Help with Obtaining Road Load Coefficients

[fekkin]

Hello all,

I have done a number of coastdown tests on the flat in our test vehicle from 30m/s to near rest and have attempted to obtain the road load coefficients using Simulink, but the values returned from parameter estimator were out of range of typical values (Drag coefficient of 1.5). When fixing the vehicle's known drag coefficient and calculating the rolling resistance force as coefficient A/f0, the value for B/f1 was returned as ~25Ns/m, is this typical for a vehicle with significant drivetrain losses and inertia?

Using these coefficients on the dyno, we found the load was too high, and the vehicle was topping out at less than half the top speed it can achieve on the road. Should I be conducting tests from different speeds?

I would much prefer to calculate these coefficients from vehicle parameters, but everywhere I look I find conflicting information on the process. Is there an accurate calculation that I can perform to obtain these coefficients?

Also, the dyno we use is a Maha MSR500/2 and all three entry fields for the coefficients use newtons as the unit. Is this a potential issue?

I would much appreciate any advice and look forward to your responses,

Tom

 

[GregLocock]

As a diagnostic do a coastdown on the dyno. If it matches the track result then the problem is not the coefficients.

More likely is that you are doubling up, most of the losses on the track are still present on a chassis dyno, so you don't need to account for the diff for example, it is still running (and is a very large part of your mechanical losses). Tire rolling resistance is actually higher on the roller than on the road.

Coastdown tests are very hard to do, between wind and gradients.

 

[fekkin]

As a diagnostic do a coastdown on the dyno. If it matches the track result then the problem is not the coefficients.

More likely is that you are doubling up, most of the losses on the track are still present on a chassis dyno, so you don't need to account for the diff for example, it is still running (and is a very large part of your mechanical losses). Tire rolling resistance is actually higher on the roller than on the road.

Coastdown tests are very hard to do, between wind and gradients.

Thanks for your response Greg,

That makes sense. Next time I go I will make sure to do a coastdown. It did not occur to me that the coefficients that the dyno requires are adapted from the track coefficients. Having said that, I used the road load adaption within the software and it returned the A coefficient as ~-450 as opposed to ~+300 and increased the B coefficient by about 40%, regardless it still topped out at the same low speed during the drive cycle.

Any tips on obtaining the right coefficients to begin with, and is there any computational method that you know of?

Thanks,

Tom

 

[GregLocock]

No tips I'm afraid - I would have used Excel or Mathcad. The way I did it most recently on our closed oval track was to do a coastdown from 6 points around the track, in each direction, so 12 runs.