MartinS Posted October 6, 2021 Author Report Share Posted October 6, 2021 Spoke too soon, not solved, worse. ppl to follow. https://drive.google.com/file/d/1h-xr362RehfqTv8OJxnqHYwqy-ZpD2q_/view?usp=sharing Quote Link to comment Share on other sites More sharing options...
MartinS Posted October 6, 2021 Author Report Share Posted October 6, 2021 Pics of scope at idle, lower RPM with misfire, now about 1500 RPM, and surging at higher RPM (?2500). As noted, misfire now at much lower RPM becoming surging as RPM increases. I can't see any real difference with the crank trigger scope and still has some low voltages in the same places. Does this need rewiring completely? The only thing I notice is the intake cam trigger is down going at idle but flips slightly above idle to up going wave. I don't think this was occurring before, so problems appear to be multiplying. This is becoming a nightmare. https://drive.google.com/file/d/16arfOOc2uHNZUEYA1WbwYoqqnncUCW8Q/view?usp=sharing, https://drive.google.com/file/d/1QO6t9tAdk6fW1vHaR3pY5J25OuER--ZE/view?usp=sharing, https://drive.google.com/file/d/1xjKl5GY05Bx8q8KH9_H3uzfIHdnbgHWC/view?usp=sharing Quote Link to comment Share on other sites More sharing options...
Adamw Posted October 7, 2021 Report Share Posted October 7, 2021 Still looks like there is a ground missing or something. Did you do this yet: On 10/5/2021 at 9:48 AM, Adamw said: With crank sensor unplugged, measure resistance between ground pin at the crank sensor plug back to ecu ground (any ecu ground will do) and also to engine block. It should be near 0 ohms for all. Also on the scope, can you press the "CH2" button, then in the CH2 menu change the coupling to DC, and where it says "MeasCurr" or similar, change that to No. Quote Link to comment Share on other sites More sharing options...
MartinS Posted October 7, 2021 Author Report Share Posted October 7, 2021 Thanks Adam, tested the sensor, no problem identified. Did trace the ground wire, found a break, fixed it, didn't test again. Rookie mistake... Today I'm going to rewire the crank sensor as, even if I do find a wire break again, rewiring is much easier than pulling the intake manifold again. Plus, that wire's proven it's deteriorated. 9 hours ago, Adamw said: Also on the scope, can you press the "CH2" button, then in the CH2 menu change the coupling to DC, and where it says "MeasCurr" or similar, change that to No. Sorry, found the 'scope manual online and meant to do all that but sidetracked by thinking I'd found the problem. Will post more scope pics today. Any thought on the inverting cam signal? Thanks for all your help Adam, lost at sea without it. Martin Quote Link to comment Share on other sites More sharing options...
MartinS Posted October 7, 2021 Author Report Share Posted October 7, 2021 Crank trigger sensor now hardwired with shielded cable (cable cut from Link harness) from sensor to ecu, no sockets/plugs in between, just one approx 4-5cm section where wires soldered together. Continuity confirmed from sensor plug to ecu, both signal and ground. Scope pics linked, DC coupling. Cam trigger signal still inverts slightly above idle. Runs better, but still misfires above about 3000RPM and intermittent misfire at lower RPM. https://drive.google.com/file/d/168NvuAAqaoSl_m8pXPDhDJIWhrk7f3vW/view?usp=sharing, https://drive.google.com/file/d/1MsgB89saHr857SWB7EN35Y8Cx9ZuwFmz/view?usp=sharing https://drive.google.com/file/d/1qpEblmVKAXFoA0EqOzSCwpbICP3aj65A/view?usp=sharing Quote Link to comment Share on other sites More sharing options...
Adamw Posted October 7, 2021 Report Share Posted October 7, 2021 The cam signal inverting is normal for many Subarus and the trigger code is designed to tolerate that. Crank still doesnt look much different, and is still a very distorted waveform. I think it would now be worth sourcing another crank sensor to try, I cant think of much else that would be causing the distortion now with it direct wired. Quote Link to comment Share on other sites More sharing options...
MartinS Posted October 7, 2021 Author Report Share Posted October 7, 2021 Ok, another new one... Quote Link to comment Share on other sites More sharing options...
Adamw Posted October 7, 2021 Report Share Posted October 7, 2021 What do you mean, have you already tried a new one? Quote Link to comment Share on other sites More sharing options...
