On-Road Brake Testing. (Vehicle Test).
Pricol, a large automotive components supplier in India, asked Soliton Automation to develop an in-vehicle data acquisition system to test motorbike front-wheel disk brakes. Pricol needed a very versatile system that could provide multiple channels of data acquisition to record all the required signals from the motorbike during the trials and on-board analysis capabilities for the tester to see the results of the test while on the road.
The primary end result is the braking distance--the minimum distance required for the motorbike to come to rest from a given speed on a level road. The braking distance measured by conducting road trials depends on several factors other than the brake system, such as the coefficient of friction between the tire and the road, the weight of the motorbike with the rider and the weight distribution, and the skill of the rider. By keeping these factors as repeatable as possible, braking distance is a good measure to evaluate the effects of modifications made to the brake system during development.
Other than providing the braking distance, the test system needed to log the brake hydraulic pressure, the temperature of the brake pad, and the speed of the motorbike throughout the test. Since the entire test only lasts a couple of seconds, the system had to log at least 1,000 S/s of data on each channel.
Pricol also needed a compact user interface so that the rider could input parameters and view the results during the road tests. An off-line analysis routine was necessary for post-test analysis in the lab.
The system had to handle the vibrations and accidental falls that it would encounter during testing. Since the brake testing would be performed in all types of conditions, the shock and vibration requirements called for a rugged in-vehicle system.
The following sensors acquire the signals:
* Rotation Encoder--for rotational displacement and the wheel speed.
* Accelerometer--for the instantaneous acceleration of the vehicle.
* Thermocouple--for pad temperature during braking.
* Pressure Sensor--for the brake hydraulic fluid pressure.
Soliton Automation attached the 1,000-pulses-per-revolution (ppr) rotational encoder to the hub of the front wheel of the motorbike and mounted a capacitive accelerometer near the center of gravity of the motorbike to directly measure the instantaneous acceleration of the motorbike. During a brake test, the acceleration of interest is derived from the DC signal after filtering out the AC components produced by the vibrations present at the mounting point. The thermocouple is bonded to the brake pad using a high-temperature bonding adhesive.
The possibility of using an off-the-shelf datalogger was discarded in favor of a PC-based system. Off-the-shelf dataloggers did not offer the versatility that Pricol needed, but Soliton Automation could program a PC-based system to fulfill the requirements. The PC-based system also had a user-friendly interface for the rider during the testing.
The FieldWorks FW2000 Series Computer from Kontron Mobile Computing suited the application both in terms of size and ruggedness. Two PCMCIA slots were available to include a data acquisition board and an Ethernet board.
A multichannel E Series PCMCIA high-speed data acquisition board from National Instruments provided the necessary analog inputs and the counter/timer inputs for the application. The FieldWorks computer uses a Pentium processor and runs Windows 98. The application software was developed in National Instruments' Lab VIEW 6i.
Soliton Automation designed a low-cost, compact, and rugged micro-controller-based integrated keyboard display for the user interface. The unit, with 14 keys and a two-line LCD, mounts on the handlebar of the motorbike and interfaces to the computer using the RS-232 serial port.
A 12-V battery powers the FieldWorks computer, the user interface, and all the sensors and associated signal-conditioning units. A separate battery isolates the data acquisition system from the motorbike's electrical system for noise immunity. During extended field trials, the battery is switched with the one on the bike for recharging using the bike's charging system.
A typical on-road brake-test sequence involves numerous runs on a selected road at different speeds. The braking distance is calculated for each speed until repeatable results are obtainable, and then the trials are repeated at a different speed.
The rider inputs the speed for the trial via the user interface. The system monitors the speed continuously and gives the rider a prompt to start braking when the motorbike reaches the desired speed. When the motorbike comes to a stop, the software instantly calculates the braking distance, and the distance is displayed on the LCD.
Figure 1 is a block diagram of the setup for measuring the braking distance. Braking distance is obtained by counting the number of encoder pulses received from the start of braking to when the bike stops (Counter 1). This count, divided by 1,000, gives the number of wheel revolutions, and multiplying this figure by the wheel circumference determines the braking distance.
The start trigger for Counter 1 is obtained from the pressure-transducer signal. When the hydraulic pressure increases beyond a predefined threshold, a trigger signal is generated which starts Counter 1.
The encoder output also is fed to Counter 2 on the data acquisition board. Counter 2 is configured through the Lab VIEW software in the timer mode to measure the pulse width of the encoder pulses, which is used to determine the instantaneous speed of the bike. When the speed drops to less than another predefined threshold, a stop trigger signals the end of the run, and the Counter 1 value is read to determine the braking distance.
While the encoder gives the most direct measurement of the braking distance, the accelerometer also plays a role in the tests. When a wheel-lock condition occurs, the encoder pulses stop. If this is not properly detected and compensated, the braking distance obtained by counting the encoder pulses would be lower than the actual braking distance.
The wheel-lock condition is detected by looking for an abrupt drop in the speed. The accelerometer data fills in the missing pulses to obtain the corrected braking distance.
The braking distance measured solely by using the accelerometer data is not as accurate as the distance calculated by counting the encoder pulses. The braking distance given by the onboard data acquisition system is validated by directly measuring the braking distance on a road.
A set of graduated lines was drawn on a test track marking 0, 10, 15, 20, 25, and 30 meters. The rider would brake just as the front wheels crossed the start line (0 meter line). When the bike came to a stop, the actual distance traveled was measured with respect to the closest graduated line on the test track and compared with the value calculated by the data acquisition system. The value given by the system was accurate to 1% over a distance of 30 meters.
Along with the encoder and the accelerometer data, the temperature and the pressure data for the complete run is stored in the hard disk of the computer for post-test analysis. In the lab, the stored data is analyzed by connecting a monitor and keyboard to the FieldWorks computer or transferring the data to a lab PC via an Ethernet network. Figure 2 shows one of the screens of the LabVIEW application that was developed for the test-data analysis.
The On-Road Brake Tester can measure braking distance to an accuracy of 1% over a distance of 30 meters. The system also handled wheel-lock conditions by using the accelerometer data during these periods.
The small form factor of the FieldWorks embedded computer and the National Instruments PCMCIA data acquisition cards made it easy to mount the system on the motorbike. The whole system was rugged enough to handle accidental falls during testing.
The system allowed the user to set up different tests during the road trials and see the results without connecting a laptop or a monitor to the embedded computer. Since the on-road brake test system is based on open PC standards and development environment and uses general-purpose data acquisition hardware, the system is versatile, user-friendly, and expandable and was developed cost-effectively.
About the Authors
The three authors are employed by Soliton Automation Private Ltd. in India. V. Arunachalam is a senior project engineer and has a bachelor's degree in electrical and electronics engineering from Bharathiar University. Gokul Dass T.V., the technical communications specialist at Soliton, holds a bachelors degree in science from Bharathiar University. Ganesh Devaraj is the CEO and a graduate of the University of Michigan with a Ph.D. in physics and an M.S. in electrical engineering.
[Figure 1 omitted]
[Figure 2 omitted]