rusefi.com

I hope I am not bombarding too much in here. I am finally able to put real miles on the car and am working through issues as they arise.

I am working through revving the engine out to limiter now and I am getting RPM spikes. These spikes are causing the limiter to kick in early. The spikes according to the logs seem to be double the immediately previous RPM. The engine is a BMW M42 4 cyl 1.8L with 60/2 trigger wheel with the stock VR sensors on the Crank and Cam. RusEFI is set to "VW 60/2". I know there was some contention with this trigger setting, but I didn't want to change anything without some confirmation or suggestion. I used this trigger setting for first start and it has been working so far. I have attached a log where the spikes can be seen several times. Looking at the vehicle speed therea re also spikes high and low around the same times as the RPM spikes are happening.
Examples in the linked Log in seconds 95.309, 107.559, 444.964 - 448.562, 766.339 - 768.584

Proteus 0.4
Firmware:v20211008@28919
Tune:https://rusefi.com/online/view.php?msq=689
Log: https://rusefi.com/online/view.php?log=386

At the moment your tune says
* "I have 60/2 on the CAM" (that's not what you want)
* I am hesitant about CAM - I have input configured but more is inactive

Complexity of trigger configuration is a very known issue unfortunately. https://github.com/rusefi/rusefi/issues/3401 is the most recent one, some minor improvement was actually done under https://github.com/rusefi/rusefi/issues/3401 please update your bundle to see fresh project, it would hopefully make it easier to spot "60/2 on the CAM" drama

considering your inactive cam it's hard to say while we have so many issues but it could be similar to https://github.com/rusefi/rusefi/issues/3269 which was closed but who knows

Please keep those amazing reports coming!

I apologize, I linked my most current uploaded tune without checking it. I made cam changes before taking this log.

Here is the correct Tune that was loaded when taking the log: https://rusefi.com/online/view.php?msq=692

Well, that explains how you have the hiccup - you've fixed half of configuration issues.

Please consider addressing the other half of configuration issues:
please consider upgrading to latest bundle which would make your configuration issue more explicit

I've reopened https://github.com/rusefi/rusefi/issues/3269
Right now I do not have an explanation. Let's start by updating project .ini and fixing your trigger setup. I bet issue would stay but maybe it would not?

AndreyB wrote: ↑

Tue Oct 26, 2021 6:09 pm

Right now I do not have an explanation. Let's start by updating project .ini and fixing your trigger setup. I bet issue would stay but maybe it would not?

vvt position ~= double his trigger offset, and he's somehow in "4 stroke cam" mode and that works? Double his trigger offset smells funny, like the same thing nik saw.

I think we are both trying to say "your RPM seems doubled please fix and take fresh logs"

Test #1: Set "operation mode" to "4 stroke without crank sensor", re-check trigger offset with timing light. Set cam inputs to NONE. Set spark mode to wasted spark if not already. Take a log. Does the blip still happen?

Test #2: Change nothing else, but add the cam input as it's configured now. Take another log. Does the blip still happen?

mck1117 wrote: ↑

Tue Oct 26, 2021 6:22 pm

Test #1: Set "operation mode" to "4 stroke without crank sensor", re-check trigger offset with timing light. Set cam inputs to NONE. Set spark mode to wasted spark if not already. Take a log. Does the blip still happen?

Test #2: Change nothing else, but add the cam input as it's configured now. Take another log. Does the blip still happen?

I am updating bundle now. I should be able to complete these tests in about an hour and will respond with results.

I must have messed up the trigger offset set up when i set it originally(with a junky old timing light. When checking with a nice newer timing light today I had to change it from 114 to 124 and now it is all lining up.

When set to no cam and wasted spark there are no spikes and no hiccups.

When trigger is set to 4 stroke with cam, changing nothing else the spikes returned. It seems to be only when over 4000rpm. you can see in the log there are no spikes until 528.650 and that is the first time the engine is over 4000rpm.

Tune with corrected offset and trigger set back to no cam: https://rusefi.com/online/view.php?msq=694
New Log: https://rusefi.com/online/view.php?log=388

Having a look....

