FUNDAMENTALS

Tolerances
It is impossible to manufacture a dimension to an exact value. Tolerances must therefore be placed on the drawings to restrain the variations to permissible limits. The tolerances provide the zones in which the outline of the finished part must lie. A designer well aware of that the cost of a finished product can increase rapidly as the tolerances on the components are made smaller. Designers are constantly admonished to use the widest tolerances possible. Situations may arise, however, in which the relationship between the various tolerances required for proper functioning has not been fully explored. Under such conditions the designer is tempted to specify part tolerances that are unduly tight in the hope that no difficulty will arise at the time of assembly. This is obviously an expensive substitute for a more thorough of the analysis of the tolerancing situation. The allocation of proper production tolerances is therefore a most important task if the finished design is to achieve its intended purpose and yet be economical to produce. The size of the tolerance, as specified by the designer, depends on the many conditions pertaining to the design as well as on past experience with similar products if such experience is available. Knowledge of shop processes and machine capabilities is of great assistance in helping to determine the tolerances in the most effective manner. A revision of the design may be called for if the tolerances are too small to be maintained by the equipment available for producing the dimension. Unless otherwise specified, tolerances are assumed to apply after the application of inorganic coatings, such as plating, anodic processes, and so on. If the tolerances are required before the inorganic coatings, designer has to mention the same on the drawing with a note-tolerances apply before plating. Tolerances are assumed to apply before the applications of organic coatings, such as primers, paint and lacquer.

Methods of specifying tolerances

The manner of placing tolerances on a drawing depends somewhat on the kind of product or type of manufacturing operation. If the tolerance on a dimension is not specifically stated, the drawing can contain a general note that gives the value of tolerance for such dimensions. Sometimes, however, notes are not used, on the supposition that if each dimension is considered individually, wider tolerances than those called for in the note could probably be used. Tolerances may be placed on the drawing in a number of different ways. In the unilateral system (Fig.1) one tolerance is zero and all the variation of the dimension is given by the other tolerance. In bilateral dimensioning (Fig.2) a mean dimension is used with plus and minus variations extending each way from the mean dimension.

Fig 1. Unilateral tolerances

Unilateral tolerancing has the advantage that a tolerance revision can be made with the least disturbance to the remaining dimensions. Thus in (Fig.1) suppose that the fit had been originally dimensioned with tolerances h1 and b1 for hole and shaft. Suppose that after some experience with this fit it is found that larger tolerances could be used. The tolerances can be easily changed to h2 and b2 without affecting other dimensions already present. In the bilateral system (Fig.2) a change in tolerances also involves a change in at least one of the mean dimensions. Tolerances can be easily changed back and forth between unilateral and bilateral for the purpose of making calculations.

Fig 2. Bilateral tolerances

Maximum material condition ( MMC ); Least material condition ( LMC ) :

A part is said to be at maximum material condition (MMC) when the dimensions are all at the limits that will give a part containing the maximum amount of material. For a shaft or external dimension, the fundamental dimension is the largest dimension permitted, and all the variation, as permitted by the tolerance, serves to reduce the dimension. For a hole or a internal dimension, the fundamental dimension is the smallest value permitted, and the variation as given by the tolerance serves to make the dimension larger. A part is said to be at the least material condition (LMC) when the dimensions are all at the limits that give a part with the smallest amount of material. For LMC the fundamental value is the smallest for an external dimension and the largest for an internal dimension. The tolerances thus provide the parts containing larger amounts of material Maximum material tolerances have a production advantage. For an external dimension, should the worker aim at the fundamental or largest value but form something smaller, the parts may be reworked to bring them within acceptable limits .A worker keeping the mean dimension in mind would have smaller margins for any errors. The actual quantity of material between MMC and LMC parts may be of small consequence. These terms do, however,

provide convenient expressions for denoting the different methods for specifying the tolerances on drawings.
Example