MartinS Posted October 7, 2021 Author Report Share Posted October 7, 2021 Yes, is a new one in there. Follow your logic though, can't see what else it is. If new one fixes problem, I'll work on a refund for the current on, if doesn't, I have a spare. Didn't keep the original for reasons I can't recall but regret... Quote Link to comment Share on other sites More sharing options...
MartinS Posted October 20, 2021 Author Report Share Posted October 20, 2021 New trigger sensor installed today. No change in misfire. Should say, another new trigger sensor Quote Link to comment Share on other sites More sharing options...
Adamw Posted October 20, 2021 Report Share Posted October 20, 2021 Do you know the history of this ecu? Any chance it could be damaged? Can you try a scope of crank sensor with the ECU unplugged. Quote Link to comment Share on other sites More sharing options...
MartinS Posted October 21, 2021 Author Report Share Posted October 21, 2021 The ECU was included in a package of parts I bought from the original owner. It included engine mount bar and mounts, header, ECU, and ECU wiring looms. I thought the seller was a Link dealer originally as he ran a shop in Phoenix specializing in VW Vanagon Syncro's but this turned out not to be the case. He had collected the parts to do this conversion but never completed it. The engine mounts, header, and wiring had definitely not been used; the package for the ECU was open but included all the original documentation and did not appear to have been used...I have no way of confirming that except when I first accessed it via computer, it required unlocking and had no setup installed. Can't confirm it isn't damaged though. The first scope images I sent you are cranking with the ECU disconnected and they aren't significantly different from with the ECU attached and cranking. I have seen other trigger scope images people have posted with the EZ30R (essentially identical trigger setup to the EZ36) and mine do not have as consistent voltages. The only thing I can come up with is that, in Canada, the EZ36 only came with an automatic transmission. I have installed a manual transmission but I'm pretty sure the flywheel diameters are identical and the distance from the sensor to the notches on the flywheel are not different. Regardless, the sensor/flywheel distance is fixed which does not explain the different amplitudes. Otherwise, I can't come up with a reason the crank sensor 'scope looks abnormal. I have tapped into the signal and ground lines directly from the sensor less than half a meter from the sensor. Other than the stock plug at the sensor which looks in good condition and has been cleaned/deoxidized, the wires have no interruptions/plugs/sockets. At a loss... Quote Link to comment Share on other sites More sharing options...
MartinS Posted October 21, 2021 Author Report Share Posted October 21, 2021 Could a ground fault of some kind within the ECU cause the abnormal 'scope tracings? Quote Link to comment Share on other sites More sharing options...
Adamw Posted October 21, 2021 Report Share Posted October 21, 2021 5 minutes ago, MartinS said: Could a ground fault of some kind within the ECU cause the abnormal 'scope tracings? Thats why I asked for a scope with the ecu unplugged - then it has no influence. Connect the scope direct to the sensor so there is no wiring involved at all. Quote Link to comment Share on other sites More sharing options...
MartinS Posted October 21, 2021 Author Report Share Posted October 21, 2021 Will do Quote Link to comment Share on other sites More sharing options...
MartinS Posted October 22, 2021 Author Report Share Posted October 22, 2021 https://drive.google.com/file/d/1yl87ByyPPpSASyL8hdH9nj4BCK7WdFGC/view?usp=sharing Photo of scope of cranking, signal taken with ecu unplugged. Ground and signal taken directly from sensor. Did another scope with plugging new wires directly to the sensor (ie not using any of wiring or plug in the car), tracing same as above. Next shot one previously posted at idle with ecu connected (obviously), signal taken from wire from sensor, ground from common signal ground of ecu https://drive.google.com/file/d/1MsgB89saHr857SWB7EN35Y8Cx9ZuwFmz/view?usp=sharing In previous tracings, the ground for the scope signal was taken from the common signal ground of the ecu (via a free green wire in the wiring loom). I assumed this was ok as there is continuity between all the signal grounds from the ecu including the shielded grounds from the trigger sensor. Pic of cranking with scope wired directly to sensor with new wire. https://drive.google.com/file/d/1sivK7TL6yqXdlhjfZxUNPau3fPc_Yjol/view?usp=sharing Quote Link to comment Share on other sites More sharing options...
Adamw Posted October 22, 2021 Report Share Posted October 22, 2021 Your waveform looks much closer to the expected shape with ecu unplugged. Just need to figure out now if it is something internal in the ecu or some influence from a ground connected to something it shouldnt be or similar. With the crank sensor unplugged, but ecu plugged in and powered up, can you measure if any voltage is coming out of either pin at the crank sensor plug (from the ecu). Connect one voltmeter probe to any good ground, and measure if there is any voltage on either the trig1 or ground pin. Quote Link to comment Share on other sites More sharing options...
