'89 K3500 VERY rough idle, losing power while accelerating, hard to start.

Disclaimer: Links on this page pointing to Amazon, eBay and other sites may include affiliate code. If you click them and make a purchase, we may earn a small commission.

Joined
Aug 27, 2024
Messages
11
Reaction score
18
Location
Washington
Hey there everyone, first time posting here. As the title states, I have an '89 K3500 454tbi that just developed a very nasty idle and throttle issue. The truck will only start when I hit the gas pedal, and once it starts it wants to die, and then once it catches itself the idle bounces between 800ish-1500ish rpm (no tach). No perceivable change in rpm when shifting in or out of gear, but while I'm driving it falls on its face under acceleration, then will catch itself, rinse repeat.

I replaced the fuel filter, and then checked the fuel pump pressure and it's right between 5-6psi. The spray pattern from the injectors looks heavy and thick, not a fine mist, but still in a conical shape. I'm assuming it might be the FPR or possibly the coolant temp sensor? But I don't want to go throwing parts at it when I'm not sure if that's the actual issue. I did the timing a few weeks ago and it was running perfectly, until a few days ago when I was driving on the freeway and these problems appeared. I've read many forum posts from here and several other sites and I can't seem to find any concrete answers.

One last detail, the previous owner started de-smogging the engine, the smog pump, egr rails, and lines have been removed. The egr valve is still in place, but the rest of the system is gone. I have a custom chip and block off plate on the way, not sure if the egr is the culprit since from my understanding that system doesn't come into play at idle. Also, there is no codes being thrown. The check engine light did come on briefly on the highway the day things went south, but when I attached a jumper to the obd1 port I only got the 12 code.

Any help is much appreciated, I can post a video of the issue if that would help with diagnosis.

Thanks in advance,
-J
 

JDGMC

Supporting Member
Supporting Member
Joined
Jul 20, 2022
Messages
276
Reaction score
552
Location
Upstate NY
Check every vacuum line for leaks (including intake and throttle body) then check your TPS and IAC). The manual for your truck will help you work through this and many other potential problems related to your symptoms. The manuals are Here
 
Last edited:
Joined
Aug 27, 2024
Messages
11
Reaction score
18
Location
Washington
Check every vacuum line for leaks (including intake and throttle body) then check your TPS. The manual for your truck will help you work through this and many other potential problems related to your symptoms. The manuals are Here
Okay I'll go check all of the vacuum lines before it gets dark, much appreciated.
 

Schurkey

Supporting Member
Supporting Member
Joined
Jan 14, 2018
Messages
12,569
Reaction score
16,333
Location
The Seasonally Frozen Wastelands
'89 K3500 454tbi

checked the fuel pump pressure and it's right between 5-6psi.
That's half of what it should be. Maybe less.

Spec is 9--13 psi, most folks want it towards the higher end.

Is your gauge accurate? Are you trying to read a 9--13 psi spec, on a gauge calibrated to 100 psi?

I'm assuming it might be the FPR
Possible. More likely a failing fuel pump, or poor voltage to the fuel pump.

Test fuel pump voltage as close to the pump as practical--generally the most rearward connector before the harness goes up 'n' over the tank. The voltage test MUST be done with the fuel pump RUNNING. Ideally, you'd have close to battery (or alternator, if the engine is running) voltage.

Then test voltage on the ground wire, again as close to the pump as practical, and with the pump running. Ideally, you'd have less than one volt.

possibly the coolant temp sensor? But I don't want to go throwing parts at it when I'm not sure if that's the actual issue.
You're saying it's fixed, and the problem was the MAP sensor. And maybe that's correct.

I would connect a scan tool, and verify EVERY sensor and computer output. Replace the O2 sensor if you have any question about it's age/mileage.

the previous owner started de-smogging the engine, the smog pump, egr rails, and lines have been removed. The egr valve is still in place, but the rest of the system is gone.
There are no "EGR rails". I suspect you're talking about the AIR pump plumbing.

I have a custom chip and block off plate on the way, not sure if the egr is the culprit since from my understanding that system doesn't come into play at idle.
EGR is non-functional at idle. OTOH, it can leak, providing a rough idle--but that's a symptom of failure, not normal operation.

There's no genuinely good reason to get rid of EGR.

Also, there is no codes being thrown. The check engine light did come on briefly on the highway the day things went south, but when I attached a jumper to the obd1 port I only got the 12 code.
Reading "codes" is better than nothing. The REAL diagnostic power is in the data stream. You need a scan tool so you can access the data stream--including but not limited to the history codes.

Problem solved! Bad MAP sensor was the culprit.
Problem is NOT solved, if your fuel pressure is genuinely that low.
 
