must be
capable of overcoming the spring forces in all operating conditions.
Actuation Method What method is used to control servo position (mechanical lever, solenoid, etc.)?
This will usually be defined by the overall system layout and design methodology of the hydraulic
system. The actuation method also requires analysis. For example, if the servo is controlled by a
solenoid maximum and minimum available voltages need to be determined, and a solenoid needs to
be selected with is compatible with this voltage range. When a mechanical linkage is used,
characteristics of the mechanical system will be important. Characteristics to examine include inertia,
friction, maximum applied load, compliance (stiffness) of structure and components, reset forces if a
spring in the servo needs to reset a mechanism, etc.
Speed of Actuation Device this is especially important for 2 position (open/closed) servos. A fast
response servo will create pressure waves when it closes. Where a servo is used to shut off flow in
response to a detected fault in the system, there is usually a requirement on the time required to close
the servo. Many solenoids will open/close a valve in 50-100 milliseconds.
Closed Loop System Interactions when using a servo in a closed loop system, the closed loop
system interactions (both stability and closed loop performance) should be evaluated.
Friction friction of the spool may be important for mechanically actuated servos. For example,
mechanical primary flight control linkages will have minimum and maximum friction requirements. Any
servo connected to the linkage will add to the friction in the mechanical system. Friction is generally
not a lower concern for solenoid actuated servos. (See Friction Hydraulic Components)
Seals/Clearances affects overall reliability of the servo. Some servos may not use seals and will
Failure Modes the main failure mode are servo valve jamming in any position from full closed to full
open and contamination. It may also be possible for a solenoid to not position the servo properly for a
given current due to high friction or resistance in the spool or solenoid.
Chattering servo should be evaluated for potential to exhibit chattering or limit cycle behavior under
servo and the damping. This is primarily a concern for proportional servos where the servo can be
positioned in any position between open and closed.

Servo Qualification
See Qualification - Hydraulic Components for discussion on qualification of hydraulic servos and
required certification testing.

Servovalve, Hydraulic - Description
A hydraulic servovalve is a servo (see Servo, Hydraulic Description) with a device (either flapper
nozzle or jet pipe) used to position the servo. The term electrohydraulic servovalve is often used
because servovalves are controlled through an electrical signal. Servovalves are normally used when
accurate position control is required, such as control of a primary flight control surface. Position control
is achieved through a closed loop control system, consisting of command sensor, feedback sensor,
digital or analog controller, and the servovalve. Servovalves can be used to control hydraulic actuators
or hydraulic motors. When a servoactuator is used to control an actuator, the servovalve and actuator
combination are often referred to as a servoactuator. The main advantage of a servovalve is that a low
power electrical signal can be used to accurately position an actuator or motor. The disadvantage is
complexity and cost which results from a component consisting of many detail parts manufactured to
very tight tolerances. Therefore, servovalves should only be used when accurate position (or rate)
control is required. For more information on closed loop position control using electrohydraulic
servoactuators, seeServovalve, Hydraulic Position Control. Standard servoactuator terminology is
provided at the end of this section.
A schematic of a servoactuator is shown in Figure 1. The actuator is included to show how the
servovalve and actuator components work together. The primary components in a servovalve are a
torque motor, flapper nozzle or jet pipe, and one or more spools. The flapper/nozzle (alternatively jet
pipe) and the spool valve are considered stages. A stage provides hydraulic force amplification:
flapper/nozzle or jet pipe goes from low power electrical signal to spool p and the spool valve
amplifies p on the actuator. The servovalve shown in Figure 1 is a 2 stage servovalve. Almost all
servovalves are 2 stage, but some 3 stage designs exist. A 3 stage servo has an additional spool
valve between the 1
st
spool valve and the actuator. The 1st spool valve provides a spool p to the
2
nd
spool valve.
The servovalve shown in Figure 1 uses a flapper nozzle. A servovalve has a hydraulic pressure inlet
and an electrical input for the torque motor. The input current controls the flapper position. The flapper
position controls the pressure in Chambers A & B of the servo. So, a current (+ or -) will position the
flapper, leading to a delta pressure on the servo, which cause the servo to move in one direction or
the other. Movement of the servo ports hydraulic pressure to one side of the actuator or the other,
while porting the opposite side of the actuator to return. Operation of a servovalve is described in
more detail below.

Figure 1 Flapper Nozzle Servoactuator

Flapper Nozzle System
Flapper position is controlled by the electromagnetic torque motor (see top portion of Figure 1). The
torque developed by the torque motor is proportional to the applied current. Currents are generally
small, in the milliamp range. A torque motor consists of two permanent magnets with a coil winding
attached to a magnetically permeable armature. The armature is part of the flapper piece. When a
current is applied to the coils, magnetic flux acting on the ends of the armature is developed. The
direction of the magnetic flux (force) depends on the sign (direction) of the current. The magnetic flux
will cause the armature tips to be attracted to the ends of the permanent magnets (current direction
determines which magnetic pole is attracting and which one is repelling). This magnetic force creates
an applied torque on the flapper assembly, which is proportional to applied current. In the absence of
any other forces, the magnetic force would cause the armature to contact the permanent magnet and
effectively lock in this position. However, other forces are acting on the nozzle, such that flapper
position is determined through a torque balance consisting of magnetic flux (force), hydraulic flow
forces through each nozzle, friction on the flapper hinge point, and any spring (wire) connecting the
flapper to the spool (which is almost always installed used in servovalves to improve performance and
stability).
As the applied current is increased, the armature and flapper will rotate. As the flapper moves closer
to one nozzle, the flow area through this nozzle is decreased while the flow area through the other
nozzle increases. The flapper generally rotates over very small angles (~ 0.01 rad) and the gap (G in
the figure) is around 0.002 0.003 inches. If the gap, G, between the magnet and the flapper end gets
too large, the torque motor may latch and become inoperative due to limited available torque from the
torque motor.
The flapper nozzle consists of the flapper, two inlet orifices (O
1
and O
2
), two outlet nozzles (n
1
and n
2
),
nozzle backpressure nozzle (n3) and usually a feedback spring. As described above, the torque motor
positions the flapper, which in turns controls the flow through the nozzles. The inlet orfices, O
1
and O
2
,
are important as they create a pressure volume whose pressure is controlled by the flapper.

Figure 2 Flapper Nozzle Geometry

Referring to figure 2, for the flapper nozzle to control flow in a linear manner, the relationship

must be maintained. This relationship implies that the circumferential area created by the flapper
distance to the nozzle must be smaller than the nozzle diameter, such that the circumferential area
controls flow and not the nozzle diameter. In this way, the flow area varies linearly with flapper
position. Also, the torque motor materials, windings and overall design features lead to accurate
control of torque such that small movements of the flapper are possible. This leads to accurate control
of the pilot spool, which in turns provides accurate control of the actuator.
The goal of the flapper and nozzles is to control the pressure acting on both sides of the pilot spool.
When the flapper is in the neutral position, the nozzle flow areas are equal and the pressures P
n1
and
P
n2
are equal. When the flow areas and inlet nozzle pressures are equal, the flow forces through each
nozzle keep the flapper centered in the neutral position. For a zero lapped pilot spool valve, there
would be no flow into or out of the actuator chambers. As the flapper moves towards one of the
nozzles, the outlet flow area is reduced for this nozzle. Outlet flow area increases for the other nozzle.
For example, looking at Figure 1 let the flapper move towards the n1 nozzle. This will reduce the outlet
flow area and the pressure P
n1
will increase. At the same time, the outlet flow area at the n2 nozzle
will increase and the pressure P
n2
will decrease. A delta pressure p = P
n1
P
n2
will occur across the
pilot spool piston and the pilot spool will displace to the right. High pressure fluid will then flow to the
P
A
actuator chamber while the P
B
actuator chamber is ported to return. Depending on the size of the
flapper and nozzles, the p across the pilot spool is limited in magnitude (200-300 lb range for
medium size aerospace applications).
Most servovalves incorporate a feedback spring (wire) between the pilot spool and the flapper. This
wire is shown as a dotted blue line in Figure 1. Examining Figure 1, if the flapper moves to the left, the
p on the pilot spool moves the spool to the right. The feedback wire will then pull the flapper back
towards the neutral position. Hence the feedback wire provides a stabilizing force to the flapper and
helps improve stability and response of the flapper system. This same affect can be done
electronically by putting a feedback sensor (usually a linear variable differential transducer) on the
pilot spool. The output of the sensor is fed back electronically to reduce the current command and
allow the flapper to move back to the neutral position.

Jet Pipe
Another method to control the pilot spool is to use a jet pipe configuration. The jet pipe is an
alternative to the flapper nozzle system; however, a similar torque motor is used to control the jet pipe
position. A schematic of a jet pipe servoactuator is shown in Figure 3.

Figure 3 Jet Pipe Servoactuator

The jet pipe converts kinetic energy of the moving fluid into static pressure. When the jet pipe is
centered between the 2 receiver holes in a receiver block, the pressure on the servo is equal.
However, when the jet pipe is rotated toward one of the receiver holes, the pressure at this receiver
hole is greater than the other receiver hole, thus creating a load imbalance on the servo. Figure 4
shows a schematic of the jet pipe illustrating how pressure varies between the receiver holes as the
jet pipe is rotated.

(a) Jet Pipe Centered (b) Jet Pipe Rotated to Right

Figure 4 Jet Pipe Operation

The stagnation pressure at the tip of the jet pipe is given by Bernoullis equation as

This is the stagnation pressure at the midstream of the flow and would represent the maximum
pressure given by the jet pipe. From the center of the jet stream, the pressure drops off as shown in
the Figure 4(a). In the Figure 4(a) configuration, the pressures on both sides of the servo are equal
(p=0). In Figure 4(b), the jet pipe has been rotated to the right. This has the effect of increasing the
pressure on the right side of the servo and reducing pressure on the left side. The servo will then
move to the left. As a general rule, movement of the jet pipe is sufficiently small such that the
differential pressure will vary linearly over the range of jet pipe travel.
Optimization of nozzle performance is done by experimentation. There is a relationship between the
nozzle diameter and receiver hole diameters, which usually must be developed through testing. Also,
the distance from the nozzle exit to the receiver is important (L = 2 D
n
has been suggested in
literature) as well as the distance between the receiver holes. In general, the receiver holes should be
as close together as possible, to keep P
1
and P
2
as high as possible. It is desirable to keep receiver
holes as large as possible to prevent contamination issues. The goal of a jet pipe design is to achieve
the necessary maximum p across the pilot spool and maintain tight position control (no different from
a flapper nozzle design).
The biggest advantage of the jet pipe valve over the flapper valve is less sensitivity to contamination.
Jet pipe orifices are generally larger than flapper nozzle orfices at the expense of more leakage flow.
A clogged flapper nozzle orifice will cause a servovalve to go hardover in one direction. A jet pipe
valve will generally fail neutral or operate sluggish if the inlet nozzle plugs. However, both
configurations are still used today and both have proven to be reliable and accurate in service.

Servo
The servo part of the valve is exactly the same as any servo or spool valve. The function of the servo
is the same for either a flapper nozzle or a jet pipe servoactuator. The relationship between flow and
p through the servo valve is governed by the orifice flow equation. Servo position is determined by a
force balance on the spool, which includes the p created from the flapper nozzle or jet pipe, friction
forces, spring forces and flow forces acting on the spool. For a complete description of a servo,
see Servo, Hydraulic Description.
When the spool is in the neutral position, the servovalve is in the null position. In some applications, a
compression spring is installed on each side of the servo to help keep the servo centered. In other
applications (spoiler panel servovalves, for example), a spring is installed in one side only which will
push the servo in one direction. For flight spoilers the spring would bias the actuator to the retract
position. So, in the absence of electrical commands, the spring pushes the servo towards the retract
command position allowing hydraulic fluid to flow to the retract chamber. The applied current required
to overcome the spring force and return the servo to the null (no flow) position is referred to as the null
bias. The null bias current will drift in service due to changes in supply pressure, operating
temperatures, wear and other factors. Good servovalve design practice is to keep long term null bias
shifts to within 3% of rated current.
Flow characteristics and the affects of load flow and load pressure drop are determined by the servo
or spool portion of the valve. The equations that describe the relationship between these parameters
are given in Servovalve, Hydraulic Equations. To understand the theoretical performance of a
servovalve, these mathematical relationships must be understood.

Actuator
The actuator part of the servoactuator has the same characteristics as any hydraulic actuator.
See Actuator, Hydraulic Description for more information on actuator designs and characteristics.

Servovalve Flow Characteristics
Plots of typical flow characteristics of a servoactuator are shown below. As stated above, flow
characteristics are determined by the servo. Therefore, the orifice flow equation describes the flow
characteristics. Several figures are shown below to highlight the behavior of servovalves. These
figures represent a zero lapped servovalve, which are the most common in aerospace applications.
Servovalves with overlapped or underlapped spools will have different flow characteristics around null
(for further explanation see Servo, Hydraulic Description).

Figure 5 shows the relationship between control flow and load pressure. The shape of the curves is
determined by the orifice flow equation,
(1)
where i / i
max
is the applied current expressed as a ratio of maximum current, is the density, p is the
pressure drop through the servo flow ports and q is the flow rate through the servo. The servo valve
will have a higher gain around null then at the end of the spool travel.

Figure 5 Servo Flow Characteristics Control Flow vs. Load Pressure

Figure 6 shows the how control flow varies with input current. The relationship is linear for a constant
p across the servo, as can be seen from equation (1). Also shown in Figure 6 is the effect of
pressure drop through the servo. Similar to Figure 5, flow increases for higher p. Figure 6 represents
an ideal servo, with no hysteresis or friction effects.

Figure 6 Servo Flow Characteristics Control Flow vs. Input Current

Figure 7 shows the effects of hysteresis on the ideal flow curve. As current is cycled from 0 to
maximum positive value, then to maximum negative value, then back to zero, the control flow will
follow the blue curve. Hysteresis results from electromagnetic affects in the torque motor as well as
friction in the valve. The electromagnetic effects have a nonlinear characteristic where the width of the
hysteresis loop varies with the input signal. Friction behaves more like a backlash nonlinearity which
tends to be more constant over the range of movement (when moving at constant speed). Servovalve
hysteresis is measured at slow speeds where valve dynamic (inertia) effects are negligible.

