Hopeful fix for warm engine dying

[Millster]

For a long time, my truck has had a very vexing issue. It will start on the first crank almost every time (even after sitting for months) and run great... until it gets warm. And we're not talking engine warm, we're talking about ambient temperature warm. The engine can come up to temp and it'll run fine. On cold days, I can go almost 20 miles before I start having problems. Warm days, sometimes it will die just sitting in the driveway.
In any case, the symptoms right before it dies are consistent with running out of fuel and to add to that, after it has died the fuel pumps will not run. If I let it sit for about 15 minutes without trying to start it, it'll take a few cranks to get a prime and then fire right up.

Someone on here mentioned problems with the fuel pump relay. Well, I had replaced mine not too long ago so I figured that was not the issue. BUT, that relay is tied to the ECU relay. The wheels in my head started turning. I finally got a chance to take a look yesterday and sure enough, my ECU relay is hot enough after about 10 minutes of running that it's uncomfortable to put your hand on. Wrapping it up and letting it get really warm makes the engine start to run weird. Give it a little tap with the end of a wrench and the engine will stutter and then clearly go through re-reading the sensors to get the mixture right before settling back into idle. Tap it hard enough and the engine dies.

Looks like I found my problem. And to think... it's only been essentially off the road for 6 years.

 

[RyanL]

Your symptoms sound exactly like a bad TFI! That good to know. Sounds like the coil in the relay was going bad. Where is the relay located and where did you buy it? I might grab one and keep it as a spare or something or just replace it as a preventative measure. People tend not to think of relays as moving parts, but they are and they do go bad (more often it's the contacts that go bad rather than the coil part in my experience though). They, sometimes can make a chatter noise or a hum when the coil goes bad.

 

[Frank The Tank]

I was going to suggest those cold headbands you just add water to. Wrap your entire truck in them and drive through puddles, but I think your idea is more practical.

Frank

 

[Millster]

Your symptoms sound exactly like a bad TFI! That good to know. Sounds like the coil in the relay was going bad. Where is the relay located and where did you buy it? I might grab one and keep it as a spare or something or just replace it as a preventative measure. People tend not to think of relays as moving parts, but they are and they do go bad (more often it's the contacts that go bad rather than the coil part in my experience though). They, sometimes can make a chatter noise or a hum when the coil goes bad.

Actually, the TFI had gone out about 6 years ago. It died outright and completely. It wouldn't restart for anything until I learned about the TFI. I thought that I might have gotten a bad one and did end up replacing it again. Didn't fix it, obviously.
The relay is on the passenger side fender closest to the firewall. The fuel relay and ECM relay sit right there together. I had replaced the fuel relay when this issue started cropping up because I had a bad fusible link and ended up losing a starter solenoid as well. Just never even figured I'd have fried the ECM relay. I haven't purchased one just yet, but O'Reilly and Napa both have them. They're $17.
I had considered the relays, but my experience is usually that they immediately cause problems, not this fun situation where it waits until it warms up. Likewise, this would be the first time I've seen the mag-coil fail.

And Frank, HA! Even though the truck does love puddle diving, I'm thinking that might exacerbate my already existent rust issue. Sitting in my drive for so long non-operational has done my paint no favors.

 

[Halffast]

I have an 86 2.9L myself, it had same problem at 40K

My 86 2.9L that I bought new would die after warming up and I stopped at a stop sign or red light at 40K miles. I changed the O2 sensor and fixed it that time. Now I have other problems at 170k and will start looking at what you are checking.
The problem with these old ECMs is that anything out of calibration affects them and can cause a lot of the same confusing symptoms, and it is hard to figure out which sensor or whatever is doing it. Your ECM relay should not be getting, hot as you already know, I have the manual from Ford and you can't really tell much about the operation of the relay in the circuit that only controls the power to the ECM. Could the ECM be drawing that much power? Looking at the circuit, it doesn't seem possible that this relay would ever get hot, but it no doubt is! I hope you find the problem. the relay probably needs changed as you know too. You ARE going to check to see if a new relay gets hot? Good luck with it.

