Sale ends todayGet 30% off any course (excluding packages)
Ends in --- --- ---
Discuss all things tuning in this section. News, products, problems and results.
I'm having issues controlling the dual VANOS on an S50B32 E36 M3. Increasing feedfoward with no PIDS, when I reach around 85.5% I get maximum advance (inlet) and am unable to retard the cam. I can then manually retard the camshaft via an external switch, however no luck through MoTec. I've attached some screen shots, of retarding the exhaust camshaft, one photo shows through the MoTec , I see a change in Actuator duty cylce output but not change in position.
Are there specific requirement's for pin assignments different to what I currently have? see screen shot or config.
The VANOS has been overhauled and benched tested.
Any help would be greatly appreciated!
It's not clear from your pics, does the M1 low side drive those solenoids? If so, have you removed the factory solenoid connectors that have the diodes in them?
Thanks for your reply!
It's Hi side drive, using M1's half bridge outputs. I've attached another photo to hopefully make it a little clearer, after making a few wiring & calibration changes.
Green arrow - Increasing the feed-forward slowly with no PIDS, at 85.5% the intake camshaft advances to maximum position and the actuator output duty cycle also begins to change. At this point the actuator duty cycle, is very unstable and I am unable to retard the camshaft from this position.
I cannot seem to get a PWM control on the camshafts (even running on the dyno) and they just go to maximum position, and unable to revert back to zero position?
I just had a quick look at M1 tune as I haven't done any engine with cam control with one. But it looks to me that the control logic is a lot like Link so the feedforward table with a dual solenoid set-up will not behave like it would in a normal single solenoid cam control strategy. One direction will be "+ve" duty cycle and the other direction is -ve DC. So in theory normal feedforward is zero. I suspect the feed-forward probably needs a small value to compensate for the solenoid deadtime - as soon as you get over that deadtime the cam should start to move. But you will never get it to retard using the feedforward table because you would need "-ve" feed-forward value to do that and it doesn't look like it will allow you to enter -ve numbers.
I have no idea how significant deadtime is on these solenoids but if it were me I would set feed-forward to something like 20% as a starting point and start giving it some P until it starts to show response. Output frequency needs to be quite low on these, we normally use 50Hz, but up to 100Hz will probably work.
You might want to check out this thread on the MoTeC support forum:
You might want to post your support questions there. Seems like there is more VANOS experience to be found...
Here is a much older thread regarding dual VANOS that may be interesting...
That first thread on the Motec forum is for an M54 which is just a single low-pressure solenoid like most common vvt systems, so all those comments are irrelevant to this one.
The 2nd old thread about the S62 though is a very similar high-pressure vanos to that being discussed in this thread, it was nice to read my guess as to how it worked was pretty close.
Totally agree though, this probably would be solved a lot quicker on the Motec forum or by the dealer that sold the ecu.
Thanks for the input guys, great food for thought! I did read the top thread, but don't know how I missed the second one! Thanks David. I posted here because the high-pressure dual VANOS is less common and a bit more niche, figured there'd be some more specific info I could learn here. Which I have because it makes sense, currently I can advance one way but only retard the camshaft by swapping polarity. So by setting feed foward to overcome latency, I should be able move the camshaft via Proportional Gain. So I'll proceed with your recommendations Adam, I reckon that's the money shot :)
Just an update, this fixed the issue. I now have full VANOS control, Thanks again!