Cold Start Issue 914-4, Admitting I'm stuck and asking for help Options

[Superhawk996]

Good - proper starting fluid.

So about the only reason that stuff won’t burn / fire an engine is when there is no spark.

So going back to assumptions - if it won’t even cough on starting fluid - you need to be looking at why there is a lack of spark.

FWIW - using a timing light to verify spark isn’t a good plan.

When a plug is soaking wet or fouled heavily enough from carbon - it will flow electricity right out the center electrode, down over the ceramic insulator (which is no longer insulating) and to the ground electrode.

High voltage energy flows though the plug wire but doesn’t manifest as a spark. The timing light doesn’t know the the difference and will flash the timing light . . . But there is no spark jumping the gap.

[Blue Lightning]

So about the only reason that stuff won’t burn / fire an engine is when there is no spark.

So going back to assumptions - if it won’t even cough on starting fluid - you need to be looking at why there is a lack of spark.

FWIW - using a timing light to verify spark isn’t a good plan.

When a plug is soaking wet or fouled heavily enough from carbon - it will flow electricity right out the center electrode, down over the ceramic insulator (which is no longer insulating) and to the ground electrode.

High voltage energy flows though the plug wire but doesn’t manifest as a spark. The timing light doesn’t know the the difference and will flash the timing light . . . But there is no spark jumping the gap.

Agree timing light is not best, but there is no way the engine is going to fire if I pull the plugs to watch them spark. If the timing light is not running, then I'm going in after the Pertronix or the coil (which is a new Beru blue one...come to think of it, this is a part that changed between the car starting and not starting well).

That said, when I crank the engine for two or three minutes (not continuously, but additive) with it trying to catch, it does eventually start and run (idle ~1000 rpm). This makes me think the spark is okay (along with what the plugs look like when I pull them after idiling). I think this probably fits with the CHT reading high at the ECU? As the engine sorta warms up during those first couple minutes of trying to fire/skip, that might be enough to bring the CHT down to what the ECU needs?

The other reason I can think of that the starting fluid won't burn is if it is way over-rich?

Regardless, checking the CHT at the ECU is first on the list for Saturday morning. If that looks okay, checking spark at the plugs maybe should be next.

[Superhawk996]

With all due respect, you’re missing the point about the necessity of pulling plugs, and making sure you have a visual spark. You’re assuming you have spark and you’re assuming it is too rich.

You cannot verify either without pulling the plugs and checking visually.

Assumptions always slow down troubleshooting and often lead down dead end paths.

Agree checking CHT all the way to ECU ought to be your first order of business.

[T.Rick6]

(IMG:style_emoticons/default/welcome.png)

Welcome to the madness; once sorted these are pretty awesome little cars.

After reading your post I have a few questions: where on the temp gauge is the needle?
just barley reaching the red, 1/4 of the scale or pegged hot? 1 thing I've learned is I'm not smarter than the German engineers, unless you can prove the engine temp I would believe it.
I think you purchased the PO's issues of running hot; messing around with the flaps and bellows tells me this was a previous issue and they tried to fix by removing the bellows etc. Paper towel in the fan is BIG RED FLAG. If it were me I would pull the motor, pull the tin and clean and inspect; these are great little motors until they get too hot and drop a valve seat, not good. I know that sounds drastic, but you want KNOWN information; I'll bet you a beer something is clogging the cooling fins; as Superhawk said Mouse nest? While the engine is out I would replace the shelf to engine tin rubber.

Here's another drastic move that I would make if you are planning on keeping the FI (I would keep the FI system) purchase a new FI harness, at 50 years old it's starting to break down, cracks lead to corrosion inside the wire sheath which equals higher resistance. This would eliminate interment harness issues.

After engine reinstall At this point I would start troubleshooting the rest of your starting issues.

If you need a hand I'm local and would be happy to help on weekends.

Todd 404 four four 1 five seven o2 (IMG:style_emoticons/default/beerchug.gif)

[Blue Lightning]

With all due respect, you’re missing the point about the necessity of pulling plugs, and making sure you have a visual spark. You’re assuming you have spark and you’re assuming it is too rich.

You cannot verify either without pulling the plugs and checking visually.

Sorry if I implied I didn't agree with you. When debugging problems I often start with the easy stuff to check (timing light on plug wires), then work back to the more difficult stuff if I don't find the problem (pull plugs and check for spark). Moving back to the ECU for the CHT measurement instead of at the splice is another example.

[Blue Lightning]

(IMG:style_emoticons/default/welcome.png)

Welcome to the madness; once sorted these are pretty awesome little cars.

After reading your post I have a few questions: where on the temp gauge is the needle?
just barley reaching the red, 1/4 of the scale or pegged hot? 1 thing I've learned is I'm not smarter than the German engineers, unless you can prove the engine temp I would believe it.
I think you purchased the PO's issues of running hot; messing around with the flaps and bellows tells me this was a previous issue and they tried to fix by removing the bellows etc. Paper towel in the fan is BIG RED FLAG. If it were me I would pull the motor, pull the tin and clean and inspect; these are great little motors until they get too hot and drop a valve seat, not good. I know that sounds drastic, but you want KNOWN information; I'll bet you a beer something is clogging the cooling fins; as Superhawk said Mouse nest? While the engine is out I would replace the shelf to engine tin rubber.

