divil

I hope the quality of these is ok - the light is not great where I park my car, but for the really important tests I got out a flashlight so you can clearly see what I have hooked up. Also there may be some wind noises and door rattling noises, just try to ignore those!


Here's the control test - I'm just making sure my car runs consistently with my DME, and the KLR removed:





Next test - this is with the eBay DME, no KLR. I did not clearly show which DME was attached in this video, so I then did another one, where I made sure you could see.







Here's the first test of the Specialized/ECU Doctors DME, still no KLR:





And, finally, the Specialized/ECU Doctors DME and KLR:





I did not test my KLR with anything, nor did I test the Specialized KLR with the eBay DME, as there didn't seem to be any point - but I can do this tomorrow if necessary.

PerryB

Wow, that's exactly what I needed, but not what I expected. Thanks divil, that took a lot of time and work to run those tests and to post them. I am greatly appreciative and also indebted to you.

I really expected that the computers would not work, as ALL my other testing has shown that the computers are getting the correct imputs but are not firing.

So, back to a few basic questions: When I replaced the speed and reference sensors I bought generic ones from a company called FAE, and they are the longer lead ones originally for a BMW model that also uses the Bosch Motronic system, could they be the cause of this no start. They perform exactly as they should according to the factory manual.

Years ago, my brother had an Alfa GTV-6 that had a no start that was caused by a ignition switch that would crank the engine, but would not fire the ignition circuit. I've tested that the DME gets 12 DCV in the start and run positions, is there some other test for the ignition switch? It powers the DME relay correctly.

Is the speedometer sensor in the transaxel a possible cause of the no start?

I am completely stumped and am ready to send the car across the state to a repair shop for diagnosis and repair.

Perry

Van

The speedometer sensor has *no* impact on starting.

When the starter motor is engaged, a +12v signal will go to the DME on pin 4. If that's not happening, then your ignition switch might be bad.

I have suggested before swapping in another set of speed and ref sensors...

divil

I know this is a very, very long shot, but what the hell. You didn't replace your starter by any chance did you?

I began to wonder a while back if your starter could be cranking backwards due to a wiring mistake...I didn't know at the time that they can only crank one way. But I did stumble across this...

WHY DOES MY 912 CRANK BACKWARDS???!!!

Van

Wow - good thought. If you're looking at the front of the engine, the belts/crankshaft/distributor rotor/etc. should all spin clockwise.

PerryB

Van, Divil,

Thanks for the suggestion. I removed the distributor cap and cranked the engine. Rotor turns clockwise as viewed from the front of the car looking back towards the rear.

I still have no spark. Van, I appreciate your comment about the speed and reference sensors, but they still show up as completely withing all the specs when I test them. Has anyone else ever had a no start problem with the aftermarket FAE sensors that was cured when they were replaced with the OEM Bosch ones? I just wish I had some indication of what was wrong, rather than just stumbling around in the dark replacing parts because they might be bad.

I was worried that something about my car was damaging the DME or KLR but Divil's testing showed that that is not happening--some good news at least.

Again, I will appreciate any suggestions from anyone.

Perry

PerryB

Also, I retested the start signal at DME pin #4. It gets ~11v during cranking and only during cranking.

Perry

Van

Perry, just to verify, you've pulled the fuel rail with the injectors attached, and you can see fuel puffs while cranking? But you're not getting spark?

PerryB

Van,

Thanks for your reply.

No, I smell gas at the tailpipe and have wet spark plugs. For most of my testing, I have pulled fuse #34 to prevent the pump from running and washing down the cylinders with gas. Originally, I did not have spark OR fuel. Only after having the DME rebuilt did I get injection.

I currently have a spark plug jammed in the distributor end of the coil wire and resting on the cylinder head. I get no spark at this plug, although it sometimes fires when I turn the ignition key from on to off.

There is something that prevents the coil from firing, I just cannot figure out what...

Perry

Van

Have you tried another coil?

As a good habit, I always use an alligator-clip test lead to ensure the plug is grounded.

Also, when it comes to fuel, I prefer to pull the fuel rail and injectors out, aim them up, and crank the engine over so my eyes can actually see that they're working as expected (or, for any test, I want "proof" it's working - not just go off of inference). What if you had a fuel pump issue and only a little fuel was dribbling out the injectors - it'll make the plugs wet, but won't be enough fuel to combust...

