Troubleshooting - Hard starting when hot

[Van-Tramp]

Those mpg numbers seem very low. I have a dodge 360 (compared to your 351) and get 9-10 in town and 13+ on the highway at your elevation. You may want to look into the O2 sensor(s) as they too can fail and default your computer to run full rich.

Everything you said also makes sense to look into

 

[user 423]

1.Yes but it may be more involved than just being rich. Any sensor malfunction will send bogus info to the PCM so the air/fuel mix will be off. You mentioned faulty fuel pressure. That could be part of the problem.

2.The eng.coolant temp reading (plus inputs from several other sensors) affects the air/fuel mix. Here's a link that is a good resource for understanding EFI (plus other topics):

http://www.aa1car.com/library/fuel_injection_basics.htm

3. Separate problem. Once it starts, the PCM constantly adjusts the A/F mix, so it may very well run OK, as your's seems to do.

4. Here's another view:

 

[user 423]

Those mpg numbers seem very low.

Keep in mind he's driving a box van, so his MPG will be lower than my van which has the same engine. I get 10 city and 15 highway. Sounds like he may have it pretty well loaded,too.

 

[highdesertranger]

check that sensor and circuit before you willy nilly start replacing parts. a good scan gauge will show the temp the CPU is seeing. you can also check the sensor with an ohm meter, however this won't tell you if you have a bad wire or connection. I have a black box that will substitute for the sensor you can run through all the temp ranges. mine is similar to this one,

http://www.autoditex.com/automotive-resistance-substitution-box-62.html

so first off see what the CPU is seeing and compare it to actual water temp. then go from there. highdesertranger

 

[user 423]

...a good scan gauge will show the temp the CPU is seeing.

This van is OBD1 so most scanners will only give basic info. Ford dealers had a really fancy one that would do a temp check but most probably don't still have them around in their shops as everything has upgraded to OBD2. It might be possible to find a shop that has the better scanner though.

you can check the sensor with an ohm meter, however this won't tell you if you have a bad wire or connection.

The Haynes test has you checking for voltage to the sensor with the key on, so that should indicate a bad wire/connector. It also may show that a faulty PCM isn't sending the signal to the sensor. I would think that would throw a code...but maybe not.

Very good advice about checking things out thoroughly before throwing parts at it!

 

[GotSmart]

Let me see if I can summarize this idea about the coolant temperature sensor and ask a few questions to clarify this.

First, thanks for all the good info and diagrams and links. Very helpful.

Summary: Let's say this sensor / thermostat is malfunctioning. It seems that everyone is saying that would make the fuel condition too rich. Too rich when the engine is cold is okay because that simulates choke conditions and helps starting. However, too rich when the engine is already hot does something bad, presumably causes the hard starting symptoms that I have experienced.

1. Is my summary description above correct? If not, what do I misunderstand?

2. Logically, how could a device that reads engine coolant temperature affect the fuel system? My only guess is that engine coolant temperature is a proxy for engine temperature which may be needed by the computer to adjust the fuel mixture. Is that right?

3. The vehicle seems to run fine once it's going, with the possible exception that is gets worse gas mileage than I think it should be capable of. About 6-7 city, and 8+ mpg on the highway. Fully loaded. On the other hand, some people have said that's normal for this vehicle. 

That is only background to ask this question: it would seem to me that if this sensor is bad and it's constantly making the engine run rich, wouldn't that affect more than only starting when it's hot? Shouldn't it also run poorly all the time?

4. From the diagram I still can't tell where that doohickey is located. If anybody already knows where to look to find it, please let me know. Otherwise, I'll go off hunting.

Thanks a bunch for all the good info so far.

Tom

Post #18 shows the position in the diagram. On the top of the motor. 

The sensor being a part of a electronic system, alters the fuel feed according to the message received. Therefore what it does depends on what the electronically programmed doohickey that is not working properly tells it.

 

[Vagabound]

Yeah, I saw the diagram in Post 18. The problem was it didn't show enough of the engine for me to be able to tell where it actually is. However, the next diagram of the wiring harness seems to show a lot more.