Last edited:

JDGMC

Supporting Member
Supporting Member
Joined
Jul 20, 2022
Messages
276
Reaction score
552
Location
Upstate NY
Good to hear you found what may have been one of the problems, I overlooked the fuel pressure from your post and as @Schurkey already stated, the fuel pressure is too low. You should confirm that..not sure you’re in the clear just yet. Also, pull a couple of spark plugs on each bank and check the electrodes for color and condition. Btw one of the first things I did after buying each of my trucks was check components that could leave me or my better half stranded. I replaced the fuel pump/filter, clean out the fuel tank, and clean the TB/injectors. The resson for replacing the fuel pump is that every darn truck was on the very low end for fuel pressure or below. Since the trucks all had extremely inaccurate fuel gauges it’s likely the trucks could have run out of fuel several times and/or driven many miles near empty which will degrade the fuel pump. I also upgraded the fuel pump wiring to 14g - I think the OEM wires are undersized causing voltage drops etc (OEM~18g or 0.8-1.0 mm2). I definitely would check all components related to your problem and test each using the service manual if possible. You are more likely to pinpoint solution(s) and an explanation for a problem if you check all the pertinent vitals for a particular drivability issue. You will also become familiar with the tools/equip required to service the truck and get to know the health/truck as you go. If the data is difficult to understand for what’s normal and how it should behave (much like a symphony), then post all the diagnostic info on this site and someone like schurkey will surely narrow it down and recommend the next steps. Btw I mentioned the TPS in my previous response. I should have included the IAC also.
 
Last edited:
Joined
Aug 27, 2024
Messages
11
Reaction score
18
Location
Washington
I have 0 confidence in the pressure tester I'm using. It's a harbor freight special that maxes out at 10psi. It is specifically marketed at a "fuel pump tester" I hooked it up just in front of the fuel filter on the drivers side, I know 8-15 is standard but this isn't an "inline" type tester, it only has a single inlet so it only checks pressure from the pump to the filter, not all the way to the engine. I was told 5-6psi was perfect for this test. But I trust you guys more than the guy at Napa so I will try to find a true inline tester and try again.
 
Joined
Aug 27, 2024
Messages
11
Reaction score
18
Location
Washington
That's half of what it should be. Maybe less.

Spec is 9--13 psi, most folks want it towards the higher end.

Is your gauge accurate? Are you trying to read a 9--13 psi spec, on a gauge calibrated to 100 psi?


Possible. More likely a failing fuel pump, or poor voltage to the fuel pump.

Test fuel pump voltage as close to the pump as practical--generally the most rearward connector before the harness goes up 'n' over the tank. The voltage test MUST be done with the fuel pump RUNNING. Ideally, you'd have close to battery (or alternator, if the engine is running) voltage.

Then test voltage on the ground wire, again as close to the pump as practical, and with the pump running. Ideally, you'd have less than one volt.


You're saying it's fixed, and the problem was the MAP sensor. And maybe that's correct.

I would connect a scan tool, and verify EVERY sensor and computer output. Replace the O2 sensor if you have any question about it's age/mileage.


There are no "EGR rails". I suspect you're talking about the AIR pump plumbing.


EGR is non-functional at idle. OTOH, it can leak, providing a rough idle--but that's a symptom of failure, not normal operation.

There's no genuinely good reason to get rid of EGR.


Reading "codes" is better than nothing. The REAL diagnostic power is in the data stream. You need a scan tool so you can access the data stream--including but not limited to the history codes.


Problem is NOT solved, if your fuel pressure is genuinely that low.
I forgot to elaborate on why it made sense to me, the guy at Napa said the pressure in the large fuel line is always lower than what's actually going into your engine, as the line chokes down pressure increases so in that big wide line 5-6psi would equal 10-12psi in a line half the size. It made perfect sense to me when he said it but if you guys think that's wrong I will check again with a better tester.
 

Road Trip

Supporting Member
Supporting Member
Joined
Jul 8, 2023
Messages
1,695
Reaction score
4,804
Location
Syracuse, NY
Is your gauge accurate? Are you trying to read a 9--13 psi spec, on a gauge calibrated to 100 psi?

Greetings Anchor-Enthusiast,

The overall success of a sharp troubleshooter is due as much to the quality/accuracy of their
test equipment in addition to having a working knowledge of the subsystem they are working on.

Schurkey is making a very valid point about trying to get actionable info out of a 0-100 psi
fuel pressure gauge on a 'low-pressure design' TBI system. (I define 'actionable' as data that's
reliable enough that I'm willing to change parts because of what I'm being shown.)

EDIT: I see that you replied that you are using a HF 0-10 psi test gauge. Those gauges were
designed to test fuel pumps feeding carburetors, for they typically worked best at ~5-6 psi.
If it's a vintage test gauge from the last century, it could be lying to you.

For what it's worth, to me this would be the (affordable) gold standard in the 0-15 PSI fuel pressure
tester:

You must be registered for see images attach

Here's the current Summit listing for this ~$37 + shipping gauge: (OTC vacuum/pressure test gauge)

NOTE: I would be willing to use a cheaper gauge, but ONLY after I could test it side-by-side with one of these
and verify that it agreed. (Same thing for a bang-around daily use cheap multimeter -- only after it's proven
to agree with my 'reserved for reference use' Fluke. :0)

I forgot to elaborate on why it made sense to me, the guy at Napa said the pressure in the large fuel line is always lower than what's actually going into your engine, as the line chokes down pressure increases so in that big wide line 5-6psi would equal 10-12psi in a line half the size. It made perfect sense to me when he said it but if you guys think that's wrong I will check again with a better tester.