Figure 7 Servo Flow Characteristics Flow Curve with Hysteresis

The last curve, Figure 8, repeats the curve in Figure 7, but adds in a null bias offset. A null bias is
created with an unbalanced spring on the servo. The spring will force the servo to not be centered
(offset in one direction). A certain amount of torque motor current is required to offset the spring force,
which has the affect of shifting the flow curve. In the example of Figure 8, 1 milliamp (positive) is
required to hold the valve in the no flow (actuator stationary) position. The main reason for a null bias
offset is when current is lost the actuator will be powered in a desired direction (usually to retract).

Figure 8 Servo Flow Characteristics Flow Curve with Hysteresis & Null Bias

Servoactuator Terminology
In the aerospace industry, the document that governs servoactuator design is SAE Aerospace Recommend
Practice (ARP) 490. A copy of this document is available form Society of Automotive and Aerospace Engineers
(SAE). Important definitions are provided below.
Control Flow - Flow through valve ports (to actuator). Control flow is referred to as no load flow when there is
zero load pressure drop (i.e., pressure due to load on actuator piston is 0).
Flow Gain - Slope of the control flow vs. input current curve in any specified operating region (see previous
figures). Typically this refers to the normal flow region which is outside of the null
region and before saturation near the end of travel.
Hysteresis - Difference in the valve input currents required to produce the same valve output during a single
cycle of valve input current, expressed as a percent of rated current (see previous
figures). A nominal value for hysteresis is 3% of rated current.
Linearity - Degree to which normal flow curve conforms to normal flow gain line with other operational variables
held constant.
Load Pressure Drop - Differential pressure between control ports. Control ports are the ports to the actuator.
Null - Condition where the valve supplies zero control flow at zero load pressure drop (actuator is held
stationary at the null position).
Null Bias - Input current required to bring the valve to null, excluding the effects of valve hysteresis.
Null Region - Region about null where effects of lap in the output stage predominate.
Null Shift - Change in null bias due to changes in supply pressure, temperature and other operating conditions,
including wear.
Polarity - Relationship between direction of control flow and direction of input current.
Pressure Gain - Rate of change of load pressure with input current at zero control flow (psi/ma, for example).
Rated Current - Specified input current (milliamp) of either polarity to produce rated flow.
Rated Flow - Specified control flow corresponding to rated current and specified load pressure drop.
Servovalve - A electrical input, flow control valve, which can provide continuous control. Output stage flow is a
direct function of current.
Stage - Any device where hydraulic amplification occurs in a servovalve. Three most common stages are
flapper valve, jet pipe and spool valve. Servovalves can be single-stage or two-stage.
In rare cases, a servovalve may have 3 stages.
Threshold - Increment of input current required to produce a change in valve output, expresses as a
percent of rated current. A normal value of threshold is 1% of rated current.

Selecting a Servovalve or ServoActuator
Servoactuators are inherently complicated with nonlinear behavior. In addition, servoactuators are
associated with electronic controllers and closed loop position control loops. When selecting a
servovalve or servoactuator a large number of factors need to be examined. From a performance
perspective, the critical items to look at are covered in Servovalve, Hydraulic Sizing. Beyond
performance factors covered in Servovalve, Hydraulic Sizing, other items would include
temperatures
Rated Flow catalog data for servovalves will provide data for single operating condition which
consists of flow rate, inlet pressure and applied current. This data is a design operating point for the
valve and can be used for initial sizing studies.
Rated Pressure - catalog data for servovalves will provide data for single operating condition which
Rated Current - catalog data for servovalves will provide data for a single operating condition which
Closed Loop System Interactions when using a servo in a closed loop system, the closed loop
system interactions (both stability and closed loop performance) should be evaluated (see Servovalve,
Hydraulic Position Control)
Failure Modes the main failure modes are servo valve jamming in any position from full closed to full
open and a hardover command. Servovalves are normally closely monitored through the electronic
controller for any abnormal behavior.
Electrical Interfaces electronic controller output signals need to be compatible with torque motor
requirements. Electrical connectors should be high standard (such as military standard) to prevent
deterioration in service (which can lead to hardover failures).
Envelope/Mounting servoactuator must fit in the available space and have mounting features that
allow the unit to be mounted to primary support structure
Hydraulic Fittings hydraulic connections must be compatible. Inlet and outlet ports will often have
different size connectors to prevent cross connecting hydraulic lines during installation.

Applications of Servoactuators
Servoactuators will be used for any hydraulically actuated control where precise position or rate
control is required. The most common application for servoactuators is on primary flight control
surfaces. A servoactuator will be used on any primary flight control surface that is hydraulically
actuated. On primary flight control surfaces, servoactuators will be part of an overall power control unit
(PCU). More information on power control units can be found in Power Control Units, Hydraulic
Description. Other applications for servovalves in commercial aircraft would be flap or slat control,
spoiler panels and nosewheel steering. On military applications, servovalves can be used to control
many other systems on the aircraft such as thrust vectoring.

Servoactuator/Servovalve Qualification
See Qualification - Hydraulic Components for discussion on servovalve and servoactuator qualification
and required certification testing.

Mechanical Systems
Information modules are provided under this link for general aerospace mechanisms and
mechanical linkages. General mechanism design, four bar linkage, cable system, gears and power
screws are covered. Additional mechanism components, which are used to aerospace mechanisms,
are also addressed. These components include bellcrank, dwell linkage, lost motion, overcenter
linkage, pogo, and summing linkage.
Cable System - Description
This module prevents an overview for cable systems. The module only deals with aspects of cable
systems and does not address a complete flight control system. A cable system is considered a
functional element within a flight control system and as such is addressed as an individual element in
the series of Cable System modules. A simple elevator control system is shown in Figure 1. The
elevator system shown in Figure 1 shows a linkage connecting the pilot to a sector and a sector to a
flight surface. In figure 1, the cable system consists of both sectors and the cable. Flight control
systems are addressed in the Flight Control modules (see Flight Controls Systems Description).

Figure 1 Simple Elevator Control System

In general, a cable system is defined as two cables connecting into a cable sector (quadrant), T-lever,
capstan or bellcrank arrangement, plus any pulleys and fairleads. There must be two cables since the
cables can only be pulled, i.e., they can only take tension loads. This leads to a cable system as
illustrated in Figure 2 and Figure 3. When sectors are used they will usually have equal pitch
diameters. Therefore, no mechanical advantage is gained in the cable system. When T-Levers or
bellcranks are used the radius may change from one end of the cables to the other to obtain a non-
unity gearing ratio. Selection and features of sectors and T-Levers are discussed in Cable System
Sectors andCable System T-Levers. The main difference between a system with a sector and a
system with a T-lever is that the torque applied to a sector for a given cable tension is the same for all
rotation angles of the sector. For a T-Lever the torque is impacted by the angle between the cable and
T-lever arm so cable tension will vary with rotation angle for a fixed input torque.

Figure 2 Basic Cable System with Sectors

Figure 3 Basic Cable System with T-Levers

Cables (wire rope) are constructed of individual wires that are wrapped together to form a strand and
then weaving several strands together. For example, a 7 x 19 cable consists of seven strands woven
together with 19 wires woven together to from a strand. This makeup of cables gives cables good
strength characteristics while maintaining a good flexibility. However, as cable is stretched or wrapped
around a sector or pulley, the wires rub on each creating friction and wear. Therefore, cables do not
have infinite life. Cables should be periodically inspected in airplanes for wear (flattened or cold
worked wires, broken wires and excessive wear in a pulley/sector groove). Cables are discussed in
more detail in Cable System Cables.
The main advantage of cable systems is that a force (or torque) can be transferred the length of an
airplane with relative simplicity and high efficiency. Cable systems are also fairly lightweight, easy to
maintain and low cost. The disadvantage is that the systems are not rigid since a cable will stretch
under load. Also, in cable systems with small angle pulley cable wraps there may be undesirable wear
at these pulleys, leading to increased maintenance and shorter inspection intervals.
Cables are routed through an aircraft fuselage and wing using pulleys. A pulley is used whenever a
change in direction of the cable is required. Pulleys can also be used on long cable runs to prevent
excessive droop or to ensure a slack cable will not catch on some other component or structure. An
illustration of a system with pulleys is shown in Figure 4. Each pulley must lie in the plane defined by
the cable runs entering and leaving the pulley (two lines define a plane) so that the cable does not rub
against the pulley flange. Pulleys add friction in the system and introduce a cable wear location. Thus
pulleys (direction changes) should be kept to a minimum. Selection and features of pulleys are
discussed inCable System Pulleys.

Figure 4 Cable System With Pulleys

When the cables are installed a turnbuckle is used to preload the cables. Preload tensions are in the
range of 75 150 pounds. A typical reversible flight control system will have cable pre-tension in the
70-90 lb range, while an irreversible system may have cable pre-tension in the 130-150 lbs range.
Cable pre-tension are set such that the cable wont go slack under maximum expected pilot forces.
Also, cable pre-tension may be set higher to increase cable stiffness (i.e., to achieve less cable stretch
for a given load), since stiffness increases with increased cable tension. More information on cables
can be found in Cable System Cables and more information on turnbuckles can be found inCable
System - Turnbuckle. An example of determining the cable pre-tension level can be found in Cable
System Design Example.
To understand the dynamics of a cable system, refer to Figure 5. Under the applied torque shown in
Figure 5, the tension in top cable will increase while the tension in the bottom cable will decrease.
When cable tension increases, the cable stretches. When cable tension decreases, the cable stretch
(cable length) is reduced. For maximum expected pilot input, a design goal is to not have a cable go
slack. Slack cables are undesirable because slack cables are at risk for catching on surrounding
structure, fasteners or possibly other equipment. The cable preload should be set such that the bottom
cable will not go slack when the top cable is stretched to maximum expected pilot input torque. For
commercial aircraft, the maximum expected pilot input is defined by 14 CFR 25.143(d).

Figure 5 Cable System With Loads

For forces acting on a sector, the tension acts equally and opposite in direction on both sectors as
shown in Figure 6. A torque balance on the left sector includes the pilot applied torque, torque due to
top cable (tension top x pitch diameter), torque due to bottom cable (tension bottom x pitch diameter),
and friction due to pulleys/sectors (includes friction due to cable wrap and bearing friction). Cable
tension arises from the initial pre-tension and then any difference between
1
and
2
, which implies
additional cable stretch that occurs when a pilot force is applied against a reacting surface hinge
moment.

Figure 6 Cable System Showing Free Body Diagram

More detail on cable system equations and modeling can be found in Cable System Equations.
Design consideration for design and layout of cable systems can be found in Cable System Design
Considerations. A design example for a cable system is presented in Cable System Design
Example.
Cable systems are used for other functions beyond primary flight control. Cable systems are also used
to connect an autopilot servo to a sector or T-Lever, control of secondary trim systems (such as trim
wheel connection to a pitch trim surface) and other functions such as landing gear freefall cables.
Autopilot servos often utilize a cable capstan to create greater cable travel from a small diameter
sector (see Cable System Capstan). The fundamentals covered in the Cable System modules apply
equally well to these other cable system applications.
Four Bar Linkage - Description
A four bar linkage is the pre-eminent mechanism building block. Four
bar linkages are common in aerospace mechanical linkages. At times, a
mechanism may be a four bar linkage, but the fact that a mechanism is
a four bar linkage may not be obvious. An understanding of four bar
linkages is essential to understanding mechanisms in a general sense.
The mathematical characteristics of a four bar linkage directly apply to
any general mechanism. This module provides an overview of four bar
linkages. Some examples are provided. The mathematical equations
and analysis techniques are provided inFour Bar Linkage Equations.
A general four bar linkage is shown in Figure 1. A four bar linkages
consists of 4 bodies which are the three links plus ground (ground is
always considered a link when analyzing mechanisms). For analytical
purposes, four bar linkages are portrayed and treated as planar
mechanisms. However, in practice implementation of a four bar linkage
could be spatial (non-planar). Four bar mechanisms are not always
obvious in a mechanism and careful inspection may be required to
determine is a linkage is indeed a four bar mechanism. For a spatial four
bar linkage, an equivalent planar linkage can often be determined and
the analytical techniques discussed in Four Bar Linkage
Equations and Four Bar Linkage Design Methods apply.

Figure 1 General Four Bar Linkage

The parameters in Figure 1 are defined as follows
O
1
ground point for link 1
O
2
ground point for link 2
l
1
length of link 1
l
2
length of link 2
l
3
length of link 3
d
1
horizontal (x) distance between
ground points for link 1 and link 3
d
2
vertical (y) distance between ground
points for link 1 and link 3
1
angle of link 1 to x axis (measured
as shown in Figure 1)
2
3
angle between link 2 and link 3 (as
shown in Figure 1)
F
c
axial force in link 3
Typical terminology for a four bar linkage labels link 1 as the input link,
link 2 as the coupler and link 3 as the output link. Of course, link 3 could
just as easily be the input link. A four bar linkage is described by the 2
constraint equations

These equations use the parameter definitions shown in Figure 1. These
constraint equations state that the sum of distances in the x direction
and the sum of distances in the y direction around the four bar linkage
must be zero. Note that the equations are nonlinear due to the cos and
sin functions. The constraint equations hold true for all possible
positions of the linkage, hence they form the basis for mathematical
analysis of a four bar linkage. For more details on mathematical
analysis, see Four Bar Linkage Equations.
One characteristic of a four bar linkage is that there is a single degree of
freedom (1 DOF). If we let link 1 be the input link, then the position of
link 2 and link 3 are uniquely determined by the position of link 1. Thus if
the l
i
and d
i
, plus the input link angle, |
1
, are known, then the
positions |
2
and |
3
can be computed as a function of |
1
. Thus the four
bar linkage has 1 DOF. This characteristic is true of all airplane
mechanisms. If the input to the mechanism is known, then the position
of all components in the mechanism can be computed. Some
mechanisms may have 2 DOFs, using separate inputs such as pilot
stick input and autopilot actuator input.
Four bar linkages are very versatile and have been designed to do a
multitude of tasks. For example, four bar linkages can be used to have
the output travel in a straight line or do general path following. Figure 2
shows a four bar linkage that can be used to draw a straight line.

