Car vs driveline acceleration during tire longitudinal slip

[TomTork]

I'm writing a simulation program for an automatic launch control (ALC). I need to know quantitatively the dynamic distribution of vehicle acceleration decrease vs driveline acceleration increase during drivewheel traction breakaway (burn), as a function of longitudinal slip rate. Also what is the decrease in traction recovery threshold due to any hysteresis effect? SAE considers driveline ineritial mass as 3-5 percent of the vehicle weight.
Any meaningful discussion is greatly appreciated...TomTork

 

[GregLocock]

You'll get up to 180 degrees of windup in the halfshafts, and the tractive effort will peak at around 6-20% slip ratio.

In effect you are looking for the dynamic changes in the ratio rpm/actual vehicle speed.

A decent CAN Bus logger would probably allow you to plot those out, or you could intercept the timing pulse and the ABS tone wheel from an undriven wheel to do the same.

In first gear you'll find the inertial effects are much bigger than 3-5%, it depends on the square of the gear ratio.




Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[TomTork]

I was hoping that I could get an answer to my thread without instrumenting my old non-ABS '54 Healey. Seems that the relative acceleration of the driven wheels and non-driven wheels could be derived mathematically when the car is fully characterized. I apologize for the following tedious drone, but it may be necessary to get my thread fully understood.

Using my Olds V8 powered Healey with 47/53 F/R% as an example:

A 2700# car/driver accelerates from a standing start with WOT. It has 450#ft peak engine torque, total gear reduction of 8.5 and a 27.5in dia rear tire. This produces rear wheel torque or FORCE of 3400#. Since the SAE assigns driveline rotary plus reciprocating inertias as 3-5% vehicle weight or 135#. This increases the effective MASS of the car to 2835#. The potential car ACCELERATION is thus 1.2G with 100% rear weight transfer.

The DOT tire's coefficient of friction is only .8, so the tires start slipping as the engine torque goes thru 382 #ft at 2800rpm. If WOT continues, the driveline increases velocity drastically while the car decreases velocity less drastically.

Since this example is purely intuitive we could consider a ridicules extreme:
The 2835# car hits a patch of ice or the driveshaft breaks at 2800rpm in 1st gear. In both instances the coefficient of friction immediately goes to zero. If WOT is maintained, the car coasts with drag deacceleration. Since the driveline is completely decoupled from the car, all the torque produced by the engine is directed to the inertial mass of the driveline so it accelerates at 20x 1.2G or 24G!

So what is the precise distribution of potential acceleration between the 2700# car and the 135# driveline masses as functions of longitudinal slip rate?
Is there enough information given above to develope an equation that can be implemented in my computer ALC simulation?

 

[GregLocock]

I disagree with almost every number you've posted, but yes, I'd have thought that would behave in roughly the correct fashion, if you had a curve of longitudinal slip ratio vs tractive force.

Those are measured, so the problem is finding one on-line.

I suggest you look in Thomas Gillespie's book in particular, as I think he's worked through exactly what you are doing, or Reimpell and Stoll. Both books also include plots of rather generic looking friction vs slip curves. R&S shows one that is pretty linear up to 0.85 mu, 10% slip, peaks at 1.03 mu, 20% slip and decays smoothly to 90% slip 0.92 mu









Cheers

Greg Locock

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[NormPeterson]

Instead of using a flat estimate of "added weight" to account for driveline acceleration effects, why not work directly with the various rotational inertias, their respective alphas, and the relationships between those alphas' and vehicle acceleration through effective tire circumference, gearing and slip %? It'll probably end up being an iterative solution.

Norm