Here's another drastic move that I would make if you are planning on keeping the FI (I would keep the FI system) purchase a new FI harness, at 50 years old it's starting to break down, cracks lead to corrosion inside the wire sheath which equals higher resistance. This would eliminate interment harness issues.

After engine reinstall At this point I would start troubleshooting the rest of your starting issues.

If you need a hand I'm local and would be happy to help on weekends.

Todd 404 four four 1 five seven o2 (IMG:style_emoticons/default/beerchug.gif)

The first time I saw a 914 was probably late '90's, and have wanted to try one since. Finally have time and some money. Looked for about a year before finding this car. It's not perfect (it has been in at least 1 wreck), but it looked like most everything wrong with it was within my skill set.

I didn't realize there were rebuilt harnesses available. If the resistance of the CHT shows there is a problem in the wiring, then that will be the trigger to go there. Otherwise there are more pressing issues IMO (split tie-rod boots, right turn signal doesn't catch/release, lots of cosmetic stuff like splits in the seat stitching and piping/seals).

To answer your question on the temp gauge, it was on the line between the white and red. If the stuff I found to improve the cooling doesn't solve that issue, then I will need to look at the airflow (bad seals, missing grommets, etc.). I was hoping to delay dropping the engine until the fall but it may come to that sooner.

Agree that I have inherited the PO's issues, but that's half the fun of an old car (or boat, or house, or ...), as long as you don't wreck it before you solve them.

I may take you up on your offer of help, but I don't think I will have parts to play with this weekend, so would be following weekend. I'm in Johns Creek/Alpharetta area.
Need to show some love to one of the other cars this weekend, too.

[emerygt350]

Although dropping it is not at all bad, I don't think you are there yet... You could start with a cheap boroscope/endoscope for nests. Keep checking the easy stuff (don't throw parts at it, although there really are not many you can throw at it).

[Blue Lightning]

Although dropping it is not at all bad, I don't think you are there yet... You could start with a cheap boroscope/endoscope for nests. Keep checking the easy stuff (don't throw parts at it, although there really are not many you can throw at it).

I actually have a not-so-cheap endoscope that I will break out this weekend. Didn't think about using that. Thanks!

[Blue Lightning]

Went through the ECU connector checks:

- 1-13: 365 Ohms (60F air temp) (air temp sensor)
- 3-gnd: 3.0 Ohms (#1 FI)
- 4-gnd: 2.7 Ohms (#4 FI)
- 5-gnd: 3.3 Ohms (#2 FI)
- 6-gnd: 3.0 Ohms (#3 FI)
- 7-15: 88 Ohms (MAP primary)
- 8-10: 337 Ohms (MAP secondary)
- 9-gnd: 10 beeps (throttle position sensor)
- 20-gnd: 10 beeps (throttle position sensor)
- 11-gnd: yes light (ECU ground)
- 12-21: checked continuity pattern when cranking (points #1)
- 12-22: checked continuity pattern when cranking (points #2)
- 16-gnd: yes light (ECU power)
- 24-gnd: yes light (ECU power)
- 17-gnd: yes correct throttle pattern (TPS idle)
- 18-gnd: yes light (starter circuit gen)
- 19-[fuel pump in relay box #3]: 0.3 Ohms
- 23-gnd: 3.17 kOhms (CHT)

"light" meant that I used a headlight bulb to check that significant current would actually flow instead of just checking voltage or continuity.

The 3+ kOhm CHT seems pretty high still. I did wiggle the wire around and the reading did not change. Today the 3.17 kOhm is the same reading at the splice as at the ECU (with the ECU connector disconnected).

New CHT hasn't shipped yet...

[Blue Lightning]

Spark plug pics. Have not tried to start the car since last weekend, so these are after the last time it "ran".

[Blue Lightning]

Checked actual spark on all 4 plugs.

[Superhawk996]

CHT 3k at ambient of around 60F is fine . 76’ 2.0L used the 012 CHT sensor

[Superhawk996]

Two questions

1) what did you consider to be ground when making measurements? Engine case or chassis ground?

2) if you used the engine case as ground: is your transmission ground strap between chassis and the transmission/engine there and has it been cleaned and decent looking?

Note: great job taking the time to take all those measurements back at the ECU (IMG:style_emoticons/default/smilie_pokal.gif)

[Blue Lightning]

Pulled the CHT from the engine. Measured:

2500 ohms at 70F
230 ohms at 212F

These match the data here for the 012 sensor ('76 2.0): http://www.914world.com/bbs2/index.php?sho...351065&st=0

Did notice that the wire coming off the CHT has some insulation scraped away. Going to add heat shrink before reinstalling. This is one of the grommets that is missing from the engine tins.

[Blue Lightning]

Two questions

1) what did you consider to be ground when making measurements? Engine case or chassis ground?

2) if you used the engine case as ground: is your transmission ground strap between chassis and the transmission/engine there and has it been cleaned and decent looking?