Figure 2 Four Bar Linkage with Straight Line Output
Referring to Figure 1, maximum efficiency of a four bar linkage is
obtained when the angle is kept close to 90 degrees. This minimizes
the compression load, F
c
, acting on link 3. F
c
is wasted force as it does
not contribute to the output torque and adds friction and additional
column loading to link 3..
Another important characteristic to understand about four bar linkages is
that the mechanical advantage of a four bar linkage varies with linkage
position. Thus the mechanical advantage is not constant throughout the
range of movement a characteristic of all mechanisms. Mechanical
advantage is discussed inMechanisms Overview and Mechanical
Advantage. In this module, mechanical advantage is shown to be
equivalent to the velocity ratio. As shown in Four Bar Linkage
Equations, velocity ratios vary with position. The variation in velocity
ratio over the range of movement of a mechanism (i.e., the mechanical
advantage) is an important feature to understand. Large variations could
be a problem in a overall mechanism design. Generally speaking, small
movements around input link angles of 90 degrees are good to minimize
variations in mechanical advantage.
Examples of four bar linkages are shown in the figures below. Figure 3
shows a leading edge mechanism for the leading edge of an airplane.
This mechanism is a four bar linkage that raises the leading edge
section of the wing to give the wing more camber at low speed. The four
bar linkage is driven by an actuator connected to one of the links.

Figure 3 Leading Edge Mechanism

A second four bar linkage example is shown in Figure 4. This example is
from a fighter jet aircraft that utilizes a series connection of four bar
linkages. Note that the complete linkage is not shown. Also, in the
airplane this linkage is not planar, even though the picture shows a
planar mechanism.

Figure 4 Pushrod Based Flight Control System

Another four mechanism is shown in Figure 5. This mechanism shows
two four bar linkages connected by a cable system. The first four bar
linkage is the connection between the control column and the forward
sector. The ground points for this linkage are the rotational axis for the
control column and the bearing attachment for the sector. The second
four bar is the aft flight control sector to a flight control surface bellcrank
using pushrods. The ground points for the four bar linkage is the bearing
attachment for the sector and the bearing attachment for the bellcrank.
The input to this four bar linkage is the control column and the output is
the elevator position.

Figure 5 Generic Elevator Control Linkage

The last four bar mechanism is a helicopter cyclic control mechanism
(blade pitch control) shown in Figure 6. This mechanism is four bar
linkage where the coupler is short link between the two rotating shafts.
There are several four bar linkages in Figure 6. The first four bar linkage
consists of link 1, link 2 and the bellcrank. The second four bar linkage
consists of the bellcrank, link 3 and the swashplate. The third four bar
linkage consists of the swashplate, link 4 and the propeller blade. Note
that the top swashplate is always rotating, while the bottom swashplate
is stationary. This allows a stationary mechanism to drive a rotating
mechanism.

Figure 6 Helicopter Cyclic Control Mechanism

Further analytical information on four bar linkages can be found in Four
Bar Linkage Equations and design information in Four Bar Linkage
Design Methods. These modules provide the background information for
analyzing and designing the linkages shown in Figures 3 through 6.
2010 DesignAerospace LLC Contact Us
Site Designed by RealSoft
Solutions

Four Bar Linkage Design Methods
The purpose of a mechanism is transmit motion and force (or torque) from one location to another.
The challenge when designing a good mechanism deals with the relationship between position and
force (which always hold and become a constraint on the design) and the fact the equations are
inherently nonlinear (although a mechanism can be approximately linear over a short travel range).
The relationship between position, velocity and force is through the kinematic equations and
mechanical advantage relationships. In any mechanism design, there will always be some trail and
error (or experimentation) that must occur.
Design methods for four bar linkages encompass generic trial and error plus standard textbook
methods. Design solutions using any methodology are not guaranteed, so it is possible to pose a four
bar linkage problem that is not kinematically feasible. Therefore, flexibility and creativity is important in
linkage design. Intuition and insight are also helpful. Intuition and insight can be gained by
understanding the equations governing four bar linkages and by reviewing the design examples
provided here. Practice and experience is also helpful for developing insight.
Design methods for four bar linkages fall 5 general categories:
- Mathematical
- Trial & Error
- Function Generation
- Motion Synthesis
- Path Synthesis
Within each category, there are various approaches to design that can be taken. In some categories,
extensive methods are developed in mechanism textbooks. Some methods are mathematical based
and some are graphical.
The reason for presenting all methods is to provide a good overview of the approaches and how they
work. When designing a four bar linkage mechanism, any combination of these methods or other ad
hoc method that an engineer chooses to use is acceptable for designing a linkage to meet the
requirements. Requirement verification should always be completed prior to finalizing a design.

Mathematical Approach
A purely mathematical approach is good for analysis but does not account for geometric constraints
so a 3D layout should be part of a mathematical based design process. The mathematical approach
uses the equations from Four Bar Linkage Equations module. The iteration procedure using the
position and velocity equations is used to determine the velocity ratios and thus the gearing ratio over
the range of movement. An example is provided in Four Bar Linkage Mathematical Design.

Trial and Error
Trial and error relies on a basic understanding of constraints involved in four bar linkage operation and
is the most common method used in industry. Usually in aerospace, many parameters are defined
(such as ground points on structure) which simplifies the trial and error process. Knowing which points
are fixed in a mechanism (such as ground points) and understanding how motion is constrained (such
as movement constrained to an arc), trial and error can be quite fast. The best way to demonstrate the
trial and error process is through an example. An example is provided in Four Bar Linkage Trial and
Error Design.

Function Generation
Function generation is a methodology to design a four bar linkage such that the relationship between
input and output motion follows a given mathematical relationship. For example, if the input motion
(rotation) is represented by x and the output motion (rotation) is represented by y, the function
generation provides a method to design a linkage such that y = x
1.5
holds over a range of input link
rotation. Both mathematical and graphical approaches have been developed to accomplish function
generation. An example of a mathematical approach (Freudensteins Method) is provided in Four Bar
Linkage Function Generation. Graphical methods also exist. The most common graphical method is
the Overlay Method. An item of note with function generation is that since the method applies to
position relationships and not velocity relationship, the method does not directly allow a person to
design for both travel and gearing ratio.

Motion Synthesis
Motion synthesis address the problem of designing a four bar mechanism to place a body in desired
locations during rotation of the linkage. The body is the coupler component of the four bar linkage.
Three desired positions of the body are required to design the linkage. These locations become the
inputs. Two basic design methods exist; both are graphical methods. The first method is to select
attach points for the linkage on the body (coupler) and then determine the location of the ground
points. The second method is to select locations for the ground points and then determine the location
of the input link and output link attach points on the body (coupler). Examples of these methods are
provided inFour Bar Linkage Motion Synthesis Ground Points Fixed and Four Bar Linkage Motion
Synthesis Ground Points Variable.

Path Synthesis
Path synthesis is a four bar linkage design method to design a linkage such that a point on the coupler
passes through selected points. At least three points must be provided. The method uses kinematic
inversion to design the linkage. Inputs to this design methodology are the three coupler points and the
location of the ground points. The output is the definition for link 1 and link 3 as well as the attach
points on the coupler. An example of path synthesis is provided inFour Bar Linkage Path Synthesis.

Gears - Description
Gear trains and gearing linkage exist in almost all flight control applications and are therefore
important to understand. Furthermore, an understanding of gear train fundamentals provides the
framework for analyzing all mechanical system linkages. In aerospace, gearing systems are used in
electromechanical actuators (refer to Actuator, Electromechanical Description), hydraulic pump and
motor gearboxes, nosewheel steering drives and other applications. The typical purpose of the
gearing system is to take a low torque, high speed motor and gear this down to a high torque, low
speed at a flight surface, flap torque shaft, nosewheel steering strut, etc. In effect, speed is exchanged
for an increase in torque.
For gear terminology, refer to Figure 1.

Figure 1 Gear Terminology

The circular pitch is the arc distance along pitch circle from a point on 1 gear tooth to the same point
on the next gear tooth (from pt A to pt B in the figure). The circular pitch is computed using

where D is the pitch circle diameter and N is the number of gear teeth.

Figure 2 Two Gear Mesh

Figure 2 shows two gears meshed together. In Figure 2,

The gear ratio relationships are given below. Derivation of these relationships can be found in Gears
Equations.

An examination of the above gear ratio relationships shows that if T
1
is low and
1
is high, then T
2
can be high
and
2
can be low. This is referred to as gearing down of the motor output, which allows a lighter, smaller, high
speed, low torque motor to be used to apply larger loads at a slower rate to, say, a spoiler panel or trim tab.
Effects of inertia, friction and efficiency through a gear train are discussed in Gears Equations.
The goal of gear design is normally to match a desired angular velocity ratio. A velocity ratio can be
specified directly or via a force/torque transfer relationships (see mechanical advantage discussion
in Mechanisms - Mechanical Advantage). A designer can choose to use 2 gears, a multi-gear train,
compound gear train or a planetary gear train to achieve a desired velocity ratio. The choice depends
on desired velocity ratio, space constraints and general design guidelines for gears and gear teeth.
Further details on gearing systems (multi-gear trains, compound gear trains and planetary gear trains)
can be found in Gears General Gearing Arrangements.

Gear Backlash
Backlash and deadzone are characteristics of gear trains and general mechanisms. From an analysis
point of view, backlash is a nonlinear behavior and excessive backlash can reduce mechanism
performance and cause closed loop instability. Physically, backlash is caused by our inability to
manufacture perfect parts. As a result, there must be some clearance between gear teeth, as shown
in Figure 3. In Figure 3, the input gear must travel a distance b/2 before contacting the output gear to
impact motion. When the input gear has contacted the output gear and reverses direction, the input
gear must travel a distance b before the output gear will start to move.

Figure 3 Gear Backlash

Anti-backlash gears are available and have been used in some aerospace applications. One method
for a anti-backlash gear design cuts a gear into two halves, which each half being of the thickness
of the original gear. One of the gear halves is fixed to the shaft and the other gear half is allowed to
rotate on the shaft. Several coil springs are then used to push the 2 gear halves apart (radially) so that
the gear teeth fit tight between the teeth of the mating gear. The coil springs are sized such that
compression of the springs is above the normal operating torque of the gear. An example of an anti-
backlash gear set is shown in Figure 4.

Figure 4 Anti-Backlash Gear

Design Considerations
The selection of gear materials is very important, especially in applications where the gear train is part
of a critical load path. In aerospace, materials for gears are normally a case hardened steel. Steel
generally provides the best strength properties and it is durable from a fatigue standpoint. Case
(surface hardening) of the gear teeth surface improves wear characteristics and reduces chances of
galling in service. For critical applications, freeplay requirements are very tight and wear
characteristics are important. Hardening is often used to ensure the gear teeth dont wear in service.
In non-critical applications, which also are low load, other materials may be used. In some
applications, plastic gears may be used. Any choice of material should be examined from a static
strength, fatigue, wear and environmental standpoint. The main environmental concerns are thermal
affects, vibration and corrosion. Bearings used to support gears should also be of high quality, such as
military or aerospace standard.
Surface or case hardening of gear teeth surfaces can be used to maximize wear characteristics,
provide good surface stability and reduce backlash. Surface hardening is a material processing
technique for increasing the (Rockwell) hardness of gear teeth surface. Surface hardening normally
involves introducing carbon (at high temperature) into a low carbon steel thereby increasing the
hardness for a thin material layer around the part. The purpose is to improve the wear characteristics
of the gear teeth. The process for hardening is usually done after heat treat. Surface (case) hardening
processes should be examined closely as there is a potential to induce hydrogen embrittlement in the
material. It is important that the surface hardening process does not affect the heat treat and material
properties of the base metal. Two methods for gear teeth hardening are carburization (carbon is
introduced into the material surface) and nitriding (nitrogen is introduced into the material surface).
Other surface hardening methods exist.
During manufacture, a good practice for critical load path gears is to send a material coupon through
with a batch of gears during heat treat and case hardening. The coupon blank metal about the same
size of the gear can then be inspected to ensure the heat treat and surface hardening processes
were done properly. Destructive inspection methods on the coupon are used to verify proper material
properties for the batch of gears.
Sizing of gears from a stress and fatigue point of view should take into account (i) heat generated
during operation, (ii) static strength margin of the teeth, (iii) fatigue capability, (iv) abrasive wear
characteristics of the gear material over the expected life, and (v) noise/vibration affects during
operation. Gear teeth can be analyzed structurally treating each tooth has a cantilever beam. Special
formulas are available in various gear design handbooks and AGMA standards.

When evaluating a gear train for a given application the following parameters should be considered in
the evaluation.
Torque, Limit (or Max Operating) Limit torque is the maximum torque the gear train must be able to
provide at the output. Limit load is used as a static load to size the gear teeth. Ultimate load will be 1.5
times limit load. In some cases, specifications will give a limit load that is above max operating load by
a certain factor (10-25% higher). Therefore, it is important to ensure limit load, maximum operating
load and ultimate load are clearly defined when specifying or analyzing a gear train.
Endurance and Fatigue Loads Endurance and fatigue loads represent the normal operating loads
that a gear train would be expected to see over its operating life. Endurance and fatigue loads are
usually the same spectrum. Endurance is associated with the gear train operating (rotating) and
fatigue loads are associated with the gear train held statically. Like static loads, fatigue loads are used
to size the gears through fatigue analysis. Gears with high load and low cycles will likely be sized by
the static load while gears with requirements for high operational cycles will likely be sized by
endurance loads. Endurance loads are normally associated with cycles (i.e., so many cycles at one
load, so many cycles at another load and so on). Endurance tests are normally done with the gear
train (or actuator) operating and fatigue loads are applied with the gear train static (not rotating).
Fatigue requirements will specify the number of lives that a gear train must be tested. For example, if
1 life of endurance and 4 lives of fatigue testing are required a common practice is to do 1 life of
endurance (gear train operating with endurance load spectrum applied) and 3 lives with the
endurance/fatigue load applied with the gear train not operating (held statically). During fatigue testing
the same gear teeth should be used (loaded) through the 3 or 4 lives of fatigue testing.
Bearings Gears are mounted to shafts through bearings. The bearings should be rated for the
expected loads (both static and fatigue) to be seen in service. High quality bearings, such as military
or aerospace grade should be used in critical applications. In addition, the bearings will be more
susceptible to environmental conditions than the gear. Therefore, careful assessments followed by
appropriate environmental tests should be conducted to validate the bearing over the expected life of
the gear train.
Gear Train Speed The speed of gear train is not normally a big concern in aerospace gear train
applications. However, gear trains have an upper RPM limit above which gear meshing may not occur
properly and there may be excessive vibration leading to premature gear failure. Gear materials, gear
teeth profiles, size of gear teeth, and amount of lubrication affect the maximum RPM limit.
Gear Material Gear material should be chosen carefully. Steel is the usual choice for aerospace
gears. The material is heat treated to achieve a desired hardness range (Rc 30 and above). For gear
trains used in critical flight components, the material and process specifications should be clearly
specified and the manufacturing process tightly controlled. In some cases, it may be necessary to
specify material properties tighter than the general material or heat treat specifications.
Surface (Case) Hardening Surface hardening will likely be required for critical gear train
applications. The 2 most important aspects are to ensure proper specifications are in place to define
the surface hardening process and that proper checks are done during manufacture to ensure the
surface hardening process only hardens the outermost surface (usually around a few thousands of an
inch) of the gear teeth. Sending test coupons through the hardening process with a each batch of
gears is a good method to allow inspection of the process without having to destroy an expensive
machined part. A qualified metallurgist should review all material process specifications.
Surface Finish For gear trains to work at highest efficiency, the running surface finish should be as
smooth as possible. Surface plating of gear teeth surfaces is not a good idea because of the potential
for flaking of the finish, which will lead to increased and accelerated wear on the surfaces as well as
bearing contamination.
Temperature In aerospace applications, gear trains operate over a wide temperature range. Material
properties can change over the full range. In addition, thermal expansion and contraction may lead to
binding during operation. Both of these aspects should be validated in a gear train application via test.
Backlash (Freeplay) A gear train will always contain a specification for backlash or freeplay.
Backlash or freeplay is determined by applying a small torque in one direction and then measuring the
distance the output gear moves when applying the same torque in the opposite direction. In
measuring freeplay, the torque should be small enough so that structural deflections are not occurring
(a few percent of maximum load is usually sufficient). Also, freeplay is a measure of control accuracy.
Anti-backlash gears can be used to reduce freeplay.
Lubrication Most gear trains and bearings are lubricated and require periodic lubrication when in
service. Both the type of lubrication (normally grease) should be examined for compatibility with the
materials. Periodic lubrication intervals should be established on the basis of test or service
experience with similar gear train applications. Endurance testing is good approach for validating a
chosen grease and a lubrication interval.
Life - Gear trains are designed for a certain life at a given load. Validation of the life capability of a
gear train should be validated by endurance and fatigue tests.
Efficiency Efficiency affects the input power requirements for a power screw. Efficiency requirements
of the gear train will affect the design and sizing of the gear arrangements and the number of gear
meshes. More gears lead to lower overall efficiency, as there is a 3-5% power loss for every gear
mesh interface due to bearing friction and friction between the gear teeth. Bearing characteristics at
temperature and under load should be evaluated to ensure bearing friction is low under all expected
operating conditions.

