Can't get 951 running
 

Hi all,
I've got a 951 that I've been working on for about a year now. I finally have the motor all back together, however I can't get it to run. The starter spins just fine and the RPM sensor is outputting a 200RPM+ signal, so the DME should be, and is, telling the plugs to fire, but the car just won't start.

I can't for the life of me figure out a possibility, given that I have fuel pressure, I have spark (although apparently not at the right time -maybe a distributor rotor set screw has backed out, I'm going to check this tonight).

I did have the car running for about 5 minutes 2 weeks ago. The AF ratio was very lean so I shut the car down. It hasn't run since, although it occasionally sounds like it is trying to fire 2 cylinders. When I pulled the plugs, cylinders 2 and 3 were blackened while cylinders 1 and 4 looked brand new. I've cranked the motor over about 100 times (good oil pressure), but I think that I might have washed my bores because I have a heavy fuel smell in my oil now. No oil in the coolant or vice-versa.

I have checked the output of the reference and speed sensors with an oscilloscope -- as far as my novice eyes can tell, they are outputting the correct waveform and voltage.

I have fuel pressure at the fuel rail, and I pulled my injectors to verify they work. They do. The cam timing is verified good. I changed the plugs out twice now, fearing that all the fuel and no ignition would foul them. No change in condition. I shipped my DME and KLR off to a friend who tried them on his car - the car ran with no problems.

A few notes on some modifications I've made, thinking it will probably help troubleshooting -
Vitesse Stage 2
- Turbo
- MAF
- SMT6 piggyback computer/datalogger
Zeitronix ZT2 Wideband AF meter/datalogger
- A/F gauge
- Boost gauge
- EGT gauge
Milledge Engineering cylinder head
Milledge Engineering Cam

Does anyone have any suggestions for me? Do you have questions for me that might give you better information to base an idea from? I'm thinking my choices are spark timing, fuel timing, or compression. I don't have a leakdown tester but would sure love to borrow/rent one. None of the local parts stores had one to rent.

I'm at my wits end here. I have followed every wire from the DME to their connectors in the engine bay. I have continuity where I need it, no circuit where I need it, and I have fixed any shorts that were in the harness (Check your injector harness! Most likely bad!)...

I really have no clue where to go now and I know there are some 944-951 experts here, so I'm just hoping someone can provide a little guidance or maybe point me in a direction on some test points. I also would be more than open to having some help, so if someone feels like joining me, I would provide all the food you can eat, beer you can drink etc. I debated trailering it to Hershey and setting up a sign that said "$250 for anyone who can get this car running."

Its just so frustrating. I have spent every free moment I've had in the past year working on this car. I just want it to run! In June, I was patient and had accepted the long down-time in front of me. In October, I started getting a little impatient and wanted the car to run before 2005. In January, I thought the end was in sight and remained patient. Now, when I have no chance of bringing the car to Hershey like I had hoped, no leads on how to fix it and gorgeous Porsche driving weather, I'm at the end of my rope.

Awful thoughts of insurance claims or 'parts-car' have invaded my dreams lately...PLEASE HELP!

Shaaa. After all that work and parts added could be many things. Start by splitting it from fuel or spark. Get some starter fluid and, using it according to directions, spray some in the intake. If the motor starts and then dies it is likely fuel related. If it does not start it spark. (Fuel is not ruled out yet but spark is indicated. Let us know what you find. Kinda sounds like a spark timing issue. Plug wires on the right plugs? Cam timed? Dizzy timed right? Do you have an alarm system?

Tony, watcha think?

Flip the Speed/Ref sensor plugs if you are not sure that you have them on correctly.

Worth a try, the car will not start if they are backwards.

Michael

Flip the Speed/Ref sensor plugs if you are not sure that you have them on correctly.

Worth a try, the car will not start if they are backwards.

Michael

Paul,

You said that your AF ratios were real low, but I didn't see you say that you readjusted your fuel ratio setting. Given the number of performance mods that you did, you probably won't be anywhere near the fuel ratio setting that you need to be at. I would suggest that you adjust your AF settings first as that will definitely prevent your car from starting. Also, you should double-check for any air leaks in your intake system, these can lead to false air readings.

David

Paul,

Don't give up....there is a solution....believe me I feel your pain. I spent forever tracking down a similar no-start problem that ultimately was proven to be a failed enrichment program in the DME.