A hole is dimensioned 25 25.040 mm; the shaft is dimensioned 24.060 - 24.040. Change these dimensions into bilateral form and also into maximum material dimensioning. Solution Bilateral: Hole mean = (25.0 + 25.04) 2 = 25.02 The dimension with bilateral tolerance = 25.02 0.020 Shaft mean = (24.06 + 24.04) 2 = 24.05 The dimension with bilateral tolerance = 24.05 0.010 Maximum material condition: Hole dimension is 25.00 Shaft dimension is 24.06
+ 0. 040 - 0. 000

+ 0. 000 - 0. 020

Cumulative and noncumulative tolerances

Consider the details for the parts of Fig 3. The designer no doubt believed that the dimensions of the Fig 3(a) would give satisfactory parts that would assemble with each other. However, as shown in Fig 3(b) it is possible for parts to be made in accord with (a) and yet interfere on assembly. The difficulty can be easily corrected if the dimensions for all surfaces extend from the left side, as shown in Fig 3(c). The left edge, from which all the dimensions in Fig 3(c) originate, is accordingly called the datum and is so marked. Datums are usually marked with a letter of the alphabet and placed in a box attached to the edge view of the surface. The drawing may of course contain many unimportant details, which have nothing to do with functioning and assembly. The dimensions for these need not, of course, originate at the datum.

Fig 3. Effects resulting from cumulative dimensioning

Redundant dimensioning
In a given direction, a surface should be located by one and only one dimension. Much confusion and expense can arise from violation of this rule. For example, consider the horizontal dimension of the part shown in Fig.4 (a). For a part made as in Fig.4 (b), lengths AB and AC are in accord with the drawing, but BC is not. Perhaps length BC is the important one for proper functioning, but a production man could argue that technically he had followed the drawing by making A B and AC correctly. Similarly, in Fig.4(c), length BC and AC are in accord with the drawing, but AB is not. The difficulty can be corrected simply by omitting one of the dimensions in the Fig.4 (a). The two dimensions that should be retained depend on manufacturing convenience or the functional requirements of the part. From the discussion above it is obvious that only sufficient dimensions should be placed on a drawing to form the requisite tolerance zones. Any additional dimension will nearly always result in parts that are out of tolerance.

Fig 4. A redundant dimension can be the cause of out-of-tolerance

Features & Reference dimensions

A feature is a specific characteristic portion of a part, such as a surface, hole, slot, screw thread, or profile. While a feature may include one or more surfaces, the term is generally used in geometrical tolerancing in a more restricted sense, to indicate a specific point, line, or surface. Some examples are The axis of a hole The edge of a part A single flat or curved surface, to which reference is being made, or which forms the basis for a datum. 6

A reference dimension, marked REF, is intended for information purpose only, and is usually used for checking the calculations. It is not to be used for manufacturing or inspection. Typical dimensions and In-situ dimensions A dimension marked TYP on the drawing is applicable to all the similar entities of the drawing other than that specified. The dimension marked In-situ on the drawing implies that the given dimension to be machined / maintained / verified / inspected on the spot / at the site of the operation before disturbing its position / location. Datum All bodies are three dimensional, and on engineering drawings, a body can be assumed to be placed in a system of three perfectly smooth planes oriented exactly 90o to each other. Such planes are called datums. Theoretically, perfect planes cannot be produced. However surface plates, angle plates, machine tables, and other equipments used in manufacturing and inspection are usually sufficiently accurate that they may be considered theoretical planes and thus used as datums. An edge view of a datum plane appears as a line on a drawing and is marked by letters such as A, B, C,------ placed in a box attached to the line representing the datum. If a plane in a frontal view is a datum, it can be marked if desired by placing its letter in a box and running a leader to an oversize dot placed on the plane. Datum should be actual physical surfaces, from which measurements can be made. Abstract concepts such as centre planes, centre lines or axes of rotation or symmetry should be used as datum.( Fig 5 )

