Study of Wear in Conditioned Granular

Soil by Using a New Test Device

Cristina Gabriela Oñate Salazar*, Daniele Martinelli*, Carmine Todaro*, Daniele Peila*, Alessandro
Boscaro **
* DIATI, Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129, Torino, Italy;
**UTT, Mapei SpA, Via Cafiero 22, 20158, Milano, Italy

The wear phenomenon in TBM machines is not a simple issue, as many parameters are playing an
important role on the consumption of the tools during a tunnel excavation. The action of the wear due
to the contact with the soil can occur on the excavation tools, on the cutterhead, in the excavating
chamber and on the screw conveyor.
The test developed at the Laboratory “Tunnelling and Underground Space Center” of the Politecnico
di Torino is aimed to reproduce the interaction between the tools and the excavated material,
simulating the excavation conditions in terms of soil conditioning and water content. The test, which is
not standardized, is studied to evaluate the wear in a global point of view. The present work starts by
testing the influence of the use of chemical agents during the conditioning phase on the drop of the
wear phenomenon, in terms of metal weight loss.

1. INTRODUCTION
The wear phenomenon of the excavation tools used in TBM tunneling machines is playing recently a
major role in the design stage of full-section tunnels in both soils and rocks. Concerning EPB excavation
technology, the main issue regarding productivity is the wear of its components (Peila et al.,2012),
which increases the so called dead time in the whole process. The friction generated between the
components and the ground causes the consumption of the tools and other metallic components in
contact with the ground, described as primary and secondary wear by Nielsen et al. (2006a). This leads
to downtime, with risky and complicated maintenance operations (Peila and Schulkins, 2009), as
occurred in recent tunneling works in Liguria (Italy).
The study of this argument in granular soils is very delicate, as it encloses not only the properties of
the excavation tools, but also the characteristics of the ground to be excavated (hardness, mineralogy
composition, shape and dimension of the particles and moisture content). The penetration mechanism
is based on the scraping of the ground, by mean of cutting tools, with its consequent consumption.
Concerning the rocks, the excavation principle is characterized by crushing the rock mass matrix by
applying high compressive stresses through cutting disks.
In such a way, to avoid frequent downtimes and their consequent economic losses, it is very important
to study this subject and therefore understand and forecast the durability of the components, in order
to try to maximize the productivity of the machine by reducing the frequency of maintenance stoppings.

1
 Although this problem is crucial in TBM tunneling, there is a lack of standardization regarding the
assessment of abrasivity potential. The available tests from literature (Nielsen et al., 2006a) follow:
 Cerchar test, giving Cerchar Abrasivity Index (CAI);
 LCPC test, giving the LCPC abrasivity index (ABR);
 NTNU abrasion test, giving the abrasion value (AV–AVS);
 Hardness Vickers test, giving the hardness value (HV).
Regarding NTNU tests, Nilsen et al (2006b and 2007) and Jakobsen and Dahl (2010) proposed some
modifications to these in order to make them suitable for granular soils. Thuro et al. (2007) adapted
LCPC abrasivity test, which is originally conceived for rocks, to soils.
In every test mentioned before, there is an inherent weakness that reduces the effectiveness in
predicting the abrasiveness potential, which is the fact that they do not consider the content of water
does not allow a proper assessment of the wear, as discussed by Gharahbaghet al. (2011). Also Jakobsen
et al. (2012) and Rostami et al (2012) remarked the strong influence of the water content of the soil on
its abrasiveness.
Gharahabagh and Rostami(2010) and Rostami et al. (2012) have developed a new test with the
purpose of simulating the reciprocal interactions between soil particle sand excavation tools in
conditions similar to those found in-situ. This test allows analyzing granular material samples coming
from the site, after being suitably conditioned by adding water and/or foam in a pressurized tank.
The aim of this research is to show the influence of water content on the wear phenomenon based
on tests developed with a device specially designed by TUSC (Tunneling and Underground Space Center)
DIATI, Politecnico di Torino. Another variable considered is the addition of wear reduction chemical
products capable to reduce the consumption of cutting tools.

