bob944

I'm posting here with the hope that you all can help me with any sort of advice....I'm out of ideas.

I have been battling a NO-cold-start condition in my '84 944 for many months now, and I still can't dig a fix out of all the archive info. No amount of cranking will get the cold engine started, but it WILL run after three or four tries with a shot of ether to assist. Once running, however, even the slightest hint at pressing the gas petal after the car barely sputters to life on ether will immediately kill it, and will continue so for about 4 or 5 minutes until the engine begins to warm up. It runs rough during this phase, but once the engine is fully warmed it runs, and idles just fine. The engine will restart with no problems for up to several hours afterward when engine has cooled, but after that time I must use a shot of fluid again. Note, the problem is completely independent of ambient temp/season, only the temp of the engine itself seems to matter.

Things I have checked (in somewhat random order):
-- DME relay and fuses
-- Brand new DME engine temp sensor
-- AFM meter readings seemed ok, but swapped with 2 different (used) parts to diagnose potential bad air temp sensor, but no change
-- Swapped TPS with used part and should be adjusted properly according to resistance readings.
-- Pulled cold-start-valve - seemed to work fine in freezer/12V test, blasted with lots of carb clean and reinstalled
-- Injectors were professionally rebuilt about 1.5 years ago and do not SEEM to be the problem - no big puff of smoke at start up to indicate a slow injector leak
-- Fuel pressure check - normal pressure with cold engine and pump jumpered, and passed 20 min leakdown test pressure (obviously fuel pump runs), and appears that check valve is not failing over 20 min. period anyway.
-- Verified definite fuel spray from each injector with "fuel rail in bucket" test to physically see each one spray while cranking. The fuel in the system is approaching 6 months old...so perhaps it's starting to slow the flow beginning to gum up....but I tend to think that getting it running on ether would disprove this.
-- Attempted to start cold engine by first jumpering fuel pump for ~10sec to be certain of rail pressure, but no luck
-- Magnecor plug/coil wires are fairly new and in good cond.
-- Verified definite spark to each plug by grounding against manifold and cranking engine
-- Verified that reference sensor plugs were not inadvertently reversed and both read proper resistance values (no scope to verify wave) - tach IS jumping slightly when cranking.
-- Pulled speed/ref sensors and cleaned particles but did not move bracket to disturb gap - seems like car would never start with fluids if either one of these sensors were the culprit? Should I go through the troubles to reset the gap anyway?
-- Grounds "appear" to be in good shape and were cleaned in the last year, but haven't revisited them....again, wouldn't this cause problems starting/running on a warm engine also?
-- Tried starting with one injector wire pulled in case over pressurized from bad FPR - nada
-- Inspected rotor and dist. and appear to be in pretty good shape, cleaned slight carbon from cap contacts
-- Vac lines appear to be in good order and routed correctly according to the under hood decal
-- Have NOT swapped DME or opened to look for bad solder joints,etc.

I JUST CAN'T FIGURE THIS OUT! It seems to me that the engine must be extremely lean if it requires extra "fuel" in the form of fluids to cold start, and any added "air" from the throttle causes it to die until warmed up. Am I completely wrong, and there is an electrical problem sending bad signals to the DME? I guess it's possible that I have verified existence of spark and fuel, but one or both is not adequete ENOUGH....except that I don't even get a slight sputter on cranking without using ether.

Please please feel free to offer any other ideas you might have. I've spent hours and hours researching and testing this problem and nothing has worked. I have been pretty careful, but I certainly may have screwed up any one of the above tests. Obviously I hesistate to start buying expensive new parts for no reason until I can isolate the problem area. It seemed that some logical testing would determine the source of the problem, but I just can't seem to figure this one out.

Thanks VERY MUCH in advance for your help!!!
Andy

Macfreak007

Sounds exactly like the problem I was having, sounds like a BAD DME, have you ever jump started the car? There must be a component that goes bad in these DMEs because mine did this, and I resoldered the whole thing, still did it, but when I swapped DMEs it was perfect. I would be willing to bet this is what is wrong with yours. I would like to figure out what component it is that goes bad...
do you have a DME that you could try hooking up to troubleshoot this? 90% sure thats what the problem is for you!

AznDrgn

That is really interesting. If you have fuel and you have spark the only other things you need for ignition are timing and compression. Those two things normally don't change much wether the engine is warm or cold. I'm stumped but maybe someone else on the board will have an idea. Where in Haymarket are you? I goto school down in Harrisonburg so I'm not all that far away from you.

Lorenfb

If ether results in a good cold start, then the ignition function is O.K.
You need to get an injection light (small light inserted into injector plug)
and check for an injection pulse while cranking. The light should pulse
within 1-2 seconds after the start of cranking. The light should be brighter
if the temp sensor is disconnected (next test). This checks the enrichment
function of the DME when cold.

