Development of a housing for on-line-wear-analysis of clutch lining pads under dry friction on test benches.

Albers, Albert ; Kelemen, Simon

1. INTRODUCTION

The Institute of Product Development at the University of Karlsruhe analyses among other things the tribological behavior of clutch lining pads under dry friction on test benches. 20 mm pellets were cut out of the friction pads, which were bonded on a plane steel-disc and reamed on a cast iron disc. The challenge was to analyze the abrasion of the pads in-situ. In order to understand the tribological behavior of the coupler coating, it is important to know the abrasion process itself. The abrasion can be measured by using RNT or with optical devices. Using RNT the pads need to be radioactive irradiate. The amount of the microscopic abrasion particles can be measured accurately by its radioactivity (Shakhovorostov, 2005).

Compared to other methods of abrasion measurement the RNT allows a continuous, non-contact on-line analysis under real operation conditions. In order to capture and measure the particles it is necessary to build a housing around the friction contact to isolate it from the environment and keep all particles inside. This is important because of the particle size range (1 nm to 100 um) and the radioactivity. In dry friction systems a gas is needed to transport the particles to the sensor. Here dried air is used to sweep the abrasion particles of the housing walls and transport them to the sensor.

To solve the problem of adherence different technical solutions can be apprehended. One solution is to glaze the wall inside (electrolytic polishing) or to coat the wall with teflon or nano-particles to avoid adherence. These procedures are expensive and the surfaces are easily damaged.

Alternatively adherence can be avoided by using tribo-electrical effects (Polzer et al., 1982), which appear during the friction in contact. The same polarized and charged particles push themselves off resulting in a minimized adherence.

2. TRIBO-ELECTRICAL CHARGE IN FRICITION CONTACT

The effect of electric charging during friction is often to be avoided in the industry. Electrically charged machine parts may lead to the destruction of electrical components. The tribo-electrical effect appears when the surfaces of two tribological friction partners of different materials are in contact. When disconnected, the electrons migrate from one surface to the other and become positively charged, while the other surface absorbs the electrons and becomes negatively charged.

[FIGURE 1 OMITTED]

If the disconnected partners are electrical isolators or not grounded, the different poles adhere. Basically every body, gas or fluid, conductor or insulator can be charged. The charging is highly restricted by the following factors:

* Condition of Surfaces

* Size of Contact area/friction area

* Conductivity of air

* Electron affinity

* Pressure between surfaces

* Velocity of separation

* Velocity of friction

The voltage in the single discs can be some megavoltage after the disconnection. The chemical material identity essentially defines the voltage generation. An important tool to electrically characterize a material is the tribo-electrical array. The array describes which of the two materials at friction conduct positive, i.e. gives electrons away or adheres electrons, i.e. becomes negative. The bigger the distance of the friction-partner to each other, the easier is the electron diffusion.

3. TRANSFER TO A HOUSING FOR TESTING DRYLY FRICITON PADS

The paramount disadvantage of coupler friction pads is the unknown composition of the pads. A rough overview can be obtained with an EDX-analysis. Figure 2 shows a list of possible ingredients of friction-pads in dryly coupler friction pads (Severin et al., 1999).

Figure 3 shows the distribution of different materials in accordance to their electrical behavior. Materials on top of the array are more likely to charge positive, while materials on the bottom are more likely to charge negative (Simco, 2009).

Since the composition of the friction pads is unknown it is not possible to determine the electrical potential of particles while flying out of the friction contact towards the housing walls. They can be positive or negative, depending on the position of the particle material in the tribo-electrical array.

[FIGURE 3 OMITTED]

To charge particles with a defined potential it has to be a voltage at the cast iron disc, which charges explicit the part in friction contact.

4. TECHNICAL REALISATION AND RESULTS

The pellets and the friction discs were electrically isolated and positively charged. All components between pellet body and surface friction pad were forced a positive charge. If the housing is also positively charged, the abrasion particles are slowed down and do not adhere to the housing surface, irrespective of form or surface roughness.

As a consequence the surface of the pellet carrier must be isolated. The surface is negatively polarized, because of the positive charge at the cast iron disc. Figure 4 shows the housing with its electrically charged components.

[FIGURE 4 OMITTED]

The first experiments verified the function of the presented housing. The friction disc and at the electrically isolated housing wall were charged with 600-1400 V. The pellets rotated at 800-950 rpm with a torque of 5 Nm (Gauger, 1998) against the cast iron disc from 2-8 h (see Tab. 1).

The amount of particles shown in Tab.1 was not the complete amount of abrasion. In the test the housing was not completely charged positive. One half was not electrically isolated to give a comparison between the both sides.

At a charge of 1200 V a reduction of the particles about 20- fold could be noticed. Before and after the test the isolated part of the housing was measured. The difference between these measurements is the mass of the particles on the housing. The amount of 2mg is the minimal tolerance of the balance. After the test with a charge over 1200 V nothing was seen on the wall. In the wipe test no particles adhered on the white paper.

5. CONCLUSION AND OUTLOOK

It has been shown that using electrical charge is a simple and effective way to avoid adherence of particles to a housing wall. By applying a positive potential at the cast iron disc it was possible to charge the abrasion particles to a certain electrostatic level which guaranteed that the particles push themselves off the wall.

The next step is to design a flow optimized housing, which combines the effects of optimized particle flow with an electrically charged housing and pellets for measuring on-line dryly continuous coupler friction systems.

6. REFERENCES

Gauger, Dirk (1998) "Wirkmechanismen und Belastungsgrenzen von Reibpaarungen trocken laufender Kupplungen (Mechanism and maximum loads of friction couples in dryly running clutches)" VDI-Verlag, ISBN 3-18-330101-6, Dusseldorf, Germany

Polzer, Gottlieb, MeiBner ,Franz (1982): "Grundlagen zu Reibung und Verschleifi (Basics of friction and abrasion)", Deutscher. Verlag fur Grundstoffindustrie, Leipzig, Germany

Severin, D, St. Dorsch.;.(1999) "Mechanismen im Reibkontakt trockenlaufender Bremsen und Kupplungen (Mechanism in friction contacts of dryly running brakes and clutches)", Technische Universitat Berlin, Germany

Shakhovorostov, Dimitry (2005) "Untersuchung der Dissipationsmechanismen der Metall/Metall-Reibpaarung mit Hilfe der Radionuklidtechnik und faseroptischen IR-Temperaturmessung (Verification of dissipations-mechanism with radio-active tracer technique and temperature measuring)", Verkehrs- und Maschinensysteme Technischen Universitat Berlin, Germany

Simco B.V. (2009) Available from: http://www.simco-elektrostatik.de/, Accessed: 2009-04-30

Tab. 1. Results of test with electrical charge
and non electrical charge

Rotation Torque Time charge Particle
speed [Nm] [h] [V] amount
motor [mg]
[rpm]

800 5 2 1300 <2
800 5 2 0 80
900 5 2 600 10
900 5 2 0 50
900 5 23 1200 <2
900 5 3 0 50
950 5 3 1400 <2
950 5 3 0 110
950 5 8 1350 <2
950 5 8 0 60

Fig. 2. Examples of components in dryly friction pads

Binder Fibers Mineral filler

Phenolic resin Steel wool Mica
Mela mine resin Brass wool Slate powder
Natural rubber Fibre glass Kaolin
 Basalt fibre Barium sulphate
 Wcllastonlte Chalk
 Araroid

Oxide Metals Antiseize

Zinc oxide Iron Graphite
Magnesium oxide Copper Sulphite
Feric oxide Zinc Coke
Chromium oxide Brass
Aluminium oxide Bronze
Silicon oxide