Difficult "extreme cold" starts, 1984 CFI

[kishy]

The Townie has stayed faithful to me and continued to - knock on wood - run great. Fuel economy is looking a bit bleak but that roughly coincided with the drop in temperature so I don't really suspect much is wrong there. If I do the math and it's <10MPG I'll revisit that though.

A new problem has surfaced and so far also coincides with temperature. The engine will crank, sputter like it it's trying to start (so has air fuel and spark, just not in a mixture that it likes), but not actually fire up and continue running when the key is released. In order to start it in these cases, I've had to crack the throttle open to perhaps double cold idle while cranking and it starts without issue then. The same applies to holding it WOT and immediately releasing as soon as it starts, starts no problem.

It has happened 3 times: yesterday morning, this morning, and today after work. -11, -14, -9 Fahrenheit respectively, which are roughly the lowest temps this car has yet experienced with me.

Any thoughts on what's going on here?

FYI my starting procedure is:

• punch the accelerator once to set fast idle
• turn to run, let the fuel pump prime (at least halfway through the prime, I don't just go straight to start)
• crank engine, normally starts no problem. Crank time will increase inversely proportionate to how short I let the fuel pump prime go for.
• if it did not start, repeat with throttle cracked open as described above, starts and runs

Any thoughts? Is this typical of the CFI cars?

This car came to me with a remote start system installed, but I opted to remove it as I didn't have a remote for it. I'm having a hard time seeing how it could possibly work since there doesn't seem to be any way to set the fast idle cam when leaving the car...input there is also appreciated since I might choose to install one again.

FYI - not related to starting I'm pretty sure: my exhaust heat riser flapper valve thing is disconnected and fully open. The vacuum system to pull hot air into the intake from the exhaust manifold is capped off due to leaks. I realize these are probably causing it to take longer to warm up and probably part of my fuel economy issue since more of my (very short) commute is spent <130 degrees. On these cold days it's not even hitting 180 by the time I'm at work.

[VicCrownVic]

The first winter I had The Ice Car it would run really rough in cold weather starts if I could get it to start at all. My symptoms were symptoms were pretty much what you describe with yours. I ran a bottle of iso-Heet fuel line antifreeze/water remover with every fill up and after a while I noticed that it was taking colder temps to get it to act up and eventually it had no cold start issues.

Since I had the fuel lines disconnected for a while when I was swapping the engine I figured it would be good to run that stuff again. This time I've been putting a bottle in every other fill up and no starting issues so far.

[kishy]

The first winter I had The Ice Car it would run really rough in cold weather starts if I could get it to start at all. My symptoms were symptoms were pretty much what you describe with yours. I ran a bottle of iso-Heet fuel line antifreeze/water remover with every fill up and after a while I noticed that it was taking colder temps to get it to act up and eventually it had no cold start issues.

Since I had the fuel lines disconnected for a while when I was swapping the engine I figured it would be good to run that stuff again. This time I've been putting a bottle in every other fill up and no starting issues so far.

I had wondered about fuel line antifreeze. I actually have some...claims to be "for carburetors", I'll do some research on the bottle contents before assuming it's cool for injection.

Thing about it is, it seems to me that would mean a partial restriction in fuel because of frozen moisture in the line. Then when the fuel flows past it, it melts. But why would giving the engine more air (throttle open) encourage it to start in a condition where fuel flow is reduced?

It's probably a smart idea to use the stuff anyway, so I will...just not sure what to think of the cause of the issue.


Edit: Thought those temps seemed too low. Website I was using to check the temperatures said it was in Fahrenheit but numbers make more sense in Celcius...
So still cold, but not as cold as suggested in the OP.

[kishy]

So, looks like the gas line antifreeze I have on hand (a pile of bottles I got for 50 cents each intended for the Ranger, which has a carb) is Prestone (Recochem) 35-353PRES. MSDS shows it is 100% Methanol. Seems that isopropanol is considered better in FI, but can't really find the 'why' behind that. Methanol vs Isopropanol seems to be the same comparison as Heet vs Iso-Heet, ignoring additives.

Methanol itself seems to be harmless to the injectors, O2, and cats...so I don't see why I wouldn't be able to use it, especially at <1% of the tank content. Will see if it helps with my problem.

[DerekTheGreat]

Vic could be onto something with water in the lines.