Qualification testing for gears and gearboxes should include all of the mechanical environmental tests
required by RTCA/DO-160 or Mil-Std-810 or other appropriate environmental specification. Normally
environmental tests will be done at the actuator or component level. For example, if a gearbox is
incorporated into an actuator, qualification will include other components such as a power screw,
gearbox, any no-back device and any electronics installed in the actuator.

Gears General Gearing Arrangements
Gear trains and gearing linkage exists in many aspects of aerospace mechanisms. Gearing systems
are also part of any electromechanical actuator design. Gearing systems at their simplest consist of 2
gears. When 2 gears are not sufficient due to size, volume or strength constraints, a multi-gear train or
a compound gear train is required. When a large gear ratio is desired within a small volume, then a
planetary gear arrangement is required. Designs of these types of gear trains are discussed below.
Other gearing arrangements rack & pinion, bevel and worm gears are also presented below. The
same principles for spur gears apply to other gear arrangements.
When designing gear trains a preliminary structural guideline regarding gear teeth size is given by the
relationship
(1)
This is a basic guideline on gear ratio limitations that is driven by gear teeth size and ensuring gear
teeth have sufficient structural strength. The smaller gear is usually limited by physical size limitation
and by ensuring gear teeth have sufficient strength. The larger gear is usually limited by physical size
and ensuring gear teeth are not too small. A textbook rule of thumb for the face width of a gear tooth is
that the face width should be 3 to 5 times the circular pitch of the gear. More specifically, sizing of
gears from a stress and fatigue point of view should take into account (i) heat generated during
operation, (ii) static strength margin of the teeth, (iii) fatigue capability, (iv) abrasive wear
characteristics of the gear material over the expected life, and (v) noise/vibration affects during
operation. Gear teeth can be analyzed structurally treating each tooth has a cantilever beam. Special
formulas are available in various gear design handbooks and AGMA standards.

Two Gear Train
A two gear train is shown in Figure 1.

Figure 1 Two Gear Train
For a desired velocity ratio, gears are selected using the relationship
(2)
However, N
1
and N
2
must be integers and therefore
2
/
1
can only be adjusted in discrete
increments. If N
1
and N
2
can be found such that equation (1) is satisfied, then a workable gear design
is achievable. If this relationship is not satisfied, then a multiple or compound gear train may be
required. If a gear relationship exists that satisfies the constraints, then the pitch diameter can be
chosen using N
1
and N
2
, followed by the remaining gear parameters. Stress analysis must also be
performed using expected loads in service to validate gear teeth sizing.
Results can be used to select gears from vendor catalogs or gear drawings can be created for
manufacture. Using non-standard gears, it is possible to have non-integer values of gear teeth per
inch.

Multi Gear Train Arrangement
Multi-gear trains are used when input and output shafts are far apart and space/design considerations
do not allow for 2 large gears. Multi-gear trains are also used when more than 1 shaft must be driven
at different speeds. Multi-gear trains are required to provide a + or direction of the output gear
relative to the input gear (2 gear trains always have input and output gears rotating in opposite
directions). For a design example, consider a 3 gear train shown in Figure 2.

Figure 2 Multi Gear Arrangement
At each gear mesh,
(3)
and for the gear train
(4)
or
(5)
Velocity ratios for a multi-gear train are computed in the same manner as a two gear train. However,
the only 2 gears that matter when computing the overall gear ratio is the first and last gear [see
equation (5)]. Intermediate gears do not have any effect on the overall velocity ratio and are called
idler gears. If an intermediate gear was used to drive a 2
nd
shaft, the 2
nd
shaft would have a different
velocity ratio. Design principles for each pair of gears in a multi-gear train are the same as a two gear
train where tooth ratios should be within the constraints of equation (1).

Compound Gear Train Arrangement
A compound gear is a gear where 2 or more gears are rigidly attached, such that they rotate at the
same speed. Compound gear trains are used when the velocity ratio is outside of the 0.2 5 range
and space is limited. An example compound gear train is shown in Figure 3.

Figure 3 Compound Gear Train Arrangement

(6)
An iterative process is required to design a compound gear train. To design a compound gear train,
the general steps are as follows:
- Select number of stages, n
s
. A stage is the number of gear meshes between compound gears.
For example, for the compound gear train shown in Figure 3, the number of stages is 3.
- Using the desired overall velocity ratio , compute the velocity ratio for each
stage using

If the p
s
ratio is outside of the 0.2 to 5 range, increase n
s
and recompute p
s

- Choose a ratio for the number of teeth for each stage such that the ratio is close to p
s
. Referring
to Figure 3, the teeth ratios (N
1
/N
2
), (N
3
/N
4
), and (N
5
/N
6
) would need to be chosen.
- Choose gear tooth numbers for other stages such that the gear tooth ratio is close to p
s
and the
total velocity ratio is equal to p
t

- Choose a diametrical pitch for each gear pair (preferably the same for all gear pairs). Compute
the pitch diameters using

(7)
- Iterate as required until results are satisfactory

Bevel Gears
A bevel gear arrangement is shown in Figure 4. Bevel gears provide a change in direction between
the input and output shafts. Figure 4 shows a 90 degree change in direction, but other angles are
possible.

Figure 4 Bevel Gear Arrangement

The kinematic nomenclature for a bevel gear arrangement is shown in Figure 5. This figure shows a
bevel gear arrangement that has a shaft angle different than 90.

Figure 5 Bevel Gear Kinematics

In Figure 5,
p
and
g
are the pitch angles for the pinion and output gear, respectively. The shaft
angle, , is the sum of ,
p
and
g
. The pitch radiuses are given by r
pinion
and r
gear
. For bevel gears, the
pitch radius is normally specified at the larger end of the teeth. The pitch cone for each gear
represents the area between the root and end of the teeth. Similar to spur gears, the velocity ratio is
computed as
(8)
When designing bevel gears, the shaft angle and number of teeth on each gear are usually known up
front. The corresponding pitch angles can be computed using
(9)
and
(10)

Rack & Pinion Gears
A rack and pinion gear arrangement converts rotary motion from a pinion to linear motion of a rack.
Figure 6 shows a picture of a rack and pinion.

Figure 6 Rack and Pinion Gear Arrangement

The kinematic relationships for rack and pinion gear arrangements are similar to spur gears. Figure 7
shows the basic definitions for a rack and pinion gearset.

Figure 7 Kinematic Definitions for Rack and Pinion Gearset

For a rack and pinion to mesh together properly, the pitch of rack and pinion must be equal, i.e.,
p
pinion
= p
rack
. The gear ratio equation is given by
(11)
Note that the number of teeth on the rack is not relevant to the velocity ratio. The linear speed of the
rack is simply a function of the pinion pitch radius and the angular velocity of the pinion.

Worm Gears
A worm gear provides a 90 degree change in the direction of rotation while also providing a large gear
ratio. A worm gear consists of a helical gear and a worm as shown in Figure 8. The worm gear threads
are shaped like a power screw (see Acme thread screws in Power Screws Description). The threads
on the helical gear are not straight (like a standard spur gear) but have a helix angle that matches the
lead on the worm gear threads. Worm gear arrangements are non-backdriving, as the helical gear will
not be able to back drive the worm gear. Hence worm gears are self-locking.

Figure 8 Worm Gear Arrangement

The kinematics of worm gear arrangement is shown in Figure 9.

Figure 9 Worm Gear Kinematics
When specifying a worm and worm gear pair, the axial pitch of the worm and the circular pitch of the
worm gear need to be specified. The gear ratio equation is given by
(12)
where the radius terms are of the corresponding pitch dimensions. The pitch diameter of the worm
gear is given by
(13)
where N
wg
is the number of teeth on the worm gear and p
wg
is the circular pitch of the worm gear. The
worm may have any pitch diameter since there is no relationship between the number of worm gear
teeth and the worm diameter.

Planetary Gear Trains
A sample planetary gear train is shown in Figure 10. This planetary gear consists of 2 gears and an
arm. Gear F is called the sun gear and gear L is called a planet gear. The input gear is gear F. The
output rotation is the arm rotation. The planetary gear arrangement shown in Figure 10 has 2 degrees
of freedom (DOFs) since the arm can rotate with or without input from gear F. Either spur or bevel
gears can be used in planetary gear arrangement.

Figure 10 Planetary Gear Arrangement

Another example of a planetary gear train is shown in Figure 11. The input gear is gear 2 and the
output is the arm rotation. This arrangement is similar to the arrangement of Figure 10 except that an
intermediate gear (gear 3) has been added and gear 4 is connected to a ring gear (gear 5), which is
fixed (non-moving). This arrangement has 1 DOF since the output arm will only move when the input
gear (gear 2) is rotating. All realistic planetary gear arrangements will have 1 DOF.
Figure 11 also provides some insight into the operation of planetary gear arrangements. Note that 1
rotation of the output arm occurs when gear 4 travels one full rotation around the ring gear. For this to
happen the input gear, gear 2, will have to rotate many, many times. Hence a high-speed input is
geared down significantly at the output. An example is provided later to show this gearing down affect.

Figure 11 Planetary Gear Arrangement, 2
nd
Example

The number of degrees of freedom in a planetary gear train is computed using
(14)
where
F = number of degrees of freedom
L = number of rigid bodies (gears, arms and ground)
B = number of revolute joints (bearings)
M = number of meshes between gears

For the planetary gear train shown in Figure 10
L = 4
B = 3
M = 1
so that
(15)
which verifies the planetary gear train in Figure 10 has 2 degrees of freedom.

Torque Summing and Speed Summing in Planetary Gear Arrangements
For input redundancy purposes, gear trains may be driven by separate drive shaft (motors). When two
separate drive shafts are used, the gear train will either be a torque summing arrangement or a speed
summing arrangement. In a torque summing arrangement, the torques are added (at the same speed)
and in a speed summing the speeds are added (at a constant torque). Thus a torque summed gear
arrangement cannot increase speed over a single drive motor speed and a speed summed gear
arrangement cannot increase torque over the torque of a single drive motor.
A torque summing arrangement is shown in Figure 12.

Figure 12 Torque Summing Gear Arrangement

In Figure 12, either gear 1 or gear 2 or both could drive the output gear, gear 3. If gear 1 is the driving
gear, then gear 2 must be freewheeling otherwise the gear train will be lockup up. Likewise, if gear 2
is driving, then gear 1 must be freewheeling. If gear 1 and gear 2 are driven by an electric motor with a
brake then this implies the brake must be open and the motor not powered. The driving motor must
then overcome the residual friction in the opposite gear train and motor plus the back electromotive
force in the opposite motor. If both gear 1 and gear 2 are driving gear 3, then gear 1 and gear 2 must
be going at the same speed. In this case the output torque on gear 3 will be the sum of the applied
torques at gear 1 and gear 2 (adjusted by the gear ratio).)
A speed summing arrangement is shown in Figure 13.

Figure 13 Speed Summing Gear Arrangement

In a speed summing arrangement the speed of the output arm is twice the speed of both input gears
multiplied by the corresponding gear ratios. Generally the gear ratio from gear 2 to the arm is equal to
the gear ratio from gear 6 to the arm. Like a torque summed gear arrangement, either input gear in a
speed summed gear arrangement can drive the output gear. However, unlike a torque summed
arrangement, when one gear is driving a speed summed gear arrangement the other input gear must
be held fixed. Otherwise, the single driving gear will backdrive the other input gear and there will be
little or no motion of the output gear (in effect, the gear train would have 2 DOFs). If motors having a
brake drive the input gears, then the brake would need to be closed to hold the motor fixed when the
other motor is driving. So, the brake control logic between a torque summed and speed summed
arrangement is opposite. Assuming equal gear ratios from each motor, when one motor is driving the
output gear the gear moves at of full speed. Full speed is obtained when both motors are driving.
The output torque is equal to the torque of a single drive motor multiplied the corresponding gear ratio.
For example, consider a speed summed gear arrangement with equal gear ratios of 10 (from each
motor to the output arm). Let the motor torque be 2 in-lbs with a nominal speed of 300 rad/sec. When
a single motor is driving, the output torque will be equal to 20 in-lbs with an output speed of 30
rad/sec. When both motors are driving, the output torque will be 20 in-lbs and the output speed will be
60 rad/sec.
Power analysis for either torque summed or speed summed gear arrangements are based on the
relationship
(16)
where T is the torque and is the angular velocity. Obviously the power can be measured at any gear
in the gear train. At the output shaft (arm) of the planetary gear train the total power output would be
the output torque multiplied by the angular velocity of the arm.
In any gear train, there will be torque (power) losses due to friction at occurring at each gear mesh
(approximately a 3% loss for each gear mesh for hardened steel gears). The basic equation is
(17)
As stated above, is approximately 0.97 for a single gear mesh.

