ETC oscillation issue with BMW S54B32.

[PeepPaadam]

Project is customers race car, BMW E92 with S54B32 engine. Factory harness, plug and play adapter and S54 VEMS ordered from webshop with double-vanos and ETC support.

Problem is oscillating throttle when TPS is connected, when disconnected oscillation stops (TPS shows ~60%) and obviously motor opens throttle only when pedal is pushed past 60% and then it opens to full.

All the connections have been verified (polarity of TPS/PPS GND and 5v), analogue input pins etc. Should TPS2 (on the throttle motor) show ~4.8volts when throttle is closed?

Here's current config file (please review ETC part and VANOS part): http://www.vems.hu/files/PeepPaadam/BMW_S54B32_pnp.vemscfg

And here's video of the issue: https://www.youtube.com/watch?v=487Q0X98_5I

Please could you take a look at this as the 1000 km endurance race is already after a month and we really need to do some testing too.. Thank you in advance!

[MWfire]

Put I active range to 0 or new firmware.

[VEMS]

Hello Peep,

A lot of changes have been applied to ETC code to make it easier to tune since 1.2.23 (see: http://www.vems.hu/wiki/index.php?page=GenBoard%2FUnderDevelopment%2FFirmwareChanges) for proper (and easy control) it is imperative you upgrade to 1.2.31 or later.

If after this you still have tuning problems, please upload a vemslog  (or create sharing) of ETC operation doing a few slow pps 0-100% sweeps and link it here i'll review and provide hints.

Best regards, Dave

[PeepPaadam]

I can flash 1.2.32 tomorrow, but would anyone have 1.2.32 ETC portion to share for S54B32? I can make calibrations and corrections myself, but would really appreciate if someone have working settings so I can eliminate the tuning issue here.

At the moment I am just not sure whether it is tuning issue or something else (though all connections were correct afaik, made per instructions and verified later).

[Kamuto]

Project is customers race car, BMW E92 with S54B32 engine. Factory harness, plug and play adapter and S54 VEMS ordered from webshop with double-vanos and ETC support.

Problem is oscillating throttle when TPS is connected, when disconnected oscillation stops (TPS shows ~60%) and obviously motor opens throttle only when pedal is pushed past 60% and then it opens to full.

All the connections have been verified (polarity of TPS/PPS GND and 5v), analogue input pins etc. Should TPS2 (on the throttle motor) show ~4.8volts when throttle is closed?

Here's current config file (please review ETC part and VANOS part): http://www.vems.hu/files/PeepPaadam/BMW_S54B32_pnp.vemscfg

And here's video of the issue: https://www.youtube.com/watch?v=487Q0X98_5I

Please could you take a look at this as the 1000 km endurance race is already after a month and we really need to do some testing too.. Thank you in advance!

will meet you there

[PeepPaadam]

Lol, hopefully, if we get the car running But yes, it's the Palanga 1000km race. BTW, car is from Latvia and the drivers have participated many times before.

[PeepPaadam]

I uploaded 1.2.32, calibrated PPS/TPS/1/2 and inserted recommended PID values from help file. Throttle behaves normally now, but only issue is that it does not open fully. Pedal is pressed to 100%, but TPS and TPS2 show 98% and I see that throttles are not fully open.

Also is there any way to verify VANOS solenoid operation? I mean to verify that outputs are in fact correct (which output and polarity, e.g. i4/5 or 4/i5).

Here are config and log files:

http://www.vems.hu/files/PeepPaadam/BMW_S54B32_1.2.32.vemscfg

http://www.vems.hu/files/PeepPaadam/v3.3_u010280-2015.06.11-14.56.08.vemslog

[VEMS]

Hello Peep,

Verifying output polarity is easy, is the cam control regulating in the right direction ? E.g. when target is higher than measured position does cam position move up ? If not invert polarity.
A small manual engagement test (disconnected from ecu) will show you if you have the right wires (e.g. tap intake solenoid control wire to +12v intake cam should move, tap other intake solenoid wire to +12v intake cam should move in other direction).