As far as comparing temperatures, the OBD1 reader that I have doesn't show temperature. Not sure what else to do about that other than go back and check that sensor with a multimeter.

Slow2day, thanks for putting so much effort into helping me understand this. I'm making progress here, at least mentally. And that has to happen before I get the wrenches out.

Tom

 

[user 423]

^
I don't mind since whatever you find will help me when I get around to making these same checks on mine. It's 18 degrees now in KC, so it will be a while before I can mess with mine and get it running well enough to tow a trailer.

Need to get it running better (and maybe a frontend rebuild) as I hope to head for South Dakota sometime in May or June.

BTW: On that 2nd. diagram I circled the IAT (Intake Air Temp sensor). Depending on if it's accessible enough, it would be a good idea to check it with a meter also while you're in there.

PS: Not to muddy the waters too much but I didn't think of it before. Anyway, if the check valve in the FP was bad, you would have to prime the fuel system in the morning to get it to start. If it starts right up on the first try in the morning, I don't see how the check valve could be bad.

 

[B and C]

All good information here. Your problem describes mine exactly.

I had the fuel pump replaced two years ago with a Bosch fuel pump as a precaution as I was going to Alaska the next year and did not want to have to worry about it going out (over a 100K on a used vehicle). While I was in Alaska, it started behaving as you described if I stopped to eat or a short sight-see. I started holding the accelerator pedal down to the floor to help with the hard start (flooded). Black smoke would pour out for a couple of seconds but it started a lot faster than hanging on the key longer. My temperature gauge was registering normally so the thought of a temperature sensor gone bad was not even on the radar. I thought it only had one (I am old school). Now to wait for it to dry up and warm up so I can get it checked. My van is in the backyard and will get stuck several times before I can get it out :-(

Thanks for bringing up this issue. I have been just living with it for a year as I normally only stop the engine long enough to fill the tank. Otherwise the stop is for far longer.

 

[Vagabound]

...
PS: Not to muddy the waters too much but I didn't think of it before.  Anyway, if the check valve in the FP was bad, you would have to prime the fuel system in the morning to get it to start. If it starts right up on the first try in the morning, I don't see how the check valve could be bad.

That was my thinking as well, but I don't know enough about it to feel certain.  However, I was unsure enough about it to pause my plan to drop the fuel tank and put in the new FP that I've been lugging around just in case.  An added twist of irony regarding the FP is that about the time I thought I was up to speed on how to do that repair myself, I reached the point where I doubted it as the problem. Figures.

EDIT: Also, thanks a bunch for the fuel injection basics link.  Found it very enlightening.

======

B and C / Brian,
You're most welcome.  I've been tolerating / battling this longer that reasonable, and much longer than I wanted, so if you eventually learn anything in your situation, please share.  It might still be useful! ;-)

Tom

 

[Vagabound]

Hey, new piece of info ...

Regarding the idea of a bad engine coolant temperature sensor, I just ran across an opinion that the cause is unlikely to be that sensor _IF_ the temperature gauge in the vehicle rises / reads normally, which mine does.  And then if not that, maybe a fuel relay.

What do you all think?

Tom

 

[user 423]

^
Was this online info? Have a link?

There is a separate temp sender for the gauge, so I'm curious as to how that can affect the signal from the ECT to the PCM...

What relay do they refer to? As far as I know, a bad relay would affect all fuel delivery and not just during hot-starts. But then again if the relay fails because it is overheated after running the engine, that may make sense.

 

[highdesertranger]

yep 2 completely separate circuits. one has nothing to do with the other. highdesertranger

 

[Vagabound]

^
Was this online info?  Have a link?

...

What relay do they refer to?  ..

Yes, it was online, but only in the comments under a YouTube video. Nothing very authoritative.

The reference to the relay was not very specific, but I've heard it discussed in the past and the term fuel pump relay was used.


===========


A strange thing happened on the way to the market ...

This short story is intended to provide clues in this mechanical mystery, hopefully leading to more clarification about which of these avenues to pursue.

I said earlier in this thread that the truck runs well. Generally it does. Very rarely have I noticed a problem with it running, and when I did, it was mostly while idling and very briefly. There was one exception to that.