Mechanisms Bellcrank
A basic bellcrank is shown in Figure 1. Bellcranks are the most common component in mechanical
linkages. Bellcranks are simple devices that are used to change the direction of movement. In Figure
1, the input and output direction of movement differ by 90 degrees. Bellcranks can also create
mechanical advantage when L
1
L
2
.

Figure 1 Bellcrank Geometry

Referring to Figure 1, the kinematic relationship for a bellcrank is (assuming small angular
displacements)
(1)
(2)
where the terms are defined in Figure 1. The basic force (or torque) relationship is
(3)
(4)
Thus when L
1
= L
2
, there is no gain through the bellcrank. If L
1
> L
2
, then F
2
> F
1
, so the gain is greater
than 1. If L
1
< L
2
, the gain is less than one. Equations (2) and (4) assume the angle between the
bellcrank arms are 90 degrees. When the input/output arms are not 90 degrees, the forces will act an
angle. This angle needs to be taken into account. For example, consider the bellcrank shown in Figure
2.

Figure 2 Bellcrank with Angles 90
The force equations for the bellcrank in Figure 2 are
(5)
Assuming no frictional losses,
(6)
This leads to the relationship
(7)
From equation (7), the force relationship equation for the bellcrank shown in Figure 2 is
(8)
The relationship between input and output motion for the bellcrank shown in Figure 3 is the same as
equation (2).
Bellcranks come in many different designs, shape and sizes. An example of manufactured bellcrank is
shown in Figure 3.

Figure 3 Actual Bellcrank Picture
Mechanisms Design Considerations
This module identifies the design considerations that should be considered when designing a
mechanism. This module is intended to apply to all aerospace mechanisms, including flight control,
door mechanisms, throttle mechanisms, landing gear, etc. For more details on mechanical flight
control system design see Flight Control System Design.
When designing a mechanism from scratch without any prior system or experience to guide the
design, the design task can be initially overwhelming. Additionally, in literature, there is not much
information on how to approach mechanism design. A basic approach to mechanism design is shown
in Figure 1. Following this flowchart will help keep the design process on the right track. The amount
of time spent in each phase of the process depends on the expected complexity of the mechanism,
similarity to other designs and the amount of information that is initially available.

Figure 1 Basic Mechanism Design Process

Following Figure 1, the design steps are described as follows.
Step 1 The purpose of Confrontation/Requirements Definition is to take a big picture view of what
you are trying to accomplish with the mechanism and define a basic approach (or the general nature
of the mechanism) to accomplish the objective. At the same time, the design requirements of the
mechanism should be determined. Requirements will include the kinematic relationships for position,
velocity and force/torque transfer. Requirements should also include weight, limit load, duty cycle,
backlash, freeplay, stiffness, friction, accessibility, manufacturability, environmental and so on.
Requirements will likely evolve as the design matures.
Step 2 Information Gathering can be done in parallel with Confrontation/Requirements Definition.
The type of information that would be gathered would be information on similar designs to include
performance characteristics, reliability, weight, etc. Vendor catalogs for such things as bearings,
cables, gears, etc can also be used to obtain information. The intent of the obtain background
information activity is to obtain sufficient information to make good decisions and accurate tradeoffs to
support the development of the viable design concepts.
Step 3 At the Design Concepts phase, basic design concepts are defined and tradeoffs done. Here
you would decide whether to use linkages, gears, cams, cables, etc. in the mechanism and where
they would be located in the mechanism. Also, if lost motion, dwell or sequencing is required during
some portion of the mechanism motion, this aspect should be identified during this phase. Experience
and intuition with mechanisms play a large role when generating concepts. General CAD layouts can
be done at this stage to support development of design concept(s). Conceptual evaluation tools such
as the Pugh Concept Selection methodology may be helpful for evaluating different concepts. As a
general comment, better designs are obtained when multiple concepts are developed and a concept
selection process is used. At the end of Design Concepts phase, basic layouts, either in CAD or via
sketches should be prepared for the selected design concept that is expected to meet the overall
requirements.
Step 4 In the Synthesis phase, the performance (kinematic) aspects of mechanism are determined.
Recall that kinematic techniques are used to design mechanisms. In this phase, the kinematic
characteristics of each portion of the linkage and the overall linkage will be determined. Hence, the
gearing ratios will be computed for the range of motion of the mechanism. Cable stretch
characteristics, cam shapes, gear sizing, etc. are aspects that need to be defined and included in the
Synthesis phase. Mechanisms software to evaluate and understand mechanism characteristics would
be very helpful here. Certainly any type of mathematical model of the mechanism will be helpful. In
some cases, a mathematical model of the mechanism is required. During synthesis, the mechanism
design should be evaluated against the design requirements from Step 1.
Step 5 Optimization involves detailed analysis of all aspects of the mechanism and refining the
design to meet all design requirements. During this phase, items like a undesirable characteristic in
the mechanisms during a portion of movement (from a nonlinearity for example) would be examined
and any improvements implemented. All parts should be evaluated for loads using stress or finite
element analysis. Freeplay, stiffness, weight, friction, reliability are other characteristics that should be
evaluated. Clearances and accessibility when installed on the vehicle should also be assessed.
Step 6 Finalization involves completing the drawings and any analysis/reports required to document
the final mechanism design. Also, during finalization necessary performance, load and environmental
qualification testing should be performed.

When designing a mechanism for a given application the following parameters should be considered
during the design process. Depending on the nature and function of the mechanism, the importance of
each item on the list below will vary from one mechanism to another.
Velocity (Gearing) Ratio The primary performance criteria for mechanism is the gearing ratio.
Gearing ratio is the same as the velocity ratio. Since mechanism behavior is nonlinear in general, the
gearing ratio varies throughout the movement of the mechanism. Therefore, gearing ratio should be
evaluated at all possible positions of the mechanism. Gearing ratio can also be affected by structural
compliance (stiffness) in the mechanism and thus structural compliance should be considered when
determining gearing ratio under load conditions. If cables are used, cable stretch should be accounted
for when computing gearing ratios.
Force/Torque Transfer Force and/or torque transfer through the mechanism is related to gearing
ratio. Usually the maximum input force is specified and the corresponding maximum output force is
specified. This actually determines the minimum required gearing ratio (over the full range of travel)
that is used in design of the mechanism.
Load, Limit (or Max Operating) Maximum operating or limit load will size components through a
static strength analysis. Limit load may be same as maximum operating load or may be slightly higher
than maximum operating load, depending on how the terms are defined in the specification. Ultimate
load is 1.5 times limit load. At limit load, the mechanism must operate properly and meet all
performance requirements. At ultimate load, the mechanism must remain intact but some permanent
deformation is acceptable. The mechanisms does not need to operate after application of ultimate
load.
that a mechanisms would be expected to see over its operating life. Endurance loads are normally
much less than limit load. Endurance and fatigue loads are usually the same spectrum. Endurance is
associated with the mechanism operating and fatigue loads are associated with the mechanism held
statically. Like static loads, fatigue loads are used to size mechanism components through fatigue
analysis. Mechanisms with high load and low cycles will likely be sized by the static load while
mechanisms with requirements for high operational cycles will likely be sized by endurance loads.
Endurance loads are normally associated with cycles (i.e., so many cycles at one load, so many
cycles at another load and so on). Endurance tests are normally done with the mechanism operating
and fatigue loads are applied with the mechanism static (not operating). Endurance testing is done to
prove out wear characteristics and the ability of the mechanism to not develop excessive freeplay in
service. Fatigue requirements will specify the number of lives that a mechanism must be tested. For
example, if 1 life of endurance and 4 lives of fatigue testing are required a common practice is to do 1
life of endurance (mechanism operating with endurance load spectrum applied) and 3 lives with the
endurance/fatigue load applied with the mechanism not operating (held statically).
Input/Output Travel Limits Mechanisms are designed to operate over a required operating range.
Stops must be installed at either the output linkage or the input linkage or both to ensure the
mechanism cannot be operated outside of the design travel limits. The stops should be sized for limit
load and ultimate load capability.
Friction Friction occurs whenever there is a rotating component (such as a bushing or bearing or
cable rubbing on a fairlead). Friction should be minimized as friction leads to inefficiency. However,
some friction is often desirable to hold a mechanism in place and to provide some force feel when
operating a mechanism under no load conditions. Selection of low friction bearings, eliminating any
possible rubbing or chaffing and minimizing the number of components in the mechanisms are the
best methods to keep friction low. Operating loads also affects friction (higher loads result in higher
friction).
Stiffness Includes stiffness (or rigidity) of mechanism components, the mounting brackets and
attachment structure. Stiffness affect gearing ratio and can also affect total travel. Cable stretch will
have a significant impact on stiffness characteristics. Stiffness should be included in gear ratio
computations.
Clearances Clearance of all mechanism components to other components and structure throughout
the full range of motion is critical. Normally some margin is used, such as maintaining 0.25 inch
clearance to structure and other components. In areas where structure is rigid, clearances below 0.25
inch can usually be tolerated.
Environmental Considerations All environmental qualification categories apply to mechanisms. Key
categories are temperature, vibration, icing, and fluids susceptibility. The relevant industry guidelines
are RTCA DO-160 and Mil-Std-810.
Freeplay/Backlash Any mechanism will have some amount of freeplay or backlash in the system.
Obviously freeplay and backlash should be kept very small. Depending on the function of the
mechanism, freeplay may be very critical. In freeplay/backlash is critical, tight tolerance parts, zero
backlash gears, etc. should be used to keep freeplay/backlash to a minimum.
Bearings Mechanisms typically use bearings in pushrods, bellcranks and other components.
Bearings come in many different designs (ball bearings, roller bearings, thrust bearings, etc.) and are
manufactured from a variety of materials. Selection of high performance bearings with good reliability
is important. Military quality bearings are a good choice. Spherical bearings are also normally chosen
to prevent binding during mechanism movement. The bearings should be rated for the expected loads
(both static and fatigue) to be seen in service. High quality bearings, such as military or aerospace
grade should be used in critical applications. Appropriate environmental tests and endurance tests
should be conducted to validate the bearing over the expected life in the mechanism.
Weight Weight should also be minimized to the extent possible. Weight efficiency is always
important. During the Design Concepts phase, the mechanism design(s) should be assessed for
weight efficiency. Mechanism concept, mechanism simplicity, loads and selection of materials are
primary drivers of weight.
Efficiency Efficiency is determined by dividing the power out by the power input to the mechanism.
Power losses occur from friction and compliance of structure and components. If cables are used,
cable stretch will also absorb energy through friction between the cable strands (strands rub together
as the cable is stretched).
Complexity/Number of Components As the number of components increases in a complexity and
cost increase. Reliability will also decrease. Optimum designs in terms of simplicity and weight are
always desirable. During the Design Concepts phase, the mechanism design(s) should be assessed
for complexity. Simpler designs, which meet the design requirements, should be selected.
Failure Modes - Failure modes should always be considered in any design. The two most important
failure modes are a jam in the mechanism and a disconnection of a component in the mechanism.
Another important failure mode is high friction levels high friction could lead to pilot induced
oscillations in a flight control system.
Materials Materials must be capable of handling limit and ultimate loads plus the endurance/fatigue
load spectrum. Manufacturing processes for critical (primary load path) parts should be examined
closely to ensure parts will be manufactured with the desired material properties intact. Materials
should also be chosen that have well-established material properties. Special attention should be
given to new or unique materials. Non-critical (non primary load path) mechanism parts may not
require as much scrutiny as critical parts.
Component Design Design details for each component in a mechanism should be examined
carefully. Parts can have (and often do have) a unique shape driven by requirements to carry the
necessary loads, provide adequate clearance between structure and other moving parts, and to
minimize weight. Some parts may require tight tolerances. Methods of staking bearings are important.
Pushrod end clevises must be threaded sufficiently in the rod so sufficient threads are available to
carry the load. Cable selection and installation characteristics must also be done properly. Standards
exist that cover many detail design features for mechanism components. Military and Aerospace
Standards or existing company standards should be followed whenever possible.

Qualification testing for mechanisms should include all of the mechanical environmental tests required
by RTCA/DO-160 or Mil-Std-810 or other appropriate environmental specification. Normally
environmental tests will be done at component level. For example, bearings and cables can be
evaluated separately. Analysis can be used in lieu of some tests. For example, fungus resistance for
metal parts can normally be done by analysis. For some mechanisms, environmental qualification will
rely on service history of similar mechanisms used in other flight vehicles. Use of service history will
reduce the amount of qualification testing that would be required. However, endurance tests and load
tests should always be done on new mechanism installations. When conducting endurance and loads
tests, the use of actual mounted hardware (brackets and such) is preferred.
Mechanisms Dwell Linkage
Dwell is used in a mechanism when sequencing is required. From a theoretical standpoint, dwell
occurs when the output velocity is zero, which implies very little movement of the output. In practice, it
is not possible to achieve zero velocity at a dwell position so there will always be some movement of
the output. Figure 1 shows a simple and common means to achieve dwell.

Figure 1 Dwell Mechanism

In Figure 1, as the crank is rotated from point A to point B, the slider will move to the right. However,
as the crank is rotated from point B to point C, there is very little movement of the slider. The total
travel of the slider as the crank goes from A to B is given by
(1)
The velocity is given by
(2)
As long as is small xdot will be small. Dwell occurs or nearly occurs during the rotation from A to C.
In practice, the crank/slider mechanism shown in Figure 6 will be one part of a mechanism. Thus as
the crank goes from A to C, other portions of a mechanism will be moving.
A practical example of a dwell linkage is shown in Figure 2. Figure 2 shows a latch pin and a rotating
latch. The mechanism in Figure 2 is similar to the general mechanism of Figure 1 in that the latch pin
will move very little during a portion of the input bellcrank rotation. During the initial input link
movement, the latch rotates approximately 90 to encapsulate the latch pin. As the input link continues
to move there will be very little movement of the latch. The portion of the motion where dwell occurs is
indicated in the figure.