Fig.5. Datum

Specification of values Values shall be specified on the drawing for every dimension necessary to define the size or location of each surface, line, point, or feature of the item delineated thereon. It should not be necessary for an essential dimension to be deduced from other dimensions, or from reference to other documents, nor for a drawing to be scaled. Independency of requirements Every requirement on the drawing is intended to be applied independently; without reference to other dimensions, conditions, or characteristics unless a particular relationship is specified. This rule applies to the size of dimension, and to each limit of size, size tolerance, geometrical form or positional tolerance, as well as to any physical, chemical, electrical, or other requirement. In a particular relationship to another requirement is desired, it must be specifically indicated by a note or symbol on the drawing, or in a related specification or standard. Implied geometrical form Every part or feature is intended to have the geometric form, which the drawing represents. It should not be necessary to specify the geometric shape of a feature, unless some particular precision is required, lines which appear to be straight imply straightness; those that appear to be round imply circularity; those that appear to be parallel imply parallelism; those that appear to be square imply perpendicularity; center lines imply symmetry; and features that appear to be concentric about a common centre line imply concentricity. Therefore it is not necessary to add angular dimension of 90 to corners of rectangular parts, or to specify that opposite sides are parallel. However, if a particular departure from the illustrated form is permissible, or if a certain degree of precision of form is required, it must be specified. If a slight departure from the true geometrical form or position is required, it should be exaggerated pictorially in order to indicate clearly where the dimensions apply. Dimensions, which are not to scale, should be under lined freehand as shown in Fig.6.

Fig.6.Exaggeration of small dimensions

Angular dimensions Angular dimensions are intended to control the general orientation of lines and surfaces, rather than individual points on the lines and surfaces. If two plane-gauging surfaces are now brought into contact with the surfaces the angle between them must be within the angular limits specified on the drawing. If control of individual points on the surface is required, an angularity tolerance should be used.( Fig 7 )

Fig.7.Angular dimensions

Linear dimensions (without datums) When datums are not specified, linear dimensions are intended to apply on a point-to-point basis, either between opposing points on the indicated surfaces, or directly between the points indicated on the drawing. The following examples should help to clarify this principle of point-to-point dimensions. Fig 8.shows a circular feature with a diameter shown as dimension D. This mean that the diameter, when measured between any two opposing points around the circumference, such as at a-a, b-b or c-c shall be within the specified limits of size.

Fig 8. Application of diameter dimensions

The diameter of a cylindrical part, such as diameter D in Fig.9, applies at any opposing points along its length, such as at a a , b b , c c , or d d .

Fig 9. Application of diameter dimensions

The rule applies whether or not there are obstructions in between, as shown in Fig 10, where diameter D applies, at a-a, b-b, or c-c. However if there is any doubt in such cases, such as when surfaces are widely separated, it is preferable to repeat the dimension

Fig 10. Application to interrupted surface

In applying the rule to rectangular parts, measurements for thickness are made normal to the centerline between the surfaces. For parts of uniform thickness this is equivalent to making measurements normal to the surface. A sufficient number of such measurements are made at various points on the surface to ensure that thickness limits are met for the entire surface. This applies equally to parts, which are bent or bowed, as shown in Fig.11. 10

Fig.11.Thickness of thin part

If thickness is not uniform, that is, if the surfaces are not parallel, measurements theoretically are made normal to the centerline, as shown in Fig.12 and may not be quite normal to the surface.

Fig.12. Thickness measurements

11

The same interpretation applies to length measurements, which ordinarily would be made normal to the end face if they were parallel and square with adjacent faces, as shown in Fig.13.

Fig.13.Length measurements

Fig 14. Length measurements

If the end surfaces are not square with adjacent surface or parallel with one another, precise measuring requires the measurements be made normal to the center line or center plane as shown in Fig.14. In spite of the above rules, measurements must be kept within the confines of the part, if so shown on the drawing, and must not be made to a point in space, as shown in Fig.15.in an attempt to measure normal to the faces. For this reasons, measurements on very thin parts effectively become measurements parallel with the surface.