But lets take it one step at a time, i reviewed your vemslog (v3.3_u010280-2015.06.11-14.56.08.vemslog) perhaps you missed some things so i'll point them out:

1) there are secondary trigger errors, solve those first no need to test cam control when the engine is not running without trigger errors.
2) third trigger position rolls over most likely bad measure tooth configured

Please supply a triggerlog of engine cranking 8-10 seconds without INJ and IGN fuse to verify correct trigger polarity and secondary/third trigger operation.

Than to the matter of ETC control:

1) apart from applying some default values it does not seem tuned at all for proper operation tune it.
2) ETC integral scaling curve needs to be filled according to default in help for proper operation.
3) tps2/pps2 curve is setup non linear is that intentional ?
4) ETC reference curve does not make much sense, tune it with PID control disabled.

I hope this is enough to get you started; If you supply the triggerlog i'll look into your trigger issues.

Best regards, Dave

[PeepPaadam]

PPS2/TPS2 curves are what the voltages were according to TPS/PPS positions, I did not make them non-linear intentionally.
I got the throttles to work satisfactory after tuning ref DC table, now it responds ok and opens fully.

But I still have sec trigger errors. When cam sensors are disabled, I can start and run the car with crank sensor only (and coils in wasted spark mode) perfectly, no errors but obviously that is not an option since VANOS does not work then.

Here are current config, log and triggerlog files:

http://www.vems.hu/files/PeepPaadam/BMW_S54B32_1.2.32_trig_err.vemscfg

http://www.vems.hu/files/PeepPaadam/v3.3_u010280-2015.06.15-13.55.13.vemslog

http://www.vems.hu/files/PeepPaadam/v3.3_u010280-2015-06-15-12.58.triggerlog

[VEMS]

Hello Peep,

I have quickly reviewed your triggerlog, you have swapped secondary and third trigger. The 8-1 needs to go to the secondary trigger, the even 6 pulse needs to go to the third trigger this most likely also means you need the swap the output channels for intake and exhaust side cam control.

Best regards, Dave

[PeepPaadam]

Thanks a lot for the prompt response!

Lol, this is confusing Because in the cam control dialog it says intake is second and exhaust is 3rd trigger, today I verified that they are and thought it is ok

I will get to the car on wednesday, will report back after that. Anything else I could check while there?

[VEMS]

Hello Peep,

I can see how it can be confusing, it is annotated in the S54 trigger configlet in VT (but you would need to open the file with text editor to see it):

# Trigger settings for, missing tooth 60-2 primary, 8-1 secondary and 6 pulse third trigger for 6 cyl BMW s54 engine double vanos with full camsync
# Primary trigger VR
# Secondary trigger HALL: Exhaust camshaft (8-1 pulses) needed for cam sync
# 3th trigger HALL: Intake camshaft (6 pulses), needed for campos for intake cam
# Coils are dumb coils, need internal IGBT in v3.
# Dual VANOS needs 4 PFETS v3 inside to control it.
# Firing order: 153624

Reason for this requirement, since the intake cam has 6 even spaced pulses (3 in both exhaust and compression phase) there is no way to sync to that, the exhaust cam has a pattern which allows us to differentiate between exhaust can compression phase (3 pulses in one, 4 in the other) hence it is used for camsync.

Best regards, Dave

[gunni]

Shouldn´t the ECU have selectable choices somewhere?

I.e I tell it I have Intake cam on 36-13 and exhaust cam on 18-x , then Intake is intake and exhaust is exhaust, I should also be able to just fit the intake or camsync to the 18-x input and use that for camsync rigth?

[VEMS]

Hello Gunni,

Shouldn´t the ECU have selectable choices somewhere?

I.e I tell it I have Intake cam on 36-13 and exhaust cam on 18-x , then Intake is intake and exhaust is exhaust, I should also be able to just fit the intake or camsync to the 18-x input and use that for camsync rigth?

I think you are stating some nice to have here, as you know this is not how it currently works.

Best regards, Dave