Figure 2 Mechanism with Dwell Component

Figure 3 shows another means to achieve dwell in a mechanism. The mechanism in Figure 3 is called
a hypocycloidal mechanism. In this mechanism, the planet gear will roll along the inside of the ring
gear as the input crank is rotated. Point A will then follow the dashed lines shown in the figure as the
input crank is rotated. When Point A is in the position shown, there will be very little motion of the
output slider. This is where dwell occurs in the mechanism. In practice, a mechanism of this nature will
have a input crank motion around 60 degrees maximum where a portion of the stroke will cause slider
travel and another portion will cause the slider to be in dwell.

Figure 3 Hypocycloidal Dwell Mechanism

Mechanism textbooks provide many other examples of dwell mechanisms. Many involve four bar
linkages. Dwell in four bar linkages can be designed into linkage motion using the methods discussed
in Four Bar Linkage Design Methods.
Mechanisms Lost Motion
Lost motion provides sequencing in a mechanism. There are many means to achieve lost motion in a
mechanism. Lost motion is similar to dwell in that both methods allow sequencing in mechanical
mechanisms or linkage (see alsoMechanisms Dwell Linkage). The difference between lost motion
and dwell is that with lost motion there is no motion during sequencing down the lost motion path while
for dwell there is always some motion of the mechanism (no motion on any path is exactly zero).
Figure 1 shows one method to achieve lost motion in a mechanism.

Figure 1 Lost Motion Through Slotted Pulley Arrangement

In Figure 1, a cable passes over a pulley, which is held in one end of a slot via a spring. Assuming a
reaction force at point B on the cable, when a force is applied to cable at point A (in the direction of the
arrow), the pulley will be pulled to the left. When the pulley travels the length of the slot and hits the
left side of the slot the cable at point B will start to move. Thus the pulley axis can be connected to a
component, which will be moved first, and the cable can be connected to a second component which
will be activated second.
An example of the Figure 1 mechanism would be a landing gear freefall cable. In this case, the pulley
will be connected to a hydraulic valve. When the pulley moves in the slot, the spool in the hydraulic
valve is moved such that the actuator ports are opened to return. Once the spool is moved and hits
the end of travel in the spool housing, the cable (at point B) moves releasing the landing gear uplocks.
The gear will fall out of the wheel well through gravity and the hydraulic fluid in the landing gear
actuator flows to the return as the gear extends.
When analyzing a mechanism with lost motion, there will be two (or more) operating modes for each
output. The mechanism would need to be analyzed for each operating mode. Note there would be two
operating modes for the lost motion mechanism in Figure 1. In general, there could more than two
operating modes depending on the number of lost motion devices in a mechanism.
Another method to achieve lost motion in a mechanism is shown in Figure 2. Figure 2 illustrates lost
motion through a tab in a slot for rotational motion. As shown, the input link is splined to a shaft so that
it rotates with the shaft. When the tab contacts the outer link, the outer link will start to rotate. The lost
motion occurs when the input tab is not contacting the output link. Note that lost motion could also be
achieved by switching the input and output link, so that the output would occur at the splined shaft.

Figure 2 Rotational Example to Achieve Lost Motion

Alternative methods to achieve lost motion can be found in the many textbooks on mechanisms
available in engineering libraries. The textbook examples are too numerous and too varied to list here.
The general idea is for lost motion is to have a connection path disengaged (non-moving) while other
parts of the mechanism are moving. Any means to achieve lost motion in a mechanism is considered
a lost motion device or lost motion mechanism.
Mechanisms Overview and Mechanical Advantage
The purpose of mechanism is transmit motion and force (or torque) from one location in a flight vehicle
to another. The challenge when designing a good mechanism deals with the relationship between
position and force (which always hold and becomes a constraint on the design) and the fact the
equations are inherently nonlinear (although a mechanism can be approximately linear over a short
travel). The relationship between position and force is through the mechanical advantage
relationships.
Kinematic analysis deals with position and velocity analysis. However, the mechanical advantage in a
mechanism is usually of design importance. Using kinematic velocity analysis a relationship between
kinematics and force/torque transfer can be developed.
The fundamental work and power relationships are
Work = F * d
Power = F * d / t = F * Vel
If the input to the mechanism is a force, F
in
, with a velocity component, V
in
, then
P
in
= F
in
* V
in

or for a torque input,
P
in
= T
in
*
in

Likewise, for power out
P
out
= F
out
* V
out

or for a torque output,
P
out
= T
out
*
out

Note that an input force or torque can produce either an output force or an output torque.
Therefore, since P
in
= P
out
(ideally), the mechanical advantage (MA) can be written as

or

These relationships are dimensionless.

For cases where an input force produces an output torque (or vice versa)

or

In English units, these dimensions are rad/sec to ft/sec and lbs / lb-in (or the inverse).
Therefore, a power or force or torque requirement can be converted to a velocity ratio, and velocity
analysis can be used to design a mechanism for a required mechanical advantage. However, keep in
mind that velocity ratio is normally a function of mechanism position and hence mechanical advantage
will be function of mechanism position.
From a mechanism design process, when given a mechanical advantage requirement, convert this
relationship to a velocity ratio and use kinematic analysis to design the mechanism. Kinematic design
approaches are discussed in the gear and 4 bar linkage modules.

Mechanisms Overcenter Linkage
An overcenter linkage or device is used to provide a mechanical stop in a linkage and prevent any
backdriving of the mechanism. Figure 1 shows an overcenter mechanism. The depiction in Figure 1 is
simple in nature, but is adequate for explaining the principles of an overcenter mechanism. Overcenter
implies a rotational mechanism.

Figure 1 Overcenter Mechanism

In position A, as the crank is rotated clockwise, the slider will move to the right. When the slider gets to
its maximum stroke position (where the link is horizontal) the mechanism is considered to be in the
center position. As the crank continues to rotate clockwise, the link moves past the center position,
i.e., overcenter. In position B, the crank has rotated to the point where the moving stop has hit the
fixed stop. In this position, any force that pushes on the slider cannot backdrive the mechanism. The
function of the spring is to hold the crank against the stop so that vibration or any freeplay in the
mechanism will not cause the mechanism to rotate back out of the overcenter position. The spring
does impact the driving force for the mechanism, as the spring force must be overcome to move the
mechanism. Setting the spring forces is a tradeoff between the impact on the driving force and holding
the moving stop tightly against the fixed stop. When the crank goes overcenter, the slider is effectively
in a dwell. Thus overcenter and dwell can be incorporated into the same mechanism motion.
Overcenter mechanisms are used to hold a mechanism at one end of the mechanism travel. Two
overcenter devices can be put into a single mechanism to hold the mechanism in either end of the
mechanism travel. The spring(s) will add to the normal operating load in the mechanism.
Mechanisms Pogo (Load Limiter or Bungee)
A pogo are also referred to as a load limiter or a bungee. The function of a pogo is to allow operation
of mechanism when a jam occurs downstream of the pogo component. A pogo is shown in Figure 1.
When a pogo is assembled, the spring compression is adjusted to a desired force level. When a force
less than the spring force is applied to either end of the pogo, the pogo acts like a rigid link. When a
force greater than the spring force is applied to the pogo, the spring will compress. An applied greater
than the spring force implies a jam has occurred on the output side of the pogo.

Figure 1 Pogo

Under normal operation, the spring force is large enough (relative to normal operating loads) to keep
the pogo stick rigid. If one side jams, then it is possible to move the other side at an operating force
large enough to compress the spring. To visual jam operation, refer to Figure 1 and let the outer rod
be fixed and push on the inner rod so that the pogo is in compression. The inner rod lip (right hand
side in Figure 1) will push on the slider, which will compress the spring. The spring force is reacted
into the lip on the outer rod (left hand side in Figure 1). If the inner rod is pulled, so that the pogo is in
tension, the behavior is similar except the applied force and spring force reaction point switch sides.
Force transmitted through device will not exceed the spring force as long as no internal hard stop
within the pogo is contacted.
The function of a pogo in a flight control system is to allow other portions of a mechanism to continue
operation should a jam occur in a mechanism downstream of a pogo. However, because the spring
must be compressed, the operating force in the mechanism will be higher. The increase in force is
normally high enough to be quite obvious to a pilot.
An application of a pogo in a flight control system is shown in Figure 2. Figure 2 shows a pogo on the
input link to the servo on each rudder PCU. If a rudder PCU servo should jam, the rudder pedals could
still drive the other rudder PCU by compressing the spring in the pogo of the jammed servo. Clearly,
the force level would higher but the function of one rudder PCU would be retained. Without the two
pogos in the system, a jammed servo would jam the complete rudder control system.

Figure 2 Simplified Powered Flight Control System with Pogo
Mechanisms Pushrod
A pushrod is a rigid structural link used in flight control mechanisms. A pushrod typically connects to a
bellcrank and/or forms the coupler link of a four bar linkage assembly. A generic pushod is shown in
Figure 1. The pushrod consists of a center body and two adjustable rod ends. The rod ends are
adjustment to support rigging of all or part of a mechanism. Since the rod ends are threaded and
screw into the center body, a means for ensuring sufficient thread engagement is required. Quite often
this is done through a witness hole (see Figure 1) where the threads must be engaged up to the
witness hole as a minimum. Thus the witness hole provides a means to verify proper engagement.
The rod ends are held in place through various locking means that include a jam nut or tabbed washer
or safety wire or some combination of these. It is critical to ensure the rod ends will not move once
adjusted and locked into place.

Figure 1 Generic Pushrod

Pushrods are commonly made out of steel but also can be manufactured out of aluminum or
composite material. The rod ends are generally manufactured to a MS specification. Spherical
bearings are located in each rodend. The threads, which are defined by the MS specifications, will
usually be small pitch to ensure small adjustments are possible rod ends can only be adjusted in
increments of 180 degrees.
Figure 2 shows further detail of the eyebolt connection to the control rod. Figure 2 shows how the
eyebolt threads into control rod and the location of jamnut. Once the eyebolt is threaded into the
control rod to the witness hole, the jamnut is tightened on the control rod and then safety wired to the
locking tab. The locking tab is held in place using serrated washers. In some pushrods a bent tab
washer is used where the washer has tabs that can be bent along the flat spot of the jamnut and along
a flat spot on the control rod (this configuration is now shown in the figures). Note that the eyebolt can
be positioned at any angle and then locked into place at the desired angle. Hence the eyebolts at
each end of the pushrod can be at different angles (e.g., one straight and the other rotated 90
degrees). Also, the pushrod length is adjusted on one end by rotating the eyebolt 180 degrees. The
change in length will be number of turns x the thread pitch. All pushrods will have a spherical bearing
in each eyebolt. The spherical bearing accommodates installation tolerances and ensures that the
pushrod load will only be axial.

Figure 2 Eyebolt / Control Rod Connection

Figure 3 shows a more complicated eyebolt to control rod connection. In this arrangement, the eyebolt
threads into a vernier sleeve, which is then threaded into the control rod. There are two jam nuts and
two sets of serrated washers now because of the two threaded attachments. The jam nut and vernier
sleeve could be locked wired together or locked wire to a locking tab (if installed), or even wired to the
locking tab shown. This arrangement with the vernier sleeve would provide for finer adjustments in the
pushrod length. The arrangement could also be used to lengthen a pushrod if a particular pushrod
was a little short for a given application.

Figure 3 Eyebolt / Control Rod Connection

Besides have the proper length, the primary design criteria for a pushrod is strength. Pushrods carry
an axial load only and must be sized for the expected maximum loads expected in service.
Traditionally, pushrods have been manufactured out of steel, both the control rod and eyebolt. In low
load applications, aluminum pushrods have been used. More recently, composite control rods are
being manufactured and used in some applications. Composite control rods should be thoroughly
tested in a given application to prove out the strength and durability of the composite material and the
threaded connection between the eyebolt threads and control rod threads. Means of locking
jamnuts, safety wire, locking tab, bent tab washers, etc. should always be included in pushrod.
Another means of locking protection would be to use LH threads on one end of the control rod and RH
threads on the other end of the control rod. The torque applied to a jamnut should never be
considered a locking means.
Secondary design considerations are adjustability and the locking method for the threaded rod end. In
some applications, a very fine adjustment of the pushrod length is required. Fine adjustments can be
done by using a small thread pitch, using a vernier sleeve arrangement as shown in Figure 3, or by
adjusting both rod ends simultaneously.
Spherical bearings can tolerate some amount of mis-alignment (3-5 degrees) during installation. If the
mis-alignment capability of the bearing is violated, then a torque is put on the bearing and a bending
stress on the pushrod. Obviously this situation should be avoided. Also, there should be no freeplay in
the connection between the pushrod and connecting component. Washers should be chosen as well
as a bolt clamp up torque to ensure there is no side way movement at the connection between the
pushrod and attached component.

Mechanisms Summing Linkage
The purpose of a summing mechanism is combine two input motions into one output motion. A
common application for summing link is on all mechanically actuated hydraulic power control units
(see Figure 2). A summing link is shown in Figure 1. This a linear summing link. Summing can also be
done rotationally.

Figure 1 Summing Mechanism

For the summing linkage shown in Figure 4, the two inputs are x
A
and x
B
. The ouput is x
C
.
(1)
Combining
(2)
(3)
(4)

As a side note, the linkage in Figure 4 becomes a subtracting mechanism if point A is the output.
Letting d
1
=d
2
, then, x
B
and x
C
are combined via subtraction according to

Two applications of a summing mechanism are shown in Figure 2. First, the output of the dual
automatic flight control actuators are summed together to drive the linkage controlling the power
control units. The second example is the summing link that drives the servovalve of each rudder
power control unit (PCU). In both examples of a summing link, the inputs are at the two ends of the
link and the output is in the middle.

Figure 2 Dual AFCS Actuator Summing Link Mechanism

An example of a rotational summing link design is shown in Figure 3. The mechanism shown in Figure
3 has two inputs, which are the speedbrake input and roll input. The speedbrake input provides a
symmetric spoiler panel deployment command and the roll input provides an asymmetric spoiler panel
command. There are two outputs, one to each spoiler panel. The two inputs are summed together to
provide a combined output command to each spoiler panel.

Figure 3 Rotational Summing Mechanism

Power Screws - Description
The function of a power screw is to transform rotary motion into linear motion. The three types of
power screws are square thread, acme thread and ball screw. Square threads and acme threads are
similar in design, where the nut slides along the screw threads (like a traditional screw/nut interface).
The ball screw thread uses ball bearings between the screw thread and corresponding thread on the
nut. In aerospace applications, the two most common power screws are the acme thread design and
ball screw design.