12

Fig 15.Measuring correct length of thin parts

Fig 16.Measuring of formed parts

It may be argued that the kind of precision of measurement shown in these illustrations is purely theoretical, and that it is not necessary to find the direction of the centerline before measurements can be made. This is quite true in most cases, but the rules do become significant when dealing with some intricate shapes, and especially with parts, which may be over or under formed. For example, if the part shown in Fig.16 (A) is under formed as in Fig.16 (B) its length L would be measured at L1. Similarly the height of the leg H would be measured at H1, not H2. If a measurement at L2 is not with in limits it does not indicate an error in length, but rather a probable angular error. When applied to positional dimensions, as shown in Fig.17.dimension D applies to a measurement from the axis of the hole to a corresponding point on the edge of the part, perpendicular to the edge, as at a. Thus, if the part is off-square, dimension D would be measured as shown atb, which is the shortest distance between the axis and a point on the edge of a part

13

Fig 17.Measurement of location

Where there are several features controlled by one dimension, such as the series of holes in Fig18.The dimension applies individually from the axis of each hole to the corresponding point on the edge. If for any reason the part were bowed the series of holes would be located on a similar curved centerline, so that the location of each hole would meet the drawing limits when measured at a, b and c.

Fig18.Measurement of bowed part

14

Location dimension (with datums) When location dimensions originate from a feature or surface indicated as a datum, measurement is made from the theoretical datum, and not from the actual feature or surface of the part.

Fig 19. Dimension referred to a datum

There will be many cases where a curved center line, as shown in Fig.18.would not meet functional requirements or where the position of the hole in Fig.17.would be required to be measured parallel to the base. This can easily be specified by referring the dimension to a datum feature, as shown in Fig.19.This will be more fully explained in subsequent units, when the interpolation of co-ordinate tolerances is compared with geometrical and positional tolerances.

Assumed datums There are often cases where the basic rules for measurements on a point topoint basis cannot be applied, because the originating points, lines, or surfaces are offset in relation to the features located by the dimensions. It is then necessary to assume a suitable datum, which is usually the theoretical extension of one of the lines or surface involved. The following general rules cover three types of dimensioning procedures commonly encountered, but it should be emphasized that if any doubt is likely to exist the required datum feature should be properly identified. This is especially important when dimension tolerances are small in relation to possible form variations.

15

1. If a dimension refers to two parallel edges or planes, the larger edge or surface is assumed to be the datum feature. For example, if the surfaces of the part as shown in Fig 20 were not quite parallel, as shown in lower view, dimension D would be acceptable if the top surface was within the limits when measured at a and b, but need not be within limits if measured at c .

Fig 20.Assumed datum

2. If only one of the extension lines refers to a straight edge or surface, the extension of that edge or surface is assumed to be the datum. Thus in Fig 21 measurement of dimension A is made to a datum surface shown at a in the second view.

16

Fig 21.Assumed datum

3. If both extension lines refer to offset points, rather than to edges or surface, it should generally be assumed that the datum is a line running through one of these points, and parallel to the line or surface to which it is dimensionally related. Thus in Fig 22 dimension A is measured from the center of hole D to a line through the center of hole C which is parallel to the base line, as at a.

Fig 22.Asummed datum

Number of significant figures The precision of limits and tolerances is not affected by the number of decimal places involved, nor by their expression as decimal or common fractions unless specifically modified by a Note on the drawing. 17

Therefore a limiting value of 1.5 implies exactly the same accuracy as 1.50 or 1.500. This should not be confused with cases where a general tolerance notes specifies different tolerances for dimensions with different numbers of decimal places. For example, the drawing might specify such as 1- Place Decimals 0.1 2- Place Decimals 0.04 A dimension expressed as 1.5 would then have limits of 1.6 and 1.4 and a dimension of 1.50 would have limits of 1.54 and 1.46. However, if the first set of limits were expressed 1.600 and 1.400 it would not imply any change in accuracy, and should not affect the acceptance or rejection of parts in the slightest degree. Single limits When a requirement is specified as a single limit, labeled maximum or minimum, it means that any value less than the maximum, if only the maximum is specified, or greater than the minimum, if only the minimum is specified, is acceptable. This means that if a drawing specifies MAX R 0.4 for the edge of a part, any radius from zero to 0.4 is acceptable. However, although the unspecified limit is theoretically zero

Fig 23.Single limit dimensioning

or infinity, it is usually controlled by other dimensions or tolerances. For example, the length of a screw thread may be specified as a minimum dimension, as in Fig.23. This means that the part would be acceptable if threaded up to the shoulder.

