Predicting toughness.

Predicting toughness

At last fall's Rubber Division meeting, a paper was presented by John Vicic of FMC's Corporate Development Center in San Jose that was quite interesting. The paper, "Dynamic and mechanical analysis of track pad elastomers before and after service," presented some new information on track pads used on tanks that was quite interesting. Even more interesting, however, was the scientific process that was behind the work that was done. This process and the methods used appears to have potentially a much wider range of application than just tank pads. Specifically, virtually any rubber product that must withstand severe stresses in its application can benefit from the process.

Let's review

Rubber track pads used on various military tracked ordnance vehicles are critical parts of the suspension system. These rubber pads are bonded to metal shoes and, in use, provide protection to various road surfaces in addition to reducing noise and vibration in the vehicle and improving traction on some surfaces.

As used, the rubber pads must withstand very demanding conditions. This includes low and high ambient temperatures (arctic to desert conditions), high surface abrasion, very high loads (both static and dynamic) and penetration by sharp objects. Typical failure modes for track pads include cutting and chunking, blowout and abrasion. As is pointed out, one of the primary contributors to failure is high dynamic heating.

If pads fail, damage to roads can occur, handling and mobility of the vehicle are reduced and vibration within the vehicle can reach the point where electronic systems are damaged.

The work done in this paper centered around the M-1 Abrams tank. This vehicle can weigh in excess of 120,000 lbs. and is capable of traveling at speeds of 44+ miles per hour. Because of the high speeds involved and the weights being carried, existing rubber compounds have had problems meeting the service life requirement (approx. 2,000 miles). In some cases, track pads performed adequately, while in others, pads would fail prematurely. Overall performance was erratic. Typical physical properties (tensile, elongation, etc.) were not picking up the differences between "good" and "bad" pads.

In order to overcome these problems, development work proceeded along a variety of directions. This included work on existing types of formulations as well as development of new formulations using new elastomer systems. This paper reviewed the results of some of these newer formulations. In addition, it looked at new ways of testing and evaluating performance in the laboratory.

What was done?

Work in this study centered around three compounds. One was the standard compound that meets the requirements of MIL-T-11891 (control). The other two were experimental formulas developed by Ft. Belvoir using hydrogenated NBR as the base polymer. Previous work on these new compounds had shown significantly improved performance over the standard compounds. In some cases, mileage life for the new compounds was 2-3 times that of the standard compounds.

Mechanical properties of the test compounds were similar, both at room temperature and at elevated temperature (see table 1). However, analysis of the dynamic mechanical properties plus tensile and strain energy properties showed the HNBR compounds to have better hot tensile and strain energy properties as well as generally lower dynamic compliance than the control. Additionally, the HNBR compounds showed improved thermal stability compared to the control.

Figure 1 shows some of the factors that interact to control performance of elastomeric parts when subjected to dynamic conditions in application. If design parameters are held constant, the dynamic mechanical properties of an elastomer will control ultimate deformation, heat build-up and stored energy which will cause fracture.

Previous work has established that rubber components that are subjected to sinusoidal dynamic stress will generate heat according to the following equation: (1) Ed = (w/2) [(So.sup.2)] J"(w) where

Ed = energy dissipated per second

w = frequency of the applied stress (radians/sec)

So = amplitude of the stress

J" (w) = loss compliance The maximum stored energy in the same part is described by (2) Es = (w/4) [(So.sup.2)] J"(w) where

Es = energy stored per second

J' (w) = storage compliance

Equation (1) can be restated as (3) Ed = (w/2) [(So.sup.2)] tan (d) J' (w) with

tan (d) = J"(w)/J'(w) In practical terms, w relates directly to the speed of the vehicle while So relates directly to the weight of the vehicle. From these equations, it is eady to see that heat generation and stored energy increase proportionally to the square of the increased weight. While speed has a linear relationship to both heat generation and stored energy, its effects are complicated by time and the fact that the compound's dynamic properties are affected by temperature. For example, as frequency is increased, it is possible for loss compliance to increase initially, pass through a maximum, then decrease.

Temperature increase in a material is directly related to hysteresis. Previous work done by Mead and Pattie (see "Elastomers and rubber technology," ed. by Singler and Byme, US Government Printing Office, Washington, DC, 1987, p. 273) shows that hysteresis is directly proportional tan (d). This relationship can be shown as (4) H = ([S1.sup.2 - S2.sup.2]) [Pi] tan (d) / K(*) where

H = hysteresis

K(*) = complex spring rate

S1 = maximum load

S2 = minimum load

These equations are fair representations of what is happening in a rubber component. However, it should be kept in mind that, for accurate results, the equations may need to be adjusted for non-linearity of various factors. Information on this is available in published literature.

Analysis of these equations also shows that controlling the ultimate deformation of the component and the input energy is critical to the life of the component. This control can be effected by either controlling the design parameters or the material's dynamic properties. Excessive deformation of the component will lead to an increase of available fracture energy as well as heat. Heat will degrade an elastomer's long term load bearing capability as well as fracture resistance.

