Cold idle problem gets worse and worser..

E

Elessee

Well-Known LVC Member
Joined
May 19, 2007
Messages
350
Reaction score
6
Location
Modesto
'88 LSC, all stock, ran fine in all respects.
-----------
Previously, this car starts and idles just by turning the key. About 6 weeks ago, the cold engine died at idle.

Upon starting, RPM dropped a bit too low, coughed a bit, and threatened to die... and then died. On the second or third try, it idled OK and ran great afterwards.

Jump ahead a few days, and it was evident that it was no isolated incident. The cold-start idle thing was becoming a problem. In order to keep it running, I had to press the gas pedal a few times and keep RPM up there. Once the engine was warmed up, it idled OK and generally ran fine.

As the problem progressed, the engine needed to be completely warmed up before the car could be driven.

Finally, even with the engine at normal operating temperature, the engine is likely to die while stopped. The throttle must be pumped to keep it running. Sometimes I need to brake while throttling (in D at a stop sign) to raise RPM and get enough engine momentum to take off.
When cruising down a street, steady throttle, the engine sometimes "hunts", with RPM rising and falling, and the throttle has a delayed, or no effect. It's becoming a potentially dangerous thing..
----------
[[ One peculiar thing that might be a clue: While the cold engine is warming at idle, and the throttle pressed to keep RPM high, RPM will rise to about 2500, drop to 1500 and back up to 2500. The foot on the throttle is steady. RPM rises and falls, rises and falls about every two seconds...

As the engine warms up, that 1000 range gradually shrinks. Then RPM is perhaps 2100 and falls to 1900.... up and down, up and down. ]]
----

I did some research and ran codes, and came up with a few ideas.

Vacuum leak: None is evident. I physically and visually checked all lines (used an inspection camera under the intake manifold) and everything looks good to me. Strong vacuum. Vacuum solenoid connections are good. After a lot of checking, my instincts tell me vacuum is OK... but who knows for sure..

IAC- idle air control motor. It was mounted tight, clean inside and sealed just fine. I took a lot of time on this, cleaning and testing it however I could, but in light of its being such a likely suspect, it was replaced anyway.
No effect. Wasted $60.

TPS- Throttle position sensor.
I got a code on this one. Code 63. Boy, it sure would be nice if this was the problem. Two screws and I'm done. But, I put an ohm-meter on it and it seemed OK. Being such a likely suspect, it was replaced anyway.
No effect. Wasted $35.

EGR- exhaust gas recirculation.
This one, Code 34, was in the computer's "Continuous Memory", and popped up again after I erased that memory.
The EGR seems fine to me. The diaphragm is OK, and the port / valve are clean and air-tight. I can see the diaphragm move when vacuum is applied to the vacuum port. I have watched and seen it operating with engine running.

Engine coolant temperature sensor- (Code 51) Mounted near distributor, fuel rail, passenger side.
I pulled its connector off while the engine was running, and engine died. So, it seems to be functioning (or at least communicating with the computer).
I removed it, connected an ohm meter, and heated the element with a flame, and resistance fell smoothly. (Or did it rise? I forget)
Anyway, this one seems OK.

I didn't test the other coolant temp sender, mounted on the other side of the distributor. That one looks like a pain to remove and coolant might need to be drained first. (Anyway, the Code 51 specifically points to the other one.)
-----

Understand that at this point, the engine barely runs. I managed to do one KOER (key on engine running) but it was a struggle.
Codes said things like "Idle RPM too low or too high", and "Engine is running lean".. Anyway, without fixing KOEO stuff first, KOER codes are pretty much useless.

Codes I got were mostly KOEO, and they pointed to things like the EGR, the engine coolant temperature sensor, and the TPS.
----

I'm out of ideas and after wasting $100 on parts, it's going to the repair shop on Wednesday... unless someone can think of something else I can check.

=============
[edit] I just called the part store and they have an EGR "valve position sensor" in stock.. $25.. That's the thing mounted on the rear-end of the EGR valve. So, I'm gonna go get it.
[Edit #2] No luck with the EGR sensor... I should'a known better.. I tested the original.. ohm meter.. and it did seem OK. It's just a potentiometer (like a volume control) with 3 pins. The plunger in-and-out movement is the "volume knob".

With plunger movement, resistance was ~ 4K ohms, smoothly falling to ~zero between two of the pins, and rose from zero to 4K between two other pins. Oh well.. throwing new parts at a problem is normally unsuccessful.
============

Might it be a fuel delivery problem? I don't think so.
The car ran great even after the symptoms began, as long as the engine was warm. I even took it to the snow knowing about the cold-start problem. Once warmed up everything was OK up in the mountains (a few weeks ago).
 