MartinS Posted October 23, 2021 Author Report Share Posted October 23, 2021 Oh, that's promising! Will put loom back together tomorrow, measure, and look for ground problems. Thanks Adam Quote Link to comment Share on other sites More sharing options...
MartinS Posted October 23, 2021 Author Report Share Posted October 23, 2021 At the plug to the sensor, Ign on, 32 mV on the signal socket, 18 mV on the ground socket Quote Link to comment Share on other sites More sharing options...
MartinS Posted October 27, 2021 Author Report Share Posted October 27, 2021 On 10/23/2021 at 11:42 AM, MartinS said: At the plug to the sensor, Ign on, 32 mV on the signal socket, 18 mV on the ground socket Not sure how to interpret these results. They are reproducible and consistent. Seem to imply there is very little to no leakage into the ground but there is some at 18 mV. The source could be any of the sensors grounded through the ECU which have a voltage supply I suppose. These are the cam position sensors, MAP sensor, Wideband O2, and foot pedal and throttle position sensors for the e-throttle. Only other source the ECU. Nothing else signal grounded which has power supply I can think of. Is it worth disconnecting these to see if one of them is causing the small voltage in the ground? The small voltage leak (32 mV) in the crank sensor signal line is from the ECU as there are no other connections in the wiring loom. Is this normal? Are either of these voltages or the differential of 14 mV between signal and ground enough to cause the interference with the crank angle sensor signal seen in the oscilloscope tracings? Do either imply an internal fault in the ECU? Not sure how to proceed from here... Quote Link to comment Share on other sites More sharing options...
Adamw Posted October 27, 2021 Report Share Posted October 27, 2021 Sorry, I have been meaning to dig a G4 out to confirm what is normal. I just done that and I have about 120mv on the trig 1 pin so yours sounds normal. I dont think the 18mv ground offset is unusual. This is not a conclusive test but it kind of suggests to me the distortion on the trigger waveform shape is not coming from within the ecu. Not too sure what to suggest next. One test that would be valuable would be if we could get a scope capture of it cranking with only the crank sensor, power and ground connected to the ecu. So leave B plug unplugged, and depin all the others from the A plug. So only 12V, ground, trig1 and trig 1 ground connected to ecu. Quote Link to comment Share on other sites More sharing options...
MartinS Posted October 27, 2021 Author Report Share Posted October 27, 2021 Trigger oscilloscope screen, cranking with plug A trigger1, trigger-ve, 12v, and ecu ground only attached, plug B unplugged. https://drive.google.com/file/d/19JVboSeOYaNUWV11ACwa5hPYOfZdm2U7/view?usp=sharing, https://drive.google.com/file/d/1pkz7g8J9eEn-iBSVnwiwz4k5NsfI4nUP/view?usp=sharing Sorry, pics not centred well Different part of cycle. https://drive.google.com/file/d/1RA2_nGb2opIreS1V91bdVEgmwogJ79eS/view?usp=sharing Quote Link to comment Share on other sites More sharing options...
Adamw Posted October 27, 2021 Report Share Posted October 27, 2021 Ok, that was a good test, waveform still looks good. So the distortion is coming from some other wiring. My thought is most likely one of the grounds. Lets try the same test with some other stuff plugged back in. I would probably start with the other sensor grounds and 5V wire reconnected. Just so you know what Im looking at, here is a bad one, the amplitude is much higher above the zero volt line and the waveform especially in the missing tooth areas is not symmetrical: And here is your most recent one in comparison: Quote Link to comment Share on other sites More sharing options...
MartinS Posted October 27, 2021 Author Report Share Posted October 27, 2021 Ok, I see what you mean now. I was never really sure I understood what you were seeing. I was looking more at amplitude not symmetry around the zero V point. Looks much more obvious to me now. I already ran a scope with B loom plugged back in. Here's the link. https://drive.google.com/file/d/1Iu_FizZq7sCGTXvrg4SfjBh46LLGNnKn/view?usp=sharing I think it looks ok which leaves me with the grounds on A loom. Do you agree? Quote Link to comment Share on other sites More sharing options...
Adamw Posted October 27, 2021 Report Share Posted October 27, 2021 Yep, still looks ok, so start plugging stuff back into A plug. Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.