 

[Millster]

My 86 2.9L that I bought new would die after warming up and I stopped at a stop sign or red light at 40K miles. I changed the O2 sensor and fixed it that time. Now I have other problems at 170k and will start looking at what you are checking.
The problem with these old ECMs is that anything out of calibration affects them and can cause a lot of the same confusing symptoms, and it is hard to figure out which sensor or whatever is doing it. Your ECM relay should not be getting, hot as you already know, I have the manual from Ford and you can't really tell much about the operation of the relay in the circuit that only controls the power to the ECM. Could the ECM be drawing that much power? Looking at the circuit, it doesn't seem possible that this relay would ever get hot, but it no doubt is! I hope you find the problem. the relay probably needs changed as you know too. You ARE going to check to see if a new relay gets hot? Good luck with it.

Certainly, my 02 sensor is needing replaced. The harness is melted beyond return and I'm certain the sensor itself is not in tip-top condition. But I've not been too worried about that one. No codes being thrown and I'm thinking it's been that way for most of the life of the truck.

I agree that it doesn't seem as though the relay should get hot and that's what leads me to believe it is the switch coil that is failing and not the actual circuit itself. If you think about it, you've got an electromagnet that has to stay energized for as long as the vehicle is running in order to provide power to the ECM. Its relative proximity to the exhaust system provides extra heat and in the case of mine, being exposed to the gap in the hood where water can seep down, i'm sure did not help. I've got a relay on order (they're surprisingly in short supply, here) and once it comes in, I'm going to run the truck until it dies and then replace the relay. Normally, it takes about 15 minutes for the truck to start back up (assuming cool-down). If it starts right away, I know I've nailed it. If not, it's back to the drawing board. I'll definitely monitor the temp of the new relay, as well. My guess is, it'll run perfectly cool.

 

[Millster]

Thought I would post a quick update:

I took the truck out to get some fuel today, thinking it was low... as it turns out, my gauge has just stopped working. I'll add that to my growing list of pains.

Anyway, got the fuel and proceeded to run it around for a bit hoping I could get the relay to heat up and die. It took forever and ended up happening with it sitting back in my driveway idling. Anyway, it died as it usually does. The idle slows way down and the engine shakes to high-heaven. A slight touch of the accelerator and it bogs out, backfires and dies. This time, I was able to get it to restart immediately a couple of times before it finally died and would just crank. I hopped out and plugged in my new relay, excited as could be... crank, crank, crank... nothing. Plugged the old one back in (which had now cooled off) crank, sputter, VROOM. WTF?

So, I've proven that it is definitely a tired relay overheating and losing contact at least. What I cannot understand is how in the world I got a total bum brand-new part. I put it to a multi-meter and got about 5K ohms for 2 seconds before it spiked to about 40K. If I pulled the meter and put it straight back, it would go to 30K and sag back to 40K. Leave it off for 30 seconds or more and back to 5K before jumping to 40K. Apparently the coils in these ECU relays are just crap? Going to contact NAPA tomorrow and hope they'll let me return it since it was straight defective. Not sure what their return policy is on relays.

 

[RyanL]

Thought I would post a quick update:

I took the truck out to get some fuel today, thinking it was low... as it turns out, my gauge has just stopped working. I'll add that to my growing list of pains.

Anyway, got the fuel and proceeded to run it around for a bit hoping I could get the relay to heat up and die. It took forever and ended up happening with it sitting back in my driveway idling. Anyway, it died as it usually does. The idle slows way down and the engine shakes to high-heaven. A slight touch of the accelerator and it bogs out, backfires and dies. This time, I was able to get it to restart immediately a couple of times before it finally died and would just crank. I hopped out and plugged in my new relay, excited as could be... crank, crank, crank... nothing. Plugged the old one back in (which had now cooled off) crank, sputter, VROOM. WTF?

So, I've proven that it is definitely a tired relay overheating and losing contact at least. What I cannot understand is how in the world I got a total bum brand-new part. I put it to a multi-meter and got about 5K ohms for 2 seconds before it spiked to about 40K. If I pulled the meter and put it straight back, it would go to 30K and sag back to 40K. Leave it off for 30 seconds or more and back to 5K before jumping to 40K. Apparently the coils in these ECU relays are just crap? Going to contact NAPA tomorrow and hope they'll let me return it since it was straight defective. Not sure what their return policy is on relays.