For a power screw with square threads, the threads follow the profile shown in Figure 1. Figure 2
shows the thread profile for an acme thread power screw. The mating nut for the screw profiles is not
shown in either Figure 1 or Figure 2.

Figure 1 Square Thread Profile

Figure 2 Acme Thread Profile

In Figures 1 and 2, the distance p is the distance between the same points on adjacent teeth. p is
often referred to as the screw lead, which is the distance a nut would travel for one complete rotation
of the screw. The remaining dimensions of the thread are based on the dimension, p. For an acme
thread screw, the thread profile is not normal to the screw centerline but is at an angle,o. The
angle, o, is called the thread angle. For a standard acme thread, o = 14.5. Also illustrated in Figures
1 and 2 is the definition for the major diameter, mean diameter and minor diameter of the screw. The
mean diameter is often referred to as the pitch diameter. The pitch diameter is the center location of
applied forces for a screw.
A ball screw thread profile is shown in Figure 3. For a ball screw, the thread profile is cut to
accommodate the ball bearing. The grooves will be machined to match the ball bearing as close as
possible while maintaining some clearance at the deepest part of the groove. The ball bearing
contacts the thread along a contact line, with a contact angle as shown in Figure 3. For a ball
screw/nut combination note that the threads between the screw and nut will be separated by a small
gap (i.e., they dont mesh together as they would for an acme thread screw/nut). Forces between the
screw and nut are applied along the line of contact in the threads. Lead would also be measured at
the contact point between 2 adjacent teeth.

Figure 3 Ball Screw Thread Profile

An acme thread screw is characterized by low efficiency due to the friction between the screw and nut.
Because of the friction, an acme thread screw is not easily back drivable. Ball screws are
characterized by having a high efficiency (due to the low friction of the balls). As such ball screws are
better suited to high speed applications. Because of the low friction, ball screws are easily
backdrivable. Consequently most ball screw actuators contain no back devices or motor brakes to
prevent backdrive. Common applications of power screws in aerospace are flap, slat and stabilizer
actuators.
A screw thread may have 1 or more starts. A single start screw will have a single thread on the screw
and mating nut. A two start screw will have two threads on the screw and the mating nut. The threads
will start 180 degrees apart on the end of the screw such that every other thread will be associated
with 1 start. For a two start screw, the lead is measured between every other thread. For a single start
screw Lead = Pitch. For a multiple start screw Pitch = Lead / number of starts. The definitions for pitch,
lead and starts, and the relationships between these parameters, are the same for an acme thread
screw and a ball screw.

For a ball screw, two other definitions are important:
Ball Return Paths: As a ball screw rotates, friction between the ball bearings and threads tend to
push the ball bearings in one direction. A return path is thus added to all ball screws to allow the
balls to cycle from 1 end of the thread back to the other end. This has the affect of reducing ball
screw friction and increasing efficiency. Without a return path there may also be a tendency for
a ball screw to bind. All ball screws will have a ball return path. See Figure 4 for a example of
power screw ball return path.
Circuits: A circuit is the complete path for a set of balls. The circuit consists of the path between
the screw and nut threads as well as the return path. For a given start within a ball screw/nut
assembly, there may be 1 or more circuits. Multiple circuits are desirable for redundancy.
Assuming a single circuit can carry maximum operating loads, then loss of a one circuit in a 2
circuit ball screw allows the ball screw to continue operating. Figure 4 shows an example of a
ball screw circuit.

Figure 4 Ball Screw & Nut Showing Ball Return Path and Circuit

The selection of screw and nut materials is very important. In aerospace, power screws materials for
the screw, nut and ball bearings (for ball screws) are normally a case hardened steel. Steel generally
provides the best strength properties and it is durable from a fatigue standpoint. Case (surface
hardening) of the screw and nut surface improves wear characteristics and reduces chances of galling
and binding over long periods of time in service. For critical applications, such as horizontal stabilizer
control, freeplay requirements are very tight. Hardening is commonly used to ensure the freeplay
requirements are maintained in service. Another means used in manufacturing to maintain tight
freeplay requirements is to machine the screw and nut to tighter tolerances and then use lapping
techniques to open up the freeplay to the correct requirement. This approach leads to a matched
screw/nut combination. In non-critical applications, which also are low load, other materials may be
used. In some industrial applications, plastic screws and nuts are used. Any choice of material should
be examined from a static strength, fatigue, wear and environmental standpoint. The 3 main
environmental concerns are thermal affects, vibration and corrosion.
Surface or case hardening of thread surfaces are common in power screws to maximize wear
characteristics, provide good surface stability and reduce backlash. Surface hardening is a material
processing technique for increasing the (Rockwell) hardness of ball screw and ball nut wear surface.
Surface hardening normally involves introducing carbon or nitrogen (at high temperature) into a low
carbon steel thereby increasing the hardness for a thin material layer around the part. The purpose is
to improve the wear characteristics of the screw and nut. The process for hardening is usually done
after heat treat. Surface (case) hardening processes should be examined closely as there is a
potential to induce hydrogen embrittlement in the material. It is important that the surface hardening
process does not affect the heat treat and material properties of the base metal. Two methods for
screw hardening are carburization (carbon is introduced into the material surface) and nitriding
(nitrogen is introduced into the material surface). Other surface hardening methods exist.
During manufacture, a good practice for power screw parts is to send a material coupon through with
a batch of screws and nuts during heat treat and case hardening. The coupon blank metal about the
same size of the screw or nut can then be inspected to ensure the heat treat and surface hardening
processes were done properly. Destructive inspection methods on the coupon are used to verify
proper material properties for the corresponding batch of screws and nuts.

When evaluating a power screw for a given application the following parameters should be considered
in the evaluation.
Lead Lead represents the distance the nut will travel (linearly) during a single screw rotation. Pitch is
defined as the distance from a point on 1 thread to the same point on adjacent tooth. Lead is also
referred to as pitch for a single start screw (a screw with 2 starts will have a Lead = 2 * Pitch). Since
actuators are usually specified through a linear (nut) velocity, the chosen lead (or lead angle) will have
implications for motor and gear train design. For example, a motor will have a certain speed (rad/sec),
which when translated through a gearbox may or may not drive the nut at the required velocity
(in/sec). Hence the motor characteristics, gearing through any gearbox arrangement and the screw
lead must be compatible to achieve desired overall performance. Lead may also affect accuracy. A
larger lead (or lead angle) will require finer rotational position control to maintain accurate nut position
control. A smaller lead angle will require a larger screw rotation to produce the same output
translation. Lastly, for an ACME thread screw if the lead angle is small enough an ACME screw will be
self locking (see Power Screws Equations).
Number of Starts Power screws may contain a single screw thread (single start thread) or multiple
screw threads. For a 2 start screw, the 2 screw threads will start 180 degrees apart on the screw shaft
and run intertwined along the screw. For a multiple start screw the screw lead would be measured
between threads associated with a given start (for example, on a 2 start screw the lead would be
measured between every other thread). Multiple start screws are often used to provide load-carrying
redundancy since each start (or thread) would be sized to carry the maximum load. Also, for a ball
screw this would imply that the balls could be lost for one thread and the screw would still be able to
operate properly.
Number of Ball Circuits For a ball screw/nut, there may be 1 or 2 ball circuits. Two ball circuits would
provide redundancy if each circuit is designed to carry ball loads. In this case, if one circuit should fail
such that all of the balls were lost, the other circuit would maintain operation of the ball screw.
Normally, failure of a ball circuit is considered a latent failure. In aerospace, it is unlikely that a two
start screw would have two circuits for per start (4 ball circuits total) as this amount of redundancy is
not generally required. However, for a single start screw the use of a 2 ball circuit design should be
evaluated from a redundancy perspective.
Return Path Design Characteristics The design of the ball return path for each circuit should be
carefully evaluated. The circuit should be designed to withstand the repetitive impact loads of the balls
flowing through the circuit. In addition, the ball circuit cover or support should have sufficient
fasteners such that the loss of one fastener will not degrade operation of the ball circuit. Fasteners
should have appropriate locking features to prevent the fasteners from backing out. A tube mounted
on the side of the nut can be a problematic design as it can be difficult to provide proper structural
support for the tube. A plate with machined ball path is generally a more robust design.
Load, Limit (or Max Operating) Limit load is the maximum load the power screw must be able to
move. The static load is used to size the screw threads and is also used to evaluate column load
characteristics of the screw. Limit load is used in static load analysis. Ultimate load will be 1.5 times
limit load. In some cases, specifications will give a limit load that is above max operating load by a
certain factor (10-25% higher). Therefore, it is important to ensure limit load, maximum operating load
and ultimate load are clearly defined when specifying or analyzing a power screw.
that a power screw would be expected to see over its operating life. Endurance and fatigue loads are
usually the same spectrum. Endurance is associated with the power screw operating (rotating) and
fatigue loads are associated with the power screw held statically. Like static loads, fatigue loads are
used to size the screw through fatigue analysis. Power screws with high load and low cycles will likely
be sized by the static load while power screws with requirements for high operational cycles will likely
be sized by endurance loads. Endurance loads are normally associated with cycles (i.e., so many
cycles at one load, so many cycles at another load and so on). Loads may also vary with a function of
nut position on the screw. Endurance tests are normally done with the power screw (or actuator)
operating and fatigue loads are applied with the power screw static (not rotating). Fatigue
requirements will specify the number of lifes that a screw must be tested. For example, if 1 life of
endurance and 4 lives of fatigue testing are required a common practice is to do 1 life of endurance
(power screw operating with endurance load spectrum applied) and 3 lives with the endurance/fatigue
load applied with the power screw not operating.
Side Load A power screw application may have a side load imparted to the screw. Side loads should
be eliminated or kept to an absolute minimum. Side loads will impact the life of the screw and can also
lead to a larger screw to account for column loading in combination with side loads. Sides loads can
be minimized by proper actuator installation. Using spherical bearings for attach points will keep side
loads minimal.
Screw Material Screw material should be chosen carefully. Stainless steel is the usual choice
aerospace power screws. The material is heat treated to achieve a desired hardness range (Rc 30
and above). In some cases for acme thread screws, the Rockwell hardness between screw and nut
will be different. For power screws used in critical flight components, the material and process
specifications should be clearly specified and the manufacturing process tightly controlled. In some
cases, it may be necessary to specify material properties tighter than the general material or heat treat
specification.
Surface (Case) Hardening Surface hardening will likely be required for critical power screw
applications. The 2 most important aspects are to ensure proper specifications are in place to define
the surface hardening process and that proper checks are done during manufacture to ensure the
surface hardening process only hardens the outermost surface (usually around a few thousands of an
inch) and not the complete part. Sending test coupons through the hardening process with a each
batch of screws and nuts is a good method to allow inspection of the process without having to
destroy an expensive machined part. A qualified metallurgist should review all material process
specifications.
Surface Finish For power screws to work at highest efficiency, the running surface finish should be
as smooth as possible. Surface plating of power screws surfaces is not a good idea because of the
potential for flaking of the finish, which will lead to increased and accelerated wear on the surfaces.
Temperature In aerospace applications, power screws operate over a wide temperature range.
Material properties can change over the full range. In addition, thermal expansion and contraction may
lead to binding during operation. Both of these aspects should be validated in a power screw
application through testing.
Balls (Ball Screw) The balls in a ball screw are generally hardened steel ball bearings manufactured
to ball bearing specifications. The controlling specifications for the ball bearing should be reviewed for
compatibility with the load and life expectancy for the balls. Additionally, each circuit will have a fixed
number of balls such that the balls completely fill up the circuit with minimal freeplay in the circuit.
Speed Rating (Ball Screw) Speed rating generally applies to a ball screw configuration. A acme
thread power screw is not suitable to long, high speed duty cycles. For a given ball screw design,
there will be a critical velocity of the balls. Above a certain speed a resonance vibration will occur in
the power screw assembly. The vibrations result from the repetitive force of the balls hitting structure
in the return path and hitting each other as they flow through the ball circuit.
Backlash (Freeplay) A power screw will always contain a specification for backlash or freeplay.
Backlash or freeplay is determined by applying a small load in one direction and then measuring the
distance the nut moves when applying the same load in the opposite direction. In measuring freeplay,
the load should be small enough so that structural deflections are not occurring (a few percent of
maximum load is usually sufficient). Freeplay can lead to flutter when a power screw is connected to a
flight surface. Also, freeplay is a measure of control accuracy. Anti-backlash nuts can be used to
reduce freeplay. However, use of anti-backlash nuts on horizontal stabilizer and flap actuators is not
used in practice.
Lubrication Most power screws are lubricated and require periodic lubrication when in service. Both
the type of lubrication (normally grease) should be examined for compatibility with the materials.
Periodic lubrication intervals should be established on the basis of test or service experience with a
similar power screw application. Endurance testing is good approach for validating a chosen grease
and a lubrication interval.
Life Power screw are designed for a certain life at a given load. Life is generally given in terms of
inches of travel at a given load. For a power screw designed to a new specification, the power screw
will be designed for desired load and life. If an off the shelf power screw is used, the life rating will
need to be adjusted for any application.
Efficiency Efficiency affects the choice of a acme thread or ball screw. When a high efficiency power
screw is required, a ball screw actuator will be required. Efficiency of the power screw will affect the
design and sizing of the power drive (either electric or hydraulic motor).
Left Hand or Right Hand Thread The thread orientation determines which direction extends the nut
on the screw and which direction retracts the nut on the screw. This will affect the drive gear box
design and motor polarity so the proper command signals match extend and retract directions of the
nut.
Other items associated with power screw applications include no backs devices, load path
redundancy, seals, scrapers, drive torque and gearbox interface. These items are covered in the
electromechanical actuator module (see Actuator, Electromechanical Description).
Qualification testing for power screws should include all of the mechanical environmental tests
required by RTCA/DO-160 or Mil-Std-810 or other appropriate environmental specification. Normally
environmental tests will be done at the actuator level, which will include the power screw, gearbox,
any no-back device and any electronics installed in the actuator.

Flight Control Systems
An overview of aerospace mechanical flight control systems is provided in several modules within
this link. Reversible and irreversible flight control systems are covered. An overview of a fly by wire
flight control is also included.
Flight Control Systems - Description
The purpose of a flight control system is transfer motion/force input from a pilot to a flight control
surface. In a traditional aircraft, flight control systems are broken down by axis of control: pitch, roll
and yaw. Flight control systems can be either reversible or irreversible.
A reversible system is a flight control system where movement applied to the control surface moves
the control in the flight compartment. A simple example is shown in Figure 1. Reversible flight control
systems are used on smaller aircraft where the hinge moment (surface) loads are small enough that a
mechanical linkage system is adequate.

