Acceleration and Yaw Measurements

Figures 1 and 2: Yaw Rate (right) and Acceleration (left) Sensors

 The CXL04LP3 Crossbow accelerometer (figure 2) was chosen for the application. This is a general purpose, low cost, low-power, sensor commonly used in automotive applications. It can measure + 4 G with a 5 volt input.  For measuring yaw, the CRS03-04 (figure 1) from Silicon Sensing fit our application while still maintaining low cost and rugged packaging.  The sensor can read yaw rates of up to 200 degrees/second.

For this system, a lowpass filter was designed to attenuate any unwanted signals such as engine vibrations. An Analog Active Lowpass Butterworth filter was chosen for the design. An active filter is a type of filter that uses one or more active components that provide some form of power amplification. This provides a steeper roll-off which allows for a quicker attenuation. The Butterworth was chosen because it has a smooth passband and a very smooth increase in stopband attenuation. It also has very low phase distortion.

The filter design software, WFilter, was used for initial design and testing. Computer and hand calculations can be found in the appendix. The filter has an attenuation of 12 dB per octave and was tested with a cutoff frequency of 10 Hz. Results from this testing can be seen below in Figures 3 and 4.

Figure 3: Magnitude of Filter Output

 Figure 4: Phase Output

After the design was completed in WFilter, the circuit was modeled in PSpice as shown in Figure 5.

Figure 5: Circuit Diagram

After the circuit was verified using PSpice, the circuit was physically  built on a breadboard to validate design. The circuit was tested over a broad range of frequencies to show proper attenuation. The oscilloscope measurements from this testing are shown below in figure 6.

Figure 6: Experimental Filter Response

An aluminum housing was also constructed to insure proper protection and integrity of the circuit board.

Steering Position Measurement

Figure 7: Steering Position Assembly

The steering position will be recorded by measuring the voltage difference across a rotary potentiometer.  A small mounting was cut from 0.07” steel sheet metal using the Learning Factory’s water jet cutter and was then welded together.  This mounting attaches to two ties at the rack of the car.  The potentiometer is held parallel to the steering shaft.  A friction disk is attached to the potentiometer that is in contact with the steering shaft (see assembly in Figure 7).  When the steering turns the resistance of the potentiometer changes, this analog signal is then recorded to the ADL.  A simple linear equation then relates the voltage change to the number of degrees the steering moves.

Shock Position Measurement

Figure 8: Linear Potentiometer

To record the compression and extension of the shocks while cornering, four linear potentiometers were used.  In a practical application, a linear potentiometer is simply a voltage divider.  As displacement, changes, the slider moves along the resistor. Potentiometers are cheap and fairly accurate, but wear out eventually due to the physical contact at the slider.  Celesco linear potentiometers were used, as shown in figure 8.  These potentiometers were developed specifically for the auto racing industry, have high life expectancy and 0.1- 0.2% linearity.
 

Brake Pressure Measurement

Figure 9: Brake Pressure Sensor

There will be two brake pressure sensors (figure 8) installed on the formula car.  These sensors will measure the brake line pressure in the independent front and rear systems.  The sensors will be installed in the foot box of the car down stream from the front and rear master cylinders.  The sensors will be installed with a 1/4 to 1/8 inch NPT thread adapter and a 1/8 inch NPT to 3/8 X27 T-adaptor that will attach the sensors into the brake lines.  The brake pressure sensors will be tested by applying pressure to the brake pedal and reading the output from the data logger.  Validation of readings can be performed by applying a known force to the brakes and calculating theoretical line pressure with basic fluid principle equations. 

Wheel Speed Measurement

Figure 10: Hall Effect Sensor

These sensors will also be utilized to run the active traction control systems on the car during competition.  The data from the sensors will be interpreted by the onboard computer of the car as it is logging data for our tuning use.  This will allow the MoTeC computer to continually update the cars traction control. 

The main constraints and considerations for the wheel speed sensors and pickups are that they have to be compact, lightweight and robust. 

The group had several options pertaining toward the monitoring of individual wheel speed measurement.  Optical sensors and magnetic sensors were the two categories that the team considered to use.  It was concluded early in our project that the magnetic sensors would be more cost effective, more robust, and more lightweight than their optical counterparts.   

The magnetic sensors operate on an induced magnetic field from the wheel speed pickup as it passes in front of the sensor.  This induces a current within the sensor that will in turn relay that signal to the computer to be processed as data.  The group was able to utilize this method due to the fact that all of the components of the wheel packages on the FSAE car are made of non-ferrous metals.  Therefore, the pickup will effectively be the only magnetic field that the wheel speed sensors will see as the car runs.  For this task, the group chose to use a Cherry GS100701 magnetic sensor. 

 The pickups will be located particularly close to other dynamic elements of the car including the brake calipers, rotors, and hubs.  On top of these space restraints, the pickups and sensors themselves must operate within approximately 1/32^nd of an inch of one another. 

A large concern of the FSAE team is that of weight.  Great lengths have been taken in every aspect of this car to shave ounces everywhere that is possible.  Due to the permanent residence of the sensors on the car even after initial tuning, DaqTron took great consideration when choosing sensor types and designing pickups and mounting solutions.  The pickups and mounts were designed to be as small and lightweight as possible while sill allowing for their required stiffness and strength.  Steel with a 0.030” thickness was used for all of the pickups on the wheels.  Several redesigns were needed to minimize the materials needed to reduce the rotational inertia of the pickups.

The front wheel speed sensors were designed to be mounted directly to the front hub via pre-existing mounts that are also used to secure the brake rotor in place.  This means that the front sensor is very close to a rotor that will incur extreme heat under hard braking and to the moving caliper itself.  This setup uses the same existing bolt design as the rotor itself.  A Pro-E design of this pickup is shown in figure 11.

Figure 11: Front Wheel Speed Sensor

The rear wheel speed sensors are connected to the rear hubs via plugs that were machined out of aluminum stock and mechanically pressed into weight reduction holes within the rear wheel hubs.  This setup requires that the plugs be tapped and a small set screw/washer combination be utilized to secure the sensor into place.   

Figure 12: Back Wheel Speed Sensor

The group will be testing and calibrating this system by using measured distance for the car to travel.  This will be confirmed by the ratio of pickups to one rotation of the wheel and the radius of the tire on the car at that time.

Data Recording

To record the outputs of the analog sensors the project includes a MoTec ADL2.  This instrument has a large amount of memory and 28 inputs and power outputs for the sensors.

A wiring harness was fabricated to interface all the sensors with the ADL2.