Early 944 Temp Sensor Issues

[Tom]

Have you tried a spare DME?

[delgadostarside]

I dont actually have a spare DME i wish i did. I would consider trying ebay for another but not sure the reliability on a "working" one from ebay. I thought about getting a F9 tech so i could have OBD and upgrading the MAF but the cost for performance is just not worth for a daily driver.

[delgadostarside]

So we have a warm weekend here in NY finally so i can get her running and test this out.

I checked first on my Camry and In the video below you'll see I get ~756RPM on the laser reader (I tested multiple times after recording to confirm this because it's hard to hold my phone and get the laser dot on the reflective material even though the reading looks inaccurate in the video, i can confirm afterwards it is roughly the same within +-10 RPM). With the Oscilloscope tied on as well I see ~40Ms between pulses which gives exactly 750RPM. On my tachometer I see about the same result looks just over 750RPM. (Turn down you volume a bit for these two)



Now on the 944, I re-do the same testing where it's much easier to reach the crankshaft pulley too so the laser reading is actually way more consistent. This time on the laser reader I'm getting about 1035RPM. The Oscilloscope reads about 58Ms between pulses which gives 517RPM. And confirming on the tachometer at the end of the video reads ~500RPM.



So this definitely confirms tachometer operation is following the ignition pulses however the ignition system is running at half the actual Crank rotational speed. What a head scratcher. I'm guessing all i can do from here is send the DME out or find a working replacement then? I cant see shorting causing half of the pulses to not fire and the other half to consistently fire.

I just now bought a new distributor cap and rotor as a last ditch here even though spark testing shows all 4 are firing. I ended up grabbing a DME off ebay for a good price too but if both of these don't work i'm not sure what to do here honestly.

[Tom]

Does the car drive poorly, with low power, etc?

[delgadostarside]

I haven't gotten her on the road yet, I don't think shes been driven in 10+ years. I've taken her up and down the driveway though. I'm worried about stalling the car in traffic if i got it road legal since she doesn't like to start unless i pull the temp sensor first. As well as the low compression causing that carbon buildup. I'm sure it'll misfire once it gums up the plugs again. If i could get it starting and running normally without any extra input on my part i would definitely be comfortable with driving her. I would say there is a lot of vibration and after reaching operating temp it smooths out some but i've heard the early cars just happen to because of the early DME's.

[delgadostarside]

Im back with another round of testing. Just got the replacement DME in the mail and same thing, only started with the temp sensor unplugged. Once unplugged it immediately started though. I shut it down, grabbed my oscilloscope to see if the pulses were in line now. Threw it on started her up and... Same exact scenario. Pulses match up with half of engine RPM aswell as the tachometer reading what the pulses suggest rather than the actual engine speed on the laser tach.

Now this time i remembered to finally pull the plugs while it was running. Thankyou for the suggestion again, I dont know how it kept slipping my mind every time i went to test.
Cylinder 1: Bogged down when pulling plug
Cylinder 2: No change
Cylinder 3: NO Change...
Cylinder 4: (I Skipped since this showed enough at this point and it definitely wasn't running on 1 cylinder)

Well there we go that was way simpler than throwing the scope on there as it shows exactly what the scope is seeing. So from here i decided to change the plugs again. No change when pulling the wires 2 & 3 with the new spark plugs. Swapped the spark plugs on 1 & 4 with 2 & 3 and... same thing. Pulling the wires on 2 & 3 had no change here either. So i know its not the plugs. I also replaced the distributor cap and rotor last week with an OEM Bosche replacement so its not that. The plug wires show within spec for resistance and does not cut out when moving them around while testing the resistance so its not that (plus they're also brand new within the past 6 months). Coil checks out with resistances. DME is rebuilt. I am totally lost here.

Any suggestions to look into? I can only think now to run new wires for the coil and splice it off of the DME. How the **** could both DMEs be sending half the signal... is she FUBAR? I cant believe its only been running on 2 cylinders this whole time.

