Hard-start becomes no-start

[simssj]

I’ve been chasing a hard cold start issue for about 10 days. It hasn’t been a top priority for me because, with a lot of grinding on the starter, the car has eventually started every time. But now that the weather’s getting colder the problem is getting worse and this morning she left me stranded because she’d crank and crank and kinda-sorta-wanna-try to fire, but wouldn’t.

Recently, as this starting problem has been getting worse, I’ve noticed that once she has started she runs really rough, stumbling and bucking for a few moments while I nurse the throttle. But, once run for ~30 seconds or so, she’s as silky smooth and powerful as ever.

Subsequent warm-starts have been a little bit troublesome, but not nearly as bad as the cold-start problem.

Last weekend I started with the trivial stuff: New cap & rotor, fresh plugs. Replaced the DME relay. Visually checked all the vacuum lines for obvious problems. Checked and cleaned the reference sensors (which had the normal amount of filings and crud on the ends, but no problems… Confirmed that I have the “Tach Bounce” while cranking (but not sure what that indicates). Compression (cold, ~50 degrees) was around 140-ish in all cylinders. I re-seated and/or cleaned every electrical connector or ground I could find. None of these steps made any difference whatsoever.

But I didn’t do any particular harm that I could detect…. She still started only after 2 or 3 long bouts of cranking….

This morning, when she would not, no matter what, start, I dove a little deeper into the investigative process to confirm she was getting the Holy Trinity of Internal Combustion: Air, Fuel and Spark.
- I pulled the air filter and covered the intake with my hand, confirming that she was gulping air. (Poor man’s choke plate!). No help, so…
- I shot a little starting fluid down her throat to see if she would fire with that (pointing, maybe, to a fuel delivery issue). This was no help whatsoever. “AHA!” says I, “Sparking issue!” so….
- I pulled the wire from the coil to the distributor cap and confirmed a GOOD spark into the cap.
That’s as far as I got before running the risk of being late for school, so I slammed the hood and decided to deal with it ‘later’.

What I’m thinking is, with the starting fluid failing to help, I’ve provisionally pointed to an issue with spark delivery to the plugs. But I can’t understand how that would happen because the spark from the coil into the cap is (ahem) “impressive”. (Yes, I should have tested spark at the plugs, but it was cold and I was late…)

Admittedly, I haven’t been following a structured, linear diagnosis and troubleshooting technique. I’ve been wistfully hoping to discover some thingamabob disconnected from a whatchamacallit and – voila! – all would be resolved. That approach hasn’t panned out.

What’s the best advice from here? Start with the Clark’s Garage troubleshooting steps and churn through them one-by-one? Or, just start throwing parts at the problem?

From the symptoms I’ve described, any advice from you guys?

PS: Anyone here anywhere near the Central Virginia area have a spare DME I could swap for a quick comparison?

[944 Ecology]

Use a spare spark plug in the #1 plug wire to confirm that the spark is indeed getting through the distributor and rotor and plug wires.

If you're getting a tach bounce and no start with starting fluid, you've pretty much eliminated everything else.

[mytrplseven]

Go down to the parts store (or online) and get a replacement DME temp sensor. They're relatively cheap and easy to install. It's the one with the blue cap next to the water neck on the front of the engine. It's usually the first thing to consider in cases like this and easy to shotgun.

[simssj]

Thanks, guys!

@944Ecology: I have a spark plug tester with the adjustable gap and all; I plan to give it a go this evening after work. Thanks for the tip. (One thing I cannot fathom is how the DME controls spark / advance, etc... It would seem to me that, once the coil shoots a spark into the distributor, it's gotta come out almost instantaneously. I guess the rotor could have come off / loose and is spinning on the camshaft. My reasoning is, what goes in must come out, but that is about the limit of my understanding...

@mytrplseven: Thanks for the advice. This is on my short list of things to do. I hope I can get the DME Temp Sensor quickly. As much as I love the host of this forum, I'm 6 - 8 days away, shipping time, and I can't afford to be without the car that long!