It does sound like it is indeed something weird with the cam sync, then.

Can you describe your hardware setup? The crank signal all looks fine to me. Where is your cam sensor? What kind of sensor? What is the target the sensor is sensing?

Can you share a snip of the rusEFI console's engine sniffer, just running at idle? I'm interested in what your cam signal looks like in relation to the crank.

mck1117 wrote: ↑

Tue Oct 26, 2021 10:56 pm

It does sound like it is indeed something weird with the cam sync, then.

Can you describe your hardware setup? The crank signal all looks fine to me. Where is your cam sensor? What kind of sensor? What is the target the sensor is sensing?

A few things to try:

Can you share a snip of the rusEFI console's engine sniffer, just running at idle? I'm interested in what your cam signal looks like in relation to the crank.

The cam sensor is VR and is on the exhaust cam. you can see the side of it in the second picture. The target is a single peg sticking out of the cam gear. both cams have the peg but only exhaust has a sensor. TS would not allow me to set the intake cam sensor to none so i have it set on the intake cam.

I can take a console log here shortly. I am eating dinner right now.

Cadence wrote: ↑

Tue Oct 26, 2021 11:24 pm

The cam sensor is VR and is on the exhaust cam. you can see the side of it in the second picture. The target is a single peg sticking out of the cam gear. both cams have the peg but only exhaust has a sensor. TS would not allow me to set the intake cam sensor to none so i have it set on the intake cam.

It's no big deal if it's not *actually* on the intake cam, since you don't actually have VVT. Mine is configured to intake cam, but it's a single cam pushrod engine, so it's *also* the exhaust cam.

maybe a bad sensor?

what engine sniffer looks unusual. why does it not show legend?

This is from the bundle I downloaded today. The Legend shows only when I mouse over it. Could be potato computer.

Cadence wrote: ↑

Wed Oct 27, 2021 12:13 am

image.png

maybe a bad sensor?

Hmm, maybe. It's possible that the VR interface on Proteus is unhappy about your single tooth VR setup. Possible to replace with a hall sensor in the same hole? Alternatively you could just run wasted spark and not worry about the cam sensor at all. My approximately similar setup on my racecar (4cyl turbo, 60-2 crank, individual coils, etc) runs great with no cam sensor whatsoever.

AndreyB wrote: ↑

Wed Oct 27, 2021 12:15 am

what engine sniffer looks unusual. why does it not show legend?

he's also zoomed suuuper far in on the cam tooth

Kind of side note - note your warning count / last warning.

Ideally we want to address all of these one by one

I have seen pictures of a cherry hall adapter. I may try my hand at making one. In the meantime I will set it back to wasted spark. My goal is to take advantage of all of the features I possibly can. And hopefully provide good information back.

I will take a look at the warn count and last warn when I get back to it tomorrow and address those as well.

AndreyB wrote: ↑

Wed Oct 27, 2021 2:08 am

Kind of side note - note your warning count / last warning.

Ideally we want to address all of these one by one

These are the warnings on these logs.

CUSTOM_VVT_SYNC_POSITION = 6675,

CUSTOM_RE_ADDING_INTO_EXECUTION_QUEUE = 9011,
/**
* This indicates an issue with coil control - pin was not high when we were trying to set it low.

CUSTOM_OUT_OF_ORDER_COIL = 9012,
/**
* Commanded fuel exceeds your fuel injector flow


It looks like all these are related to the cam sync issues. The fuel injector duty cycle is spiking over 100% when the rpm is spiking. I will take a log on wasted spark tonight and see if these return or if there are any new or other warns.

As another side note. Is it possible to add a coin cell to Proteus 0.4 for real time stamps on the logs?

batter would be great question for proteus thread

back to trigger error codes are at https://github.com/rusefi/rusefi/blob/master/firmware/controllers/algo/obd_error_codes.h

The period of a single tooth cam would be over 85ms on low rpms so could this be a case of the dreaded watchdog?

Στάλθηκε από το VOG-L29 μου χρησιμοποιώντας Tapatalk