2014-01-2521 Influence of Test Procedure on Friction Behavior and its Repeatability in Dynamometer Brake Performance Testing
Copyright © 2014 SAE International
Abstract
The efforts of the ISO “Test Variability Task Force” have been aimed at improving the understanding and at reducing brake dynamometer test variability during performance testing. In addition, dynamometer test results have been compared and correlated to vehicle testing.
Even though there is already a vast amount of anecdotal evidence confirming the fact that different procedures generate different friction coefficients on the same brake corner, the availability of supporting data to the industry has been elusive up to this point. To overcome this issue, this paper focus on asssing friction levels, friction coefficient nsitivity, and repeatability under ECE, GB, ISO, JASO, and SAE laboratory friction evaluation tests. With multiple companies (or programs) developing and asssing the friction coefficient and friction behavior under different methods, it is inevitable to avoid conflicts of performance requirements or lack of reproducibility or correlation of test results under different test methods.
In order to provide an evaluation consistent with previous phas of the Task Force activities, the same brake corner asmbly and same friction material is ud for this study. The study is comprid of three main steps:
1.Conducting tests under veral test procedures.
2.Asssing and comparing friction levels, in-stop friction
behaviour, repeatability, as well as friction coefficient
nsitivity to the test conditions, including the friction-
couple thermal history.
3.Prenting benefits and limitations of each procedure as-
written along with a simplified comparison of the test
quences and its main test conditions.
Similar to previous phas of the project, the study us statistical tools for the multidimensional co
消防安全教育教案mparison. Introduction
Laboratory testing to determine the friction coefficient of a friction couple (friction material and its mating rotor or drum) or a given friction material on a standard mating surface is by far the most common test procedure conducted by the automotive brake industry. Therefore, understanding the similarities and differences among them is crucial to the entire supply-chain.
For this pha of the project, all dynamometer tests were conducted using the same dynamometer to test under repeatability conditions and reduce the test system contribution to test variability. The non-dyno test methods, using coupons extracted from the friction material ud specific test systems as indicated in the test procedure.
Project history
This pha of the ISO Test Variability task-force is the fifth project which spans five years of continuous work. The friction coefficient measurement and evaluation is a critical metric for the entire brake supply-chain to support key decisions such as:
∙formulation design, raw materials lection, and
manufacturing process conditions
∙lection of friction material supplier along with the specific manufacturing site(s), and
∙boundary conditions on the friction material which will determine other brake behavior such as NVH,
durability, structural integrity, and physical properties.
Over time and as the result of the interaction of regional, business, and technical conditions, there is a large number of test procedures (including regulations, standards, and corporate or proprietary), multiple test systems designs, vintages, and providers, and multiple foundation brake types, designs, and sizes to suit different applications and needs.
Since part-to-part, test-to-test, dyno-to-dyno, and lab-to-lab variability are interconnected, the brake engineering community has been asking itlf about the best method to asss the expected level of test variation, how to better quantify it using industry-approved methods and tools, and how to reduce it in a systematic and economical way [1]. Thus far, the ISO task force has completed four phas:
∙Pha 1: accuracy study involving one method, five dynamometers, and two tests on each to quanti
fy
Page 1 of 15
Page 2 of 15 repeatability, reproducibility and to detect potential
sources of variation [2] using ISO statistical methods [3], and industry-accepted software tools and methods [4]. ∙
Pha 2: Design of Experiments to asss main effects of critical factors bad on the sources of variability and the deviations from reproducibility conditions by some of the laboratories during the pha 1 of the project [5]. ∙
Pha 3: documenting findings and potential
enhancements to international test standards using the ISO 26867 method as an example [6]. Some elements and findings from this pha were already implemented during the revision of the SAE J2521 noi squeal test procedure [7].
∙
Pha 4: Vehicle-to-dyno correlation using a common vehicle-bad test procedure. This pha also ud a DOE approach and exhibited better repeatability and correlation on the front brakes v. the test brakes [8].