I appreciate the information guys; I'll post back with more results.

[simssj]

I didn't have a chance to look at the car yesterday except to note an interesting data point...

When the car wouldn't start at all in the early morning the ambient air temp was around 30F. Later in the afternoon I tried it again and after a long crank or two it did manage to fire up. At that time, ambient was about 60. So that seems to point to the DME temp sensor or something in that neighborhood. The good news was that I found a replacement DME Temp Sensor at my local foreign parts house for less than $20. As much as I hate to throw parts at a problem, this was a no-brainer and I'll throw it in this weekend.

I was surprised to discover that my local AutoZone had, in their "Free Tool Rental" program, both a Fuel Pressure tester (with metric fittings!) and a set of NOID lights, so tomorrow morning I'll put the thing in the garage and go thru it in a linear, detailed fashion.

More to come!

[simssj]

Update:

Spent some quality one-on-one time with the car last evening and this morning. Not sure anything has improved, nor that I am closer to a solution.

What I've done:
1) Methodically diagnosed fuel delivery from tank to injectors. With FP running I've got ~40PSI in the rail. 30 minutes after shutting down the FP there's still ~30PSI so I've ruled out the Fuel Pump, the FPRs and the check-valve as causes for hard cold start.

2) Put the noid light on and checked for firing. No problem; the light flashed as expected as I cranked the motor. So I'm ruling out "injection" as a cause for hard cold start.

3) Put a spark tester in-line between the distributor and the plugs. Good, hot bright spark. (My tester is adjustable and I could run it to a gap of almost 0.5 inches before I couldn't get spark, so I'm ruling out spark as a cause for hard cold start.)

4) As long as I was in there, I replaced the DME Temp Sensor that was suggested by mytrplseven. The old one ohm'd out "pretty close" to spec (both at the sensor terminals, and at the DME harness) but I had a replacement in hand, and it was cheap, so I swapped it and threw the old one into my "used, but suspected to be OK" parts bin. I'm ruling out the DME temp sensor as a cause for hard cold start.

After all of this diagnosis (which, basically, went nowhere) I left the car outside overnight and this morning, at about 40F ambient, she ground for a little while, but fired up OK. No worse than before. Arguably better. But still not "right."

I'm feeling like I am at the point where the DME itself is the only thing that seems suspect. If I had a known-good substitute I'd like to swap it in and see if that makes an improvement.

But, on the other hand, the DME that's in there seems to be capable of firing the injectors, the plugs, measuring RPM, enabling the fuel pump to run and I don't know what all else. So if it's sick, it's not *very* sick, is it? Maybe I'll pull it and hit the solder joints, just for good measure.

I'm starting to feel stumped.

Somebody, please, throw me a fresh idea.

[tamathumper]

Sorry if I missed it... What did you do, specifically, to test the injectors?

[simssj]

Well, you bring up a good point. I didn't test the injectors, per-se. I did test each injector connector with a noid light the ensure it was getting the firing signal. (Each was.)

The PO replaced the injectors with a set of rebuilts about a year ago and the car has been running fine for most of the time since then. I still have, in a box somewhere, the original injectors he pulled out. Do you think it make sense to have them sent out for a rebuilding and swap them with what's in there?

Do you have a suggestion on how to specifically check the injectors? (I sense I'm getting in over my head.)

[tamathumper]

Injectors can be tested a couple ways, but yours are hard-fixed to a fuel rail IIRC, so you can't do some of the tests you could if they were connected by hoses. There's total flow testing either by measuring into a container over time, or there's measurements of the drop of pressure in the fuel rail, both of which are somewhat involved, and your best bet is to probably put in known good injectors if all your other efforts have been for naught.

[guru944]

Simssj:

Check your Engine Ground. Run a Thick 10Ga wire from Engine Block to (-) Battery Terminal and the to Chassis Ground. Also verify DME Ground. Ground somewhere maybe floating. Verify (-) on Coil, it should go to 0.01 or lower when you crank engine. I see this all the time on the German cars...Porsche, BMW, Benz, Audi and VW. This is a real problem...

