Brake System Performance at Higher Mileage.
Simulating the effects of corrosion exposure, debris exposure, and brake component wear simultaneously throughout an entire brake corner lifecycle is a difficult task, and one often left to vehicle level durability type evaluations. As a result, detailed bench level measurements of key performance characteristics with all three of these inputs are frequently unavailable. In order to bridge this gap, brake corner hardware from an array of different vehicle types were collected following a 10 year accelerated corrosion exposure durability test, and subjected to a number of bench level brake corner dynamometer performance evaluations to capture caliper fluid consumption, residual brake drag, apparent friction, and brake torque variation. By first comparing this measured data to the population averages from new condition product validation data sets, an assessment of performance degradation can be made. Following this, predictive simulation tools can be used to assess the impacts to customer facing attributes - namely pedal feel and fuel economy. Additionally, subjective evaluations of cosmetic corrosion and service considerations are made.
The authors will note that the study represented in this paper results from review of internal test reports published over a span of about 4 years. Vehicles were included in this internal benchmarking activity somewhat opportunistically. The present study attempts to tie together and draw comparisons between results from what had been a series of self-contained measurement programs.
A cursory internet search on corrosion related brake complaints yields a number of examples of recent model vehicles generating significant customer dissatisfaction - one does not have to look hard to identify exasperated vehicle owners with corrosion related brake complaints.
As such, an investigation into various design strategies' efficacy in preventing corrosion related brake failure modes is of crucial importance.
For the benefit of the reader, an index of topics covered and their organization within this paper is noted below:
1. Description of Samples and Test Exposure
2. Summary of Findings and Discussion
a. Comparison of NAO and Low Met Friction Behavior Following Corrosion Exposure
b. Analysis of Pad to Caliper Bracket Corrosion Binding Effects in Sliding Calipers
c. Comparison of Brake Caliper Corrosion Binding With and Without Regenerative Braking
d. Effect of FNC rotors on corrosion of braking and mounting surfaces
e. Effect of Regenerative Braking on Rotor Surface Corrosion Cleaning (Vehicle G to I)
f. Effect of Usage on Brake Hose Expansion
g. Different Effects of Corrosion Binding on Front and Rear Axle Hardware
h. Different Effects of Corrosion Binding on Aluminum vs. Cast Iron Calipers
i. Comparison of carbon ceramic to traditional cast iron rotors in a high performance system
j. Assessment of Cosmetic Corrosion Performance (zinc vs. zinc-nickel vs. paint)
k. Effect of Long Term Usage on Rotor Lateral Runout (LRO) and Disc Thickness Variation (DTV)
l. 5 year vs. 10 year corrosion exposure
3. Test Case: Corrosion Induced Fuel Economy Degradation through Vehicle Life
4. Test Case: Corrosion Induced Pedal Feel Degradation through Vehicle Life
1. DESCRIPTION OF SAMPLES AND TEST EXPOSURE
Brake corner hardware from a variety of vehicle types were evaluated, including cars and trucks, hybrid and traditional powertrains, and different corrosion protection design strategies. A high level summary of the vehicles from which hardware was retrieved is shown in Appendix A at the end of this paper.
Each set of brake corner hardware was subjected to vehicle level general durability testing including accelerated corrosion exposure. The test cycle simulates the effect of 10 years of use and corrosion exposure in a 95th percentile severity corrosion environment. Vehicle E is an exception, in that only 5 years of equivalent corrosion exposure was simulated. Following durability testing, the brake calipers, pads, and rotors were retrieved from the vehicle for component testing. Where possible, rotor LRO and DTV measurements were recorded prior to component removal, and other relevant disassembly information was noted.
Vehicle A front calipers have an anodized and painted aluminum body, and use brake pads with phosphate treatment (NAO only) and a powder paint. One test began with low met brake pads, and completed the entire schedule using the original pads. The second test began with low met brake pads, but required a pad replacement mid test, at which point NAO pads were used. The front and rear rotors are cast iron and treated with ferritic nitrocarborization (FNC). Rear corner hardware from this vehicle was not available for analysis.
Vehicle B front calipers have an anodized and painted aluminum body, and use brake pads with powder paint only. Rear calipers have an anodized and painted aluminum body, and use brake pads with powder paint only. The front and rear rotors are standard cast iron.
Vehicle C front calipers have an anodized and painted aluminum body, and use brake pads with powder paint only. The front rotors are standard cast iron. Rear corner hardware from this vehicle was not available for analysis.
Vehicle D opposed piston front calipers have an anodized and painted aluminum body, and use brake pads with powder paint only. Rear opposed piston calipers have an anodized and painted aluminum body, and use brake pads with powder paint only. The front and rear rotors are standard cast iron, with cross-drilling.
Vehicle E front and rear calipers are identical to vehicle D, though they are paired to a carbon ceramic front rotors, and use pads with carbon ceramic compatible friction material. Brake pads are protected with powder paint only.
Vehicle F twin piston sliding front calipers have a Zinc plated cast iron body and bracket with clear coat sealer, and use brake pads with phosphate treatment and a powder paint. The front rotors are cast iron and treated with ferritic nitrocarborization. Rear corner hardware from this vehicle was not available for analysis.
Vehicle G single piston sliding front calipers have a Zinc plated cast iron body and bracket with clear coat sealer, and use brake pads with phosphate treatment and a powder paint. Single piston sliding rear calipers have a Zinc plated cast iron body and bracket with clear coat sealer, and use brake pads with phosphate treatment and a powder paint. Rear calipers have a mechanical integral park brake feature. The front and rear rotors are treated with ferritic nitrocarborization.
