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

[Blue Lightning]

Car: 1976 2.0, FI

Background: bought in November last year. I am least the 4th owner. Drove a couple times, then plugged up the fuel filter (? no fuel pressure when the fuel pump was running, but see note below on replacing the fuel pressure regulator), so spent a couple months with the gas tank out and replacing the fuel lines, pump (and yes, filter). Read a lot in the meantime, so corrected vacuum line issues (vac advance was connected, replaced all the lines that didn't already look new, added hose clamps). But where the car started and ran well last year, I can't get it to happily start any longer. While cranking, it will skip and catch. After a few minutes of cranking (and skipping and catching) it will finally start and run fine for a few minutes but only idle around 1000 rpm (until the AAR warms up the rest of the way?), then dies. Starter fluid has no effect, so guessing it is running rich? Pulling the air hose off the cold start injector while running causes the idle to increase to maybe 1400 rpm (but with hunting, as you might expect with that big leak). There are a lot of threads on what to look for when the engine is running lean, but few on it running rich.

Some things I have checked:
- AAR works (https://youtu.be/YmO1ZCLDB7g)
- Compression is good (130-140 psi cold)
- Replaced the MPS last weekend with new one from autoatlanta, after finding the previous one would not hold a vacuum
- Replaced fuel pressure regulator after finding it bad (yes, I've been told these never go bad, but this one would not hold any pressure). Fuel pressure set to ~29 psi.
- Timing is good (~27 degrees at 3500 rpm with vac retard disconnected)
- Vac retard mechanism is working (checked with vacuum pump)
- Spark on all 4 cylinders (checked by timing light)
- Smoke tested the intake...only leaks now are the bearings on the throttle plate
- CHT reads 2k at cold, ~250ohms at warm/hot (at splice connector)
- Plugs look okay (see attached)
- Fuel injectors are all spitting fuel (checked into bottles...all about the same) and have new seals
- idle adjust screw has some control, but I cannot get the idle above 1000 rpm with it
- valves adjusted back in December (my first time, but the car runs so I assume I didn't do too bad of a job?)

I'm at a loss of what to try next. (IMG:style_emoticons/default/hissyfit.gif) I'd like to consider myself fairly good with cars (we have a 1966 Mustang, as well as a Mazda RX-8 that I have kept running for >20 years), but I'm about ready to throw in the towel and have the 914 towed to someone smarter than me!

Thoughts?

[Lockwodo]

This resource from the 914 Tech Notebook is a really helpful step by step troubleshooting guide:

https://bowlsby.net/914/Classic/zTN_Man06.pdf

Also, looks like you're located in Atlanta. Perhaps get help from the Auto Atlanta folks to sort it out?

[emerygt350]

If they sold you an MPS then hopefully he can help.

Original distributor?

[Blue Lightning]

This resource from the 914 Tech Notebook is a really helpful step by step troubleshooting guide:

https://bowlsby.net/914/Classic/zTN_Man06.pdf

Also, looks like you're located in Atlanta. Perhaps get help from the Auto Atlanta folks to sort it out?

Thanks for that article link. I thought I had been all through Bowlsby's website, but hadn't come across that one yet.

And yes, AA is the place with smarter people I would have it towed to. (IMG:style_emoticons/default/dry.gif)

[Blue Lightning]

Original distributor?

Yes, with a Pertronix (the 1847v module that belongs in these cars). Odometer shows 80k miles, and based upon the records I have I don't have any reason to believe it has rolled.

I've looked at the 123Ignition replacements, but didn't want to change too much all at once (for the same reason I'm having problems now).

[emerygt350]

Yeah, that's a tough decision.

A leak will not cause hunting on a djet. Hunting is normally due to a lean mixture. I wonder if it isn't the mps. This could be a wiring issue.

Just to be clear, it ran fine, then didn't, and then you changed all those things?

[Superhawk996]

Pulling the air hose off the cold start injector while running causes the idle to increase to maybe 1400 rpm (but with hunting, as you might expect with that big leak).

What does this mean? The cold start injector has a fuel line going to it not an air line.

Did you mean AAR?

[emerygt350]

I haven't looked in a while but doesn't the aar air to the plenum go through the cold start body?

[Blue Lightning]

Pulling the air hose off the cold start injector while running causes the idle to increase to maybe 1400 rpm (but with hunting, as you might expect with that big leak).

What does this mean? The cold start injector has a fuel line going to it not an air line.

Did you mean AAR?

Yes, the hose that runs from the AAR to the cold start injector.

[Blue Lightning]

Yeah, that's a tough decision.

A leak will not cause hunting on a djet. Hunting is normally due to a lean mixture. I wonder if it isn't the mps. This could be a wiring issue.

Just to be clear, it ran fine, then didn't, and then you changed all those things?

It ran "fine" (as in it would start and I could drive it ~15 miles, but the idle would be around 300 rpm at that point and the temp gauge in the console pushing the red zone...not that this gauge is all that accurate).

Then I plugged the fuel filter up (fuel pressure of 2-3 psi at injectors), so that started with the litany above (gas tank, fuel pump, filter, lines), which led to the car not wanting to start again, so starting poking around at the vacuum lines then the FI where I found the bad MPS and fuel pressure regulator, and mis-plumbed vacuum lines). And now the car is less than interested in starting when cold (60F outdoor temp).

[Superhawk996]

It ran "fine" (as in it would start and I could drive it ~15 miles, but the idle would be around 300 rpm at that point and the temp gauge in the console pushing the red zone...not that this gauge is all that accurate).

Has your oil temp gauge previously worked properly?