Scott at Team Harco

I have not read through the whole thread and hopefully my "tip" is not out of line. I discovered a very simple mistake when I did a clutch job on Herbie about 10 years ago. It was also the same mistake made by another lister.

There is a connector with a blue and red wire that I believe does exactly what Van refers to above. The connector is right near the brake booster. I don't recall if the starter will or will not turn over if this is disconnected - but I know the car will not start.

Hopefully it is something so simple - but then again, this is so simple - one will definitly do the palm plant if it turns out to be the problem.

Good luck no matter what - this has gone on for far too long. I hope the solution is found soon.

OK - I just found that old thread - the connector is to get power to the starter. So forget this suggestion.

specsalot

Perry - Sorry to hear the saga continues, but I do like the good news that the computers function. I've been away from these forums distracted by the holidays. Time to re-engage. Divil - Great Job testing the computers.

I have used longer lead BMW sensors in the past with no issue, except organizing the longer wire. One of the things I think these cars are sensitive to (reading thread here, and my own experience) is the crank sensor sometimes picks up cross talk and fires out of sequence. When my cables were a "bit of a mess" under the hood, I would occassionally get the same "Ruh" hard crank sounds that Divil recorded in his videos. I'm fairly certain when these occurred there was some kind of mis-firing taking place. It was not common occurance, but with cabling squared away, this issue has never happened again.

I found the wiring repair harness I used on my car had incorrectly labeled leads for the speed reference sensor. The wires were the correct colors, but the leads had been incorrectly connected in the plug. Although I don't think the car is sensitive to that I swapped leads at the DME anway. This implies that after market harneses can sometimes be wired backward. I think you've already verified this is not the case for your crank reference sensor wiring.

You repeatedly showed that you were not getting a square wave output from your DME, but this DME has shown itself to be working now (as does your KLR). By implication this points to either voltage supply problems to your DME or issues with the sensor inputs preventing the DME's from operation in your car. I think you've proven proper voltage supplies, so this sounds like some kind of sensor issue that is disabling the DME from operating in your car.

I'm at work (night shift) but don't have any diagrams here at the moment. I've heard several suggestions pointing toward this potentially being an ignition switch problem. Perhaps someone has a spare they could loan you for testing. If I were in your shoes, that would be my next move.

Keep at it. This can be solved. You've already dealt with the most expensive aspects (computers), so its time to retrace a few steps or look in different directions. It is likely a wiring issue.

divil

Just checking in again

Sorry if you've covered this already, but if I remember correctly, you didn't want to remove the DME connector cover before - so I'm guessing your measurements of the crank sensors were with the sensors unplugged, or maybe with the DME connector unplugged? The reason I mention this is, I have noticed a big difference how my crank sensor signals look when unplugged vs. probing the back of the DME pins with everything hooked up. For instance, the amplitude of the ref sensor was way lower when it's actually hooked up to the DME. In addition, there was a huge amount of noise in the signal when cranking with the sensor unplugged, which disappeared when it was connected.

None of that prevents my car from running, but I think it does show that you never really know what signal the DME is seeing when it's actually trying to start, unless you have everything hooked up.

PerryB

Van, Divil, Specs,

Thanks for your ideas and support. I continue to test the ignition/computer systems on my car in the hope that I will discover the source of this no start.

I have done all the testing of the speen and reference sensors at the DME connection in the passenger footwell. I took all the resistance and oscilloscope readings at this main connection to ensure that there wasn't a problem in the wiring harness between the DME connection and the connection with the sensors.

Currently (no pun intended) I have some strange resistance readings for these sensors when testing them in ways other than between pins #8 and #27 for the speed sensor and #25 and #26 for the reference sensor. I don't know how to interpret my results yet, but I will post results as they become more comprehensive.

Again, thanks for all the support.

Perry

Van

I seem to recall that you mentioned if you used the same flywheel, or a different one... But I don't remember your response. Is there any way that the orientation of the flywheel is wrong? So the TDC signal is at the wrong time?

I'm wondering if you're getting the spark trigger at a time when the rotor isn't actually pointing at one of the terminals in the cap. This would give you the erratic spark you see, and it wouldn't run, because the spark event wouldn't be at TDC.