Vehicle H single piston sliding front calipers have a Zinc-Nickel plated cast iron body and bracket with clear coat sealer, and use brake pads with powder paint only. Single piston sliding rear calipers have an anodized aluminum body and Zinc-Nickel plated cast iron bracket with clear coat sealer. Rear calipers have a motor on caliper electric parking brake feature. Brake pads with Zinc-Nickel plating and a powder paint are used in the rear. The front and rear rotors are treated with ferritic nitrocarborization.
Vehicle I single piston sliding front calipers have a Zinc-Nickel plated cast iron body and bracket with clear coat sealer. Front pads are protected with powder paint only. Single piston sliding rear calipers have a Zinc-Nickel plated cast iron body and bracket with clear coat sealer. Rear calipers have a motor on caliper electric parking brake feature. Brake pads with Zinc-Nickel plating and a powder paint are used in the rear. The front and rear rotors are treated with ferritic nitrocarborization.
2. SUMMARY OF FINDINGS AND DISCUSSION
2a. Comparison of NAO and Low Met Friction Behavior Following Corrosion Exposure
In an attempt to contrast friction behavior differences in NAO and Low Met formulations following 10 year equivalent accelerated corrosion, data from vehicle A front corners is analyzed. Vehicle A offers both NAO and Low Met friction variants, with otherwise the exact same opposed piston aluminum front caliper design. The rotors paired with both friction sets are also identical, and FNC treated.
As can be seen in Figure 1, the general tendency for both friction material types is to experience a nominal drop in apparent friction with corrosion accumulation, which varies depending on the type of stop conditions in question (pressure ramp and FMVSS135 style stop conditions are evaluated). A combination of taper wear and corrosion binding of the pads within the caliper assembly work to reduce the overall efficiency of the caliper as a clamping device, reducing the net apparent friction of the brake corner. Additionally, accumulated corrosion on the braking surfaces reduces pad to rotor surface contact, and the ability of the friction material to form and maintain a stable tribofilm, reducing the friction output of the brake pads. The result is a net decrease in apparent friction which will be evident in vehicle level pedal feel, though in this example, little difference between the NAO and low met friction is evident.
Figure 1. Apparent Friction of NAO and Low Met Linings on Vehicle A following corrosion exposure. % Change in Apparent Friction Following Corrosion Exposure Cold Ramp Warm Ramp Cold Effectiveness Failed Power NAO -8% -7% 0% -15% LOW MET -7% -3% -3% -18% Note: Table made from bar graph.
2b. Analysis of Pad to Caliper Bracket Corrosion Binding Effects in Sliding Calipers
In order to quantify the effects of different brake pad and caliper bracket corrosion protection strategies, we can compare fluid consumption and brake drag results from vehicle G and vehicle H rear hardware. These particular corners offer a unique look at corrosion effects, as vehicles G and H have similar strong hybrid powertrains, leading to greatly reduced foundation brake duty cycle throughout the durability test. It is clear in visual analysis of the hardware that corrosion growth in the brake pad to caliper bracket contact area is present, and may lead to increased slide forces. In the new condition, drag torque values between vehicle G and H calipers are quite similar. In Figure set 2, we show post 10 year equivalent corrosion vehicle G front and rear corner drag results as a percentage of drag exhibited by the respective vehicle H corners. It should be noted that in all cases, the data set is an average of both LH and RH samples, excepting vehicle G rear hardware, as one corner was fully seized, and exhibited drag torque exceeding the maximum torque capabilities of the test machine (20 Nm).
As one may expect, the increased corrosion protection performance of the Zinc-Nickel caliper bracket plating over the Zinc only configuration leads to a considerable improvement in drag torque performance after 10 years equivalent corrosion exposure. Additionally, the addition of Zinc-Nickel plating on the vehicle H rear pads demonstrates considerably more contrast with the rear hardware from vehicle G. It should be noted that the front brake pads and rotors in vehicle H were replaced at 80% of completion of the durability test (thus the tested hardware had 20% corrosion durability exposure on the pads and rotor, and 100% exposure on the caliper).
Next, we will analyze fluid consumption in a similar manner. Only the front corner was included in this analysis, as the rear data are skewed by having one caliper completely seized (and thus nonfunctional for testing), and presenting only the functional caliper for comparison would present misleading conclusions.
Additionally, we can evaluate vehicle F to assess performance degradation following corrosion exposure. This vehicle's brake corner loading, as a heavy truck, is considerably higher than vehicles G and H which were previously compared. Note vehicle F was tested only to 5 years equivalent exposure. As no other corners presented an interesting comparison within the same vehicle segment, Figure set 4 displays the average degradation in drag and fluid consumption for these front brake corners from this design's new state. Again we see calipers with zinc only plating and brake pads with no additional corrosion protection beyond phosphate treatment and paint resulting in significant binding, and degraded performance as a result. This observation is key, as drag torque degradation is present and severe after only 5 years of corrosion exposure - well within the typical life of a single set of brake pads for many vehicles in the field.
2c. Comparison of Brake Caliper Corrosion Binding Corrosion with and without Regenerative Braking
We can next analyze vehicle H (strong hybrid) and vehicle I (traditional ICE) data to compare the effects of reduced friction braking duty cycle on corrosion binding effects. As can be seen in appendix A, these two vehicle applications have very similar brake corner content, so the addition of regenerative braking capability on vehicle H over vehicle I poses an interesting comparison.
