GPS speed VS Wheel speed - Traction control

[Slim47]

Whatsapp guys 

I have a drift car. In general, traction control is not very relevant for us . But at the start for the first two gears, this feature is extremely useful.

I have rear wheel speed data and high frequency GPS data. Can I set the traction control function to a reasonable difference between these values that the reference to hold the motor will target? 

P.S. As you understand, the meaning of the rotation of the wheels from the front axle will not be reliable and are not provided for by the suspension design

[dx4picco]

a guy made a post not long ago, can't find it back, he was explaining that even the high freq gps, you get a speed error proportional to the acceleration.

if you got that in check like he did with a math formula, then yes I guess

[Adamw]

GPS is pretty hopeless from standing start, quite a bit of lag before it even notices the car is moving.  The higher end systems combine accelerometer data to correct the low speed errors.  

[Slim47]

Thanks for answers. Perhaps then it is possible to set the limiting acceleration of the rotation of the wheel?

[Adamw]

Yes, this type of function is called "Power Management", it is in the motorsport folder.  This is where you specify a target wheel or driveshaft speed, usually against elapsed time from launch (race timer function).  The ecu will apply cuts or retard to keep the wheel speed at the predetermined speed.

[Slim47]

Thank you, Adam.

But I deliberately used "wheel acceleration" (read the legend) in the graph, and not absolute speed.

That is, we understand that in the presence of adhesion, wheel acceleration under load will have a limited value, and if adhesion is lost, it will grow exponentially in a fraction of a second.

Power management is more suitable for driving through the pitlane with a speed limit or full drag race. Perhaps I misunderstand how power management works.

 

[Adamw]

Your graph shows wheel speed on the Y axis vs Time on the X, which is typically how you do standing start power management.  Yes the gradient is accel but you have your exponent working in the wrong direction, accel will decrease as speed increases.  

You can try controlling wheel acceleration if you wish but I dont think you will find it useful.  A log from a drag car below to illustrate, wheel speed on the top trace, wheel accel on the bottom.  The two large spikes are gearshifts.  Wheel speed was logged at 500Hz and the accel calc is deriving over 10ms (100Hz).  To make acceleration useful you would have to derive over a relatively long time period which would mean the ecu wont be able to take corrective action until after the slip has already happened.    

[Slim47]

On 8/20/2022 at 1:32 AM, Adamw said:

Your graph shows wheel speed on the Y axis vs Time on the X, which is typically how you do standing start power management.  Yes the gradient is accel but you have your exponent working in the wrong direction, accel will decrease as speed increases.  

You can try controlling wheel acceleration if you wish but I dont think you will find it useful.  A log from a drag car below to illustrate, wheel speed on the top trace, wheel accel on the bottom.  The two large spikes are gearshifts.  Wheel speed was logged at 500Hz and the accel calc is deriving over 10ms (100Hz).  To make acceleration useful you would have to derive over a relatively long time period which would mean the ecu wont be able to take corrective action until after the slip has already happened.    

Thank you so much. You are the best

In your example, it is clearly visible from the graph below that there is a normal level of wheel acceleration and it is clear that in the middle of the race there is a exponential wheel acceleration. It can be argued that at this moment there was slippage. It remains to figure out how to set up LINK ECU to limit torque in this case). To make the wheel acceleration math block I have to take the speed of wheel 2 subtract the speed of wheel 1 and divide by the time between changes in wheel rotation. From the manual, I understand that the calculation speed is 1 kHz, that is, 0.001 second. But I do not understand how I can take the speed from the last calculation. I don't have enough imagination ))))

In my opinion, the schedule is correct, I considered one gear. Starting in first gear, let's say from 2000 rpm having a turbine spool at 4000 and picking up speed up to 8000, the wheel speed will increase exponentially. But it doesn't matter, your answer completely exhausts my question.

Another short question so as not to open a new topic.  VQ35 engine.  2 wideband lambdas for each half block.  One is connected directly to the Link ECU, the second to the AEM UEGO (+5V data connect to ECU), there is also a back pressure sensor in front of the turbocharger.  Can the AFR backpressure correction be applied by to both sensors?

Can I be sure that both AFR gauges show comparable values?