Figure 1 Simple Reversible Flight Control System

Another reversible flight control system is shown in Figure 2. This is a 2D representation of a system
that shows some typical components in a reversible flight control system. The systems shown in
Figures 1 and 2 both contain a cable system, however, reversible systems can also be designed using
pushrods and bellcranks without cables.

Figure 2 Reversible Flight Control System

An irreversible system is a flight control system that utilizes powered controls so that movement of the
surface will not move the control in the flight compartment. An example of an irreversible flight control
system is shown in Figure 3. Figure 3 shows a mechanical system connected to a hydraulic actuator.
The linkage positions the servo within the actuator that controls which side of the actuator sees high-
pressure fluid and which side is ported to return.

Figure 3 Irreversible Flight Control System

Irreversible systems are required when the maximum pilots input force is not sufficient to drive the
surface loads. In irreversible systems, the mechanical linkage will drive a hydraulic power control unit
(PCU). The linkage controls the spool position in a servo that applies hydraulic pressure to a hydraulic
actuator. For traditional aircraft, the transition from reversible to irreversible occurs in the 25,000
30,000 gross weight range of the airplane. The rudder surface is typically the first surface to require
powered control, where the critical flight condition is an engine out, rejected takeoff occurring just prior
to wheels leaving the runway. In some airplanes, a boost system or a hydraulic assist system is used
to supplement pilot rudder forces. An assist system is usually less expensive and less weight than a
full powered control system. Crosswind landings also result in high rudder hinge moments and must
also be considered.
Generally speaking, flight control systems are simply mechanisms that can be broken down into basic
components (see Mechanisms modules for the basic components) plus four bar linkages and cables.
For example, the system shown in Figure 3 is consists of four bar linkages connected in series until
they reach the flight control hydraulic servoactuator. The reversible system shown in Figure 1 consists
of 2 four bar linkages separated by a cable system. Therefore, any flight control system can be
evaluated and analyzed as discussed in the Mechanism and Four Bar Linkage modules. In fact, the
basic principles of a four bar linkage apply to any basic mechanism and are key to understanding
general mechanism characteristics.
Beyond the mechanism fundamentals, flight control systems can also include hydraulic system and
hydraulic actuators, electromechanical actuators (including power screws), cable stretch, structural
stiffness, springs, load variation with flight condition and surface deflection, and dynamic effects.
As stated above, the purpose of a flight control system is to transfer motion and force from the flight
compartment to the surface or actuator. To design a flight control system, the required movement of
the surface or actuator servo plus the required load must be known. Input motion is limited by human
factor considerations and maximum human force capability.
The approach to design a mechanical flight control system follows the same process and
considerations covered in Mechanisms Design Considerations. A design example for a reversible
flight control system is presented in Flight Control Systems Design Example. Additional
considerations for flight control system design can be found in Flight Control Systems Design
Considerations, Flight Control Systems Design Requirements, Flight Control Systems Reversible
Systems, Flight Control Systems Irreversible Systems.

Flight Control Systems - Reversible
A reversible flight control system is where there is a direct mechanical linkage connection between the
control lever in the flight compartment and the flight control surface. In a reversible system, if the
surface is moved the control lever will move. In reversible systems, the pilot directly feels hinge
moments applied to the surface. A simple example of a reversible flight control system is shown in
Figure 1.

Figure 1 Simple Reversible Flight Control System

A simple schematic of a reversible flight control system is shown in Figure 2. Schematically, the
system shown in Figure 2 is identical to the system shown in Figure 1. The representation shown in
Figure 2 is used for analysis. In Figure 2, the control stick, pushrod and O
2
sector are a four bar
linkage. The O
3
sector, pushrod and O
4
bellcrank are a four bar linkage. The two four bar linkages are
connected using a cable system. Analysis of the system would require an analysis of both four bar
linkage arrangements plus the cable system. The cable system acts like springs that connect the two
sectors.

Figure 2 Analytical Representation of System Shown in Figure 1

Reversible flight control systems are used on smaller aircraft where the hinge moment (surface) loads
are small enough that a mechanical linkage system is adequate. For a reversible system, the
maximum force input is limited by the capability of a pilot and the range of control input. For example,
rudder pedal travel is limited to 5 inches in each direction. If the rudder must move 30 degrees, then
the gear ratio is 5/30 = 0.167 inch / degree = 9.55 inch / radian = 9.55 in-lbs / lb (see Mechanism
Mechanical Advantage). Using FAA guidelines per 14CFR 25.143(d), the maximum short-term rudder
pedal force is limited to 150 lbs. For this example, the maximum rudder hinge moment is then (150
lbs) x (9.55 in-lbs / lb) = 1432.5 in-lbs. If the required hinge moment is greater than 1432.5 in-lbs,
some type of power assist or irreversible system is required.
In some reversible systems, high surface hinge moments are only required under certain conditions
usually a failure condition and some type of boost mechanism is implemented for this certain
condition. This saves the cost and weight of going with an irreversible system. A good example is the
increase in rudder control force required to overcome an engine out scenario. An engine out failure
creates a yaw moment on the aircraft that must be offset with rudder deflection. For example, the
maximum normal operating forces for a rudder system may be 600 in-lbs (83 lbs rudder pedal force),
but for an engine out situation the rudder hinge moment might increase to 1200 in-lbs (166 lbs rudder
pedal force). Since the 1200 in-lbs exceeds the maximum allowable force of 150 lbs rudder pedal
force, the basic design would not be acceptable. Some methods to overcome this situation without
implementing a fully powered system include (1) rudder bias system, (2) hydraulic power assist, (3)
geared tab and (4) split rudder.
Depending on the design or regulatory requirements applicable to a given airplane design, reversible
flight control systems may have single or dual mechanisms. The example shown in Figure 1 has a
single flight control mechanism (i.e, single flight control run). Smaller general aviation airplanes
designed to 14CFR Part 23 will have single flight control runs for each axis (pitch, roll and yaw).
Reversible flight control systems in larger airplanes, such as corporate business jets, designed to
14CFR Part 25 will have dual flight control runs or have a separate backup system. This is due to the
more stringent failure requirements required by regulation 25.671(c) in 14CFR Part 25. Dual flight
control runs in each axis (roll, pitch and yaw) provide protection a jam or disconnection failure in any
single run.
Design considerations for reversible flight control systems can be found in Flight Control System
Design Considerations. A discussion of design requirements can be found in Flight Control Systems
Design Requirements.

Flight Control Systems - Irreversible
An irreversible flight control system is where there is not a direct mechanical linkage connection
between the control lever in the flight compartment and the flight control surface. In an irreversible
mechanical system, the control lever in the cockpit moves a spool valve on a hydraulic power control
unit (PCU). A mechanical linkage drives the spool. The mechanical linkage will consist of a
combination of bellcranks, pushrods, cable systems, pogos, summing linkages, etc. The design of a
mechanical irreversible system is similar to the design of a reversible flight control system. The main
difference will be the spool forces are better defined than surface hinge moments. Also, an artificial
feel system is required for irreversible flight control systems. Variation in cable stiffness characteristics
over environmental and operating ranges can also be critical since spool movements will be small. As
a general rule, cable pretension values will be higher for an irreversible system (some commercial
aircraft are in the 150 lb pretension range).
Irreversible flight control systems include both mechanical linkage controlled hydraulically powered
PCUs and fly by wire systems. Today, most new irreversible systems are fly by wire. A description of
fly by wire systems can be found in Flight Control Systems Fly By Wire.
A simple example of a mechanical irreversible flight control system is shown in Figure 1. The system
in Figure 1 shows a rudder system with bellcranks and pushrods connecting the rudder pedals to the
hydraulic servoactuator (PCU). The bellcranks and pushrods form a series of four bar linkages. The
artificial feel system shows a common approach spring-loaded cam - for producing force feedback to
the pilot. Since cables are not used in this example, the system will have good stiffness. Of course,
system freeplay will need to very small. A summing link is used at the connection to the PCU.

Figure 1 Irreversible Flight Control System

Irreversible flight control systems are used on larger aircraft where the hinge moment (surface) loads
are large and a person cannot supply enough force. Keep in mind that the input travel in the cockpit is
limited by the amount of cockpit control travel and this becomes a limit on the maximum hinge moment
available without additional power input.
Pilots require feedback forces when they fly (to emulate reversible system stick force per g
characteristics and other natural aerodynamic force feedback). Therefore, artificial feel systems are
installed in all irreversible flight control systems. Artificial feel systems are done through springs,
pneumatic actuators or, in a few cases, hydraulic actuators. Springs loaded in compression are the
most reliable and hence the most common. Artificial feel systems are always redundant (sometimes
triple redundant) so that artificial feel is not lost after any single failure in an artificial feel system. A
spring-loaded cam artificial feel system is shown in Figure 1.
Irreversible systems require hydraulic or electromechanical actuators. However, electromechanical
actuators are only used in fly by wire systems (i.e., they are not mechanically controlled through
linkages as can hydraulic actuators). The choice and number of actuators in turn drive the design of
the hydraulic or electric power systems. The end result is a combination of increased size, cost and
complexity of these systems. Hence, irreversible systems are generally more costly and complex than
reversible systems. Assessment of system hazards and failure modes are therefore more involved for
irreversible systems.
Hydraulic PCUs in powered flight control systems will typically have different operating modes. Some
examples of different modes and features in PCUs can be found in Power Control Unit, Hydraulic -
Description. Generally, these modes will address issues such as loss of a hydraulic power source,
loss of a partner PCU (where 2 or more PCUs are connected to a single surface), loss of electrical
control signal, jammed spool or other detected fault within a PCU. Typical modes are normal
operation, bypass (where actuator follows surface with minimal resistance and no flutter damping
protection), damped bypass (follows surface movement but with a hydraulic damping orifice for flutter
protection), and locked or centering (actuator piston is held in failed or neutral position).
The most critical failure conditions within a flight control system are a surface runaway, jam,
disconnection of a mechanical linkage, or loss of electrical/hydraulic power. To allow operation after
these failures (perhaps degraded operation), various features are often included to allow some type of
system operation after these failures. These include pogos (load limiters), shearouts, jam overrides,
multiple hydraulic actuators with independent hydraulic sources and disconnect clutches.
In irreversible flight control systems, jam protection is provided by pogos, jam breakout mechanisms,
shearouts, and disconnect clutches when dual flight control runs are used.
Pogos (or load limiters or bungees) are a very stiff spring installed in the system. Under normal
operating loads, the pogo acts like a stiff pushrod in the system. However, if there is a jam
downstream of the pogo, then with an increase in applied force to the mechanism the pogo will
compress and allow other portions of the system to move. For example, if the cockpit control column
provide control to dual elevator mechanisms and a jam occurs in one side of the dual elevator
mechanism, then the pogo will compress in the side with the jam and the other side can continue to
operate (albeit with higher forces applied at the control column). More details on pogos can be found
in Mechanism Pogo.
Jam breakout devices function similar to a pogo. A jam breakout device is shown in Figure 2. In Figure
2, the cam and link 1 are splined together and from a rigid part. The rigid cam/link1 part can rotate
relative to the grey plate but is held in place by the spring force applied to the cam roller (through lever
1 and lever 2) that pushes the roller against the cam surface. In normal operation, the spring force will
be sufficient to maintain the roller in the position shown with normal operating forces applied to the
input pushrod. Under normal operation, the entire mechanism shown in Figure 2 - link1, cam, plate,
lever 1, lever 2 and the spring cartridge - form a bellcrank and rotate together about point 0. If a jam
occurs in either of the output links, the plate will be held fixed. When the plate is fixed, the mechanism
upstream of the jam breakout device can move by applying enough force to push the roller out of the
detent and compressing the spring. Another means to override a jam is to have a non-jammable spool
in the PCU. This is usually done with a spool within a spool design, where the inner spool will move an
outer spool if the inner spool becomes jammed within the outer spool. Higher forces will be required to
move the outer spool in its sleeve. A spool in a spool configuration is a type of jam breakout
mechanism.

Figure 2 Jam Breakout Device

As the name suggests, a shearout is a structural element, such as a pin or fastener, which is designed
to fail (usually in shear) at a given load. Shearouts are normally used to protect against an overload in
a system or part, but may be used to breakaway when a load occurs. Thus a shearout is used as
means to limit the maximum load that may be seen in a system and allows a lower limit load design
criteria to be used when structural sizing of parts.
Means to protect against disconnection failures include using a dual (redundant) mechanism, using
multiple actuators on a surface or using lost motion devices. Using multiple actuators on a surface with
dual control runs provides disconnection protection. For example, if one of the dual runs has a
disconnection the other control run would be operational and would provide control to one of the
actuators. Another feature that may be used for disconnection failures is a lost motion device. In a lost
motion device, relative motion is allowed between 2 parts. Under normal operation, the two parts will
not interact, but if a disconnect failure occurs then after an initial input motion to take up the slack, the
two parts will engage and the system output will be driven (see Mechanisms Lost Motion).
A disconnect clutch is mechanical clutch that connects two independent flight control runs. If a jam
occurs in one of the flight control runs, then the disconnect clutch is pulled (opened) through either a
handle or switch in the flight compartment which allows the non-jammed side to operate.
Figure 3 shows a generic, single axis flight control system with built-in redundancy features to address
several critical failure conditions. The system shown in Figure 3 has a disconnect clutch at the control
column and a pogo connecting the aft quadrants, which protects against a jam in either the pilot or
copilots control runs. In addition, there are pogos in the linkage to each hydraulic actuator. These
pogos would still allow control with a jammed servo in a hydraulic power control unit. Protection
against the loss of a single hydraulic source is provided through separate hydraulic sources for each
hydraulic actuator. Protection against a disconnection anywhere in the system is accomplished
through dual systems dual quadrants, dual cables runs and dual actuators.

Figure 3 Generic Flight Control System with Redundancy Features

Design considerations for reversible flight control systems can be found in Flight Control System
Design Considerations. A discussion of design requirements can be found in Flight Control Systems
Design Requirements.

Pump

Загружено:

Сведения о документе

Исходное описание:

Оригинальное название

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

Pump

Загружено:

Авторское право:

Доступные форматы

Pump, Hydraulic Description

Вам также может понравиться