Due to the logistics, and the engineering resources associated with this project, only one brake design and size and one friction material have been ud for all the phas of the project. Hence, the reader needs to exerci caution when applying the finding of this project on other testing activities, especially when using numerical ranges or numerical values
indicated on the different papers previously published.
工业制硝酸Secondly, the industry is challenged to develope and allocate
resources to conduct similar projects using other test
procedures, different friction materials, and testing on different brake designs.
Test system lection, test procedures, and test parts
In order to characterize the friction coefficient levels,
repeatability and correspondence to other methods, the test procedures lected reflects the following categories (a) full inertia-dynamometer using vehicle-specific test conditions and test hardware, and deceleration-type of braking during the test (b) inertia-dynamometer using vehicle-specific test hardware and brake applications at constant speed during the test(s), and (c) special-purpo test systems at constant speed
rubbing the sample (coupon) against a drum or a disc of known and standard configuration and specifications.
Test system lection
Different test procedures u a different layout, different type of sample (out-of-the-box pads, single coupon, or double coupon) and mating component (vehicle brake disc, standard drum, or standard disc) as a function of the type of test. See Figure 1. The figure also depicts an illustration of the actual test system. Zimmer [9] provides a valuable summary and a brief history of the evolution of test systems covering Dynamometer, Cha Link, FAST, Ranzi Cuna, Ranzi LCR, and the Friction test machine from ATE.
The inertia-dynamometer ud for the dynamic braking tests (using corner asmblies) was lected in order to ensure a broad range of testing capabilities and able to accommodate all the test conditions required by the different test methods ud for the project. The main specifications of the inertia-dynamometer ud are the following: ∙ Test inertia (prior to inertia simulation): 200 kg·m² ∙ Torque: 5600 N·mcorel
∙ Brake pressure: 21 0000 kPa ∙ Rotational speed: 2000 RPM ∙ Brake temperature: 1300°C
∙
Full environmental control for cooling air temperature and relative humidity at the brake
For the coupon test procedures (SAE J661 and GB5763), the specific procedure determines the features and full-scales required.
Inertia-dynamometer
Special-purpo test system
SAE J2521, J2522,
and J2784 ECE R90-A3, -A9 ISO 26867, JASO C406
SAE J661 GB 5763
Figure 1. Test system, specimen type, and number of specimens as a function of the test procedure. Inertia-dynamometer tests u two pads (inner and outer) per test. The orange shapes indicate the specimen type and location. The orange arrows indicate the location of the specimen relative to the mating surface
Test procedure and test specimen lection
The process to lect the test procedures to conduct as part of this pha of the project was straightforward. As the objective was to provide an evaluation of friction coefficient under
different test methods around the world per publicly-available procedures, a arch of the different international standards commonly ud (or mandated as part of a given regulation or requirement) yielded a list comprid of three groups: (a) inertia-dynamometer tests with deceleration braking,
(b) inertia-dynamometer test with drag braking, and (c) coupon testing also using drag braking.
Inertia dynamometer test procedures using deceleration braking
Different laboratory friction material characterization methods have been developed over the last decades worldwide with commonalties and also with differences which do not allow a simple and straightforward comparison to other methods. Being the most common laboratory test for friction coefficient
characterization, the following international standards using a full brake corner asmbly and vehicle-specific test conditions were ud. Depending upon the test procedure, two, three, or ten repeats were performed to quantify test repeatability.
ISO 26867: 2009 – Road Vehicles – Friction Behaviour Asssment for Automotive Brake Systems [10]. This test method was developed as part of the industry-wide effort culminated on the ISO 15484 [11] for quality assurance of disc and drum friction materials for pasnger cars and commercial vehicles. The ISO friction behavior test is a cost-effective dynamometer test to support development, benchmarking, and product monitoring which also provides the key metrics indicated on the ECE R90 annex 9 [12] for nominal coefficient of friction, minimum coefficient of friction, and maximum coefficient of friction as part of the Conformity of Production process. This procedure was conducted on two samples.