How long did it take to get that one in? Does it have to be the exact same one? I mean there's probably only a handful of different sized relays. I'm thinking that you might be able to match it up with something similar to it (as long as it's normally open like the factory one). You might be able to get one which contacts are rated for higher current. Many smaller parts stores let you go into the back with them and look through their stuff. I know that it's saved me a handful of times when they didn't have what I needed or can't even get it in some cases.

 

[Millster]

It didn't take long at all to acquire this relay. It would be best if it's the same one because obviously that makes it easier to just plug into the harness. However, I suppose I'm not opposed to hacking the harness apart (some of the wires look a bit gnarly anyway) and just using a horn relay or the like. I'm not so sure it's the contacts that are the issue. It seems to be the mag-coil. I considered, last night, that I might want to make sure I'm not somehow getting a higher voltage on the coil side of that harness, but now the rain has started and it won't be letting up until Tuesday, so I'm stuck until then.

 

[Millster]

Alrighty... anyone want to tell me this is a stupid idea before I do it so I can get the truck back on the road while I fight with NAPA over whether or not I cooked a relay that was shorted straight out of the box?

I'm planning to tap the relay harness with a couple of spades, some wire and a switch. I'll be the one controlling power to the ECU, but I cannot see a problem with this setup unless, as I say, there is something I'm stupidly overlooking. If I don't hear from anyone before I'm done with the project, I'll report back on how it goes.

Plain and simple, the truck isn't worth me buying another relay while I'm fighting with NAPA only to possibly find it has OTHER problems aside from just the relay. At this point, if I can't fix it with a few cheap parts, the thing's worth more as scrap and parts. So, I have a heavy-duty 12V switch, some heavy-gauge wire and plenty of crimp-on spades. This should let me run the beast and see if the problem lies there or elsewhere and once I get NAPA to listen and give me my money back, I'll go buy a FoMoCo relay and fix it right.

 

[Earl43P]

http://www.therangerstation.com/tec...es/Diagrams_ElectronciEngControls2_9_1of3.JPG


Seems simpler to just add a better ground to the BK/W wire for the relay coil ground (or clean up G101 on the radiator support).

My reasoning is that your relay coil heats up.

Only thing that will cause that is low voltage being supplied to that relay coil (bad ignition switch contacts) or a poor ground (G101).

A volt/ohm meter will tell you which is the problem.

 

[Millster]

I'll posit this, though... previous failed components that have damaged the coil windings could cause heating as well. A damaged relay will heat up.
Still, I'm going to check this out. I have NOT thought of it. Which also begs the question of where in the world you've been throughout this whole thread.

Thanks for the tip. I'll try it before hacking.

http://www.therangerstation.com/tec...es/Diagrams_ElectronciEngControls2_9_1of3.JPG


Seems simpler to just add a better ground to the BK/W wire for the relay coil ground (or clean up G101 on the radiator support).

My reasoning is that your relay coil heats up.

Only thing that will cause that is low voltage being supplied to that relay coil (bad ignition switch contacts) or a poor ground (G101).

A volt/ohm meter will tell you which is the problem.

 

[RyanL]

It couldn't hurt one little bit to check out your grounds, but I'm having trouble wrapping my head around that scenario though. I don't see how having lower voltage would cause the coil part of a relay too draw too much current and overheat. According to Ohm's law, if you increase resistance or decrease voltage, power (wattage) decreases. If anything I would think that a bad connection would not let the relay draw enough power to fully engage the set of contacts causing chatter (or make contact at all). Maybe that is what is going on and the relay getting warm is normal (just a slightly misdirected diagnosis). Maybe the ECU is drawing too much power and the relay is the "weak point" and it gets hot. There's a bunch of possibilities. Fixing and diagnosing mechanical issues are so much easier than messing with this old crusty wiring. I don't see the harm in putting a switch in place of the relay unless you mix up the low current side with the high current side. Sometimes the only practical way to figure out these electrical problems is to jumper them out and see what's going on when the circuit is hot. There is even the possibility that they gave you the wrong relay (wouldn't be the first time someone has been given the wrong part from the autoparts store) It might look the same, but have a different pin out. I wish you luck man.

