2

I want to calculate the tyre pressure of a car from OBD parameters. I don't want to take TPMS or any custom PID for this. One of the indirect method is the calculate:

((w_LF + w_RR) - (w_RF + w_LR)) / ((w_LF+w_RR+w_RF+w_LR)/4)

where w_LF is the angular velocity of the left-front wheel. Now the problem I am facing is how to calculate w_LF, w_RR, w_LR, w_RF and how to find what is the actual angular velocity at that instant while driving.

I want to point that there is a reduction in tire pressure by this method

  • How does angular velocity tell you the tire pressure? – costrom Dec 16 '15 at 19:25
  • v= rw. Now velocity (v) is the same for all the wheels But with reduction in pressure in the tyre, the volume, hence radius (r) decreases of that tyre. Thus the wheel will rotate faster. Hence we can point a reduction in the tire pressure. – Soumya Sen Dec 16 '15 at 19:28
  • What if all wheels deflate by approximately the same amount, like when going from hot to cold weather? Are you sure that you can get the velocity of each individual wheel via OBD anyways? – I have no idea what I'm doing Dec 17 '15 at 16:55
  • Sure, that's a flaw. Another flaw is if the rear or front both wheels change by the same amount – Soumya Sen Dec 17 '15 at 17:07
3

This approach is fundamentally flawed for a few reasons:

  • There's a circular definition here; to know angular speed you need to know wheel radius, but you're trying to infer the wheel radius from the angular speed.

  • I don't know of any OBD-II PID which will tell you wheel radius.

  • Even if such a PID exists, how will you cater for vehicles where a custom wheel size has been fitted?

  • For v = rω to work for each wheel, you need a measurement of speed for each wheel. This data might be available if the vehicle is fitted with ABS and has 4 wheel speed sensors. Don't assume that all vehicles are like this though; some vehicles rely on just a single speed sensor embedded in the transmission housing to measure vehicle speed.

  • The formula shown is an attempt to evaluate differences in angular speed that are normalized by the average of the four readings. It might be an acceptable qualitative check, but it wouldn't tell you which wheel is low on pressure.

  • I agree completely with all your reasons. I am assuming a car with a single speed sensor and assuming the 4 wheels have the same speed. I need a qualitative analysis only; just to point out that my tires are loosing pressure. I need a way to calculate the actual rotation value of the wheels – Soumya Sen Dec 16 '15 at 19:56
  • @SoumyaSen but what will you do for radius then? You still need radius – Zaid Dec 16 '15 at 19:57
  • Assuming a car manufacturer recommended value, for the time being, I mean the tyres with which the car comes to the showroom – Soumya Sen Dec 16 '15 at 19:58
  • 1
    @SoumyaSen I fail to see how this approach will work. You're basing it on a single speed sensor, so all the v's are the same. You're assuming the same value for wheel radius, so all the r's are the same. This will mean that all four values of ω will be the same (v/r), so your formula will always evaluate to zero. Am I missing something? – Zaid Dec 16 '15 at 20:01
  • 1
    As long as you have the independent wheel sensors, then yes you can get the information you seek. This is the way some car manufacturers determine if you have a low tire (low tire will roll faster than the full tire). I believe in their measurements, they are just looking at the sensor speed for any given tire and matching it against the others. If there is a sustained period where a single tire is spinning faster, the low tire indicator is tripped. Other than that, not a chance in heck. – Pᴀᴜʟsᴛᴇʀ2 Dec 16 '15 at 22:23

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.