Approximately 3 weeks ago, after driving across country most of the day, I noticed that the truck was beginning to idle rough. I came into the next city at the end of the day, and it was running rough and didn't have much power. I was quite worried, so I limped into the Walmart parking lot and stayed there for a few days to talk to mechanics, etc. Nothing much was done while I was there with the exception of some diagnostics and the reconnection of one loose vacuum line. Even while driving in town before I left, the rough idle / rough running problem seemed to be gone. It was certainly gone by the time I left a few days later. Spontaneously resolved.

Since that time, I have not experienced that problem again ... until today.  After letting the truck
sit, engine off, for about an hour, I started it to show someone the hard starting problem. Boy, did he get a demonstration! That rough idling problem came back with a vengeance, and continued throughout the 5 minutes or so that we were listening to the engine.

When it was time to leave that area, I drove through the parking lot with the engine running pretty rough. I pulled out and up to a stop light. Very suddenly, with the speed and distinctness of flipping a light switch, the rough running / idling immediately stopped. The truck started running and idling just as smooth as you please. And then while I was still sitting at the red light, the rough running and idling returned, just as suddenly as if flipping a light switch again.

To me, there are very few things that could spontaneously switch engine performance that quickly and completely, and then switch it back just as quickly and completely. I don't know what those things are, but it seems to me that they must be digital / binary type components. They either work or they don't. They are either on or off. They do not operate in an analog way. 

And all of that makes me think about sensors that for one reason or another send one type of signal one minute, and then turn around and send an entirely different kind of signal the next minute even though the truck itself hasn't changed.

Shed any light?

Tom

 

[user 423]

...Spontaneously resolved.

Since you had a disconnected vac line that was reattached, I'd say it wasn't 'spontaneously resolved'...

To me, there are very few things that could spontaneously switch engine performance that quickly and completely, and then switch it back just as quickly and completely. I don't know what those things are, but it seems to me that they must be digital / binary type components. They either work or they don't. They are either on or off. They do not operate in an analog way. 

Several of those sensors are thermistors that vary resistance with changes in temp. The TPS changes resistance throughout the movement of the throttle. Those are analog readings that are read by the PCM which adjusts the A/F mixture by changing the pulses sent to the injectors. That is the digital part.

Anyway, I don't believe it's correct to say that the sensors have to be either on or off. They can be out of range and sending bogus info to the PCM. The fact they don't set a code makes things difficult for us. The more sophisticated scanners for OBD1 that the dealers have (had) or the ones you can buy for $500 (http://www.totalcardiagnostics.com/shop/C168-obd2-scan-tool) will give you a better picture of what's going on, like HDR said and it may be worth it to find a shop that has one. Checking the sensors that you can with a meter can at least eliminate those from the picture. While you are doing that, you are removing the connector, checking the wire for breaks in the insulation, cleaning the contact areas and applying dialectric. All that can hopefully let you cross that sensor off the suspect list. In the past I spent many years working on electronic equipment and so many times (in vehicles too) it's the little crap like that messes things up!

The van is 25 years old and all kinds of gremlins can cause intermittent rough running including bad gas or a momentarily bad electrical connection or even a vacuum leak that is intermittent because of a small crack in a line that is opened up by vibration.

Did you check for codes after the rough running today? I'd be curious to know what kind of readings you would get with a vacuum gauge (a good diagnostic tool).

Finally, depending on your tonnage and the differential ratio, highway mileage of only 8 MPG may indicate that it's not really running that well...even though there's no obvious misfiring.

BTW: How many miles on this van?

 

[Vagabound]

Digesting and working on it. Thanks. More later.

Miles: Approaching 150K, but that's misleading because the engine and transmission are newer.

Generally speaking, if I had this mysterious vacuum gauge of which you speak, where would I be connecting it to do this test?

Tom

 

[user 423]

You can put a T-fitting between the fuel pressure reg. and the vac line that goes to it. You have to be very careful not to break the nylon line though. They are brittle and can break.

