cj10jeeper said:
If you find anything where it is triggered into giving random tiny left and right inputs then do let me know
Not sure how it works yet, but the EPS has a self-centring mechanism, so it sounds like an artificial force is applied to help right the steering to dead ahead. From early reading it looks like this is measured/applied relative to the steering lock rotation sensor.
My concern was that if the sensor is sending signal 'centred' at 0 lock, yet due to alignment being done non-symmetrically (imagine turning the nearside track rod end adjustment 5 turns one way, and then the offside one 5 turns the other way, the tracking would be 'the same' but the rack would be off-centre, and the steering wheel would look 'off'), the actual natural position where the wheels want to track is maybe not where the rack and so sensor is 'centred'
So, are the wheels settling at equal toe, as they naturally do, and positioning the rack/steering wheel at say 1 deg lock, and the EPS is saying "1 deg off centre" apply some self-centering force, and doing it, and then it gets to 0 deg on the sensor, so it relaxes the force, and the steering then pulls again to equalise the toe, and so on, a circular feedback loop.
The logical fix would be to check that the alignment people set the steering rack centred properly (must be a way to check that the rack is actually dead centre, not the steering wheel, or the sensor, but the actual rack itself)...
Also, the sensor centre position can be calibrated/reset... but I'd check that what it thinks is centre is actually rack centre. Resetting it to cure the problem when the rack is offset from dead centre isn't ideal due to ackerman and camber/toe with steer geometry kinematics.
That is all just a theory for now, I'm not sure how it works in detail until I read more of these documents. But it would make sense and would generate the issue being noted.
Hmmm
Dave