Keep us updated.

Later...

[simssj]

Thanks, @guru944; this seems like a simple check to perform and one I haven't done yet. I'll try this in the morning once it's cold again.

Judging by the theads in the forum, it's getting to be hard- or no-start season; I see at least three threads about 9x4s that don't want to fire. Boy, I wish I had the silver bullet for this! I see a couple of tasty cars on CraigsList that have the same problem; I could make a bundle if I had the secret to solving this problem!

For what it's worth, my problem really seems to be temperature-dependent; last night and the night before our low was only ~50F and both mornings the car fired up on the 2nd attempt.

But I haven't checked the full battery-to-engine-chassis ground loop, so off I go....

PS: For checking the ground to the DME, is the easiest way to confirm ground for the aluminum case of the DME to chassis ground? Or is there a different ground path?

[iplagolf]

Have you tried the DME 3 pronged jumper to rule that it is or isn't the problem. Do yourself a favor, keep a new one in your glove box. You will need it. Fritz

[guru944]

simssj;

The quickest way approach is to just run a 12 or 10ga. Multi strand Copper ground wire to the (-) of the battery and the engine Intake manifold, motor mount/Transmission bolt - works good (Home Depot).

Check Voltage at (-) of Coil with a DMM Digital Multimeterit should go to 0.01Vdc when engine is cranking (Harbor Freight has them for under $10). Most of the Signal issues i see are directly related to the engine harness.

There should also be a ground on the exhaust system near the rear B Pipe before the muffler.
***No one thinks about this one at all***

This is why a Active Ground Kits work/sell and when you install them properly you + hp gains.

[simssj]

@iplagolf: I'm not sure I'm following you... Are you asking if I've 3-wire jumpered the DME relay (socket)?

I haven't. I have replaced the DME relay with a known-good (new) one. No difference. I did jumper the DME Relay socket to confirm operation of the fuel pump.

Are you suggesting something different?

[simssj]

@guru944: Thanks for the details. This seems easy, quick and cheap. I'll get on this ASAP and report back!

[simssj]

(Sorry about the delay in maintaining this thread to its conclusion; life got in the way)

I'm still unable to get my 924S to fire up when it's cold. I've tried almost every suggestion offered to no avail. But I have a new data point that I hope will spark someone’s thinking. (See what I did there? ;-)

A few days ago, when the weather was cool (around 40F) my father-in-law drawls to my son, “I ain’t never seen a car that I couldn’t get started.” (Or words to that effect; I was out of town at the time, so I’m getting this all second hand.) So Paw-paw and Junior head out to the driveway, Paw-paw has a can of starting fluid in hand.

(Caveat: I am not a fan of the starting fluid, but – again – I was out of town, so don’t get on my back about it!)

As they attempt to get the car to fire they discover an interesting thing: As long as they squirt little blasts of starting fluid into the intake the car will run at idle, but as soon as one touches the throttle to increase speed the engine cuts off like it’s been switched off.

Now, this makes no sense at all to me, but maybe it does to one of you? About the only things that I haven’t diagnosed or verified during this journey are the TPS and MAF side of the equation. Maybe something in there????

Quick re-cap of this saga thus far:
Problem: Car starts and runs normally when temperature is above around 50F ambient. Car will crank but not start when temperature is less than around 40F. Starts are iffy between 40F and 50F.

Things checked or done; no problem found:
Got good tach bounce, but cleaned the sensors just to be safe
Cleaned and checked all engine grounds
Checked Fuel Pressure and leakdown at the rail
Checked injectors firing (with noid light)
Check for good spark
Re-soldered all DME connections
Replaced DME relay with known good unit
Replaced Temp Sensor
A thousand things I can’t remember right now.

I’d really appreciate some guidance on resolving this problem; Spring is in the air and I really need to get this car back on the road!

Thanks, again, for all the help you folks have offered!