18

Geometrical Tolerancing
Definition A geometrical tolerance is the maximum permissible variation of form, orientation, or location of a feature from that indicated or specified on the drawing. The tolerance value represents the width or diameter of the tolerance zone, with in which the point, line, or surface of the feature shall lie. From this definition it follows that a feature would be permitted to have any variation of form. Or take up any position within the specified geometrical tolerance zone.

Fig 24. Tolerance zone for straightness of line

For example, a line controlled by straightness tolerance of 0.15 mm must be contained within a tolerance zone 0.15 mm wide.( Fig 24 ) Geometrical charecteristics: GEOMETRICAL CHARACTERISTICS Form of a line Straightness, roundness, and profile Form of a surface Flatness, cylindricity and profile Orientation Angularity, parallelism and perpendicularity Location Position, concentricity and symmetry Run out This is a special composite characteristic, which will be explained later
Table 1

19

Feature control notes and symbol

Some geometrical tolerances have been used for many years in the form of notes such as: PARALLEL WITH SURFACE A WITH 0.001 STRAIGHT WITHIN 0.005 LOCATE WITHIN 0.005 WITH RESPECT TO ONE ANOTHER While such notes are now obsolete, the reader should be prepared to recognize them when found on older drawings. The modern method is to specify geometrical tolerances by means of Feature Control Symbol. A feature control symbol consists of a rectangular frame, having a suitable height to accommodate the lettering used, and a length as required to contain the necessary information. This frame is divided into two or more compartments. ( Fig 25 )

Fig 25.Feature control symbol

The first compartment contains a symbol representing the geometrical characteristic to be controlled. The second compartment contains the required tolerance value. When necessary other compartments are added to contain datum reference. The frame height should be twice the height of characters used for dimensioning, tolerancing and notes on the drawing. NOTE: this frame height is based on an American, British, Canadian agreement as showing the current CSA standard. However the British standard now specifies a frame height of 2(h+1mm), where h is the character height. The international organization for standardization (ISO) has not yet standardized the size of symbols but generally shows a character height of about 60% of frame height. The ANSI standard recommends a minimum frame height of 0.3 inch (7.6mm) with 1/8 inch (3.2mm) characters. Geometrical characteristics symbol used for form of a line are shown in table 1. Other symbols will be introduced as and when required.

20

Geometrical characteristic Profile (of a line) Straightness Roundness

Symbol ( |

Size of symbol length equal to frame height Length equal to frame height Diameter equal to 75% of frame height

Table 2 Geometrical symbols for from of a line

Application to drawings The feature control symbol is normally connected by means of a leader line from either end of the symbol to the feature to be controlled. This leader terminates in an arrowhead, which is positioned as follows: 1. On the outline of the feature, when the tolerance refers to the line itself or to the surface represented by the line, as shown in Fig 26 (a) 2. On an extension of the outline, but not in the line with a dimension line, as shown in Fig 26 (b) this method applies to line or surface represented by the extension line. 3. On an extension (projection) line from the feature, in line with the size dimension, when the tolerance refers to the axis, center line, or median plane of the feature, as shown in Fig 26 (c) 4. On the center line itself, when the tolerance refers to the common axis, center line, or median plane of all features lying on that center line, as shown in Fig 26 (d) . This method is not shown in ANSI standards. 5. The feature control symbol may also be associated directly with the size dimension of the feature being controlled as shown in Fig 26 (e).

21

Fig 26.Application of feature control symbols

22