It shouldn’t be getting near red zone in a 15 mile drive. Not so sure I’d discount that symptom.

[Superhawk996]

In photo of the plug - why does the compression washer look all wet and oily?

You say starting fluid doesn’t work when it stalls and dies. Elaborate on what you’re using (are you using starting fluid or some other crazy aerosol people seem to call starting fluid). Have you checked spark when the starting fluid isn’t working (ideally by pulling a plug). If it’s rich and flooded the plug should be soaking wet with gasoline. Is this the case when it won’t start at all?

[emerygt350]

What hawk said.

[Blue Lightning]

Has your oil temp gauge previously worked properly?

It shouldn’t be getting near red zone in a 15 mile drive. Not so sure I’d discount that symptom.

That 15 mile drive to work (then another 15 home) are the only real trips I’ve taken in the car since I got it, so don’t have anything else to base the temp gauge on.

I was looking at the temperature issue when I plugged up the fuel filter. I did find that the thermostat for the baffles was broken (which I replaced and is now working), and that there was a paper towel in the fan impeller (which I removed). I’m hoping that these address the potential overheating, but without the car running I have not been able to confirm.

The heat has been removed along with the associated air guide plates (some aftermarket exhaust is on the car). Many of the engine tin grommets are missing, and the engine tin to body/shelf seals are shot, so there are plenty of other things to look at if the above hasn’t at least improved the cooling. It’s an air conditioning car, so there has been some metal modifications in the engine bay, so it is never going to be “factory” in that sense. Current plan is to look at the heat (and associated air guide plates) this fall/winter.

[Superhawk996]

Got it - agree; set aside cooling issues for the moment as not likely related to the cold start issue.

What I’m trying to get a sense of is whether there is some underlying fast overheating (ie mouse house in the tin) that would quickly confuse the Fuel Injection by maybe starting too rich but very quickly going too lean to re-start if the 3/4 side head when the CHT is overheating.

[Superhawk996]

Would be useful to see condition of all 4 spark plugs.

[Blue Lightning]

Got it - agree; set aside cooling issues for the moment as not likely related to the cold start issue.

What I’m trying to get a sense of is whether there is some underlying fast overheating (ie mouse house in the tin) that would quickly confuse the Fuel Injection by maybe starting too rich but very quickly going too lean to re-start if the 3/4 side head when the CHT is overheating.

I have not seen any indication of debris that I would expect a mouse house to leave on the floor under the car.

I did find (what I consider) a decent set of videos on debugging a won't-start-914:

https://youtu.be/-bMBoADt7Fg?si=wxh7gFKWUtGtuKN8
https://youtu.be/MSQWQ2Nd2Ck?si=ATNib8WZ99NExASH
https://youtu.be/fbmanDTIun8?si=ukzTAFeUDxrAAm5_

I mention this as when going through their ECU connection debug in January, I noted that the CHT was reading 4.95 kOhm at the ECU, where it reads ~2 kOhm at the splice today. At the time this didn't seem unreasonable given the data at http://www.914world.com/bbs2/index.php?sho...351065&st=0

but the Kjell Nelin article linked above brought me back to those videos where I saw the resistance measurement I made at that time. Not sure if I buggered that measurement, but want to go back and double-check it. In the video series, the root problem ended up being a disconnected CHT, so it might be a wiring problem making the CHT look (close to) open at the ECU that is preventing the car from starting. Hopefully nothing wrong with the wiring harness (IMG:style_emoticons/default/smile.gif)

I did order a new CHT. The current plan of attack is:

- Check CHT value at ECU (while wiggling wire, etc.)
- Replace CHT with new, if necessary (I have one Brad Mayeur's spacer/standoffs on order, too, so would like to add this if I pull the CHT, which I probably will anyway since I bought a new one)
- while ECU is disconnected, verify the coil resistances of new MPS at the ECU
- probably check all the leads at the ECU connector, since I have disturbed quite a bit of stuff over the last few months
- Crank engine for a minute or so, but pull spark plug(s) before it starts running and check for gas/wet
- Remove fuel line going to cold-start injector and plug, to verify this isn't flooding the manifold
- Disconnect and plug decell valve hoses (this was suggested in the Kjell Nelin article)

[Superhawk996]

How do you know you didn’t buy it with a mouse house - just sayin’ (IMG:style_emoticons/default/poke.gif)

The fatal mistake I keep seeing guys make when trying to troubleshoot problems is getting past their own assumptions.

Your to do list isn’t too bad - you especially need to follow up why the resistance to CHT differed between the engine connector and the ECU

I’ll caution you about shotgunning parts by swapping CHT and adding in the extension. adding more variables into the mix rarely helps and more often than not makes things worse.

Troubleshoot what you have. Test and verify existing parts. Avoid the path of least resistance that is shotgunning parts.

[L-Jet914]

If you plan on removing the CHT sensor be sure to get the proper socket that allows the wire to pass through it.

[Blue Lightning]

In photo of the plug - why does the compression washer look all wet and oily?

You say starting fluid doesn’t work when it stalls and dies. Elaborate on what you’re using (are you using starting fluid or some other crazy aerosol people seem to call starting fluid). Have you checked spark when the starting fluid isn’t working (ideally by pulling a plug). If it’s rich and flooded the plug should be soaking wet with gasoline. Is this the case when it won’t start at all?

Unsure why the compression washer looks wet and oil. The engine itself is pretty dirty/oily (weather has just gotten good enough to clean it). I have had the plugs out, but not replaced, since I bought it in November.

The starting fluid I'm using is attached (heptane and ethyl-ether).