A solution requires focusing on the fuel function of the starting process, i.e.
extra fuel makes for a good cold start. Using the cold start injector as with
the ether should also make for a good start. Bypass the thermo-time switch,
so that the valve sprays whenever cranking. This should yield the same results
as the ether.

nine-44

Hmmm, sounds like possibly a DME issue as Carl said. I'm wondering. My G60 corrado had issues(along with others) with static burning an opamp in the control unit. It amplifies the signal from the temp sender. The issue with the Corrado resulted in very high cold idle, like 3000-3500 if you touch tha gas when cold.

bob944

Thanks for the input so far guys. I'm really trying to *think* logically about this problem that is driving me insane. You all present valid points, but I'm still confused becasue certain things don't seem to add up. I have not "jump started" the car, but I'm not sure why I would need to since the Optima is charged, and I know it WILL start on ether.....am I missing something.....maybe you are suggesting I fried the DME during a jump?

It certainly could be a bad DME-$$ which is why I've left that as a last resort. It does seem, however that if the DME were truly bad that the car would NEVER run....although perhaps the circuits pertaining to cold enrichment are all that has gone bad???

Likewise goes for the injectors.....if I can physically verify with my eyes that the injectors are squirting while cranking with the fuel rail in a bucket, isn't that proof enought that they are indeed FIRING during cranking, and that SOME amount of fuel is getting to the cylinders? Also, if timing or compression were bad, I definitely suspect it would cause major running problems on a warm engine too.

Although I have verified fuel, and spark.....perhaps one or both is just not STRONG enough. Maybe my plugs are too fouled/weak to ignite, maybe my injectors aren't squirting enough fuel, maybe my coil isn't putting out enough voltage.......maybe maybe maybe. I simply can't understand why the presence of fuel and spark (hopefully at the right time and quantity) won't force even a sputter of ignition.....not without an ether kick.

I will try and get my hands on a working DME to test that idea, but it may take a little while to fine one.

Still baffled........keep the ideas coming. Thank you everyone!
Andy

Lorenfb

"isn't that proof enought that they are indeed FIRING during cranking, and that SOME amount of fuel is getting to the cylinders?"

No, it may not be ENOUGH fuel when cold.

Again:

"You need to get an injection light (small light inserted into injector plug)
and check for an injection pulse while cranking. The light should pulse
within 1-2 seconds after the start of cranking. The light should be brighter
if the temp sensor is disconnected (next test). This checks the enrichment
function of the DME when cold."

Macfreak007

someone in VA just needs to come on here and let this guy borrow a DME, the problem you are having is the EXACT problem that I was having, you describe it exactly the same. Everything else seems normal, which was the case with my car. Swapping the DME made the car run perfectly, instantly. Trust me on this one, it is DME related, find someone to borrow a DME from in VA. I'd send you one to try, but the shipping back and forth probably wouldn't make it worth your while. I'm sure someone a bit closer wouldn't mind you borrowng their's for a sec. all you would have to do is pull the plug on it and plug it in the other one. I'd wager to say that I am 100% sure this is your problem, as it is EXACTLY the same as mine. In my opinion, don't send more money on an ignition light, when you can solve the problem without any other fooling around by borrowing another DME. someone in VA help this guy out!

Lorenfb

"In my opinion, don't send more money on an ignition light,"

Not ignition light, injection light/LED - < $5.00

Macfreak007

yeah, i guess if its about $5.00 go for it if you want, but I will wager alot that the DME is the issue. at least you'll have it for troubleshooting in the future.

SoCal Driver

Reset the clearence on the sensors. Porsche has a bulliten out on this problem with deep cold starting. You need a scope to adiquately check the pulse coming off of the sensors.

Please note injectors are not "rebuilt" just back flushed and cleaned including changing out the fine mesh screen in the intake side.

What cold start valve? Do you mean the idle control valve?

Lorenfb

"Please note injectors are not "rebuilt" just back flushed and cleaned including changing out the fine mesh screen in the intake side."

Right on! Look at the way they're contructed, they can't be really rebuilt.

Stan944

A good engine temp sensor doesn't necessarily mean there is voltage on it, i.e. the wireing or the DME guts could cut out the (source of) current.
The engine won't start then (as verified by Danno and me).
You might want to semipermanently hook a voltmeter to confirm it.

AznDrgn

If you think it might be a bad DME I have one in my '83 that I can bring down and let you try. I'll be heading back down to Harrisonburg from Northern Va either Tuesday or wednesday. PM me if you want me to bring my DME with me.

Mike S

Does the tach jump when you are attempting to start? If so, then the reference sensor is picking up a signal. You can also try tapping on the DME box. My solder joints recently broke and i had an intermittant start. I figured it out by tapping on the box one day.