In figure 5 fluid consumption post corrosion is shown as a percentage increase from each analyzed corner's new condition performance. We find that the increase in fluid consumption trends on the front and rear axles of these vehicles are mixed. Vehicle H front hardware exhibits a nominal decrease in fluid consumption after 10 years equivalent corrosion exposure, primarily at low pressures, whereas the vehicle I hardware exhibits a nominal increase. The resulting data provides no clear indication that the reduced duty cycle has negatively impacted the fluid consumption behavior due to corrosion. In reality, the fluid consumption changes over time are due to the competing effects of reduced lining thickness due to wear, piston seal relaxation over time, taper wear and pad shape abnormalities, and corrosion binding - thus the competing factors lead to mixed results in this comparison. More significant differences may have been apparent had the data been available to make this comparison on hardware with less robust corrosion protection available, such as vehicle G. Additionally, the pad and rotor replacement at 80% of vehicle H corrosion durability testing further complicates the data analysis.
The opposite trend is shown on the rear hardware fluid consumption data, where vehicle I exhibits more than 50% increase in fluid consumption throughout the entire pressure curve, where vehicle I hardware exhibits a nominal decrease, primarily at low pressures. Considering that rear brake duty cycle is notably lighter than fronts in standard passenger vehicles, combined with the additional effects of debris flow that will be discuss later in this paper, we see evidence of more pronounced rear caliper fluid consumption degradation in the regenerative braking vehicle analyzed due to corrosion binding.
Drag data from these two vehicles, shown in figure 6 again as a percentage increase from each analyzed corner's new condition performance, shows other differences. The front hardware from vehicle H shows a drag increase at low experience pressure, with higher experience pressures reversing the trend. Vehicle I front hardware shows a nominal increase in drag, becoming more pronounced as experience pressure increases. Similar to the fluid consumption data, it would seem that corrosion binding did not occur any more significantly on the regenerative braking vehicle H compared to vehicle I, rather the opposite may be true based on the measured drag torque. Again, this may not have been true in a comparison of vehicles with less robust corrosion protection.
Figure 5. Fluid consumption increase from new condition to post corrosion durability - vehicle H and vehicle I Pressure % Fluid Consumption Increase [kPA] Vehicle H Front Axle Vehicle I Front Axle -30% 14% -19% 27% -8% 14% Pressure Vehicle H Rear Axle Vehicle I Rear Axle [kPA] 56% -32% 58% -13% 52% -7% Note: Table made from bar graph. Figure 6. Drag Torque increase from new condition to post corrosion durability - vehicle H and vehicle I % Brake Drag Increase @ 400 RPM Vehicle H Front Axle Vehicle I Front Axle DRAG @ 1000 67% 0% DRAG @ 2500 88% 75% DRAG @5000 -19% 117% DRAG @ 10000 -72% 163% Vehicle H Rear Axle Vehicle I Rear Axle DRAG @ 1000 436% 100% DRAG @ 2500 371% 150% DRAG @5000 188% 180% DRAG @ 10000 304% 100% Note: Table made from bar graph.
The rear brake corner drag data shows the clearest evidence of reduced friction braking usage resulting in measureable consequences. Vehicle H rear hardware exhibits a drag increase between 188% and 436% of the new vehicle condition, whereas vehicle I ranges from 100% to 180%. This demonstrates the potential for rear brakes on regenerative braking vehicles to be subject to considerably more drag torque increase over time than non-regenerative braking vehicles, even with more robust corrosion protection strategies in place. It should be noted that drag data are not available on these exact brake corners in the new condition, however it is a fair assumption that new condition drag is comparable to the established averages from production validation.
2d. Effect of FNC Rotors on Corrosion of Braking and Mounting Surfaces
Often servicing brake rotors becomes difficult due to the corrosion growth that occurs between the rotor and hub mounting surfaces after prolonged exposure in a corrosive environment. The use of FNC treatment on the rotors provides a clear difference in this corrosion potential, as is evidenced by the comparison of vehicle B and vehicle D front rotors in figure set 7. While objective forces for removing the rotors was not directly measured, it can be observed in the post test photographs that the FNC rotors used on vehicle B exhibit considerably less corrosion growth in this interface following a 10 year accelerated exposure compared to the standard cast iron rotors used on vehicle D. This difference is observed despite the fact that both rotors used the same zinc rich protective top coating in this area. It can be expected that the removal forces of the FNC rotor will be considerably lower following prolonged exposure.
Additionally, differences can be clearly observed in the braking surfaces in these two rotors. Vehicle B FNC rotors exhibit small, localized spotting of red rust around the braking surface. The localized nature of rust formation indicates that the corrosion penetration into the rotor is limited. Comparatively, the vehicle D non-FNC rotors exhibits significant red rust formation and smearing, with additional pockets of uncleaned rust forming in the countersunk cross-drilled holes in the brake plate. Additionally, vehicle D brake plate surfaces have several pad "shadows" evident of prior corrosion adhesion of the pads to the rotor. The benefits of FNC treatment evident here on rotor brake plate surfaces relative to corrosion growth and pad to rotor corrosion adhesion are similar to other published investigations on these topics documented to date, and will not be explored in depth in this paper. The interested reader may find further information in the references [2, 5].