But, what RPM is your fast idle set to run at? Does the thing surge when it starts and attempts to run? My '85 did, and it was very particular about cold RPM being right as it didn't seem to add more fuel based on engine coolant temperature -it would start & idle higher, the idle would drop resulting in lower manifold vacuum, begin to sputter, computer senses dip in manifold vacuum, throws more gas at it, idle picks up and then repeat until it either warmed up or died out. My exhaust manifold thingy worked. I just think that helped with it warming up. So, you may want to try increasing cold idle/fast rpm if it's ~1,500 rpm or less. If I remember the sticker correctly it Ford wanted something like 2,200rpm, absurd really but must've kept the gas from condensing in the intake. I think I had mine set to 1,800rpm but never got the chance to test it in dead cold weather again as I sold it.

[kishy]

It's around 1900. I forget exactly where I left it. I had brought it down from where it was because >2000 seemed way too high, it screamed sitting there idling that high. It was in the vicinty of 2200 before.

No surge. Starts a tiny bit lower then kinda elevates itself up to the full idle speed within 10-20 seconds, not sure exactly why it does that, computer is evidently trying to figure out how much gas to give the large amount of cold air I guess. Could also be that the oil is just too thick and stuff doesn't want to move. If the accelerator is touched (and only if it is), it will fall off the fast idle setting to a lower but still faster than hot idle setting within 30-60 seconds, depends exactly how cold it is. There's another distinct drop in idle speed between 160-180 and one final drop in idle speed to normal hot idle right before it hits fully warmed up. I believe the last two changes in idle speed are being done by the computer via timing, as the choke heater should be long-since hot at that point, unless I'm mistaken.

At any of its idle speeds, it idles at a consistent smooth RPM with no surging.

[VicCrownVic]

I had wondered about fuel line antifreeze. I actually have some...claims to be "for carburetors", I'll do some research on the bottle contents before assuming it's cool for injection.

Thing about it is, it seems to me that would mean a partial restriction in fuel because of frozen moisture in the line. Then when the fuel flows past it, it melts. But why would giving the engine more air (throttle open) encourage it to start in a condition where fuel flow is reduced?

It's probably a smart idea to use the stuff anyway, so I will...just not sure what to think of the cause of the issue.


Edit: Thought those temps seemed too low. Website I was using to check the temperatures said it was in Fahrenheit but numbers make more sense in Celcius...
So still cold, but not as cold as suggested in the OP.

Opening the throttle helped on my '91 also. I'm not entirely sure, but does the change in TPS reading tell it give more fuel? That would be my best guess.
In some cases I had to keep the accelerator down/throttle open a bit until enough heat built up that it would run on its own.

I wondered about those temps, you're not that far from me and I was going to say we didn't get that cold here. Either way still well below freezing.

I never ran any fuel line antifreeze in my '89 when I had it way up in the UP (Houghton, MI, way up there, much colder than down here) for a few years during my college days.
The only issue I ever had with that one starting up there was solved by turning the key on multiple times to let the pump prime a few times.

[DerekTheGreat]

Opening the throttle helped on my '91 also. I'm not entirely sure, but does the change in TPS reading tell it give more fuel? That would be my best guess.

When my '85 was hot that was the only way to get it to start. Would just crank forever like someone waiting for world peace.

[kishy]

Alright, so I'm now on my second bottle of gas line antifreeze (the methanol variety) due to having filled up the tank. There's been enough driving and sloshing around that it can be reasonably inferred it's mixed throughout.

Still have this problem. Won't start unless the throttle is cracked open. When it does start, it runs at a very low RPM before it gradually comes up to high idle. The fast idle is mechanically working.

What exactly does this condition represent? Dumping too much gas for the volume of air available?

As I was typing this an idea occurred to me. Could condensation (from the hot engine when it last ran, condensing and collecting on the throttle body) be freezing in the throttle body right at the butterflies, closing the gap so there isn't enough airflow? Guess the only way to verify that is to take a look at it before I try to start it.

Unrelated: I reconnected the exhaust butterfly thing. It works, and there must be some reason it was installed, so might as well use it. Engine sounds noticeably different while it's closed, but surprisingly doesn't seem to cut down on power much (not low end, anyway). Seems to get up to 130 (the bottom of my temp gauge) slightly faster than before, however, no change for 130-180. I think the ported vacuum switch flips at 128 if I remember right? So maybe that makes sense.

PS Derek thanks for the POTM nomination. Seems people must like the ol' boat, or pity me for its powertrain...

