BMW S62B50

[Matt Cramer]

I did not see a loss of sync in the tooth log (you may need to make a longer one) but I think it's filtering out a real tooth on the crank sensor based on the composite log.

[ragepower]

I did not see a loss of sync in the tooth log (you may need to make a longer one) but I think it's filtering out a real tooth on the crank sensor based on the composite log.

I did a bigger one, I leave on rev limiter like 4 seconds and click on stop button 5 seconds latter.

I bought a picoscope and I probe gnd on v3 board. And I leave a print with vr sensor (blue) and gnd (red). It seams to much noise to me, what is your opinion?

I used the strobe, and with fixed angle on 20º. At idle it was on 20º mark, on 6000rpm it was on 15º mark. Is it normal?

[ekam99]

I did not see a loss of sync in the tooth log (you may need to make a longer one) but I think it's filtering out a real tooth on the crank sensor based on the composite log.

I did a bigger one, I leave on rev limiter like 4 seconds and click on stop button 5 seconds latter.

I bought a picoscope and I probe gnd on v3 board. And I leave a print with vr sensor (blue) and gnd (red). It seams to much noise to me, what is your opinion?

I used the strobe, and with fixed angle on 20º. At idle it was on 20º mark, on 6000rpm it was on 15º mark. Is it normal?

Normal. Just add 139 us to the spark hardware latency.

[ragepower]

I did not see a loss of sync in the tooth log (you may need to make a longer one) but I think it's filtering out a real tooth on the crank sensor based on the composite log.

I did a bigger one, I leave on rev limiter like 4 seconds and click on stop button 5 seconds latter.

I bought a picoscope and I probe gnd on v3 board. And I leave a print with vr sensor (blue) and gnd (red). It seams to much noise to me, what is your opinion?

I used the strobe, and with fixed angle on 20º. At idle it was on 20º mark, on 6000rpm it was on 15º mark. Is it normal?

Normal. Just add 139 us to the spark hardware latency.

Ok thanks

Another question.

This is my vr sinal, before and after vr conditioner circuit.

20151104-0001_01.gif

Is it normal? Would be the vr wired backwards?

[ekam99]

MAX9926 (used in ms3-pro, usquirt, and some newer conditioners) and lm1815 (old conditioners) flip the polarity of a signal on the output, i.e. the positive half-period of a sin becomes zero, the negative - 5v. If nothing else is in use, you want to trigger on the center of the tooth = rising edge of the output signal.

[ragepower]

Today I did some more road tests but the road was to slippy.

I readjust the vr pots to what is on ms manual.

It seams to be the same.

I attach the tooth log and data log.

https://www.dropbox.com/s/e0jzo895kwf5z ... 1.msl?dl=0

Any help?

[ekam99]

Do you have a VR conditioner?

[ragepower]

Do you have a VR conditioner?

Yes. The v3.0 vr conditioner circuit. And I have a 10k resistor in line on vr+.

[ekam99]

looks like the conditioner does not "see" the missing tooth. Though it can be confirmed with a tooth logger, try 20k in line instead 10k.

[Matt Cramer]

Today I did some more road tests but the road was to slippy.

I readjust the vr pots to what is on ms manual.

It seams to be the same.

I attach the tooth log and data log.

https://www.dropbox.com/s/e0jzo895kwf5z ... 1.msl?dl=0

Any help?

That tooth log looks pretty clean. Can you get a composite log showing the sync loss? Usually they're not ideal here, but at least they have a flag for showing the sync loss.

[ragepower]

Could I'm doing the tooth log wrong and missing the sync loss?

I'm doing this way, rev to 5000rpm, click start, rev to 7200rpm (rev limiter, where I have sync lost) wait 5 seconds, and click stop.

[Matt Cramer]

That should get the problem.

