When you click on links to various merchants on this site and make a purchase, this can result in this site earning a commission. Affiliate programs and affiliations include, but are not limited to, the eBay Partner Network.
Looking at the first screen capture, the waveform looks good per Porsche documentation, other than the fact that the polarity is inverted.
You saw more amplitude because the sensor was unterminated; i.e. not connected to the DME, hence no load on it.
I would state that flywheel sensor is working correctly.
Has the sensor been replaced?
I'm asking because if you compare the two waveforms, the Porsche drawing shows the initial positive pulse has the highest amplitude and the negative going pulse has lower amplitude.
Yours is the opposite, as if the sensor has been wired in reverse.
I'm not saying that it won't work because it does appear that you have sufficient amplitude to trigger the DME.
It's just that your positive going part of the waveform is lower in amplitude than the Porsche drawing, and the negative part of your measured waveform has greater amplitude.
Has the sensor been replaced?
I'm asking because if you compare the two waveforms, the Porsche drawing shows the initial positive pulse has the highest amplitude and the negative going pulse has lower amplitude.
Yours is the opposite, as if the sensor has been wired in reverse.
I'm not saying that it won't work because it does appear that you have sufficient amplitude to trigger the DME.
It's just that your positive going part of the waveform is lower in amplitude than the Porsche drawing, and the negative part of your measured waveform has greater amplitude.
I did replace the sensors about 2 1/2 years ago, but I didn't mess with the gap or the wiring.
I did notice that about the negative side being lower. I thought it was strange - however, if I reverse the wires on my scope, it flips the image both vertically *and* horizontally. So I have the larger spike on top, but the pulse starts with a negative slope!
Below is a picture of it with the scope leads reversed (the scale is slightly different to the last one).
When I attached my test leads inside the DME connector I forgot to keep track of which one was which, and so this is the first image I actually saw, until I swapped the probe connections.
When you say I have sufficient amplitude to trigger the DME, is that based on the 1.2v I'm getting, or based on the fact that the engine started? Because it sometimes doesn't start, and when it does it rarely runs perfectly, and this is pretty much the only thing I can find wrong anywhere.
Have you checked continuity between each sensor's shield wire and chassis ground? Perhaps one sensor is grounded through the mount, but the other not. Maybe that's why one has noise and one doesn't?
Have you checked continuity between each sensor's shield wire and chassis ground? Perhaps one sensor is grounded through the mount, but the other not. Maybe that's why one has noise and one doesn't?
I only checked against the battery negative terminal (they were both grounded with near zero resistance). I know they are grounded differently: the reference sensor is grounded to DME pin 5, while the speed sensor shield shares pin 23 with the O2 sensor shield. Internally these grounds are connected together.
EDIT...I just had a thought. I had assumed the reason the noise went away in today's running test is because the shield was grounded in the DME. But what if the reason is, because the starter causes the noise? Still doesn't explain why it's only one sensor, but it's something else for me to test next chance I get....
The amplitude of the pulse is affected by the distance between the sensor body and the reference sensor pin and/or flywheel teeth.
If your scope is dual channel check what both sensors look like on the scope while starting and running.
I believe that one sensor gives you a pulse at TDC reference mark (the left side drawing), and the other flywheel teeth for the speed sensor (the right side drawing).
This is covered in the Porsche Workshop Manual Vol. 1 - Engine, 8 valves in section 28 page 25-26, which is a good reference for the troubleshooting that you are doing.
I'll see if I can take some scope shots this week for comparison. In the meantime, have you peeled back the rubber boots on the connectors at the back of the intake and inspected for frayed wires? Intermittent problems are tough to diagnose with the scope like that -- because you don't really know how it looks when it's not working... The speed and reference sensors feed into a proprietary chip (S100) on the DME that produces digital 5v pulses -- one for each sensor -- on the hardware interrupt pins 12 and 13 of the 8051 processor. If you're interested, you could watch those pins on a scope to see if the 8051 is getting those digital pulses. As long as the 8051 is getting a pulse, that's all that matters, and it's a yes or no thing with no room for interpretation. Here is a scope shot of he signals once processed by the S100.
Thanks for the pic! I have not peeled back the boots. Their peeling days are over I'm afraid. I think I will replace the harnesses no matter what at this point. But the signals are very consistent at the sensor connector vs. the DME connector.
What I'm concerned about now with the low amplitude is that it might not be high enough to reliably trigger the converter, as described on Rogue's website here. I was also worried that the severe noise might make it trigger slightly too early or late, but seeing that the noise is not there during running, I'm not too worried about that now (but it could still be contributing to my occasional no start). I am tempted to open the DME and take a look at the output of the S100 just out of curiosity.
Thanks for the pic! I have not peeled back the boots. Their peeling days are over I'm afraid. I think I will replace the harnesses no matter what at this point. But the signals are very consistent at the sensor connector vs. the DME connector.
What I'm concerned about now with the low amplitude is that it might not be high enough to reliably trigger the converter, as described on Rogue's website here. I was also worried that the severe noise might make it trigger slightly too early or late, but seeing that the noise is not there during running, I'm not too worried about that now (but it could still be contributing to my occasional no start). I am tempted to open the DME and take a look at the output of the S100 just out of curiosity.
If you crack it open, you want to check on pins 12 and 13 of the processor itself, as the S100 has a few passive components supporting its output. Having said that, if the motor is running, then it's a near certainty that the 8051 is getting a pulse from the S100. I suppose it's possible that the S100 is dropping the pulse every now and then from a weak sensor waveform, but not sure how you'd capture that on the scope. 99.537% of the time, intermittent speed/ref sensor problems can be traced to frayed wires at the connectors on the back of the intake.
11-23-2014, 07:19 PM
