Starting problem, HELP!!!

[LedFloyd900]

I've got a '92 3.0L 4x4 that is on and off used as a daily driver. I completely rebuilt the engine myself (I'm an aircraft mechanic so I'm not just a backyard joe-schmoe) but now it takes FOREVER to start. And it gets worse as the temperature drops or if it sits for a longer period of time. I've replaced the fuel pressure regulator with a new one and even tried a salvaged one, piddled with the temperature sensor, and have replaced the cold start valve with 3 different salvage yard ones (didn't want to throw a $135 new part I wasn't even sure about on it). It will just crank and crank and crank and once it finally does start to ignite it'll sputter and die over and over and eventually it will sputter and the whole car will shake quite violently for a good 5-10 seconds and then all of a sudden the RPM's will shoot up to a good 1,500-2,000 depending on how cold it is and idle very smoothly afterwards. Maybe I'm just completely over looking something or over thinking this but I have absolutely no more ideas and I'm tired of throwing parts at it. If anyone has had this problem or knows of any possible fixes or solutions PLEASE let me know, its driving me crazy!!

 

[mholme]

When you say it gets harder to start when it sits for a longer period of time-do you mean sits for days or after it has been run and you shut it off and go to start it a while later?

Have you tested fuel pressure while cranking? Verified the CSI fires? Honestly, mine is the same way when it gets cold out and I've verified that everything is in spec per the manual. Old cars that sit can be hard to start and I just deal with it. I don't daily drive mine and it's parked out in the elements. I leave the battery in the garage on a charger until I need it. It still takes about 3-5 attempts to start with a fresh battery.

 

[LedFloyd900]

When you say it gets harder to start when it sits-do you mean sits for days or after it has been run and you shut it off and go to start it a while later?

Well both, the longer it sits the longer it takes to start. If it sits for 8 hours it's harder to start than if it sits for 4 but not as hard if it sits for a day. And I know older cars take longer to start sometimes but this isn't that, this is obviously something wrong. I mean it almost completely drains my battery every time. It was 5 degrees one day and yes that's cold but it literally took about 5-7 straight minutes of cranking to actually start. It's almost as if the fuel pressure drops but why would temperature have anything to do with it? Over summer it would only take about 2-3 attempts to start it after sitting for a day but when it's 25 in the mornings now it takes at least 10-15

 

[identity]

Is your cold start injector opening?
If your cold start injector is good but not opening, check the cold start injector time switch. It's kind of difficult to reach - in that coolant manifold behind the lower intake manifold - but if you can get the connector off, there should be 30-50 ohms across the two pins at or below 54* and 65-90 ohms above 86*(per the FSM). On mine, it's the two-wire connector toward the passenger side, although apparently there is great variation in the arrangement of sensors in that coolant manifold across the years.
Also, you mentioned the coolant sensor. Do you mean the one-wire sensor for the gauge, or the two-wire sensor for the ECM? I am having very similar issues to yours(I'm the no-start thread right below yours) and believe the ECM coolant temp sensor to be my problem.

Good luck.

(Edited for clarity)

 

[mholme]

Well both, the longer it sits the longer it takes to start. If it sits for 8 hours it's harder to start than if it sits for 4 but not as hard if it sits for a day. And I know older cars take longer to start sometimes but this isn't that, this is obviously something wrong. I mean it almost completely drains my battery every time. It was 5 degrees one day and yes that's cold but it literally took about 5-7 straight minutes of cranking to actually start. It's almost as if the fuel pressure drops but why would temperature have anything to do with it? Over summer it would only take about 2-3 attempts to start it after sitting for a day but when it's 25 in the mornings now it takes at least 10-15

Yeah, like I said I keep my battery on a charger, because I know it's going to get a workout trying to start in the cold, as in almost kill the battery, and I can't find anything wrong with mine. Other than it's 29 years old and sits a lot. And it's a 3VZE. Lets just throw that one into the equation.

Also gasoline evaporates slower in the cold and won't ignite as easily(I'm sure you know this, but something to consider), the rough running you described might be raw fuel making an overly rich condition for a few seconds.

I'd still go ahead and verify your fuel pressure while cranking and make sure your CSI fires too.

 

[identity]

I'd still go ahead and verify your fuel pressure while cranking

Is there a simpler means of testing fuel pressure than buying an adapter or installing a permanent elbow to the rail? I'd kinda like to check fuel pressure on mine as well, since I replaced the fuel pump and filter after the truck sat for a couple of years.

Edit: also, is it possible that LedFloyd900's truck has a small leak in the pressurized fuel system which causes a loss of pressure over a matter of hours, and would that cause a significant delay in startup as the fuel pump rebuilds pressure?

 

[LedFloyd900]

Is your cold start injector opening?
If your cold start injector is good but not opening, check the cold start injector time switch. It's kind of difficult to reach - in that coolant manifold behind the lower intake manifold - but if you can get the connector off, there should be 30-50 ohms across the two pins at or below 54* and 65-90 ohms above 86*(per the FSM). On mine, it's the two-wire connector toward the passenger side, although apparently there is great variation in the arrangement of sensors in that coolant manifold across the years.
Also, you mentioned the coolant sensor. Do you mean the one-wire sensor for the gauge, or the two-wire sensor for the ECM? I am having very similar issues to yours(I'm the no-start thread right below yours) and believe the ECM coolant temp sensor to be my problem.

Good luck.

(Edited for clarity)

Sorry I meant the one wire (believe the P/N is 83420). When I pulled the engine apart it had a resistor in the line for some reason and I left it off when I put it all back together but I just recently added it back in the line but no difference. I don't think I have a two wire one but it's been almost a year since it was rebuilt so I could be wrong. Are you talking about the one that has the P/N 89462?

EDIT: Sorry now that I look back on pictures I do have a two wire one and it is the CSI switch. I just forgot because it has the conduit around it. I will test it tomorrow, I really appreciate the suggestion!

 

[mholme]

Is there a simpler means of testing fuel pressure than buying an adapter or installing a permanent elbow to the rail? I'd kinda like to check fuel pressure on mine as well, since I replaced the fuel pump and filter after the truck sat for a couple of years.

Edit: also, is it possible that LedFloyd900's truck has a small leak in the pressurized fuel system which causes a loss of pressure over a matter of hours, and would that cause a significant delay in startup as the fuel pump rebuilds pressure?

I made an adapter out of a banjo bolt from the junkyard. Just tap it and screw in a fitting and replace the banjo at the rail. Then you can clamp the gauge to it.

 

[identity]

EDIT: Sorry now that I look back on pictures I do have a two wire one and it is the CSI switch. I just forgot because it has the conduit around it. I will test it tomorrow, I really appreciate the suggestion!

There are two similar-looking two-wire sensors in that area(at least on my engine): one is the CSITS, the other is the ECM temp sensor(the single-wire gauge temp sender is between them). Both are apparently common causes of starting and driveability issues and if you're going to be scraping your hands up back there you might as well test them both. The FSM has a graph of acceptable ECT values on what looks like a log scale so I can't give you precise numbers, but if you have access to one, it's page EG1-209. Let us know what you find out

I made an adapter out of a banjo bolt from the junkyard. Just tap it and screw in a fitting and replace the banjo at the rail. Then you can clamp the gauge to it.

Right on. That's probably something worth doing next time I'm in that area.