2. TUSC SOIL ABRASION TESTING APPARATUS

Based on the bibliography, TUSC laboratory at Politecnico di Torino developed a new experimental
procedure and apparatus (Figure 1 and Figure 2). This device allows the evaluation of wear phenomenon
and its reduction with soil conditioning, and it is based on experimental procedures from literature
modified in order to make them simpler, more reliable and faster (Barbero et al., 2012).
The developed apparatus admits the evaluation of the abrasivity of different types of soil, with grain
size of maximum 2 cm, keeping constant the parameters of the device (material of the rotating
propeller, rotation speed, duration and the confining pressure) and varying the conditioning dosages.
More specifically, the wear test consists in determining the weight loss of a metal disk rotating
around its axis at a constant speed of 160 rpm inside a cylindrical container filled with the soil. During
the test, the torque provided by an electrical motor and required for the rotation of the disc is recorded.
This value is different from the one present at the cutter head and at the screw conveyor, but anyway
provides a quantifiable wear measure compliant with the service life of the chemical products used for
soil conditioning. Test campaigns have been carried out using aluminum and steel discs characterized
respectively by Vickers hardness values of 116 and 189 respectively. To ensure a continuous contact
between the disc and the ground, a normal confinement pressure of about 2 kPa is applied. At the end
of the test, the wear of the disc can be quantified by weighing the weight loss due to the friction with
the soil and thus correlating this value to the average torque applied during the execution of the test.

2
 Figure 1. Scheme of the proposed test

Figure 2. Cross section and pictures of the test device and of the wear disk.

2.1 TESTING PROCEDURE

The developed wear testing procedure can be summarized with the following steps:
 sample preparation: 20-25 kg of granular soil is mixed with water, foam and/or eventually other
chemical additives depending on the set of conditioning to be tested ;
 weighing of the disc: calculation of w1, which is the initial weight of the disc;

3
  assembly of disc on drive shaft: the shaft is inserted assembled with the disc inside the tank and
the removable bushing is locked on the tank by mean of two bolts;
 filling of the first layer of ground: the tank is filled from the bottom to the height of disc, in order
to ensure the right contact between the particles of the ground and the lower surface of the
disc;
 completion of filling of chamber: all the sample is placed inside the tank and the confinement
pressure is applied;
 start of the test (10 minutes) with real time torque measurement;
 weighing of the disc: calculation of w2 (final weight) and calculation of weight loss (Δw).

2.2 USED SOIL

In order to study the wear phenomenon, a test campaign was carried out by using a quartz coarse sand
coming from a quartzite quarry (suitable for our purpose after crushing process), which has a natural
water content of 2% by weight.
In Figure 3 a sample of quartz sand can be observed, while its grain size distribution is showed in
Figure 4. Quartz grain properties are listed in Table 1, where d10 and d60 are two meaningful diameters
while U is coefficient of uniformity. The mineralogy properties needed for wear study, were obtained
through microscopic analysis (Figure 5 and 6) and the results show that quartz is the main mineral
composing the sand, with a content of 98%, while the remaining 2% consists of feldspars and iron
impurities.

Figure 3. Quartz sand sample.

Table 1. Main properties of used soil.

Quartz sand properties

d10 0,9
d60 6,5
U 7,22

4
 100
90
80
Passing (%) 70
60
50
40
30
20
10
0
0.001 0.01 0.1 1 10 100
Φ (mm, log scale)
Figure 4. Grain size distribution of quartz sand used.

Figure 5. Quartz sand (10x): limpid grain of quartz sand (left) and detail of feldspar grain (right).

Figure 6. Limpid grain of quartz sand; phase contrast viewing. Detail of feldspar grain on the top of
picture.

5
2.3 DISCS
Test campaign was carried out by using discs made of aluminum and steel (Figure 7). In Table 2 the
results concerning the Vickers tests are shown, obtained by using HV 10 program (applied load of 10 kg).

