RomRaider

Hi. I wanted to change some things in a friends 1996 M52 320i. It had non-flashable MS41.0 from factory. Therefore he bought a newer flashable one (MS41.0 Index 04), that's from 320i too. We've successfully turned off EWS, changed rev limiter, etc. The car starts and runs but CPS doesn't work with the new DME. Engine acts like if CPS was bad and it gets better if it's disconnected altogether. If we swap in the original DME, CPS works just fine. Any idea what's going on?

Thanks in advance.

Are the sensor part numbers different? Have you cleared adaptations?

Some cheap aftermarket (non BMW) crankshaft position sensors are more prone to not working properly on some MS41. There has to be some slight differences in the hardware that make aftermarket sensors not work correctly.

That sensor need to be original. Original sensor will work every time with every MS41.

Wow, I didn't expect such quick reaction. Serial number of CPS is the correct one. I've not cleared adaptations yet, I'll try that first. If it doesn't help, it's probably what Enabled says. The CPS is relatively new but not a BMW one. Thanks for the tips, I'll report back after we try.

Also big thank you to everyone who participates in any way to development! It's incredible what MS41 is capable of thanks to you.

EDIT: Actually, the crankshaft position sensor is BMW one. Maybe the cable is bent and that causes the voltage to drop more and makes the less sensitive DME deaf. I never understood why crank and camshaft position sensors tend to break so often when they're such a simple thing.

These ms41 units are starting to get old and there seems to be all kinds of problems with the position sensors. Normally the ecu just complains about crank or cam sensor fault and when you replace it, the engine will work fine for moment. But short after the fault will come back.

I have tried to do investigation about what causes the problems, but I still don't have very good idea about what causes these problems. First thing you could try is to measure the +5v supply from that non working ecu for cps. It may have gone down from that which causes cps errors. Just disassemble the ecu connector and measure from the pins with ignition on. The lowered 5v supply seems to be some kind of common problem. Then you could try to measure that the cps sensor gives +5v back when the tab on trigger wheel meets the cps sensor. It's hakl type so this can be done with regular multimeter. And then make sure the sensor gives 0 volts. When the sensor is in the middle of two tabs. But this tells only about the sensor and/or car wiring.

Cam sensor seems to be maybe vr type which is driven with unknown frequency from ecu and then the half moon on cam FM modulates the signal. I don't yet know way to check this.

Well if nothing else helps, I'll surely play with the "broken" DME. It's a shame they might be loosing crucial functionality after the years.

I've not tested anything mentioned yet since we haven't met with the friend since. But I started reading as much info as possible about MS41 modding since there are no step-by-step single purpose guides and you either know a lot or can do only very basic stuff. I dream of running turbo setup with MS41 management so all knowledge might come in handy one day.

Anyway I've stumbled upon some files in the wiki. One of them was PDF probably meant for US mechanics (it refers to MS41.1 and MS41.2 and is in English). It contains information about all things connected to DME. Here's what it says about cam shaft sensor. Probably won't be useful for anything in particular right now but it might be interesting for your mastermind Pazi .

Thanks that is quite interesting. And knowing that frequency makes diagnosing the cam signal fault easier.

Some more info. The DME comes from a car that was very poorly swapped to M54B22. It was running with M52B20 intake manifold and this MS41.0 DME.

We've cleared adaptations and flashed new partial BIN from this forum just to be sure. It didn't change anything. Then we've cleared engine codes through INPA and run the engine. Only code that reappeared was camshaft sensor as seen on the picture bellow. After swapping in the original ECU and clearing the codes, nothing reappeared. My theory is that M54 damaged the camshaft sensor circuit of the DME in some way. As said before, the camshaft sensor is OEM BMW and works fine with original ECU.

For now we're probably done with it. The only thing my friend wanted was higher limiter with higher hysteresis and it's not worth the effort for him.

Thanks again for your help and work.

ms41 is too sensitive to sensor's quality. My guess is that if you replace CPS with latest original one the problem will disappear.

Very interesting.

I converted from MS41.0 to MS41.2 software and have the same problem with the camshaft position sensor, even though I can't tell if the problem existed before the conversion.

In any way, I noticed that I when I drive around and log my data, the value "Ignition Retard Current" always goes to maximum ignition retard (-7° to -12°) once the I reach the load where knock control activates, so even very mild loads...

I guess, that this is a safety measure as the ECU now doesn't know which cylinder is firing and which is at valve overlap.

I already ordered a new position sensor (Original BMW part... around 125€...Jesus Christ thats expensive) and hope that this fixes my problem.
I digged through the net and it seems like this sensor likes to break quite often. Also it seems like its defective out of the box sometimes... oh well, lets hope for the best.

I will report back.

The fault probaply has been there before the conversion. It has no effect on that. And yes the ecu goes into some kind of safe mode when the cam sensor doesn't work.

I have done some investigation and seems that the non-OEM sensors basically work, because there is clear signal coming back from the sensor and the phase shift is there. But the amplitude of the returning 120kHz signal from sensor isn't as high as the ecu input. With genuine OEM SIEMENS sensor the amplitude is virtually same. There are some sellers that claim to have oem sensors but those are really some hella units or something like that. My car had Delphi sensor even thought the recipe from repair shop said that they had installed oem sensor (and also billed according to that). (I bought the car as non working). But needles to say that the delphi sensor was the reason that the cam signal didn't work. Now with genuine siemens sensor, all works. I tested with four different ms41 units and non of them worked with the delphi one.

And about the cps. In my testing even all the cheap cps sensors did work just fine. But it isn't big suprise, because the cps sensor is just regular 5v hall sensor. It's very common type and only puts out 5volt square wave.

Okay, problem solved. The expensive new original part seems to do the job.

No more ignition retard and proper readings of the VANOS-angle. The car pulls like it never pulled before and the ECU finally adapts fuel and knock tables.

Out of interest I pulled back the rubber sealing at the connector of the old sensor and as it had to be expected the isolation has fallen off from the cables. This explains some things...

Yes this seems to be the reason why old sensors fail. Basically you could try to replace the wires, but i don't know how well it would work on the sensor end.

Edit: removed misleading info. Didn't fully know how my scope functions.

Pazi, could you maybe make some resistance measurements on a known working OEM/Siemens sensor?

When you were looking at it on the scope, was the returning 120kHz sinewave signal inverted in relation to the ECU output signal?
By phase shift, do you mean:
- a constant (~180*) phase shift that is present all the time between the ECU output and the sensor output
- phase shift generated on the signal by the camshaft "lobe" in front of the sensor?