Hot Braking Topics
As a professional motor-sports engineer who specializes in data acquisition, I’m often asked about what sensors to run on a car. In general, I won’t rattle off a list of expensive sensors, but in return ask what sort of questions are they looking to answer. Any engineer can spend a lot of money, but a good one is responsible with their resources.
Backing up a little bit, I loosely describe “Racecar Engineering” as developing and using available resources to out-perform the competition. Some days it’s using advanced mathematical solvers on a powertrain simulation, sometimes it means jumping on a lathe to make parts, and other days it’s just making sure the fire bottle button is labeled clearly. But for this case, we’re going to assume that we’re an average FSAE team with a car that runs competitively but we’re also not consistently making design finals.
So why would we want to put a data acquisition system with sensors on our car?
Because everyone else has it?
It costs a lot of money, so it must be important.
Because the design judges said so.
Obviously, none of these are valid reasons. Instead, I suggest that the following steps are used before spending any money on testing.
Step 1: Come up with your question, theory and/or model.
Step 2: Brainstorm how you would measure and/or calculate it.
Step 3: Choose the appropriate path based on your available resources.
Let’s work through an example. At some point you’ll need to pick out brake pads for your car. Obviously they’ll need to fit in the calipers you’ve chosen, so it’s probably a good idea to pick your calipers and brake pads out at the same time. Most reputable brake pad manufacturers will provide a graph with the coefficient of friction vs temperature for their range of pad compounds (along with a lot of other marketing info).
Figure 1: Fictional Brake Pad Coefficient of Friction vs Temperature Data
So the next question is, what sort of operating temperatures am I expecting, and correspondingly what coefficient of friction should we expect/should I use in my calculations. At this point we can try to calculate what our brake temperatures would be, or if we have a car we can measure it directly.
Step 1: Come Up With Your Question, Theory or Model
Let us assume we don’t currently have a running car to test on. We can fairly easily estimate the energy going into the brakes.
M(car) x A(decel) = Energy(brakes)
SI units make this really easy… if we integrate the energy put into the brakes over the course of a lap we can then determine the amount of power being put into the brakes. From there, you’ll need to split the energy into the front and rear braking systems based off of the forces the tires are seeing. After that it shouldn’t be too difficult to estimate the mass of the brake rotor, caliper and pads and get a reasonable guess for the temperature you’ll see. However, the cooling aspect is much harder to calculate. The area inside the wheel is very turbulent, there are lots of different materials with different heat transfer rates, varying vehicle speeds, interaction with the front wing, a rotating and squishing tire, radiation, convection, conduction… hmm, maybe it’ll just be easier to get the old car up and running.
Figure 2: Calculated Braking Power From Data on a TCR Car
Step 2: Brainstorm How You Could Measure Your System
Ok, now that we’ve got the old car running. How do we measure the brake temps?
The quick and dirty option would be to borrow a thermocouple from the lab. The lab monitor will probably ask what range thermocouple we’ll need. Luckily we’ve already done those temperature estimates.
We’ve now checked out the thermocouple, and got our car running. What’s left? We should probably make a test plan. Maybe 5 stops from 45 mph? Hmm, how about 10? Our model shows that we should reach a steady state by then.
After testing, what did we find? Looks like the brake rotor temps are lower than what we were expecting. But, should we expect that? The biggest problem here is that we don’t actually get the temperatures while braking. There is a certain amount of time that has elapsed before we’re able to stab the rotor with the thermocouple, and even then, the most braking energy goes into the system at the highest rate of deceleration. So, we might be losing temperature even by the end of the braking zone. But still, we have some actual data and we know the brakes get at least this hot.
So how else might we get temperatures while the car is actually decelerating? We could try temperature sensitive paint. Depending what you want, it can range from $30 for a bottle of a single temperature to ~$100 for a bottle of the multi temp. There are also temperature sensitive stickers which should yield similar results.
So now that we’ve put some paint on the rotors, lets run the same test… 10 stops from 45 mph and see what our results are. This number should be much closer to our calculated values since we’re getting information during the braking event, as the maximum amount of energy is being put into the system.
But our downsides are that we don’t quite have the same resolution as we did with thermocouple and this really only gives us one more data point. One the other hand, depending on the brake pads we’re looking at, this might have provided all the information we need.
…because the next step is fairly large in terms of financial and human resources required.
Professional level infrared (IR) temperature sensors can cost as high as $600 each; and while there are options in the ~$200-~$300 range, it would still be over $800 to outfit a car. Don’t forget you’d already need a data acquisition system and then have to wire and mount the sensors. On top of that you’ll need someone (or multiple people) to maintain the data system, download data, organize it, interpret it etc.
Let’s pretend that we have a rich alumnus who was willing to purchase the sensors for the team and that we have enough people interested in data acquisition to provide enough support for the project.
We run the same test as before, and discover that the temperature sensitive paint was pretty accurate. Hopefully though, we’re able to learn more about brake temperatures and shine a little more light on the brake pad selection process. Not only do we have a peak brake temp, but we’re also able to see what the temperatures are when our driver first gets onto the brakes after a long straight.
Furthermore, we see that it takes a few laps for the brakes to reach our steady state temps, but once it does, we still have a significant range of temperatures we operate in. And for some reason our rear brakes are generally cooler than the fronts. Interesting…
Perhaps these IR temp sensors brought up more questions than answers.
Step 3: Choose the Appropriate Path Based On Your Available Resources
I want to make it clear that the purpose is not to say that you need to have IR temp sensors in order to make design finals. What we’re hoping for is an understanding of the system, wise use of resources, a proper understanding of the tests/results, and a maximization of your team’s potential. A good model, some creative testing with a thermocouple and proper understanding of the results will yield better results than a poorly executed test with IR temp sensors and no model.