[VicCrownVic]

Mine took many bottles before I started to notice a difference. I don't recall how long it was before I noticed an improvement, but it certainly took a while.

I don't know enough to determine if there is some other explanation but it's very possible that there could be.
Either way I don't think it will hurt to keep running the fuel line antifreeze for a while.

One other thing, if it wasn't asked already, when was the fuel filter last changed?

Edit: Or when was the air filter last changed? If it collected humidity over the summer and the pores are freezing shut that could explain why opening the throttle to let more air in helps since not enough air is getting to the throttle in the first place. (I wasn't thinking when I originally typed this post.)

[kishy]

I'll definitely keep running it. Cheap insurance against the frustration of a freeze in the line somewhere, even if it doesn't help this problem. But maybe it will.

Fuel filter has been done twice in my ownership of the car, once at the very beginning, then again when I replaced the pump which was pretty recent. I don't have a date for that but it was in November, I think. Air filter was done while it was still hot out, yeah. So maybe try removing filter for the next cold start and see what happens.

Just eyeing your signature pictures - do you have factory DRLs? Seems to be a rare thing, but 90/91 were equipped to have them.

[VicCrownVic]

No factory DRLs, I just happened to have the headlights on in that pic. I thought Canada started to require DRLs in 1990? You'd probably know more about that than I would though.

[DerekTheGreat]

...Still have this problem. Won't start unless the throttle is cracked open. When it does start, it runs at a very low RPM before it gradually comes up to high idle. The fast idle is mechanically working.

What exactly does this condition represent? Dumping too much gas for the volume of air available?

...Could condensation (from the hot engine when it last ran, condensing and collecting on the throttle body) be freezing in the throttle body right at the butterflies, closing the gap so there isn't enough airflow? Guess the only way to verify that is to take a look at it before I try to start it.

PS Derek thanks for the POTM nomination. Seems people must like the ol' boat, or pity me for its powertrain...

You're welcome for the nomination, that's a sweet townie, I like the color combo. Love your taillights.

Ya know, throttle blade icing was pretty common with carbs back in the day. Buddy once told me he'd have to get out and clean the slush off his blades now and again. Either the fuel starts to freeze or it was bad gas with water in it that then started to ice over. But from what you've said your car idles low even after sitting all night. So either the throttle blade had gunk built up on it from before or this isn't the issue. When it's idling low does it smell rich? Do the injectors look like they're just dumping fuel down the holes? At least it does idle, if it was running lean it would want to stall. But from what you've said about when trying to start it it does sound like it's running lean. I don't think these things compensate for cold start enrichment like SEFI does, why Ford has the idle set so damn high. Because once the velocity of the gas/air mix reaches a certain speed the gas droplets wont condense on the manifold runners, which is why they stall out/need a choke.
I know we've talked about the idle before and you've got it pretty high but you might want to bump it a little more to see if that fixes your issue. I didn't get a chance to fiddle with my cold start issue once I replaced the choke-pull off mechanism on my '85. But I do remember that when I did start it on the few super cold mornings we had to let it warm up, it would start and idle pretty low. However, when I came back out to the car some 10 minutes later to drive away it was idling much higher, like you described.

[gadget73]

Ever checked the output of the air and coolant temperature sensors? If those lie, cold starts can be overly stupid.

Is the high idle adjusted right? If it doesn't bump the throttle open enough it will be hard to start too. If you're sure its actually engaging, maybe bump the high idle screw a little bit.

Carb icing really only happens while running, not while sitting. Basically you're creating a large pressure differential that sets up a condition where the air on top of the throttle blades is at a high pressure relative to underneath. The air flows past the throttle blade and expands which makes the throttle blade cold. Thats why the passages in the intake are there, it forces hot exhaust gas through the intake to warm it to compensate for the heat loss from fluid expansion. If no air is moving, nothing is expanding, so nothing is getting cold.

[kishy]

So apparently "mama Ford knows best" and I should have left the high idle up where it was. Turn and a half of the screw (back to about where it was before I messed with it) and it starts cold with no throttle now.

Still seems like a stupidly high fast idle but maybe it's how it's meant to be.

Is it supposed to fall off the fast idle on its own, or is it supposed to require an additional bump of the throttle? Thinking if I lightly sand, polish and grease the steps on the cam it might start doing that, this way if I were to start it and go into the house, it won't be at ~2300RPM when I come back out.