While checking the MAP sensor, I notice the engine "idles" much better with the MAP disconnected. "Idles" is in quotes because it doesn't idle at all unless my foot is on the gas.
However, the cold engine runs pretty smooth as low as 700 RPM with the MAP disconnected.

The MAP checks out as far as I can tell... its diaphragm is intact and its hose is OK.
---------

But a new problem with the dash temperature gauge caught my eye: It falls to zero and stays there when the engine is running. Or, when the key points to RUN / engine off..

Turn the key to OFF, or to Accessory, and the gauge reads around the 'N' in Normal, regardless of engine temp. Things weren't like this before. (Analog instrument cluster, btw)

The gauge's single-wire (red / white stripe) coolant temperature sender under the distributor is a thermistor, and as the engine cools, resistance rises as expected. So that sending unit seems to be OK.

Now I'm checking everything temperature related. There's gotta be around 5 different coolant temperature sending devices on this car.
Considering the original cold-start symptoms, and the Code 51 (ECT- Engine Coolant Temperature sensor) I'm starting to lean towards a bad wire or ground or similar involving coolant temperature.
 
DLF said:
Vacuum leak....
Click to expand...

Certainly could be.

That was the first thing I looked for.. crimping the rubber hoses, plugging the red "source" plastic vacuum lines.
Looked under the manifold with a camera. I even bought an upper manifold gasket, intending on taking a really good look around, but chickened out.
Isolated and plugged the EGR's and other vacuum solenoids.
Sprayed starting fluid in the likely spots.

What other leak search techniques are there?

But at this point the thing runs so bad that someone needs to be in the driver seat on the throttle while someone else is outside. I guess my mechanic will score on Wednesday.. unless I gets real lucky.
 
Running lean and a vacillating idle, both point to a vacuum leak.

I looked for one on my Cobra for quite a while, before finally breaking down and starting take everything apart until I finally found a split hose under the intake manifold.

Different car, of course, but what I'm suggesting is that sometimes you just have to start taking things off until you find the problem.
 
Intake removal looks simple enough. I could finally tighten those leaking valve covers... check all that ancient wiring behind the engine. (260K miles)

Admittedly, I've been going at it sorta half-assed, afraid of getting stuck for needed parts. I don't have back up wheels except for a small bike, and the wind chill factor is substantial these days..
 
If you don't have a backup, then I wouldn't suggest tackling it yourself.

I could be wrong, I'm just taking an educated guess. ;)

Does it seem to run ok at highway speeds? Usually, with a vacuum leak, once you get the RPM's up a bit, it'll run normally.
 
highway? fug... i wouldn't put it on a highway.
One short line in the first post illustrated some very weird activity.

To expand on it, I'm cruising down a fast city street, maybe 48mph in 4th gear, tack is near 1800 or so.. and something lets go.. and the engine RPM drops. Since it's in OD, the car doesn't slow much.

By reflex I step on the throttle, but nothing happens.

About two seconds later, there's power again.
 
Well, that doesn't sound like a vacuum leak, so I may have been barking up the wrong tree.

Might be time to have someone take a look at it. Could be almost anything, ignition and fuel being the most likely suspects.

Sorry I wasn't much help.
 
Vacuum is as good a guess as any, imo.

One thing against it.. I was messing around and started the car with the EGR valve body completely removed. So, there was a big open hole in the manifold.
If that made a difference, I didn't even notice until I shut it off and walked over there and saw it on my work bench..

I appreciate any input you might have to offer.

What bugging me is that new symptoms pop up as time goes on. It's as if something is gradually coming unglued.
 
A final idea.

Did all of the DTC's that you got start at once?

If so, it's possible that you may have a bad ground. Easy thing to rule out, take a battery cable, connect one end to an unpainted part of the motor and connect the other end to the negative battery terminal. If things improve, you have a bad ground cable, find it and repair it.

Once again, just guessing here, but it's something I thought about last night.
 
The codes came up all at once.
When I realized the cold idle problem was evolving into an idle-anytime problem, I hooked up the reader.
EGR was the only thing in Continuous Memory.
TPS throttle position, and ECT engine coolant temp were relatively "new" problems to the computer.

I just tried grounding Batt Negative to the body.. and to the engine block. No effect.
--------

And, I ran codes again, but on the cold engine. Since it's cold, as might be expected, codes that popped up relate to Intake Air Charge Temp, Cooling Temp sensor / ECT engine coolant whatnots are "out of range or.."