SAE 2522:2013 Dynamometer Global Brake Effectiveness Procedure [13]; developed originally in Europe as the AK Master in the late 1990s by the AK Working Group. It has become a common test procedure to determine the nsitivity of the friction material to changes in braking speed, brake pressure or deceleration, initial brake temperature, and thermal history. It is also the basis for veral vehicle manufacturers, tier-1 brake suppliers, and tier-2 friction manufacturers. This procedure was conducted on two samples.
SAE 2521:2013 Disc and Drum Brake Dynamometer Squeal Noi Test Procedure. This test procedure was originally developed by the European Working Group as the AK Noi. In addition to
becoming the de-facto standard test procedure to asss the noi propensity of brake asmblies above a threshold of 70 dB(A) in the 1.25 kHz to 16 kHz range; it includes veral friction characterization ctions uful in asssing the friction level under constant pressure along the test quence. All squeal noi matrixes above 5 bar also report the average friction coefficient. Becau of the significant differences with other test methods, ctions using brake drag applications were run to ensure proper lining conditioning and brake work history, but are not included in the comparisons for friction coefficient. This procedure was conducted on two samples.
SAE 2784: 2009 FMVSS Inertia Dynamometer Test Procedure for Vehicles Below 4540 GVWR [14]. Developed also as part of the industry discussions for the ISO 15484, this single-ended inertia dynamometer procedure replicates the vehicle test quence using vehicle loading, inertia levels, and deceleration levels mandated by the actual vehicle test quence or typically obrved on the test track. The SAE
J2784 also includes the test quence mandated by the FMVSS 105 for vehicles between 3500 kg and 4540 kg of Gross Vehicle Weight Rating. Test results from this this procedure, when combined with the appropriate vehicle information and reference torque output from the non-tested axle, can be ud to predict vehicle performance related to stopping distance. This procedure was conducted
on two samples.
JASO 406:2000 Pasnger car – Braking device –Dynamometer test procedures [15]. Developed following the main quence of the FMVSS 105 vehicle test, it allows the quantification of torque output and friction coefficient for pasnger cars under normal and failed conditions, at different load levels, and includes a water baline and recovery ction at the end of the test. The braking speeds during the test are a function of the maximum vehicle speed. This test method allows three possible test tups (a) single-ended using a single brake corner asmbly, (b) front-to-rear dual-ended, or (c) right-to-left dual-ended. This project ud single-ended testing to make it comparable to the other inertia-dynamometer tests. In addition, and different from the other dynamometer tests, the JASO C406 us the lining temperature (fixed side) to control the initial brake temperature. This procedure was conducted on two samples.
ECE 90-02:2013 Uniform Provisions concerning the approval of replacement brake lining asmblies, drum brake lining and discs and rums for power-driven vehicles and their trailers – Annex 3: item 2.2. This item on the European regulation establishes a uniform method to asss friction coefficient during cold effectiveness testing and speed nsitivity using single-ended inertia-dynamometer testing. The latest revision of the dynamometer test incorporates a burnish quence
using the structure and logic of the SAE J2522 and the ISO 26867 test procedures. This procedure was conducted on two samples.
Inertia dynamometer test procedures using drag braking
ECE 90-02:2013 Uniform Provisions concerning the approval of replacement brake lining asmblies, drum brake lining and discs and rums for power-driven vehicles and their trailers – Annex 9; Part A; item 2.2. The regulation, which is bad on the VDA 285-1 specification [16] us torque control and pressure control. The origins of this method dates back to the late 1960s as a development from Alfred Teves as the ATE Friction Test Machine using a 90 minute test cycle. It is customary to conduct torque controlled brake applications for new brake systems, while pressure controlled brake applications are rerved for vehicles manufactured prior to 1996. This project ud the latter to compare to pressure controlled braking on the SAE J2522 and the ISO 26867. This project conducted three repetitions of the test. This procedure was conducted on three samples.
