[anothermark]

I'm stumped and have run this problem by a half dozen mechanics who usually give me the expensive answer. I've got a GMC 1998 Suburban K1500 5.7 with 250,000 miles. It only starts with starter fluid now and then idles fine, maybe a bit rough, but purrs right along. So I replaced the fuel pump relay and the fuel filter but that didn't help. Then I had a mechanic replace the fuel pump and that fixed the problem. Started right up, all was good and well for a few hundred miles until it reverted back to its old self, not starting without starter fluid. (He never flushed the tanks to wash out any dirt or debris, I don't recall).


As the pump was under warranty the mechanic replaced that pump and all was well again, fired right up and I was happy, for a few hundred miles when it happened once more. It's taking longer to start now even with the starter fluid.



So, my fuel pressure gauge test measured the following. Key on, key off 50 psi where it held steady (It should be 60 - 62). At an idle it ran at 52 psi. I turned it off and pressure dropped to 48 where it held solid for 10 minutes.



From what I've read, because pressure held steady at 48 psi after I turned it off the problem isn't a leaky spider fuel injector or a leaky fuel pressure regulator under the intake manifold - and a real pain to get to. Am I right? Or, can the pressure regulator be causing problems even if it does not leak? And how would I know? Is leakage the only problem that can arise with the regulator? With two brand new fuel pumps I find it unlikely that's the problem, but what else could it be? And, if it was the spider injector or pressure regulator leaking I would think it still wouldn't have started up even with those new fuel pumps. Opinions run the course. And what are the odds of two new fuel pumps being bad right out of the box?


Maybe there's a clogged fuel line? Some obstruction? One fellow told me he had the same problem with a small piece of paper towel caught in the line. Now, here's a clue. I've driven this rig thousands and thousands of miles on dusty dirt roads in the Southwest deserts and off-road in the Rockies. Is it possible that dirt or dust or gunk is in the tank that is blocking something? I know clogged fuel filters and this isn't it. In my experience the truck usually sputters and bumps along if the filter is really clogged, but not this rig. And, I had the filter changed.


Or, is there a valve of some kind in the line that might be stuck open or shut? I have no idea, but I don't want the mechanic to throw parts at it randomly, like replacing the spider injectors or the pressure regulator, expensive as that is. Any ideas? Thanks.

 

[johnsongrass1]

Are the injectors getting signal on crank?
is the sparky thing sparking on crank?

 

[anothermark]

Thanks for your quick reply, but you lost me. Not sure what you mean by the signal or sparky thing. I took shop in high school forty years ago and my memory is not what it used to be. There must be a signal or sparky thing if it starts, right? And it runs, although I need starter fluid. And, sometimes after I run it for a while and turn it off for a few minutes it starts right up again. Does that mean I get the sparky thing?

 

[LATECH]

When the engine is cold, after setting, the system pressure goes down probably to zero.AND the oil , being warm when parked, drains back to the oil pan. Very important.
The computer is supposed to prime the fuel system when the key is turned on by energizing a relay that runs the fuel pump for 2 seconds, bringing the fule system pressure up to spec.
After the engine starts , the oil pressure switch closes and continues supplying power to the pump
So that being said ...

Verify that the fuel pump gets a 2 second pulse when turning the key to ON posisiton first. dont try to start it.
Set up to do the test first thing in the morning after it sets overnight to replicate the symptom

You can attemp to listen for it , but it is also best to have a test light hooked to the fuel pump circuit to verify that the PCM is pulsing the fuel pump relay and that it is actually powering up the circuit.

Wire colors at the pump are as follows :
Purple sending unit
Blk/white trace sending unit ground
Grey - fuel pump 12 volts +
black fuel pump ground
Sounds like you are not getting the system primed at the first start of the day. After she starts,oil pressure comes up quicker when restarted then , which activates the oil pressure switch to run the pump quicker, and the fuel system alreay has a prime ( even if it is only 40 + psi) so long crank time is probably not noticed,or present

 

[LATECH]

Fllow up---

If the fuel pump relay doent prime the system, then it is up to the oil pressure switch to run the pump. So it will take some cranking on the starter to get oil pressure, then after that it takes more cranking time for the pump to prime the fuel system .
My guess is , the first start of the day is when you have to use starting fluid?