When you boil things down to basics, if you have spark, fuel, and compression all in adequete quantities you should have ignition. If any one of those is missing you will have a problem, but also if any one of them is too weak to get the job done you won't get it running. It takes a much larger quantity of fuel to start a cold engine than you might think. The engine block temp sensor and the factory AFM temp sensor (not sure how your MAF affects this) both send signals to the DME which will map out the correct fuel curve for the current block temp and ambient air temp. If either one of these sensors has failed it can cause problems. In my case, the sensor were fine, but the cold start enrichment program in the DME was faulty. The car absolutely would not start when cold, but with a blast of starting fluid (ether) it would sputter to life. Once warmed up, the cold start enrichment is ignored, and hence my DME would work fine and the car would restart for up to several hours until cold again.

If your friend verified that your DME works correctly, be sure that he tested it on a COLD block to duplicate your problem. Then, start going through things again....back to basics.....and be sure that you have a strong ENOUGH spark getting to the plugs. Put a fuel pressure gauge on the rail and be certain that you are maintaining the proper pressure over the factory specified leak-down time. I'll assume you have already verified your fuel pump relay is working correctly. A simple cylinder compression check should be sufficient to verify if all four cylinders are near spec compression.

Your upgraded electronics complicate matters somewhat, but hopefully it isn't something related to that. If you can remove any of the electronics to simplify the testing, I would try to get back to stock electronics until things are up and running.

I'll let you know if I think of anything else.

GOOD LUCK!

Andy
'84 944 (finally running again)

Wow, thanks for the great replies guys!
I appreciate all of the suggestions.
Lets see where to begin -
Kurt,
Yeah it could be a hundred things...I couldn't afford to do the whole project all at once, so I spaced out my purchases and work over a year. Kinda wish I could have made the mods one at a time, but the finances weren't there.

I have a fuel pressure gauge attached to the fuel rail. It originally was set for roughly 30-32PSI (2bar). This was way too low (Vitesse told me to set it to 3- 3.5bar). I adjusted that and now my fuel pressure at the rail is roughly 40-50PSI.

I pulled the injectors and tested them by applying 12V to each, they all fire (they are brand new 55 lb/hr injectors.
I also checked the wiring on the injector harness to ensure a solid electrical pathway back to the DME. There was a problem here that I corrected. When finished I pinned out the connectors back to the DME. I have good continuity and no crosstalk across the injector harness.
I think the problem is more likely related to spark, or spark timing.
To that, I have triple checked my timing, it is spot on. My balance shaft alignment is also spot on, though I know that wouldnt affect starting.
My distributor doesn't have the capability to be clocked, it is stationary. Spark timing is controlled by the DME.
I have verified the plug wires are installed correctly, and I have verified that the rotor points to the #1 contact in the distro cap with the motor at TDC on cylinder 1. This is all correct.
I have some starter fluid, I was hesitant to try that but I will do so. I also am going to do a compression check. The headgasket is obviously new, so I don't think thats the problem. If I have compression problems I think they would be related to bad rings. I'm hopeful thats not the case.
There is a factory alarm system that I haven't touched. I have read some people bypass this, but I'm not sure what pins I would have to jumper to do so.

Michaelathome - The Ref and speed sensors have been scoped out and are providing what looks like the right signal on the oscilloscope. I think I mentioned earlier that the connectors for these sensors are shot, and I am seeing a good bit of AC interference on these lines. I am going to get some 3 pin indestructable molex connectors and re-pin these. Given what appears to be an intermittent spark problem (car ran once, now doesn't), and also given the fact that my tacho used to bounce occasionally (indicative of interference on speed sensor line), this might be my problem. Also considering how vital these two sensors are for starting, this is probably my best bet. Its the only thing that is obviously bad and in need of repair.

David,
My AFR was lean when I had the car running the first time. After I spoke with Vitesse, I raised my fuel pressure to the 3-3.5bar it needed to be. Vitesse says this should help alleviate the problem, at least enough to get it running. After it is running, fine tuning of the AFR is done through my computer.
Regarding Air leaks - this is definitely something I need to check. I was thinking of getting some PVC tubing, thread an adapter for th air compressor on it and try compressing the intake to listen for leaks.