2e. Effect of Regenerative Braking on Rotor Surface Corrosion Cleaning (Vehicle G to I)
Within the subset of vehicles utilizing FNC treatment for rotors, we can still observe a significant difference in rotor brake plate cleaning between vehicles with standard powertrains, and those with strong hybrid powertrains and heavy usage of regenerative braking. Figure set 8 compares photographic evidence of front rotors from vehicle G (hybrid) and vehicle I (ICE) at end of 10 year accelerated corrosion testing, and figure set 9 compares the rear corners of these same vehicles. Visual differences are apparent in the front rotor surfaces, though differences on the rear are less visually distinct.
2f. Effect of Usage on Brake Hose Expansion
On one of the case study vehicles - vehicle B - the front and rear brake hoses were measured for expansion behavior after the durability test exposure. Measurements were acquired using an inertia-dynamometer based apply system, flow meter, and data acquisition system. Hose specimens were connected via hard pipe to the dynamometer hydraulic outlet port and to a steel 'bleeder block" on the caliper end of the hose. The fluid displacement of the system without the hose was measured and subtracted from the measurements with the hose to isolate the hose fluid consumption.
These data were compared to the performance of the same brake hose construction when new (for obvious reasons due to logistic complexity, before and after durability measurements on the exact same hose specimens were not made.
Figure 10. Brake Hose Expansion Increase Vehicle B % Increase in Brake Hose Expansion Experienced Pressure (kPa) Front Hose Avg Rear Hose Avg 1000 286% 311% 2800 80% 108% 6900 104% 130% 9000 78% 107% Note: Table made from bar graph. Figure 11. Brake Hose Expansion Increase @ 80 deg C versus ambient for with mileage (vehicle B) and for new hoses % Increase in Brake Hose Expansion @ 80 deg C versus Ambient Experienced Pressure (kPa) Front Hose Avg Rear Hose Avg NEW Hose 1000 31% 35% 200% 2800 71% 41% 129% 6900 85% 64% 200% 9000 96% 70% 155% Note: Table made from bar graph.
Some interesting behaviors are evidenced in figures 10 and 11. Figure 10, which is a comparison of the with mileage parts removed from vehicle B to new hose of the same construction - shows that both the front and the rear brake hose see a notable increase in fluid consumption (expansion) with use, ranging from about 80% at moderate to high pressures, and as much as 300% at lower pressures. The increase in fluid consumption at ambient is comparable to the degree of increase noted following a history of race track testing on a hose specimen, as seen in earlier research .
Although the net fluid consumption is higher on the durability-conditioned hoses relative to a new condition, the change in fluid consumption between the ambient and a high temperature (80 deg C) conditions shows the durability-conditioned hoses exhibited more stable behavior with temperature. While new hose specimens would experience up to a 200% increase in fluid consumption when the temperature was raised from ambient to 80 deg C, the mileage conditioned parts saw no more than a 96% increase.
2g. Different Effects of Corrosion Binding on Front and Rear Axle Hardware
One might intuitively expect that that the degree of corrosion and environmental exposure for the rear brakes can be notably more severe than for the front brakes. The rationale is simple - although one might expect that the front brake may be slightly more at risk to very heavy, but very short duration, exposure from direct splash, it is actually the lighter, but constant exposure from airborne debris that proves most damaging over time. The rear brakes are in the airborne "wake" of the front wheels more or less constantly, while the front brakes may see some respite from being in the wake of another (leading) vehicle when traffic is lighter and following distances greater. Upon examination some of the case study vehicles which have both conditions of having a comparable design solution and similar levels of corrosion protection front and rear, and having retrieved and measured a complete vehicle set of brakes, one may indeed see a difference between the front and rear axle brakes, with the front brakes being in generally better condition.
2h. Different Effects of Corrosion Binding on Aluminum vs. Cast Iron Calipers
Vehicle A (the luxury sedan) and vehicle I (the family sedan) had two vehicle sets each of front brake corners included in this study, and therefore, were able to offer 4 samples of each brake corner design for comparison. Vehicle A featured aluminum calipers with phosphate washed and painted brake pad backing plates, and Vehicle I featured zinc-nickel plated cast iron calipers with painted brake pad backing plates.
Of interest in this particular comparison is to see if the caliper material choice affected the degree of performance changes incurred by each front brake corner design subjected to the durability and corrosion exposure. In the aluminum caliper, dissimilar metals (aluminum in the caliper and steel in the pad backing plate) are in very close proximity to one another, separated only by the thin and wear-able paint layer, and by a thin oxide layer (from the aluminum anodization). With the build-up of road debris and poultice in this interface, conditions that support crevice corrosion can form, which can result in an acutely virulent corrosion attack [5, 6, 7]. It is understood - cautioned, even - that the two different vehicles most likely created different operating environments for the brakes, and some of the differences in performance may be related to this. With this thought in mind, however, the data in figures 15 (depicting fluid consumption) and 16 (depicting residual drag), may be consulted, after which additional discussion will ensue:
These data show some remarkable trends - in particular, the aluminum housing caliper showed very low fluid consumption for all but one sample, and all four samples showed much higher drag. Both behaviors in the brake corners from vehicle A are symptomatic of a high level of pad to housing slide force, which was indeed confirmed on the parts when they were first removed. Under the influence of this high slide force (due to the loss of pad to housing gap due to corrosion growth of the pad backing plates), the pads most likely would gently bend, rather than slide, for most applies, until enough pad wear and/or a hard enough apply occurred to slide the pads relative to the housing and then restrain them in a new, partially applied condition. This effectively takes up much of the "running clearance" of the caliper, which lowers the fluid consumption, and increases the residual drag considerably.