The overall effect of heat, of course, is subject to an elastomer's heat capacity, thermal conductivity, as well as thermal stability. If heat is built up faster than it can be dissipated by either convection or conduction, the component will blow out.

Blowout performance is also related to the overall thermal stability of the compound. This is determined not only by the polymer, but the crosslink stability. In this study, the control compound used a conventional sulfur cure system while one of the test compounds, NBR-6, utilized an EV system and the other test compound, NBR-12, used a peroxide system. In general, it would be expected that the peroxide system would have superior thermal stability compared to the EV. The EV system, in turn, would be expected to have greater stability than the conventional cure system.

(It should be noted that, in developing actual equations to describe actual conditions, it is necessary to shift the equations to fit empirical data. This shift has been described in "Viscoelastic properties of polymers," by Williams, Landell and Ferry, published by John Wiley & Sons. For the purposes of this article, this shift does not affect the reasoning and will not be explained.)

What data was generated in the lab?

Normal physical properties were run on each of the experimental compounds. Data is shown in table 1. No physical properties were available for the control.

Thermal stability of the compounds was determined using thermogravimetric analysis, TGA. In TGA, a sample of material is heated at a constant rate and weight loss is recorded as the temperature rises. Using this technique, plasticizers come off first, followed by the polymers. After the system reaches approx. 1,100 [degrees] F (590 [degrees] C), air is introduced into the system and the carbon black is burned off. All that remains is inorganic ash.

For the test compounds, TGA showed the NBR-12 compound to have the best overall thermal stability, followed by the NBR-6. HNBR is generally reported to have significantly improved thermal stability compared to typical unsaturated elastomers such as SBR. The TGA results obtained would seem to confirm that. From this, it would be expected that the improved thermal stability of the test compounds would improve the field performance of these track compounds.

Dynamic mechanical properties were determined using a Rheometrics mechanical spectrometer, Model RMS-800. Compliance data was obtained in torsional shear at a variety of temperatures and frequencies. The data thus generated showed the loss compliance for the NBR-12 compound to be lower than either of the other two throughout the frequency range studied.

Loss compliance for the NBR-6 compound was lower than the control at high frequencies or lower temperatures. At lower frequencies or at higher temperatures, loss compliance of the NBR-6 was greater than the control. This indicates that the temperature rise in the NBR-6 compound should start off slower than the control, but should increase and exceed the control as the temperature rises.

Storage compliance for the NBR-6 was higher that either the NBR-12 or the control. Tan (d) for the NBR-12 compound was lowest of the three throughout the frequency range. The NBR-6 compound was next, followed by the control.

What happened in the field?

Field results were dramatic. The NBR-6 compound failed after only 99 miles on course. Failure was from a heat generated blowout. Actual temperature measured inside the pad at the time of failure was an astounding 1,500 [degrees] F!

The control compound completed 1,178 miles prior to failing from localized heat blowout and chunking. The NBR-12 compound performed best of all. It was still intact and running when the test was terminated at 1,643 miles.

As table 1 shows, the difference in performance between NBR-6 and NBR-12 was not evident from typical physical properties. Only when additional data was obtained, specifically loss and storage modulus, as well as tan (d), combined with an understanding of the effects of stress amplitude and hysteresis was it possible to understand what happened to the pads molded from NBR-6.

NBR-6 pads had a loss compliance consistently higher than NBR-12. In addition, loss compliance for the NBR-6 pads exceeded the control at higher operating temperatures.

In addition to the factors already discussed, an additional note should be made about strain amplitude. Strain amplitude acts as a multiplier for loss compliance. That is, under a fixed load, a highly compliant rubber compound will become even more compliant due to strain amplitude effects. As a result, the increased storage compliance will cause a reduction in the spring rate, which, in turn will contribute to heat buildup. The net result is that the NBR-6 compound may have, in fact, had a much greater compliance than indicated by the loss compliance alone. It may well have been much greater than the control, rather than similar, as the initial lab work indicated. The combination of strain amplitude and high initial loss compliance will increase the total loss compliance of the NBR-6 significantly. As a result, higher heat buildup than the control could be expected.

It should be noted that loss compliance and storage compliance for lab prepared samples of NBR-6 were very similar to NBR-12. Loss and storage compliances measured from surfaces samples of the molded track pads showed the dramatic differences from the NBR-12. It was concluded from this that there were problems in the manufacture of the NBR-6 pads, either mis-compounding or undercure.

Conclusions

While all this work is interesting for tank tracks, I think the real interest comes from the potential of expanding the scope of this work to other products. The rules, equations and factors involved in this study apply to all rubber products, not just track pads for tanks. Use of dynamic mechanical analysis, along with the determination of the various loss compliances, etc., can provide information that is invaluable in many high stress applications. This includes sealing products, solid tires, rolls -virtually any application where dynamic stress is involved and stress levels are high.

Review of the various equations certainly gives insight into the functioning of compounds in dynamic situations. While exact determination of each of the factors required to fully define a product may not always be called for, certainly review of the factors and, in may cases, determination of at least some of them may be very beneficial. Certainly, use of this analysis helps move compounding and formulations more toward "science" and away from "black art." [Tabular Data 1 Omitted] [Figure 1 Omitted]