Andy,
Wow, thanks for taking the time to write about your experience.
Regarding the intake air temp sensor - My MAF has an IAT on it. The wiring harness takes the pins from the IAT and the MAF and brings them both back to one connector that attaches to the factory AFM connector. The pins are correct back to the DME. As far as the engine block temp sensor goes, my EMS software shows a pretty much static 127 degrees farenheit, even when the car isn't running. This might be the minimum value, it was a question I had.
I actually had John at Vitesse test the DME and KLR in his car. He said the car fired right up with them.
See above re: fuel pressure. Generally once pressurized, the rail takes about a half an hour to depressurize itself. This seems fine to me. I tested the DME relay from the 951 in my 944. It fired right up. I also tested the ignition coil in this manner with good results.
The SMT6 EMS system comes with something called a shorting plug. This is installed on the end of the SMT6 harness (instead of plugging the harness into the SMT6). The point of the device is to effectively take the SMT6 out of the equation. With the shorting plug, the car operates as though no SMT6 is there, the DME functions normally. The car should start with the SMT6 or the shorting plug. It doesn't start with either, so I'm thinking thats not part of the problem. I've pinned out all of the connections from the SMT6 to the DME, all are correct and I don't have any crosstalk or shorts.

I wouldn't have any idea about fuel enrichment programs, but I assume since the dme and klr fired Vitesse's car, thats okay.
One thing I noticed last night in a last ditch attempt at getting it running for Hershey - my Knock sensor harness looks pretty bad. Practically every connector in the car was bad after 18 years of heat.
The knock harness needs to be repaired, undoubtedly, but could that cause a no-start? I would need to pull the intake to repair it.

Again, I appreciate all of the suggestions. If you read this and think of something else that I may have missed, please let me know.
Otherwise, my action items are to repair the wiring harnesses for the Speed and Ref sensors and knock sensors, to do a compression check, and to pressurize my intake to check for intake tract leaks.

Does anyone have a leakdown tester that could be loaned out?

Thanks again to anyone who offers any help, guidance or liquor.

Paul

Paul,

When rebuilding my '83 the connectors on the sensor wire were bad, shattered, broken beyond repair. What we did was just take the leads, isolate them from one another using electrical tape an plugged them in. Not knowing which was which we tried starting the car several times. Everything seemed to be working fine but no start. We unplugged the wires and swapped them. 2 cranks and we were off and running.

You will not hurt anything running them flipped but the car WILL NOT run. You mentioned "what looks like the right signal on the oscilloscope" well that's a start but it needs to be at the correct pins on the DME. Swap them real fast and give it another try.

This is what mine looked like when I got the car. EVERYTHING has been messed with.

http://members.rennlist.com/michaelathome/DSCN0567.JPG

I have since purchased a set of used/bad sensors and spliced the good conectors in it cost me $20. Lot's of people out there selling out of parts cars. You could also order a set from George at 944ecology. He will ship you tested used for around $60 each IIRC. Then you just hafe to replace your existing sensors and re-gap them. YUCK! That's why for now I spliced the connectors in.

Good luck!

Michael

Michael,
Yep, thats pretty much how they look, except a bit more wire is exposed on mine...and honestly I can't beleive I'm not seeing more crosstalk on the lines, they are pretty bad!

When I say 'what looks to me like the right signal', I'm just saying that according to the DME specs in the factory service manual, I should be seeing a repeating sine wave of 2.0V or more for the Speed Sensor, and an upward spike at TDC of 2.5V or more for the Ref sensor. I'm definitely seeing that, but I'm also seeing some low voltage interference. I was thinking that might have been muddying the signal a bit, but given that the DME needs to see voltage above a certain threshold (2V or 2.5V) it should ignore anything that doesn't exceed that threshold, so the interference wouldn't be an issue. I might be very wrong there, I'm not sure.

Its kind of funny, I called George at 944 Ecology first, when I found the problem. I've ordered this kind of thing from him in the past and I asked him if given the oscilloscope readings it made sense to replace the whole sensor. He recommended just getting a good 3 pin locking connector and re-wiring them and actually dissuaded me from buying the sensors from him. Great guy.

Anywho, I sure as heck hope this is the problem, because thats a relatively easy fix....and it makes sense to me, given these sensors roles in starting the motor.

Thanks again Michael!

Paul

Paul,
Any updates on how things are coming along? I was heading down the FFCPW the other day and got to thinking about where you were getting your beast back up and running.

-Steve