Also it sounds like the pump is holding pressure and there are no injectors leaking. Maybe your gauge reads wrong by a few PSI. check it against another if you can

 

[anothermark]

Thank you, Mr. LATech Administrator. Suddenly my problem got complicated and I'll need a competent mechanic familiar with what you said to sort this out. Yes, the first start of the day is when I have to use starting fluid. Someone told me that the problem might be electrical, like a bad ground. But here's my question: When I had both fuel pumps replaced, both times she fired up like a charm and drove wonderfully for a few hundred miles before going bad again. That to me seems dispositive, critical. And it tells me something although I'm still trying to figure out what. So, with the new fuel pumps the fuel evidently primed just fine after replacement. That's what stumps me. On another forum one gentleman was emphatic and told me it was because the tanks needed to be flushed out. Dirt and debris I assume clog up the intake port filter on the pump, and that takes a few hundred miles of driving. That's my take. He claims the problem is located before the pump based on what I wrote.

Here's his short but definitive quote: "Have the tanks cleaned out! Your description points to a problem BEFORE the fuel pump (i.e. in the tank) that repeats AFTER the fuel pump has been replaced. If what you have written is true and accurate, there is no further question."

He seems certain, and in a way it makes sense because the problem went away after both pumps were replaced, and only then did the problem reappear after driving around for a while. I'll run your suggestions by my mechanic and other tips I have been receiving. Thank you very much for you help.

Mark

 

[LATECH]

Contamination is an issue that SHOULD have been done on the first pump replacement and certainly before the second one was going in.
Did you see the pumps/screens when they came out?

 

[anothermark]

Contamination is an issue that SHOULD have been done on the first pump replacement and certainly after the second one was going in.
Did you see the pumps/screens when they came out?

No, I didn't see the pumps/screens when they came out. I trusted the mechanic, but in hindsight that trust was misplaced because he should have flushed that tank; I watched him both times carefully. Which makes me wonder, can those dirty pumps/screen on my latest, newest pump be easily cleaned or do I need another new pump installed? Also, the gas cap is original and might not fit as tight as a newer one, letting dust in. And finally, I'm not exaggerating with respect to how many miles of dirty, dusty roads my rig has traveled, especially in that fine, Southwest grit so prevalent.

 

[LATECH]

Take off the new fule filter and empty it out fom the inlet. See what is in it. If it has a bunch of good ol southwest terra firma in in then it is likely the pump is trashed from silt. Fastest way to se what is / was in the tank

 

[chasracer]

You'll have to do some research but there might be a cold start injector that is not operating or a water temperature sensor that tells the computer to richen the mixture. Either one acts like the old choke on a carburetor. I would be very skeptical that a couple of new fuel pumps are picking up enough debris to stop a cold start but allow the engine to run after it has fired up.

 

[ExcellenceAuto]

Iirc the injectors are supposed to be getting close to 60 psi @ the fuel rail. Its been about 7 years so my memory is kinda fuzzy but I can double check tomorrow.

So it needs a shot of go, go, juice every time? Or just after sitting for a few hours?

I see you said it holds pressure for 10 minutes, but you need to watch it for longer than that.
There's a lot of stuff attached to that spider, and if its leaking in there the extra gas makes it hard to start, because its flooding out inside the upper intake.
This could be why its getting harder to start.

 

[anothermark]

You'll have to do some research but there might be a cold start injector that is not operating or a water temperature sensor that tells the computer to richen the mixture. Either one acts like the old choke on a carburetor. I would be very skeptical that a couple of new fuel pumps are picking up enough debris to stop a cold start but allow the engine to run after it has fired up.

Thanks. But what tells me it might be debris in the tank is the fact that replacing the pumps - both times - fixes the problem even if for only four hundred miles. Fires right up and runs fine with a new pump. So if it's a bad cold start injector or water temperature sensor I'd still be stuck with that problem and a new pump wouldn't fix that, right? That seems key to me. If it was electrical, a bad ground or module or relay the problem wouldn't go away by merely changing the pump. New pump, all is well. The same logic goes for a bad spider injector or pressure relay. If either of those were the problem, a new fuel pump wouldn't fix that either.

 

[anothermark]

Iirc the injectors are supposed to be getting close to 60 psi @ the fuel rail. Its been about 7 years so my memory is kinda fuzzy but I can double check tomorrow.

So it needs a shot of go, go, juice every time? Or just after sitting for a few hours?

I see you said it holds pressure for 10 minutes, but you need to watch it for longer than that.
There's a lot of stuff attached to that spider, and if its leaking in there the extra gas makes it hard to start, because its flooding out inside the upper intake.
This could be why its getting harder to start.

Yes, 60 -62 psi from what I have been reading. Starter fluid only on the first start and then it might start again after a few hours without it, although that window is closing and fluid is needed sometimes repeatedly during the day. I'll look into the flooding inside the upper intake. Thanks.

 

[ExcellenceAuto]

It needs 55 PSI to open the injectors. So if your fuel pump is not putting that out when you turn the key, the truck is not going to start right up like it should.
If you deadhead it at the filter, it should be 75 PSI minimum. That will tell us if you're losing pressure inside the intake...
You can also deadhead the pump by blocking the return line. Expect 75 PSI minimum.