In comparison, the brake corner samples from vehicle I show higher-than new condition fluid consumption at all pressure levels reported, which is consistent with the normal degradation due to taper wear, cupping wear under the pistons, and doming wear under the fingers. The fact the residual drag was generally below the new condition limit (which remains proprietary, but will be acknowledged as a difficult spec to meet) shows that the pads were sliding relatively freely during this test.
Returning to the discussion of different environments for the brake corners for the two different vehicles, the authors point out now that the calipers on vehicle A were mounted behind the front axle centerline, which adds considerable protection of the caliper moving parts from the suspension and steering knuckle. In contrast, the calipers in vehicle I were mounted ahead of the front axle, where it is very difficult to 'seal' the caliper bracket to splash shield gap, and there is often a direct debris flow path around the front wheel and tire, around the caliper housing, and against the pad to bracket interface. Furthermore, the basic design of the aluminum opposed-piston caliper creates some "shielding" of the inner pad to housing interface from direct debris flow, whereas the sliding caliper is by necessity open in this area. Acknowledging the apparent advantages offered by vehicle A over vehicle I, yet noting that vehicle I was less affected by the long-term durability and corrosion exposure, strongly suggests that the dissimilar metals in the pad backing plate to caliper interface in vehicle A played a role in the realized performance.
2i. Comparison of Carbon Ceramic to Traditional Cast Iron Rotors in a High Performance System
While one would not expect many customers with vehicles equipped with carbon ceramic rotors to subject their vehicles to prolonged bouts of exposure in poor weather, it is interesting to evaluate nonetheless as individual customer behavior is frequently unpredictable. In a direct comparison of cosmetic corrosion performance between high performance vehicle D with cast iron rotors and high performance vehicle E with carbon ceramic rotors, the carbon ceramic rotor braking surface appearance is far superior to the cast iron at end of 10 year accelerated corrosion, as one would intuitively expect (reference figure 17). Interestingly, however, we do not see any evident difference in brake torque variation following corrosion exposure - see figure 18.
It is not entirely clear why the brake torque variation on the Carbon Ceramic Matrix (CCM) discs were so degraded over the new condition. However, both rear calipers on vehicle E showed symptoms of a significant increase in pad slide force due to corrosion, and the two-piece front discs on vehicle C showed a significant increase in lateral runout (see section 2k). It is plausible that with fixed calipers, very high pad slide forces, and high lateral runout that brake torque variation could result from this, even if the rotor TV was low. The rough finish and extensive cross drilling of the discs made measurement of LRO and TV too difficult to contemplate at the time. In retrospect that would have been a very useful measurement and would be a worthy activity to attempt in future studies.
2j. Assessment of Cosmetic Corrosion Performance (Various Typical Brake Caliper Protection Methodologies)
Visual analysis of the end of test parts via photographic evidence allows for the cosmetic corrosion performance of some of the common corrosion protection options in the industry to be compared. Consider the series of photographs in figures 19, 20, 21, 22 in the ensuing discussion. The authors acknowledge that the angles of the photographs may not show the most cosmetically relevant surfaces of the calipers, however, this angle was selected to avoid markings and features that would compromise the propriety of their applications.
Comparing the zinc-plated to the zinc-nickel plated caliper and support bracket in figure 19 and 20, respectively, one can note a striking difference. The zinc plated caliper is perhaps 80-90% covered in red oxidation, indicating full degradation of the sealer, full depletion of the zinc layer below, and fully developed surface oxidation of the base metal. The zinc-nickel plated caliper shows extensive "white" oxide, indicating degradation of the sealer and some depletion of the plating, but very little red oxide. There is perhaps a 10% coverage of the caliper in red-oxide.
Both aluminum calipers (the anodized and the painted) caliper show extensive build-up of road debris on the caliper surface, however, there is very little evidence of damage to the base metal. It is notable to contrast the condition of the caliper housings in both cases to the condition of the pad backing plates; there is evidence of extensive oxidation of the brake pad backing plates (which had been washed and painted, but not treated with any additional corrosion countermeasures).
2k. Effect of Long Term Usage on Rotor Lateral Runout and Thickness Variation
The lateral runout and thickness variation of the as-received brake rotors were measured on some vehicles. Logistics challenges prevented the measuring of discs on-vehicle prior to test, however, standard practice was to control the lateral runout of the brake disc and the wheel bearing assembly to less than 40 microns total for all vehicles (with a slightly higher limit for vehicles with 2-piece rotors). The post-test measurements for which both LRO and TV measurements are shown were taken on rotors only - the discs were removed from the brake corner and measured on a polar-coordinate Coordinate Measurement Machine (CMM). Consequently, the measurements were all relative to the rotor to bearing mounting face datum. Minimal clean-up was allowed to insure a good measurement (e.g. loose oxide scale and debris were removed). The measurements that show LRO only were taken by dial indicator on the dynamometer fixture, and do therefore include the wheel bearing.
Figure 23 summarizes the available measurements of rotor lateral runout (LRO) and thickness variation (TV):
Review of figure 23 points to some notable observations. Generally, most rotors showed a significant increase in LRO over what they would have been built with. Thickness variation levels were generally above the thresholds over which complaints for roughness and NVH are typically generated. Contrasting vehicle G with vehicle D is an interesting extreme, with the hybrid, NAO lining vehicle G representing the lowest possible tendency for disc wear, and vehicle D with a conventional powertrain and low-metallic linings represents the highest possible tendency for disc wear. Under the extreme conditions for corrosion and debris exposure the test vehicles were subjected to, Vehicle D was better able to keep the TV low on the front axle (at 13 and 9 microns for LF and RF, respectively) versus vehicle G (at 30 and 23 microns for LF and RF, respectively).
