[ODE] Springy AMotors?

[Brian Allen]

I'm wanting to model the behavior (trajectory) of PD-controllers using
ball-joints and AMotors.  I'm running into a previously-discussed
problem:  there's no parameter for setting ERP on a per-joint basis
for ball-joints.

I found the old patch for ODE 0.35 that was kindly posted to the list,
but before trying to get it working with the latest build, I'm
wondering why the AMotor ERP param is missing in the first place,
especially since a parameter to set CFM is available.  Is there some
fundamental issue here?  Is there a reason for the asymmetry?

Anyone have any suggestions for another (stable) way to get
damped-spring-like behavior from ball-joints?

Thanks for your help.

(ps. sorry if this is a duplicate, net connection has been flaky lately...)