 

[anothermark]

It needs 55 PSI to open the injectors. So if your fuel pump is not putting that out when you turn the key, the truck is not going to start right up like it should.
If you deadhead it at the filter, it should be 75 PSI minimum. That will tell us if you're losing pressure inside the intake...
You can also deadhead the pump by blocking the return line. Expect 75 PSI minimum.

Good advice. Thanks.

 

[LATECH]

You'll have to do some research but there might be a cold start injector that is not operating or a water temperature sensor that tells the computer to richen the mixture. Either one acts like the old choke on a carburetor. I would be very skeptical that a couple of new fuel pumps are picking up enough debris to stop a cold start but allow the engine to run after it has fired up.

No cold start injector on this rig

 

[LATECH]

It needs 55 PSI to open the injectors. So if your fuel pump is not putting that out when you turn the key, the truck is not going to start right up like it should.
If you deadhead it at the filter, it should be 75 PSI minimum. That will tell us if you're losing pressure inside the intake...
You can also deadhead the pump by blocking the return line. Expect 75 PSI minimum.

You cant "deadhead " the fule line at the filter and run the engine .Plus you need to screw a gauge in the line at the filter to check the pressure. Not what you want to do, but good that you mentioned the term "Deadhead" as it is a good tool for diagnosing internal leaks in the fuel system.
You can Deadhead the system by blocking the return line and checking the pressure at the port with the gauge,which is what you might want to do. Dont run the pump for more than a few seconds while deadheading as you can damage it
Deadheading the system at the return line will check everything except the regulator for leakage. If the pressure drops, the spider may be leaking ,or the check valve in the pump is leaking. If it holds well, but pressure drops swiftly with the system in normal fule loop( not deadheaded) then the regulator is not holding. The deadhead test will singly rule in or out a defective regulator
The OE spider had poppet style injectors and in the spider there is an electronic valve to turn line pressure on and off. The poppet is required to hold the correct pressure until the spec is breached at which time they spray
Most of the replacement spiders , if not all , have electronic injectors at the end of each spider leg. Depending on which type is in there, it may or may not exibit a pressure loss.Poppet style can leak but wont cause a leak at rest pressure of the system at the gauge, where as electronic injectors have line pressure right down to the end where it terminates at the injector.Those would exibit a loss of line pressure if a leak in a line or at the spider housing develpoed

I know this is a lot to chew on.I hope you have agood appetite..LOL

 

[anothermark]

You cant "deadhead " the fule line at the filter and run the engine .Plus you need to screw a gauge in the line at the filter to check the pressure. Not what you want to do, but good that you mentioned the term "Deadhead" as it is a good tool for diagnosing internal leaks in the fuel system.
You can Deadhead the system by blocking the return line and checking the pressure at the port with the gauge,which is what you might want to do. Dont run the pump for more than a few seconds while deadheading as you can damage it
Deadheading the system at the return line will check everything except the regulator for leakage. If the pressure drops, the spider may be leaking ,or the check valve in the pump is leaking. If it holds well, but pressure drops swiftly with the system in normal fule loop( not deadheaded) then the regulator is not holding. The deadhead test will singly rule in or out a defective regulator
The OE spider had poppet style injectors and in the spider there is an electronic valve to turn line pressure on and off. The poppet is required to hold the correct pressure until the spec is breached at which time they spray
Most of the replacement spiders , if not all , have electronic injectors at the end of each spider leg. Depending on which type is in there, it may or may not exibit a pressure loss.Poppet style can leak but wont cause a leak at rest pressure of the system at the gauge, where as electronic injectors have line pressure right down to the end where it terminates at the injector.Those would exibit a loss of line pressure if a leak in a line or at the spider housing develpoed

I know this is a lot to chew on.I hope you have agood appetite..LOL

Yikes. You just gave me nightmares, but thanks.

 

[johnsongrass1]

Are the injectors getting signal on crank?
is the sparky thing sparking on crank?

What I was trying to get at while being silly is what LATECH says, narrowing the problem down to a spark problem OR a fueling problem.
In thise case it appears to be a fueling problem. LATECH is a smart guy. I assume guy anyway....girl.....don't matter to me, the knowledge is there.

 

[ExcellenceAuto]

Also, I did test the fuel pump for full pressure at the fuel filter port one time before. Of course the engine is not going to run, but you will get a true reading what the fuel pump is putting out right there.
The pump will only run for a few seconds when you cycle the key.
Again you are looking for a minimum of 75 PSI.
This is really looking like a fuel pump is not putting out enough pressure to me. How it is only lasting a few months, that's a different story...