Scale (coupon) testing using drag braking
SAE J661:2013 Brake Lining Quality Control Test Procedure [17]. Developed into a SAE Recommended Practice in the mid-60s, this procedure us a (25.4 x 25.4) mm coupon rubbing at a
constant speed and constant normal force against a cast iron drum which rembles a typical brake drum casting. In order to achieve the target temperatures and times during the test, the system relies on a blower-heating system. For the brake size ud on this project, the test conditions equivalent to the vehicle equated to moderate braking speeds and brake pressures during city traffic. This procedure is currently ud mainly to support quality control activities and for formal certification for friction materials for a nominal friction level at normal and hot conditions per the SAE J866 [16], not for product development purpos or to predict vehicle-level behavior. It also supports the product registration to demonstrate compliance to the heavy metals limits enacted by the states of California [18] and Washington [19] in the
Page 3 of 15
Page 4 of 15 United States and effective January 1st 2014 and January 1st
2015, respectively. During this project six tests total were
conducted; three on the inner pad and three on the outer pad. It is interesting to note the fact that both ts of data produced different edge codes at elevated temperatures.
GB 5763:2008 Brake Lining for Automobiles; item 5.3 [20]. Developed in the late 1990s following the JIS 4XXX [21], this test procedure runs using two (25 x 25) mm samples running in parallel along the same rubbing track on a standard disc and with the test load applied by dead weights. The
GB5763specification establishes both, friction level and friction level variability as well as wear rates. In order to quantify the inner and the outer pad behavior parately, coupons from fives pads were conducted on the inner and five pads on the outer pads.
形容老了的句子
Deceleration v. drag braking
Different test procedures u different methods to control the brake during the friction coefficient measurement cycle. Except for the ECE R90 drag friction tests (and the drag modules of the SAE J2521), all dynamometer tests u a deceleration cycle where the torque output from the brake decelerates the inertia discs and motor asmbly. Table 1 illustrates the
method to control speed for the different procedures. It is worth noting the fact the SAE J2521 us
drag speeds of 3km/h and 10 km/h, oppod to the ECE R90-A6, GB5763, and the SAE J661 which relies on dragging speeds equivalent to
approximately 70 km/h (for the brake and tire size on this project) on the vehicle to generate the same linear speed at the friction couple interface.
Table 1. Speed control by test procedure
Test procedure
Speed control during braking
Selection of key procedure ctions for friction coefficient comparison
In order to analyze and compare test procedures veral key ctions were defined. Friction material is nsitive to load history or quence of modules. Even an identical module in a different position
will lead to different friction behaviour. Figure 2 shows the key ctions in the respective test procedure quence.
Standard inertia-dynamometer test procedures conduct a large number of different braking conditions to increa the
efficiency of the test and with repetition of certain modules to asss the effects of load history.
Specific differences between procedures and ctions
Some of the specific differences between test procedures and their ctions are: ∙ Overall kinetic energy dissipated and test duration ∙ Overall arrangement of ctions
∙ Inertia level and inertia changes (during SAE J2784 and JASO C406) to reflect different loading conditions ∙ Evaluation of brake applications and extraction of key figures from test procedures
∙
blame用法
Burnish ctions with a variety of: number of brake
applications, constant versus changing energy and power, initial brake temperature with some tests (like ISO 26867) at elevated temperatures to save test time, and
intentionally incomplete burnish to monitor the evolution (like SAE 2521)
∙ Temperatures during fade modules: coupon at or around 350°C; full scale dynamometer test at or around 550°C ∙
Test procedures with predominantly lower pressures感恩妈妈的话
show higher friction levels (SAE 2521; SAE 661; GB 5783)
Table2 2 (a) to (c) summarize test t-up, test conditions and other important information for all procedures.