[ragepower]

The car had gone to an off-road event, and on the end of it, the costumer had washed inside and offside. Inside he tried to be careful to not spray too much water to the ms box (ip 56 isolation) and to the harness. After that the car didn't start, he put a dehumidifier inside it and next day it just started but seams to misfire. Before start the coolant shows like -37ºc when was like 15ºc and the I had black smoke when accel. After cleaning some plug with electrical spray the coolant was ok but the air temp sensor was like 10ºc above. Meanwhile the afr2 gone to full lean, I test the sensor outside the pipe with a butane lighter and it was ok. Put it on the pipe again, start the engine and it was reading ok.

I checked the spark plugs and all are equal, just a bit dark.

By my analysis I notice this points:

- MAT 10ºC above room temp.
- lost sync appear sooner (5000rpm) and more times
- knock 8 to high
- engine sound like to have one cylinder less (this happen time ago at idle after reving, I touch the tps and it goes ok)
- Laptop seams to loss connection with MS sometimes

It seams to be could be a wiring problem, since after the costumer washed inside/outside the symptoms get worst.

datalog
https://www.dropbox.com/s/9vyqehkjr1n1j ... 4.msl?dl=0

tooth log:
https://www.dropbox.com/s/27t1e78lvzjxn ... 9.csv?dl=0

composite log:
https://www.dropbox.com/s/iu5zqgbutd19m ... 9.csv?dl=0

[ragepower]

I dont know if is relevant.

I have MS, iox can board, bip's373, tip125 vanos controler and a few more things inside a ip56 box. On the box face I have this molex female plug and the corresponding male plug.

[ragepower]

I decided to test the bip373 circuit. I used a 12v 55w bulb as pull-up resistor.

And on the 2 cylinders that was the highest knock levels, the ripple on scope was different (higher).

On cylinder 1,2,3,5,6,7 was this.

https://www.dropbox.com/s/k9g33t71g0eetxu/1cyl.png?dl=0

On cylinder 4 was this.

https://www.dropbox.com/s/0erzsk4flgze2d4/4cyl.png?dl=0

On cylinder 8 was this.

https://www.dropbox.com/s/7b7bd09tlzmgrwb/8cyl.png?dl=0


Could be a problem with cylinder 4 and 8 circuit?

Could this cause the misfire sound?

[Matt Cramer]

This looks like it's basically an inductive flyback spike in the output side of the BIP373. However, the BIP373 does not do any sort of clamping until the flyback spike hits several hundred volts. I don't think the spike means very much here.

I'm not seeing any loss of sync events in the tooth or composite logs - but the composite log is showing parts that are completely missing, and the data log clearly has a loss of sync problem. My guess is the loss of sync is happening at the parts where the data goes missing. I would consider scoping the power feed directly at the ECU and maybe scoping VREF too. See if there's any noise there.

[ragepower]

This looks like it's basically an inductive flyback spike in the output side of the BIP373. However, the BIP373 does not do any sort of clamping until the flyback spike hits several hundred volts. I don't think the spike means very much here.

I'm not seeing any loss of sync events in the tooth or composite logs - but the composite log is showing parts that are completely missing, and the data log clearly has a loss of sync problem. My guess is the loss of sync is happening at the parts where the data goes missing. I would consider scoping the power feed directly at the ECU and maybe scoping VREF too. See if there's any noise there.

OK.

How I identify the noise on scope?

Or is better to post where?

[Matt Cramer]

VREF should be totally flat. If it moves by more than 0.1 volt, something's very wrong.

The 12V line isn't as clean, but if it spikes by more than 1.0 volts, that also should be investigated.

[ragepower]

It can be something with VREF since the problem get worst when I plug in the laptop.

A simple capacitor would solve the problem?

[ragepower]

print.JPG

Check the measurement below on image, "pico a pico" is "peak to peak" with the engine running.
A= 12V on MS
B= 5V on MS (sometimes show 280mv peak to peak)


Also checked the voltage drop from battery to MS:

- ground: 6 mV
- 12V: 700mv

This is too much right?