Vehicle C - the only one of this set with two-piece front rotors -exhibited LRO in the range of a full order of magnitude over the level they would have been built with. One possible explanation could be that the two piece discs are equipped with a floating brake plate section, which slides in close-fitting slots relative to the hub, and are guided by springs and pins. During braking, this interface moves constantly to absorb the deflection of the disc and components in the load path of braking forces, and the spring system returns the floating plate back to its original position when the brake is released. If the ability of the brake plate to slide relative to the hub were impeded in any way - such as could plausibly occur with debris and oxide buildup - then the return-ability of the floating plate would also become impeded. In this state, the risk of the disc failing to return completing when the brake is released would increase, and the failure mode, should it occur, would be a substantial increase in LRO.
2l.5 year vs. 10 year corrosion exposure
Only one of the vehicles (vehicle F) included in this study had completed the equivalent of 5 years of corrosion exposure; all of the others experienced conditions designed to replicate 10 years. Perhaps the most relevant comparison would be to vehicle G (the hybrid sedan equipped with zinc-plated cast iron sliding calipers). Due to the many differences in caliper design (twin piston vs. single piston), in vehicle (truck vs. car), and in duty cycle (regen brakes vs. conventional), the only comparison that will be made here is cosmetic. Figure set 24 shows the most "cosmetic" region of the caliper (the customer-facing backside of the caliper fingers, and the outer tie bar) for vehicle F on the left and vehicle G on the right. Figure set 25 compares the front brake pad abutment area, and figure set 26 compares the brake rotor inner surfaces.
It comes as no great surprise or discovery that vehicle G, with 10 years of exposure, shows degradation in all surfaces depicted versus vehicle F with 5 years equivalent exposure. What is most notable, perhaps, is that the cosmetic condition of parts for vehicle F is decent, with little in the way of red oxide on the cosmetic surfaces (although as presented previously in section 2b, some performance degradation has already started to occur. This fits the understanding of how most corrosion protection options operate - the zinc plating on the caliper and bracket, and the zinc-rich coating on the rotor serving as a sacrificial finish with red rust setting in only after its depleting. The paint on the pad backing plate more or less prevents oxidation until it is removed by wear, or loses adhesion and delaminates. The point to be most wary of is the degree of degradation that occurs between 5 and 10 years exposure is considerably more than the degree of degradation between 0 and 5 years. Given that 5 years is beyond most vehicle manufacturers' published warranty periods, it is reasonable to assume that warranty data will generally not show cosmetic corrosion issues for brakes, save for the most severe of cases. That being said, it is equally reasonable to note that customers' dislike of red oxide corrosion on cosmetic surfaces does not necessary stop as soon as a warranty period expires, so for components that are intended to last the life of the vehicle (as calipers are) there has been an increasing trend towards finishes that are more effective at resisting corrosion .
3. TEST CASE: CORROSION INDUCED FUEL ECONOMY DEGRADATION THROUGH VEHICLE LIFE
A prior publication by Antanaitis, et. al.,  has calculated the relationship between brake drag and resultant fuel economy. Expanding upon this work, the same B-segment vehicle with composite fuel economy meeting 2020 regulations (41.7 mpg) will be evaluated for a case study, and the calculated value of 0.17 mpg per 1 Nm of drag will be maintained. The intent of this case study is not to relate corrosion induced brake drag to fuel economy certification in any way, rather to evaluate the potential cost to the customer over the life of the vehicle assuming vehicle exposure to a 95th percentile corrosion environment.
Using the published average distance driven of 13,476 miles per year for an American driver , the fuel consumed by our test case vehicle would be approximately 323.2 gallons per year, before any accommodation is made for increasing brake drag over time. Now, using some basic assumptions, we will calculate the differential impact to fuel economy over time had this example vehicle been equipped with marginally robust corrosion protection countermeasures similar to vehicle G, rather than highly robust corrosion protection countermeasures similar to vehicle H. We shall also assume the nominal residual drag per caliper in this example vehicle is 0.8 Nm in the new vehicle condition.
Based on the data collected on vehicle G brake corner components after a 10 year equivalent corrosion exposure, we can calculate a nominal increase in brake drag of 600% front and 2100% rear assuming the 5,000 kPa experience pressure condition as a nominal. This equates to an increase of 4.8 Nm per front caliper, 16.8 Nm per rear caliper. Total vehicle drag thus becomes 46.4 Nm per car, an increase of 43.2 Nm from new condition. It stands to reason this level of drag can be expected to increase in a relatively linear fashion throughout the vehicle life as corrosion and wear build. Thus, vehicle fuel economy for the customer in a 95th percentile corrosion environment can be calculated to decrease from 41.7 to 34.4 mpg over a 10 year period, simply due to the influence of brake drag. This calculates to a total additional fuel consumption of 345.4 gallons and a total cost to the customer of $863 over 10 years, using a nominal fuel price of $2.50 per gallon. This in turn represents approximately 6900 pounds of additional C[O.sub.2] emissions during this timeframe.