Table 2. Test t-up test conditions, and non-friction metrics for each test procedure
(a)
SAE J2521SAE J2522SAE J2784ISO 26867JASO C406ECE R90-9ECE R90-3SAE J661GB5763
(b)
(c)
The time-domain depiction of the main t-points provides a better understanding of the differences among test procedures. It also highlights the challenge it impos when trying to asss similarities or differences amongst test procedures. See Figure 3. Figure 3. Comparison of time-bad levels for braking conditions under each procedure — (a) Braking speed, (b) initial rotor temperature, and (c) brake deceleration or equivalent deceleration for the torque output Test specimens and hardware
Upon agreeing on the test procedures to perform, the project team also agreed on conducting all the tests (or extracting the coupons) using the same brake configuration and friction material ud durin
g previous phas of the ISO Test Variability task force projects. Test inertia levels, test hardware, and test conditions met the OE requirements and corresponding product specifications. As in the previous phas of the project [1,2,3], the tup was a 15 inch front brake with sliding caliper, grey iron vented rotors, and low steel brake pads.
Brake rotors were from the same batch and were lected for lowest variability regarding LRO, DTV, and first natural frequency. Low steel brake pads were from ries production, the inner and outer brake pads came respectively from the same production batch, and all pads were lected for lowest between-sample compressibility variation. Rotors and pads were prepared following the ttings and instructions for every of the procedures regarding thermocouple location and depth, thickness measurements, and total weight. For the SAE J661 coupon test, six (1x1) inch samples were obtained from three inner brake pads and three outer brake pads respectively.
Page 5 of 15
During the GB 5763 test, ten tests were conducted on five (25 X 25) mm coupons extracted from the inner brake pads, and five coupons extracted from the outer brake pads.
All dynamometer tests ud new pads, new discs, and new caliper for each test procedure. The SAE
J661 coupon ud a ground drum and the GB 5763 two-coupon testing ud discs sandblasted before each test run. Both coupon test methods u a standardized drum (or disc) with material specification and geometry different from the disc material ud on the vehicle application (and the dynamometer tests).
Different from previous phas of the project, the calipers were not fingerprinted as the fingerprinting was made in previous phas of the project [5] showing low variability for: piston moving pressure, dynamic efficiency, rollback clearance and sliding forces. The parameters are tested and controlled regularly during production.
Test results
Test results were evaluated:
∙Using standard evaluation templates for each of the procedure executed.
∙As a comparison of the friction coefficients between the single runs of the procedures (test-to-test), and
∙As a comparison of friction coefficients between procedures (procedure-to-procedure).
打扰了英语
The standard evaluations of all tests were carried out only as a control of correctness of test execution, and are out of the scope of this paper. Only “test-to-test” (repeatability) and “procedure-to-procedure” (comparability) comparisons are being prented and discusd.
Test procedures ud in this study specify calculation of friction coefficient values as “average-by-time” (SAE J2522, JASO
C406, and ECE 90 Annex 3) or “average-by-distance” (ISO 26867, SAE J2521, and SAE J2784). J661, GB5763 and ECE 90 Annex 9 are run as drag procedures at constant speeds with specific time stamps to extract the friction coefficient values. To allow comparisons of the results from all procedures it was necessary to agree on one common method of averaging: f or all comparisons “average-by-distance” values have been ud.
冬至英语
Test repeatability
Each of the dynamometer procedures (ISO 26867, SAE J2521, SAE J2522, JASO C406, ECE 90 Annex 3) were conducted twice. Annex 9 was run three times, J661 six times and
GB5763 ten times. Coupon testing ud half the samples from the inner and half the samples from the outer pad.
Figure 4 shows examples for “test-to-test” variability: two examples for dynamometer (ISO 26867 as an example of a matrix-bad test and JASO C406 as an example of vehicle quence-bad test), and the two tests using coupons. See Appendix for friction coefficient values and variability for all the other procedures.
An initial look into the friction coefficient variability (range) between runs of the procedures demonstrates for the first time the tendency between the procedure types: in general, dynamometer tests show more consistent results (lower variability).
Figure 4. Friction coefficient and friction coefficient range —
(a) ISO 26867, (b) JASO C406, (c) SAE J661, and (d) GB5763
Page 6 of 15