Note: great job taking the time to take all those measurements back at the ECU (IMG:style_emoticons/default/smilie_pokal.gif)

Engine case. On the driver's side I used the AAR valve. On the passenger I use the intake runners.

I replaced the transmission ground strap earlier this year (car had original, so just because). I think that the headlight bulb illuminates in the ECU tests above indicates that the ground is good?

[emerygt350]

I ran an extra ground from my fan case to the negative terminal. While you are troubleshooting you may want an extra.

[Superhawk996]

I think that the headlight bulb illuminates in the ECU tests above indicates that the ground is good?

Gotsta’ be careful with assumptions. When ground strap goes bad the bulb test can still work due to the tendency of the vehicle to source ground through the clutch cable. But you end up with a floating ground that is different between chassis ground and engine case ground.

I ran an extra ground from my fan case to the negative terminal. While you are troubleshooting you may want an extra.

A whole separate ground is overkill for the purpose of troubleshooting but a super quick test is to use a jumper cable temporarily between battery negative and the engine case.

This would rule out any assumption that the recently replaced ground cable is doing its job (it is possible to end up with a poor ground on body side at the weld stud — not likely but conceivable).

[Superhawk996]

Overall I don’t see any glaring red flags in your measurements that explain the starting difficulty.

So I go back to what is being assumed:

Valves adjusted properly - original post made this sound a little less than 100% certain

Static timing? Would like around 7 degrees but I’m assuming since you have proper advance at speed that static timing is very close. But I am wondering how that got set if the engine doesn’t want to start and run well. Static timing is easily confirmed & set with a volt-ohm meter.

Plugs - a bit of cold start carbon fouling is present but overall not terrible. I might start with a new set of plugs just to be 100% sure that isn’t part of the issue. Carbon fouling is insidious - It doesn’t look that bad but can lead to no-spark especially with rich mixture and/or flooded plugs (wet).

Beyond that I’d go back and re-try starting and see if the plugs get wet / flooded leading to the no-start no-run and then work from there.

[Edits below]
Do you have the means to video the start attempts and post to YouTube (which can be embedded or linked here)?

Im thinking about the injectors - you said you did the bottle test and they all flowed similarly - did the spray patterns look reasonable?

[Blue Lightning]

I ran an extra ground from my fan case to the negative terminal. While you are troubleshooting you may want an extra.

Thanks for the suggestion. I have added ground checks between battery and engine and chassis for today!

[Blue Lightning]

Overall I don’t see any glaring red flags in your measurements that explain the starting difficulty.

So I go back to what is being assumed:

Valves adjusted properly - original post made this sound a little less than 100% certain

Static timing? Would like around 7 degrees but I’m assuming since you have proper advance at speed that static timing is very close. But I am wondering how that got set if the engine doesn’t want to start and run well. Static timing is easily confirmed & set with a volt-ohm meter.

Plugs - a bit of cold start carbon fouling is present but overall not terrible. I might start with a new set of plugs just to be 100% sure that isn’t part of the issue. Carbon fouling is insidious - It doesn’t look that bad but can lead to no-spark especially with rich mixture and/or flooded plugs (wet).

Beyond that I’d go back and re-try starting and see if the plugs get wet / flooded leading to the no-start no-run and then work from there.

[Edits below]
Do you have the means to video the start attempts and post to YouTube (which can be embedded or linked here)?

Im thinking about the injectors - you said you did the bottle test and they all flowed similarly - did the spray patterns look reasonable?

Valve adjustments: my uncertainty is just because this is the first time/car I have done this. I double-checked them all twice(?) after doing them, so am pretty confident, but ??? Oddly all my other cars have hydraulic lifters (or don't have lifters!). If I can get the car started, it will idle (~1k rpm) for a couple minutes before dying, and I can keep it running by holding the throttle open a bit. What would the symptoms of incorrectly adjusted valves be, other than long-term engine damage?

Static timing: I can check this, but when I did get the engine running a week or two ago I put the timing light on it and checked the dynamic timing (3500 rpm, 27 degrees, disconnected vac retard), and it was spot-on still. I checked the vac retard mechanism with a vac pump, and it moves. The timing should be a purely mechanical thing -- other that moving the distributor or moving stuff inside the engine, I don't think there is anything that can cause it to change (well, the vac retard, but for checking the dynamic timing that is).

Checking plugs today during the doesn't-want-to-start phase is high on the list. Need to finish putting the intake runner on the passenger side back in still (removed to access CHT) -- ran out of time last night.

Injectors: I did not look at the spray patterns. My concern at the time was that I might have gotten something into the fuel lines when replacing them that might have plugged one of them, but the bottle testing indicated that they were all flowing. Further down my list is to check them more thoroughly.

Today's debug list:
- Add jumper cables/test ground between battery and chassis & engine
- crank and pull plugs if the engine still doesn't want to start to check for fuel fouling
- Plug cold-start injector fuel line to make sure the cold-start injector isn't causing problems.
- plug the decell valve hoses to make sure this isn't causing problems.
- double-check fuel pressure