Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[magnul]

As mentioned in another thread my fault code 92 don't want to go away. I hooked the sensor up to an oscilloscope this afternoon. First I checked the AC signal from the ECU with engine off:

Signal was the same with engine running and looks OK to me (with reference to http://www.picotech.com/auto/cam_ac.htm )

Then measured on the return signal (should have done them both at the same time, but only had one cable). With engine off:


And with engine running:


The return signal did not change with revs as I would expect it to do. Any ideas? Have I been really unlucky and purchased 2 non-functional sensors? Exactly what on the cam sprocket is causing the excited signal? Could this be the culprit?

-Magnus-

[Grumpy]

Exactly what on the cam sprocket is causing the excited signal?

I think it's a different height tooth on the Exhaust Cam Sprocket.
But don't hold me to that!

[Grumpy]

Been pondering your waveforms. I don't think you'll be able to tell
without producing both Primary and Secondary Waveforms together.

You need to watch for the Phase shift on the Secondary waveform
using the Primary as the base Waveform pattern. If you look at each
Waveform in Isolation, you can't tell if/when you get a Phase Shift.

So I think you're going to have to get that second lead, Magnul.

I'm as puzzled as you as to why we can't see a 'Break' in the Waveform
to indicate Camshaft position.

Do these excited Sensors use a different principle?

[magnul]

You need to watch for the Phase shift on the Secondary waveform
using the Primary as the base Waveform pattern.

You are absolutely right about that. I found out while reading:
http://opel-scan.ru/files/Opel_Simtec56.X.pdf   <-handy document for 2L owners
yesterday evening. I thought I was looking for some sort of spike for the position as well, but I guess that should show up a a lower frequency resolution (~10-100Hz).
The above run was mostly to look for 1) that the ECU is still alive and 2) that there was no interference spikes in the curves. I have already gotten another probe. I was surprised that I even managed to operate the scope:)

-Magnus-

[Dazzler]

You've lost me chaps, but if you ever find a Lamns reason for it then please let us know.

[Grumpy]

Thanks for the link, Magnul, very useful.
Nice to know we're thinking along the right lines.

Keep us posted.

[Marks DTM Calib]

Something not right there.....I calculate a 150Khz sine.

[magnul]

That's about what I calculated as well. Dunno if it means anything, as the two references above state 100-150kHz and 120-180kHz respectively.

[Marks DTM Calib]

I would infer that the ECU is working fine given the sine looks about right.

What I cant see is the signal from the cam sensor although given the time base used, you might not see it at idle. (the total scale would jsut show a single cycle at 5000rpm!)

[magnul]

I've finally found time to have another go with the scope. This time with logging both signal from ECU (blue curves) and return signal from sensor (red curves) simultaneously.

First engine off with fine time scale. There's a phase difference as it should be, and the voltaga amplitude is somewhat attenuated (scale is µs):


Looks pretty much the same with engine running:


As per advice, I reduced the temporal resolution:

This was the curve shape most of the time, but simetimes odd shapes showed up:

However, I don't know if this is real readings or just artefacts.

I don't understand much of this. One observation is that the example at the picotech page the attenuation is stronger in the return signal (to +/- 0.5V), whereas mine is +/- 1V

-Magnus-

[Grumpy]

In my opinion, Magnul, you do not have a phase shift.

Your scope readings, to me, would show that your wiring
is ok, but your Cam Sensor is either not working or not
getting the signal to cause it to phase shift.

Have had a determined Google, using various key words,
and have come up with this, which is specifically for the
X20XEV and Simtec set-up.

Look at the last 2 charts.
Notice the difference between the first, engine not running, and
the second with the Cam Sensor receiving a signal and Phase
shifting.
Notice in the first chart that he has inverted the 'B' reading
to show the difference clearly in the second chart.

If you inverted your 'B' channel reading, the 2 charts you get
from your scope are the same, ie.. no Phase Shift.

Off for a few days this evening, so if I don't reply to any of your
posts, I'm not ignoring you.  

http://www.katho.be/fddcw/simtec.htm

[Grumpy]

Will be going off-line, soon, for a few days, so a quick post of what
I would do in this situation.

Seeing as you bought after market items(?), I would be looking for
poor manufacturing tolerances in the Cam Sensor. I suspect that
the air gap between cam sensor and the target is critical.

So I would remove the cambelt cover and measure the air gap between
the new sensor and the target on the exhaust cam sprocket. As said before,
I believe that this is the teeth on the sprocket, but I am by no means certain.
It would be nice if someone could come up with the definitive answer.

I would then refit the original cam sensor and compare the position and
air gap readings with the new sensor.

I don't know how much 'wiggle room' you have when fitting the sensor, but if
enough, you can maybe get them to be the same.

Seeing as you have had a cam belt change, I would recheck the timing and
inspect the exhaust sprocket for correct fitment and positively identify what
on it provides the target for the cam sensor and that it is not damaged.
I would not take anyone elses word for this, I would want to check and see
for myself.

Hope that helps.

[Kevin Wood]

I'd say your problem is using a digital 'scope. What you are seeing with the timebase slowed down is aliasing between the (now too slow) sample rate of the 'scope and the 150KHz signal. This is probably masking any effects of the cam rotation (you should see a dip in the envelope of this signal each rotation).

If you've got access to an analogue 'scope have a look with that.

Kevin

[magnul]

Grumpy,
Thanks alot for that link. Incredible that I haven't found it myself after all these hours of browsing.  Very helpful indeed.
My signal has a 180degree phase shift (I haven't played with the curves), and I beleived that 180 was what I was looking for. Looks like it should be 90degrees even though I don't really understand whether it should be 90 only the instant the tooth passes, or constantly when engine is running.

Kevin; I suspected something like that. I don't think I have access to an analogue scope, and certainly don't know how to operate it.

At least I now know that the return signal is erroneous. Think it's strange if the sensor is position sensivtive. I have tried to move it around earlier, with no luck. Will try a couple more times and then go to a scrapper to find a working CID.

Thanks alot for all help guys.

[magnul]

Found a working CID at a scrapper today. Soldered it in, and the problem disappeared!   Bad luck to get 2 new non-working sensors. Sensor from the scrapper cost about £4, so I know where to go next time my cam sensor dies.

Thanks alot for all help. Think I'll spend the better part of the afternoon enjoying high revs now;)