Another place that is convenient is at the MAP sensor which is located at the upper left of the engine bay. In fact, the original rubber tubing on mine I found to be cracked and leaky, so it it was replaced with 1/4" fuel line. The new line is also more flexible so it's easier to put in a temporary T for the gauge. Again, you have to be careful with the line.

I have the Actron CP7803 which is about $25.

BTW: The newer engine may explain why your FPR is in a different location than mine. Do you know what year truck the engine/trans came out of?

 

[Headache]

Vagabound if you aren't far from Ehrenberg I have the factory service manuals on CD, the Ford wiring diagrams as well as the Ford diagnostics manual. These aren't the Chilton/Haynes stuff although I have the Haynes as well. It covers about a 10 year period so it might have info you can use depending how much newer your engine is. My van just turned 150k and is only a year older than yours.

Unfortunately the code readers won't give you much info as you've found out. You need a diagnostic tool such as the Snap On "red brick" which you can purchase from fleabay, preferably with the last updates before Snap On stopped supporting them. I don't have one YET. Since we have OBD 1's we don't have obvious diagnostic trees to run down. You can get one for about $200.

Good luck with your issues. As someone who recently replaced both the PCM and the TPS it was trial and error until one brief moment when the tech "caught" the PCM short out briefly, then start working again just as quickly.

 

[Vagabound]

Vagabound if you aren't far from Ehrenberg I have the factory service manuals on CD, the Ford wiring diagrams as well as the Ford diagnostics manual.  These aren't the Chilton/Haynes stuff although I have the Haynes as well.  It covers about a 10 year period so it might have info you can use depending how much newer your engine is.  My van just turned 150k and is only a year older than yours.

...

Hey Headache, how's it going?  Thanks for the offer.  I'm a few hours away at the moment.  Not heading your way for a month or so I'd guess.  If we end up in the same place, I'd really like to see those manuals.

As someone who recently replaced both the PCM and the TPS it was trial and error until one brief moment when the tech "caught" the PCM short out briefly, then start working again just as quickly.

Yeah, that's the part of my most recent description that I hoped someone would key in on.  The part that I described happening at the red light.  Even in my ignorance, I feel comfortable saying that not many things can cause that exact symptom (changing between bad and good and back again instantaneously while at a standstill).  To someone who knows much more than me about that engine type, that should mean something.

Since my last post, based on advice here and some online, I replaced my Engine Coolant Temperature Sensor.  Turns out, it was not the problem and the old one seems to have still been within specs.  At least that part is new now.

I have a few reasons to suspect the engine timing is too advanced, either causing or contributing to the hard starting problem when hot.  That's the next area I'll be exploring.

Thanks again,

Tom

 

[Doubleone]

Yes, it was online, but only in the comments under a YouTube video. Nothing very authoritative.

The reference to the relay was not very specific, but I've heard it discussed in the past and the term fuel pump relay was used.


===========


A strange thing happened on the way to the market ...

This short story is intended to provide clues in this mechanical mystery, hopefully leading to more clarification about which of these avenues to pursue.

I said earlier in this thread that the truck runs well. Generally it does. Very rarely have I noticed a problem with it running, and when I did, it was mostly while idling and very briefly. There was one exception to that.

Approximately 3 weeks ago, after driving across country most of the day, I noticed that the truck was beginning to idle rough. I came into the next city at the end of the day, and it was running rough and didn't have much power. I was quite worried, so I limped into the Walmart parking lot and stayed there for a few days to talk to mechanics, etc. Nothing much was done while I was there with the exception of some diagnostics and the reconnection of one loose vacuum line. Even while driving in town before I left, the rough idle / rough running problem seemed to be gone. It was certainly gone by the time I left a few days later. Spontaneously resolved.

Since that time, I have not experienced that problem again ... until today.  After letting the truck
sit, engine off, for about an hour, I started it to show someone the hard starting problem. Boy, did he get a demonstration! That rough idling problem came back with a vengeance, and continued throughout the 5 minutes or so that we were listening to the engine.

When it was time to leave that area, I drove through the parking lot with the engine running pretty rough. I pulled out and up to a stop light. Very suddenly, with the speed and distinctness of flipping a light switch, the rough running / idling immediately stopped. The truck started running and idling just as smooth as you please. And then while I was still sitting at the red light, the rough running and idling returned, just as suddenly as if flipping a light switch again.

