We often get asked, ‘how can we measure forces on the vehicle?’
… and ‘how easy is it to detect changes?’. This is a really difficult question to answer as each vehicle is different and there are a lot of different ways to measure, prepare and process the data. In this post we look at some of the most basic ways to capture data and compare configurations to make engineering decisions.
With a traditional wind tunnel, the environment is heavily controlled and as a result the results are highly repeatable. To do that, some sacrifices to reality are made, which can include; metal belts, non-moving ground planes, cold radiators or exhausts and unwanted tunnel geometry necessary to hold the car in place and measure forces.
Alternatively, there is outdoor testing on airfields, proving grounds, open roads or track. Here, the data is much more representative of what to expect in the real world.
However the likes of rain, wind, changes in the road surface or elevation and even other vehicles add unwanted noise to the experiment.Outdoor testing is usually reserved for the calmest of days and often involves waiting for a suitable weather window, particularly in the UK!
Catesby Tunnel offers the best of both worlds. Real world conditions – but controlled. In order to demonstrate how easy it is to find performance, with just basic instrumentation, this post will share some data from a real-world example.
“2.7km of perfectly straight tarmac, smooth to a tolerance of +/- 2mm, makes Catesby Tunnel a one-of-a-kind vehicle testing facility.”
Using just a wheel speed sensor it is possible to generate key data by conducting a coastdown test. The coastdown test involves the car getting to a constant speed before putting it into neutral and letting speed decay. The longer it takes for the car to slow down, the less drag.
This is the most basic approach and there are lot of options that can be used to improve upon the method used. In this instance we are not trying to show how accurate we can make things, but almost the opposite, ‘how easy is it to generate basic results’ to help find performance? The data below shows the speed time graph, with zero post processing.
The experiment was to see if we could tell the difference in drag between 3 different rake options, Baseline and +/-6mm in splitter height. The speed curves have been moved to start at the same speed, but apart from that, this is data straight from the data logger.
The graph shows a clear trend with increasing rake showing faster deceleration. For a race car, drag only shows parts of the story as we are also interested in downforce. Over the coming months we will share some more data using our new speed measurement system which will provide more accuracy in coastdown results.
Rake analysis in coastdown test
The most basic method of measuring load is often to measure the displacement in the damper. On a track or road, this can be incredibly difficult as the road is not flat and adds noise to the signal.
Because the road surface at Catesby Tunnel is super smooth (<+/-2mm from being planar over 2.7km) this approach becomes significantly easier.
The data below again shows minimally processed data, this time showing speed vs damper displacement, for the 3 changes in rake.
For those of you that are used to looking at these kinds of plots, you will hopefully be pleasantly surprised by the significant reduction in noise that is present. You can clearly see the different trends in displacement, with higher rake increasing load/displacement.
Impact of rake on damper displacement
Damper displacement across a speed sweep at various splitter heights.
To conclude, this set of data has shown how easily data can be gathered at Catesby Tunnel. With next-to-no instrumentation on the vehicle we were able to clearly identify changes in performance when splitter heights were altered. Although this would have been easy to spot in a traditional wind tunnel, this may not be as representative as a real car on a real road.
Later in the year we hope to share some further data, to report on repeatability and a separate linked post on hunting for performance with more controlled methods and some data processing.