There were two other codes. EGR and TPS. Since the EGR function is only around mid-throttle, I don't connect it to the cold idle thing (unless it's stuck open and leaking vacuum or similar, which it's not).

I installed a brand new Throttle Position Sensor last week.. so what's the problem with it?

" #63 - Throttle Position (TP) circuit fault, below minimum voltage. "

I'm thinking that running KOEO codes on the cold engine wouldn't matter as far as the TPS circuit goes, and that the code is valid.

Wiring to the TPS is bundled with lots of other stuff in the harness and I need to take a much closer look at it. I have the complete wiring diagrams for this car and the Electrical / Vacuum troubleshooting manual..

I can see where these symptoms could be due to a computer which has little or no idea what the throttle position is at any time.. any place.. at any temperature.

=========
[EDIT]

Well, this is interesting. I'm looking at the engine area wiring diagram sheet.

At one time or another, I've gotten codes on the ECT (engine coolant temp), the TPS (throttle position), EGR (exhaust recirculation), MAP (mass air), ACT (air charge temp sensor) ... and guess what. They all have a wire in common.

Wire number 359 - (black / white stripe) 18 gauge.

And where on the EEC does this wire ultimately terminate? Pin 46 - Sensor Signal Return

The only other thing (wrong.. see EDIT #2 below) connected to that wire is the "Knock Sensor".. but I dunno if this car even has one. Where would a knock sensor be.. pre ignition.. hmmm... Google.. piezo element.. mounted on the heads or the block.. one on each cylinder bank. I see that newer (2000+) Lincolns have them. At least one forum reply says the Mark vii doesn't have them. ( (EDIT) The diagram shows a knock sensor but the wiring dead-ends at a connector. It does not reach the EEC. )

( EDIT #2.. I just noticed that wire #359 is also one of the "V.I.P Self Test Connector" wires.)
========

OK.. suppose there is some kinda problem with wire # 359, like the insulation is worn and it's grounding somewhere. Where do I start? I don't want to unwrap all those harness branches. What's the smart way to go about it..

First see if it is actually grounded. Bingo.. it is. about 4 ohms.. that's a pretty solid ground.

jeeze.. Now what. Although the wire itself could be grounded, just one of those sensors could have an internal fault, and that would ground the entire wire.. unless none of the sensors has any grounding at all.
Back to the diagrams..

Sensors have no ground connection wiring.. not supposed to anyway.. so the wire's gotta be grounded.

(I'm thinking out loud and using the forum to keep notes.. sorry if it annoys anyone.)
 
Now what do I do?

I know that wire #359, the EEC Sensor Signal Return wire, which is mutually connected to ALL of the sensors which are coming up in the Code Reader, is somehow grounded. About 4 ohms to ground.

Put another way: If it is using wire #359, it is spitting up a code.

Must I unwrap the harness? The problem could be inside a harness... contact between 359 and a bare ground wire.

A bare ground wire inside a harness is unlikely, unless something over heated and melted. I see no evidence of that.. yet.
 
This wire tracing and search is virtually hopeless. I'll call the mechanic tomorrow and see what he advises.

Thanks to DLF for inspiring me to open the hood this morning. If not for that, I wouldn't have found (what looks like) the root of the problem as well as the solution.
 
Dropped the car at the auto-shop this AM.

Had to disconnect the MAP to drive it, which meant it was running as rich as possible. Just when I got there (about 3 miles) i see a little smoke rising from the engine compartment. And there was a single wisp of smoke in the cab, rising from near the passenger seat... hmm.. The EEC is over there by the passenger door.

He seemed interested in that grounded signal wire. Asked me if I used a digital or analog meter, and wrote everything I had already done on the work sheet.. but no comment yet.
They are busy. It might be days before I hear anything.
 
Finally got the car back after 9 days.

They changed the MAP sensor and that's the only "part" I paid for. However, there had to be complications for them to keep it for so long. I was told about the MAP sensor last week.

Registration is due, smog check required, and told them to smog it if they could get it to normal running condition. It fell just short of passing. The plan is to run some magic smog snake oil fuel additive through a tank of gas, hopefully burning away any carbon residues in the cat converters, and retest.

I suppose driving it hundreds of miles with the bad MAP and rich mixture gave the EEC troubles, and then timing plus mixture plus everything else got bent out of shape.
Seems curious that there was no "check engine" lamp at any time... nor did the test reader show any MAP code. [edit-- My notes do show a #22 code from a few weeks ago in the one pointless KOER test I managed to run. ]

oh well.. I just hope the O2 sensors and catalytic converters are not too far gone to recover.

btw, that sensor wire I suspected still reads a strong ground connection, so that was just a false trail...
 