While the test case represents the 95th percentile customer usage distribution for corrosion exposure, it should be noted that the measured drag increases are still performed at the brake caliper and rotor subassembly level. It is possible, had the opportunity to generate data including the effects of increased rotor runout, bearing deflection, etc. at end of life, that the drag increases become more profound. The point remains clear - brake design decisions for corrosion protection can result in significant vehicle fuel economy degradation over time and measureable cost to vehicle owner. What may be relatively low cost improvements at the piece cost level, such as plating strategies for components, can result in significant long term vehicle performance improvements.
4. TEST CASE: CORROSION INDUCED PEDAL FEEL DEGRADATION THROUGH VEHICLE LIFE
Logistics - proving once again to be a formidable opponent in a testing situation where many different stakeholders have an interest in the vehicles after test - prevented fully instrumented brake system tests at the conclusion of the durability tests. However, complete brake corner sets were retrieved and tested for some vehicles, including vehicle I on which this case study example is based. Using the brake corner dynamometer data, and calculation tools which have been proven to accurately reproduce vehicle level brake pedal feel behavior in the past , the pedal feel curves were reproduced by calculation based on the measured dynamometer data (the new condition was represented by validation data of the brake corners, and pedal feel curves were calculated by the same method). Figure set 28 shows the deceleration versus pedal force, and deceleration vs. pedal travel curves, for vehicle I:
A modest degradation in the deceleration vs. pedal force characteristic was noted (in fact, the approximately 2N difference calculated is well within normal vehicle to vehicle variation). This is reflective of the apparent friction levels remaining reasonably close to the new condition (and it is perhaps salient to remind the reader that vehicle I was equipped with ferritic nitro-carburized, or FNC rotors). The level of pedal travel degradation was much more pronounced, with a moderate-deceleration difference of approximately 8mm noted. This difference is at a level that even average drivers would tend to notice - if offered the ability to compare side to side. With much of this degradation likely occurring very slowly over time, it is plausible that most drivers would adjust to it as it occurred.
After reviewing the data summarized in this study, it is easy to appreciate what a demanding environment the brake corner must operate in. Intensely corrosive condition compounded by high debris flow are part of the normal spectrum of customer usage (albeit, for the samples in this study, towards the severe end of this spectrum). Behaviors such as apparent friction loss, changes in fluid consumption, increases in residual drag, and increase in brake torque variation were noted across many different brake corner designs on many different vehicles.
The with-mileage (post durability test exposure) brake corner data collected on the 11 vehicles included in this study provided opportunity through curated comparisons to see some of the effects of vehicle configuration, of caliper type and material, of caliper finish, and of brake pad and rotor type on cosmetic and performance behavior at higher mileage.
Some of the more notable observations are summarized below:
* Some loss of apparent friction was observed, with no real difference between NAO and low-met linings. A high percentage of the NAO lining variants were fitted with FNC brake rotors.
* The addition of zinc-nickel plating on the calipers and brackets, and zinc-nickel plating on brake pads, substantially reduced the degradation of fluid consumption and of residual drag torque after 10 years equivalent corrosion exposure.
* The effect of regenerative braking on long term corrosion behavior of the brake discs could be observed in one comparison (corrosion was more severe on rotors from the regen-equipped vehicle)
* Increases in brake hose expansion of up to 30% at ambient temperature were noted in brake hoses with mileage, versus the new condition.
* A difference in corrosion severity on the front versus the rear axle could be noted on one of the vehicles for which all four brake corners were tested, which may illustrate the effect of the rear brakes always being in the debris wake of the front wheels.
* Aluminum calipers with steel brake pad backing plates appeared to be more susceptible to corrosion in the padcaliper interface. For these applications, improvements in brake pad backing plate corrosion protection (such as zinc-nickel plating or FNC treatment before paint) may be especially beneficial.
* As intuitively expected, the brake plate portion of carbon-ceramic discs proved to be virtually impervious to cosmetic corrosion-related degradation. However, for reasons not entirely clear, both iron and CCM discs showed relatively high brake torque variation after durability testing (this needs more study).
* Most rotors measured for LRO and TV after mileage showed substantial increases in both relative to the new condition. The pair of two-piece rotor samples measured were almost a full order of magnitude higher in LRO than would have been allowed when they were new.
* The one comparison of 5 year vs. 10 year equivalent corrosion exposure that was possible showed - as would be expected from an understanding of the physics of how the protective finishes work - that the progression of corrosion damage was somewhat exponential with time. A much higher degree of degradation occurred between 5 and 10 years of exposure, than between new and 5 years.
For as much time and effort that went into the present study, many questions remain. Corrosion and debris-related degradation over time remain one of the biggest enemies of the foundation brake system, which exists in one of the most severe environments of the vehicle.
The ability to accelerate corrosion remains a tricky science still, and an imprecise one even at a vehicle level. There remains huge opportunities for improvement in the accurate acceleration of corrosion at a vehicle level. Lab-based corrosion testing at a component level takes on even more assumptions and risks against accuracy yet.
It should not be overlooked that mechanisms for feedback from customers in the field become significantly less available after the warranty period. However, understanding customer expectations -especially beyond the warranty period - remains critical to vehicle manufacturers' long term competitiveness and success. It may be the case that in the impending era of connected vehicles, and greatly enhanced data collection and prognostics capability, that these expectations will be understood better in the future (including customer expectations for brake service).
Specific mention will be made again of the somewhat eye-opening brake torque variation behavior of the rear CCM disc brake corners of vehicle D. Clearly the effect of environment and use over time on the CCM discs needs further research. There is no strong reason to suspect that the material itself suffered any more degradation than was expected (which would have been very little), but the effects of mileage on two-piece disc LRO and on caliper slide behavior creates a particularly complex brake corner integration to consider, with failure modes that are infrequent on other configurations that can partly if not mostly undo some of the benefits.
