It's off to track we go

It's finally track day, and there's little I like more than carrying out a good test plan with equipment we know we can trust. Once we got there, the first step was to set up the sensors so we can record data. The sensor PCBs take the voltage from the differential pressure sensors and convert it to a CAN value. We didn't code in any other conversions to keep the recorded data as "raw" as possible to make debugging easier, although we did add an averaging over 10-point intervals to limit the amount of data being sent over the CAN bus. With 25 Hz output frequency, these combined to give an effective 250hz of averaged readings. Both prototype boards plugged in to breakouts on the dashboard, keeping connections central and thus limiting the required length of both the pneumatic tubing and CAN connection cable.  

We had three tests to get through, with the on-track methods summarised below:

Straight airflow Z0.35

Target airspeed: 18 ms-1 

1. Mount S_Z0.35 rake to the vehicle

2. Record ambient pressure, density, and temperature

3. Use air pump to confirm functional readings from each Kiel probe, including yaw and pitot-static probes, using live telemetry. 

4. Driver to conduct a single run down the main straight, keeping speed below 5 ms-1 and being prepared to stop if necessary. Purpose is to confirm integrity of the rake mounting and to log sample data.

5. Driver to return to pits, recorded data to be extracted and examined to confirm it looks as expected

6. Upon satisfaction with the data, driver to conduct 12 "runs" up and back the main straight, each time accelerating quickly to the target airspeed and holding this speed for as long as possible

7. Driver to return to pits, recorded data to be extracted and examined to confirm it looks as expected

8. Unmount S_Z0.35 rake from the vehicle

Straight airflow Z-0.11

Target airspeed: 18 ms-1

1. Mount S_Z-0.11 rake to the vehicle

2. As per steps 2-7 in the previous test case

3. Unmount S_Z-0.11 rake from the vehicle

Representative corner Z0.35

Target airspeed: 14 ms-1
Corner radius: 13.2 m

1. Driver to conduct 4 hot laps to warm up tyres

2. Mount RC_z0.35 rake to the vehicle

3. Driver to return to pits, mount RC_Z0.35 rake to the vehicle

4. As per steps 2-4 in the first test case

5. Upon satisfaction with the data, driver to conduct 12 passes of the constant-radius corner at the maximum possible steady-state speed, keeping the vehicle as stable as possible.

6. Driver to return to pits, recorded data to be extracted and examined to confirm it looks as expected 

7. Unmount RC_Z0.35 rake from the vehicle

Our test plans got more detailed and precise as the year progressed and we learned what extra detail we wanted and needed from our testing. Originally, they were just instructions for setup and for the driver, but they evolved into a complex document covering the day before through to the day after. Additions included:

• Full car and equipment setup specification, including mount locations and tolerances

• Guides/processes for pre-track preparation and post-track pack-up

• Outline and justification for each test process to help in creation of on-the-spot decisions or contingencies at track should they be required (and to force the test-plan-writer to only include tests worth doing)

• Data extraction and post processing, again to help in decision-making by making interdependencies between different results clear

The plans were also accompanied by post-test reports containing an overview of key results, discussions, required future work, location of raw and post-processed data, and driver feedback. There were two main goals that justified all this complexity: Providing decision-making tools to allow us to maximise useful data even if things go wrong or testing needs to be cut short, and to provide an instruction manual comprehensive enough to allow anyone on the team to carry out the tests even if no aero team members are present. 

The yaw probe shown above is still very much a prototype design, and  its output was a little noisy. A more precise design with more rigorous wind tunnel testing will be prepared before the next aerodynamics track tests. Combined uncertainty of the yaw probe and pressure sensors using a RMS method came to 20 Pascals with a 1.25 factor of safety, which suggested that below 15 ms-1, the precision exceeds +/- 2.5 degrees. 1-degree precision is achieved at 20 ms-1. Ideally, 0.5 degree precision would be achieved at all speeds above 10 ms-1 which may require an alternative approach such as a aerofoil probe to utilise static pressure difference. The current forked design (similar in concept to a Cobra probe) only creates/uses dynamic pressure difference. 

Unfortunately due to deteriorating weather, we were unable to complete the cornering case test. This wasn't a huge loss as it would have had the least emphasis during validation anyway due to the additional uncertainty in replicating the exact driving scenario down to the steering, slip, and roll angles. More emphasis on cornering and braking will come in future tests. 

Extracting the data was easy enough as there wasn't that much to go through, nor was I interested in any relationship between different systems on the car. That said, I'd still rather not have to go through all the values and hand pick the ones we want, so I wrote a MATLAB program to do it for me. Pressure sensor data was extracted and averaged whenever airspeed (from the pitot-static probe) was within a 5% tolerance (based on dynamic pressure) of the target airspeed. The same process was then done using wheel speed to make sure the results were similar, but airspeed is more appropriate and so was used for the final results. 

The pressure data at the start of each log (when the car is stationary) was identified and averaged by the script to give a calibrated 0 reading for each sensor. Temperature effects were removed by smoothing and then subtracting the voltage offset of the spare sensor from all other sensors. Pressure values (including those from the pitot-static and yaw probe) are multiplied by a scaling value to match airspeed to wheelspeed (at instances of no wind), which is typically 0-2 percent depending on ambient atmospheric conditions. 

Validation for the rake data had two accuracy metrics:

• Total pressure magnitude at each probe

• Minimum distance from the probe location to a point with equal total pressure magntude in CFD

The second metric acknowledges the high velocity gradients in the wake, particularly around the perimeter. This means that even if the magnitudes are significantly different between experiment and CFD for a particular probe, it may only be a 5-10 mm difference in position of the wake that caused the otherwise concerning magnitude error. 

For visualisation, there isn't enough resolution of theses rakes to estimate the precise shape using only experimental data. Instead, for initial shape comparison, it was noted if the experimental total pressure suggested a freestream, perimeter, or core wake reading. This simplified "map" of the wake will guide coarse adjustments to the CFD setup before the magnitudes are used for fine tuning and quantitatively assessing CFD/track correlation.