[delgadostarside]

So today I spliced in a seperate coil wire (green) and removed the pins on any connectors with the existing wire. Sent it to both the tach and the coil and.. same thing. I took it out of the back of the tach with it now just running straight to the coil and same thing as well. I did however find this post on Pelican

http://forums.pelicanparts.com/porsche- ... 3-a-2.html

Seems a few people have had this issue but this is the only post i've found a solution in. Going to try re-gapping my Speed/Ref sensors since the resistance already measures within spec (Roughly 1040 Ohms). Im not giving up on this haha. I'm going to ignore the temp sensor issue right now and attribute it to: not enough power to turn over on 2 cylinders or the ECU not reading a high enough starting speed with a proper fuel/air mixture on 2 cylinders.

[Tom]

Before gapping, you might check the waveforms for the speed and ref sensors and see if they are within spec. Gapping won't hurt, but that wouldn't necessarily reveal harness issues (which are super common on those sensors).

[delgadostarside]

So i got her to run on all 4 cylinders and... the Culprit was the speed and reference sensor gap the whole time! I never re-gapped it after the rebuild since i had left the bracket in the same position but it must have moved somewhere along the way. Yesterday, I did testing after a long day of work and to my embarrassment, gapped the wrong sensor. The thump i got when pushing the bracket down when re-gapping was just the pivot screw bottoming out on the bracket so i ruined the speed sensor. However, it did run on 4 cylinders, backfired a few times or missed so i shut her down and realized my mistake afterwards. The tach was also now properly reporting engine speed!

All in all, it seems a Larger-than-spec gap was causing improper ignition timing from the start. It's amazing as you can almost tune the timing by loosening the screws and tightening them down on the sensors as well. It was cool to see that testing too before i realized my mistake. I heard a horrible screaching sound and pulled the sensor closest to the front of the engine (thinking this was the speed sensor at the time) to see it was perfectly fine so i assumed the other one was too (I was wrong ). So im going to order a new sensor and be more competent this time but for anyone having this same issue or may be in the future, I'd like to wrap a few points here.

Engine Running on 1 & 4 cylinders only:

Ignition:
-Check spark plug tips for oil or carbon buildup.
-Check plug wires for resistance as-well as any dropouts when bending the wires.
-Check for corrosion on distributor cap and rotor.
-Identify DME grounds behind the engine and make sure they are connected.
-Check continuity of Speed/Ref signal to back of DME harness aswell as continuity between pins
-Check for proper sensor resistance on: Speed/Ref sensors, DME temp sensor, AFM & TPS.

Oscilloscope testing:
-Check for proper waveform shape (Sawtooth) and P-P voltage on Reference sensor of 2.5 volts
-Check for proper waveform shape (Sinusoidal) and P-P voltage on Speed sensor greater than 2 volts
-Check ignition pulse signal for proper timing between pulses for idle (33.3 ms between pulses = 900RPM)

Fuel:
-Make sure you have 1/4 Tank at minimum
-Change fuel filter
-Check for fuel injector leak down or operation using a fuel pressure gauge
-Check for fuel rail leak down using Fuel pressure gauge (Attributed to faulty FPR or dirty Check valve)
-Check for proper fuel pressure when cranking, running and 20 minute leak down. (Clarks garage testing)

Compression:
-Check for even compression across the board
-Check for Oil in coolant or vice versa
-Check piston heads with a cheap cell phone endoscope (Anykit $30) and check out condition on piston heads

Thank you everyone for you help, I wouldn't have gotten this far without all of you. This community has been such a great place to browse and find like minded people who love to tackle problems and solve them in any way possible. Excited to show you all what she'll become in the next few months!

[Tom]

Great work finally getting to the cause of your bizarre issue. That one will go into my mental data base. I have to go back and look at my notes, but I might have a theory why sensors on the fringe of their range might fire every other time. In the meantime, here's an old video I did on the sensor signals that might be of interest.