Figure 7. Aluminum (left) and steel (right) discsused for the tests.

Table 2. Vickers test outcomes.

Steel disc Aluminum disc
Point tested Hardness (HV) Average Point tested Hardness (HV) Average
1 194 1 123
2 196 2 118
3 191 189,2 3 114 115,6
4 179 4 114
5 186 5 109

3. PERFORMED TESTS
As a first step, the wear phenomenon was studied by using steel and aluminum discs in rotation inside
quartz sand conditioned with different percentage of water. The used range was between 2% and 15%
(between the natural and the saturation limit). Outcomes of test campaign are shown in Figure 10: the
curve obtained will be called herein “wear bell curve”.
In the second phase of the study, the percentage of water corresponding to the highest value of
weight loss (the highest value of wear bell) has been taken into account in order to test an innovative
chemical agent, MAPEDRILL F.R.A. 02C., herein called ease FRA02C. This is an experimental prototype
developed by MAPEI SpA expressly with the goal of wear reduction and for commercial reasons,
chemical composition or more details are no available.
By using this additive, particles are covered by thin polymeric layer that should decrease friction
strengths between ground and tools. Examinations made by using a macroscope have been carried out
in order to observe the behavior of the grains when the additive is in contact with the quartz grain
surface. Some images are shown in Figure 8.

6
 The purpose of this second test campaign was to obtain the right dosage of additive for the major
decrease of wear phenomenon, with a concentration of FRA 02C ranging between 2% and 5% by the
weight on water added, as advised from Mapei technicians. Used disc was in aluminum material. Results
are shown in Figure 9.
Based on these results, with a concentration percentage of FRA02C of 2% (which is the most
economically reliable), other two different wear bells have been found, by using both the discs.
Global outcomes about weight loss, average of torque and moisture content are summarized and
shown in Figure 10 and in Figure 11.

Figure 8. Drop of the solution of water and FRA02C (2% by the weight of water) on quartz grain surface.

0.70

0.60

0.50
Weight loss (g)

0.40

0.30

0.20

0.10

0.00
0 1 2 3 4 5 6
Concentration of FRA (% by weight on added water)

Figure 9. Weight loss in function of different percentage of FRA02C used in conditioning settings. The
amount of additive was calculated with a constant total water content of 10%. Test campaign
performed by using aluminum disc.

7
 0.70

0.60

0.50
weight loss (g)

0.40
Disc: Aluminum
Ground: Quartzite
0.30
Disc: Steel
Ground: Quartzite
0.20

0.10

0.00
0 2 4 6 8 10 12 14 16
Total water content (% by weight)

Figure 10. Wear bells for quartz sand conditioned with different percentage of water obtained by
using aluminum and steel discs.

1.60

1.40

1.20
Average torque (Nm)

1.00

0.80 Disc: Aluminum

Ground: Quartzite

0.60 Disc: Steel

Ground: Quartzite

0.40

0.20

0.00
0 2 4 6 8 10 12 14 16
Total water content (% by weight)

Figure 11. Average torques depending on the water content.

8
 4. CONCLUSIONS
Understand wear phenomena and all factors that play significant role is an intricate work. Weight loss is
a parameter that depend from all these families of factor: geotechnical, chemical (ground and material
of tool), mineralogical and mutual interaction between them.
Wear bells obtained show the dependence of wear phenomenon with the moisture content.
Different percentage of water inside the ground changes in a serious way the wear attitude. Taking into
account friction between grains, if tested soil has consistence near his liquid limit (Lancellotta, 2012),
water is allocated inside gaps and friction strengths is reduced (increase of pore water pressure). We can
observe this behavior after peak of wear bells (Figure 10).
With fix setting for both discs, tests by using steel disc showed less weight loss compared to tests carried
out with aluminum disc for each percentage of water content.
Tests performed by adding the FRA 02C agent showed a significant reduction on the wear potential.
Currently research based on additive is continuing with goal of obtain additional information about its
effect on wear.
Different soils and different discs will be tested in order to understand:
 how chemical reactions (between tool material, soil and additive) could affect wear
phenomena;
 suitable metallic materials for producing excavation tools;
 main factors of wear phenomena that should be used in order to compare different test
campaigns.