To me, there are very few things that could spontaneously switch engine performance that quickly and completely, and then switch it back just as quickly and completely. I don't know what those things are, but it seems to me that they must be digital / binary type components. They either work or they don't. They are either on or off. They do not operate in an analog way. 

And all of that makes me think about sensors that for one reason or another send one type of signal one minute, and then turn around and send an entirely different kind of signal the next minute even though the truck itself hasn't changed.

Shed any light?

Tom

Hey Vagabound,  Several things came to mind when I read your symptoms.  I'm a former ASE Master Auto tech but without benefit of laying hands-on to diagnose I'm just sharing my raw thoughts... and my thoughts plus $1.65 will buy you a cup of coffee, as long as you don't forget your $1.65. 

To me, the symptoms suggest something other than fuel pump.  Here's why...  Electric fuel pumps typically either work or not; there is usually no in-between and they usually quit with little or no warning unless you know what to check for.  The main issue with older Fords and GM's are they easily plug the fuel filters and cause the fuel pump to labor and destroy itself.  Neglected and clogged fuel filters are a very common cause of fuel pump failures.  Fuel pumps can often be heard whining or laboring just before failure.  If you hear prominent whining, change the fuel filter immediately.  The way to check fuel pump health is you can remove the filter and empty the contents backwards to see if you have any metal flakes on the inlet side.  Just shake it backwards into a paper towel or your hand with the inlet down and you'll plainly see glitter if the fuel pump is worn.  Don't panic if you see some glitter, just know you need to plan a replacement real soon.  Often, a fresh filter can get you by until next payday...  You can also empty the fuel filter and try to blow through it.  Fuel injection regulators typically leak fuel when they go bad.  You can check inside the vacuum line and exterior for any signs of fuel leakage.  I've never seen one go bad without obvious leaks.

You already addressed vacuum leaks which is the first thing I'd think to look for right away, since vacuum is connected to everything.  I know you said you had it smoked, just be sure to double check the hoses at the fuel vapor can which is remotely mounted and often can't be heard or seen leaking, typically mounted under the front fender or back near the fuel tank.   Timing is controlled by the computer, the only change you might have affected by messing with the timing is "base timing" or in other words the minimum setting.  The computer still takes over control of the timing when the engine is running.  The base timing is just the full retard setting used in conjunction with knock sensor to reduce detonation (pinging).  If you advanced the base time you might hurt the engine during a detonation circumstance because it won't retard enough...

Usual culprits that wear out and cause intermittent glitches are ignition components, Throttle Position Sensor (TPS), and Idle Air Control (IAC).   Your generation van might still have ignition cap rotor wires that can deteriorate and leak electricity.  Check under the distributor cap and look closely at the rotor and cap for carbon tracking (burning and soot).  Later engines had distributorless ignition with coil packs that also sometimes had carbon tracking issues near the coil terminals.  Also look closely for whitish-gray arcing marks at the sides of the coil where it comes near the metal.  The reason ignition can affect you is because lean fuel mixtures, like when an engine is hot, will require more energy to ignite.  Rich fuel mixtures, when an engine is cold, are easier to fire off because there are more fuel molecules closer together...  When the ignition parts are compromised, electricity takes the path of least resistance, which is somewhere other than through the wires to the sparkplug.  Bad ignition can also manifest as a hiccup or slight miss under heavy acceleration.

Assuming ignition parts are good my next thoughts are TPS and IAC.  The TPS is a rheostat which has an internal sliding part that can wear out and cause a momentary open circuit that freaks out the computer.  It's very common on higher mileage vehicles and hard to identify unless specifically checking it using an old analog ohm meter.  It may not set any codes. The IAC is basically a stepper motor that controls the idle speed by metering the amount of air the engine gets at idle.  The IAC is like a motorized screw, and they wear out and carbon up inside causing them to stick, usually resulting in rough low idle and stalling right after start up.  One other thought is bad fuel.  My rig is always slow to start and idles rough on cheap 85 octane gas... but seems to run ok at speed.