(1.) Antanaitis, D., Riefe, M., and Sanford, J., "Automotive Brake Hose Fluid Consumption Characteristics and Its Effects on Brake System Pedal Feel," SAE Int. J. Passeng. Cars - Mech. Syst. 3(1):113-130, 2010, doi:10.4271/2010-01-0082.
(2.) Robere, M., "Disc Brake Pad Corrosion Adhesion: Test-to-Field Issue Correlation, and Exploration of Friction Physical Properties Influence to Adhesion Break-Away Force," SAE Technical Paper 2016-01-1926, 2016, doi:10.4271/2016-01-1926.
(3.) Antanaitis / SAE Int. J. Passeng. Cars - Mech. Syst. / Volume 9, Issue 3 (September 2016)
(4.) U.S. Department of Transportation, Federal Highway Administration. https://www.fhwa.dot.gov/ohim/onh00/bar8.htm
(5.) Holly, M., DeVoe, L., and Webster, J., "Ferritic Nitrocarburized Brake Rotors," SAE Technical Paper 2011-01-0567, 2011, doi:10.4271/2011-01-0567.
(6.) Galvele, J. R., "Tafel's law in pitting corrosion and crevice corrosion susceptibility", Corrosion Science 47 pp 3053-3067, 2005.
(7.) Abdulsalam, M. I., "Behavior of crevice corrosion in iron", Corrosion Science 47 pp 1336-1351, 2005.
(8.) Nazari, M. H, Bergner, D, and Shi, X., "Manual of Best Practices for the Prevention of Corrosion on Vehicles and Equipment used by Transportation Agencies for Snow and Ice Removal", Minnesota Department of Transportation, 2015.
(9.) Johnston, M., Leonard, E., Monsere, P., and Riefe, M., "Vehicle Brake Performance Assessment Using Subsystem Testing and Modeling," SAE Technical Paper 2005-01-0791, 2005, doi:10.4271/2005-01-0791.
(10.) Dingwerth, B., "Advanced Finishes for Brake Components and Other Castings," SAE Technical Paper 2016-01-1951, 2016, doi:10.4271/2016-01-1951.
The authors gratefully acknowledge the contributions of Mr. Robert Gauthier of the General Motors Brake Test Labs, for his enthusiastic support in developing new test set-ups and running experiments.
BTV - Brake Torque Variation
CCM - Carbon Ceramic Matrix (rotor)
FNC - Ferritic Nitrocarburization
ICE - Internal Combustion Engine
LRO - Lateral Run-Out (of a brake rotor)
NAO - Non-Asbestos Organic
TV - Thickness Variation (of a brake rotor)
Table A.1. Summary of Vehicles and Key Brake Specifications that were included in this study Vehicle Identifier Vehicle Type Front Caliper Rear Caliper A Luxury Four piston opposed, N/A Sedan aluminum body Four piston opposed, 4 piston opposed, B Performance aluminum body aluminum body Coupe (anodized) (anodized) Six piston opposed, C Performance aluminum body N/A Coupe (painted) Six piston opposed, Four piston opposed, D Performance aluminum body aluminum body Coupe (painted) (painted) Six piston opposed, Four piston opposed, E Performance aluminum body aluminum body Coupe (painted) (painted) Twin piston sliding, F Truck cast iron body N/A (Zinc+sealer) Hybrid Single piston sliding, Single piston sliding G Sedan cast iron body (Zinc + IPB, cast iron body sealer) (Zinc +sealer) Single piston sliding Single piston sliding, MOC, aluminum H Hybrid cast iron body (Zn- body, cast iron Sedan Ni+sealer) bracket (Zn-Ni +sealer) Single piston sliding, Single piston, sliding Family cast iron body MOC, cast iron body I Sedan (Zn-Ni+sealer) (Zn-Ni+sealer) Vehicle Front Rotor Rear Rotor Identifier Friction Material Type Type NAO, Painted after Cast Iron with A Phosphate wash FNC N/A Low Met, Painted Cast Iron with FNC N/A B Low Met, Painted Cast Iron Cast Iron C Low Met, Painted Cast Iron N/A 2 piece D Low Met, Painted Cast Iron Cast Iron Carbon Carbon E Low Met, Painted Ceramic Ceramic 2 piece 2 piece F NAO, Painted after Cast Iron Cast Iron Phosphate wash NAO, Painted after Phosphate wash Cast Iron with Cast Iron with G (frt); NAO painted FNC FNC (rear) NAO, Painted (frt); Cast Iron with Cast Iron with H NAOZn-Ni+Paint Cast Iron With Cast Iron With (rear) FNC FNC NAO, painted (frt); Cast Iron with Cast Iron with NAO Zn-Ni+paint FNC FNC I (rear)
David B. Antanaitis and Matthew Robere
General Motors LLC
|Printer friendly Cite/link Email Feedback|
|Author:||Antanaitis, David B.; Robere, Matthew|
|Publication:||SAE International Journal of Passenger Cars - Mechanical Systems|
|Date:||Oct 1, 2017|
|Previous Article:||A Study of the Disc Scoring Generation Principle and Reduction.|
|Next Article:||Study of Difference in Friction Behavior of Brake Disc Rotor with Various Surface Textures during Running-In by Using Simple Model.|