Sure. I'm always ready for something different. What do you have in mind?

I've already done a few things. New IAC and TPS, new MAP, new ignition wires, rotor and cap, plugs, oil and filter. Removed and cleaned the injectors / new seals for them, ran some fuel treatment in the tank.. Found and fixed a small vacuum leak under the manifold.. Cleaned out the throttle body and that general area. Did the Sea Foam treatment.. and ran some in the crankcase before changing oil.
Attempted to pass smog with a can of Guaranteed to Pass.. but it failed.

I haven't tried to smog test it again since getting it back from the shop, and doing all the above tuneup stuff, but the car does run pretty well.

Currently fixing a leaking rear caliper. I have the kit but can't rebuild this one. A rebuilt is on order and should be installed by tomorrow.

Once that brake is done, it will be drivable, and the plan is to smog it sometime this week.
 
Elessee said:
Sure. I'm always ready for something different. What do you have in mind?
Click to expand...

Start with a 5 gallon can of 89 octane. I wouldn't add it to your existing fuel as you may still have SeaFoam in there and you will want to burn that up BEFORE adding the 89 octane, hence the gallon can at the ready for when the tank empties.

Time the engine to 10 degrees BTDC. Don't forget to disconnect the spout connector BEFORE timing the engine and reconnecting it AFTER you're done.
 
No doubt there is still some SeaFoam burning.
Even if the engine is hot off the freeway, when I come to a stop there's a little white smoke drifting out the tail pipes. Head gasket? :eek: Nope. It's not steam.

First I gotta joy ride through about a half tank of gas, and then will follow your suggestion. Timing was listed in the smog test results as 10 BTDC.
 
This shop probably works about 20 cars a day, including a whole lot of Fords. Certified as a official "Test Only" California smog check station. There is just no way they wouldn't know how to properly check timing.
-------

Fixed my ATC last night. Self-Test Code #2. "Floor/Panel actuator out of position."
Removed the control box. Hooked a battery to the actuator motor pins and drove the door open/closed, back and forth about 10 times.
Checked resistance between it's other control pins and all were within specs. Everything seemed ok.
Ran another diagnostic and Code #88 (all is well). Test drive shows no problems with heat, cooling, vent controls, etc.

I remember this ATC originally failed in Reno one winter a couple years ago. Temperature was sub zero. I heard a loud scraping from the dash when I turned on the heater just after climbing into the frozen car.
I guess the door slipped on it's hinge, a gear slipped.. or something.. and the computer detected the actuator being "out of range" from that moment on, and just refused to operate it. My messing with it somehow moved it back into range.
 
Congratulations on the diagnostic and repair!

How did you get to the blend door? I have NO hands on knowledge of getting there.
 
The blend door never had a problem. My faulty memory recalled a problem with it. It is way up there by the heater core.

The actual self-test code was #2. This Floor/Panel door and it's actuator is near the center of the dash, but actuator removal requires pulling the dash... and I had unpleasant visions of being forced to do that. :(
But before assuming something is mechanically broken or electrically fried, and getting all upset about it, lets first try the diagnostic procedure..

Expose / Remove the control assembly (snap-on dash trim and then 4 screws). Pull it out as far as it goes. Un-clip it's connectors.
The manual describes which pins on it's two large connectors should be checked.
Two of the pins (6 and 9) run the door actuator's DC motor, so moving the door is just a matter of directly applying 12V (unless the motor is toast OR the door is jammed tight). Switch polarities to reverse the door direction.
The other diagnostic tests for this actuator are measuring resistances between computer feedback / drive connection pins.

I figure with a frozen door, something slipped, and the computer immediately lost track of the door position.
When I was done testing, I left the door near the middle of it's swing, hoping the computer would then detect it. And it did. But I really don't know what the actual problem was, or how it got fixed.

----------

[For those who are unaware, the ATC can run a self-diagnostic at any time. Turn the ignition to Run. Press the On/OFF and the Defrost buttons simultaneously. Release them, and within 2 seconds press the A/C button. Codes will appear. Code #88 means there are no faults (in the computer memory). Press Cool to exit.
The full list of codes and diagnostics is in the Chassis/Electrical manual Vol. B.
 
You must log in or register to reply here.

Members online

Total: 39 (members: 1, guests: 38)
Back
Top