ACKNOWLEDGEMENTS
Region Project HM-TUNES have financed this project. Different people worked with the authors to this
research. Authors thank Mapei Utt for providing chemical agent, Prof. Ugues from DISAT (Department of
Applied Science and Technology) and Prof. Clerici from DIATI (Department of Environment, Land and
Infrastructure Engineering) for operational advice and mineralogy analysis.

REFERENCES
Alavi Gharahbagh, E.,Rostami, J., 2010. Tackling a new soil abrasivity index. North American Tunneling
Journal, Sep 2010, pp16–18.
Alavi Gharahbagh, E., Rostami, J., Gilbert, M., 2010.Tool wear in soft ground tunnelling - Developing a
reliable soil abrasivity index. Proc North American Tunneling conference 2010, SME UCA, Colorado,
USA.
Alavi Gharahbagh, E., Rostami, J., Palomino, A.M., 2011. New soil abrasion testing method for soft
ground tunneling application. Tunneling and Underground Space Technology 26, 604-613.
Alavi Gharahbagh, E., Rostami, J., Palomino, A.M., Mosleh, M., 2012. Developing of soil abrasivity testing
for soft ground tunneling using shield machines. Tunneling and Underground Space Technology 28,
245-256.
Barbero, M., Peila, D., Picchio, A., Chieregato, A., Bozza, F., Mignelli, C., 2012. Procedura sperimentale
per la valutazione dell’effetto del condizionamento del terreno sull’abrasione degli utensili nello
scavo con EPB. GEAM, Anno XLIX, n.1, Aprile, pp 13-19.

9
Jakobsen, P.D., Dahl, F., 2010. Soil abrasion in TBM Tunneling. Korean Tunneling Association on
Mechanized Tunneling.
Jakobsen, P.D., Langmaack, L., Dahl, F., Breivik, T., 2012. Predicting the abrasivity of in-situ like soils.
Tunnels & Tunneling International, June 2012, pp 41-44.
Lancellotta, R., 2012. Geotecnica, Quarta edizione.
Nilsen, B., Dahl, F., Holzhäuser, J., Raleigh, P., 2006a. Abrasivity of soils in TBM tunneling. Tunnels &
Tunneling International, March 2006, pp. 36-38.
Nilsen, B., Dahl, F., Holzhäuser, J., Raleigh, P., 2006a. Abrasivity testing for rock and soils. Tunnels &
Tunneling International, April 2006, pp. 47-49.
Nilsen, B., Dahl, F., Holzhäuser, J., Raleigh, P., 2006b. SAT: NTNU’s new soils abrasion test. Tunnels &
Tunneling International, May 2006, 43-45.
Nilsen, B., Dahl, F., Holzhäuser, J., Raleigh, P., 2007. New test methodology for estimating the
abrasiveness of soils for TBM tunneling. In Proceedings Rapid Excavation And Tunneling Conference
- 2007; Toronto, ON.
Peila, D., Picchio, A., Chieregato, A., Barbero, M., Dal Negro, E., Boscaro, A., 2012. Test procedure for
assessing the influence of soil conditioning for EPB tunneling on the tool wear. World Tunneling
Congress 2012, Bangkok, Thailand.
Peila, D., Schulkins, R., 2009. Soil conditioning and backfilling grout, ITA/AITES – Training Course “Tunnel
Engineering”, Budapest.
Thuro, K., Singer, J., Kasling, H., Bauer, M., 2007. Determining abrasivity with the LCPC test. Proc 1 st
Canada-US Rock Mechanics Symposium, 2-7, pp 827-834.

10