 

[Millster]

It couldn't hurt one little bit to check out your grounds, but I'm having trouble wrapping my head around that scenario though. I don't see how having lower voltage would cause the coil part of a relay too draw too much current and overheat. According to Ohm's law, if you increase resistance or decrease voltage, power (wattage) decreases. If anything I would think that a bad connection would not let the relay draw enough power to fully engage the set of contacts causing chatter (or make contact at all). Maybe that is what is going on and the relay getting warm is normal (just a slightly misdirected diagnosis). Maybe the ECU is drawing too much power and the relay is the "weak point" and it gets hot. There's a bunch of possibilities. Fixing and diagnosing mechanical issues are so much easier than messing with this old crusty wiring. I don't see the harm in putting a switch in place of the relay unless you mix up the low current side with the high current side. Sometimes the only practical way to figure out these electrical problems is to jumper them out and see what's going on when the circuit is hot. There is even the possibility that they gave you the wrong relay (wouldn't be the first time someone has been given the wrong part from the autoparts store) It might look the same, but have a different pin out. I wish you luck man.

Agreed. The more I thought on it the less it seemed likely. However, I've had far less likely things on this truck (and on my screwy Saab) that have been the issue. So, I checked.

Here's what I found:
The grounds are good. Great, in fact. No resistance on the wires at all. They're carrying all of the voltage and current they should be.
On COULD say "crap" at this point.

I decided to push on further. I pulled the relay out of its socket and jammed one of my tester leads into the "contact-side" pins... then started the engine. Even with that thin little test lead in there, nothing got hot and the engine ran fine for 10 minutes before I just got bored. I shut it down and plugged the relay back in. With the engine off and the key simply on, the contact side of the relay was already warm. When I started the engine, the contact side immediately warmed to a point that it was uncomfortable to hold onto. I shut the engine down and decided I had nothing to lose by my next action. I opened the relay up. I found about what I expected. A nice, clean coil that was nice and cool... and some of the nastiest, most pitted, corroded contacts I've ever seen. Starting the engine with the relay opened revealed much the same. The coil stayed cool while the contacts got so hot that I didn't think directly touching them was wise.

So, I think excessive mileage, weather and no doubt the repeated failed attempts to start the engine over the years between the dead TFI and the dead fuel pump have just taken their toll on the relay contacts. The current draw from the ECU is not excessive (less than 15 amps) and a cheap test lead was able to hold the power without even heating for 10 minutes. Looks pretty straight forward.

The replacement relay from NAPA was, in fact, the correct one. I checked that out before I even considered plugging it in. The coil simply does not energize. Either there is a break in the windings or there is a bad connection from the pin to the winding (that's my guess). I just have to keep working to convince them it's defective and I did not break it. I've been working with relays longer than the parts kids have been on the Earth. I hate to pull that line out, but I've seen this before. A little break, no energy on the coil, the contacts don't close. Crap parts happen. Just own up to it, give me my money back and let's all move on.

Anyway, long story short, I'm going with my switch approach. The truck is only going to be my DD until the parts for my Saab (which suffered a catastrophic fuel system failure) arrive and then it'll be relegated back to hauling dirt, junk and things too large to fit in the other vehicles. While it's actually passable in appearance, it just has too many miles and has had too many issues for me to consider it reliable for regular transportation. Approaching 1/4 of a million for a Ranger is, I think, testing the limits of how many miles a vehicle can go before something horrible happens.
I hate bodged solutions as much as the next guy, but in this case, it's well-done and reasonable. It's not like I'm jumping the relay with a piece of coat hanger. I'm using quality parts, quality wire and doing it right.

 

[Millster]

Well folks, good news. The switch seems to have done the trick.
In fact, possibly a little better than I had hoped. It now seems to be reading all of the sensors because it's doing things it's never done before.

One of the issues was a broken vacuum cap. It ran somewhat better after that was replaced. It's now just very lumpy and running pretty rich. I know the socket on my 02 sensor is melted and perhaps that's part of the problem. Odd that it never mattered before, but now it seems to. I'm thinking the computer was never getting full power because of that duff relay and was not actually taking all sensors into consideration. It may well be that the plugs are just a bit fouled from sitting for so long, too.
In any case, it runs! I gave it every chance to die on me and aside from hitting a very nasty patch of misfiring, backfiring and belching black smoke (which was alleviated by jumping into the throttle), it never even tried.