Charitably speaking, that's a well-meaning but misleading explanation. I wouldn't recommend using that line of reasoning
while troubleshooting your fuel delivery subsystem. I'll try to substitute a concise overview of what you want to look for.

There are 2 variables that we need to be thinking about in our fuel system. The first is how high the pressure climbs to,
and the second is how much volume that the fuel pump is capable of delivering to the TBI. For example, even a good pump
can be constrained by a dirty inlet strainer, similar to a human sucking on a smashed soda straw.

Or, assuming that the inlet sock is good, the fuel filter on the pressure side between the pump outlet and the TBI nozzles
is partially plugged. Thinking about it, with the fuel pump running while the engine is stopped, you might measure 12 psi before the
filter as well as the same 12 psi after the filter. Because there is no flow (volume) ...you are simply measuring the 'deadhead' (no flow)
pressure that Schurkey has referenced.

But let's say that you start the truck, and idling there is now a 2 psi drop across the partially-plugged filter. A bit oversimplified,
but having a larger vs smaller diameter hose between the filter & TBI nozzles isn't going to make a difference at what the
injectors see/feel.

Taking this thought experiment a little further, I'm going to state that at 10 psi (nozzle) the truck will idle OK. And it will
drive during light throttle/light loads OK. But with more demand (longer injector pulsing) we could watch the pressure
drop in response, and quickly we'll reach a point where insufficient fuel will be delivered to the air rushing past, and the
bucking/stalling will begin. (NOTE: If we were to check the fuel pressure *before* the fuel filter, no doubt it would remain
steady. But the TBI shower nozzles don't care about what's going on between the fuel pump & plugged filter. They only
care about the fuel pressure at the input to the injectors. (!)

In this case, simply changing the inline fuel filter will remove the pressure drop/fuel flow bottleneck across the new filter,
and (assuming a good pump) the pressure will always be more than needed for the FPR to be able to hold the pressure
steady at the nozzles, no matter how high or low the engine demand is.

****

The reason why I dragged us through the above is that a static fuel pressure check alone is just a partial go/no-go answer.

When I see fuel pressure, this tells me that I at least have some electricity being sent down to the pump. But in order to
figure out IF we have enough fuel pressure x volume to cover all possible demands, the best answer is to (temporarily) install
a Tee that allows me to monitor the fuel pressure while the vehicle is being operated, preferably under the same loads as
when you are doing the daily driver thing.

NOTE: Since outboard fuel volume test equipment is spendy, we instead use a fuel pressure gauge AND also monitor the fuel
trims reported by the computer.


*** CAUTION CAUTION CAUTION ***

It goes without saying that safely testing the delivery pressure of a flammable liquid in the vicinity of a hot exhaust
system is
serious business. This is not a dribbly duct-tape connection & hit the road scenario. Elsewhere in this forum
members have documented some slick/safe ways to safely tee into their fuel delivery system. Deadhead testing a fuel pump
in a stone cold engine bay is both easy & gives you a quick go/no-go answer. But in order to observe the system under load
is quite a bit more involved. Be aware. Leak-free test connections and an onboard fire extinguisher is a prudent bare minimum.

****

If I were to guess, I'll bet that Schurkey's mindset is that if your system can only deliver 6 psi before the engine starts
to draw fuel, then it's not going to take too much gas pedal to force a fuel delivery error at the nozzles.

Apologies for the length, but given what the helpful NAPA dude was telling you, I wanted to try to help explain why we
need to worry about pressure and end-to-end volume delivery capability, from inlet sock > fuel pump > inline fuel filter > TBI nozzles.

And a good quality 15 psi fuel gauge tee'd in (plus fuel trim data from the computer) will tell you everything that you need
to know. EDIT: If a trustworthy gauge still tells you 6 psi, then a new pump (and/or fresh wiring) is in order since you've
already tried a new filter. Fingers crossed that the new/accurate gauge shows you 12-15+ PSI at the input to the TBI, and
that the MAP sensor was 100% of your original driveability issue.

Best of luck keeping that GMT400 on the road and earning it's keep.
 

Attachments

  • Monitoring Fiesta fuel pressure during commute.jpg
    Monitoring Fiesta fuel pressure during commute.jpg
    387.1 KB · Views: 32
Last edited:

Schurkey

Supporting Member
Supporting Member
Joined
Jan 14, 2018
Messages
12,569
Reaction score
16,333
Location
The Seasonally Frozen Wastelands
You're testing the pump dead-headed. No actual fuel flow, and the pressure regulator is not in the fuel circuit.

With fuel flowing through the throttle body, the regulator should be REDUCING fuel pump pressure to 9--13 psi, the pump should be capable of something like 15--18 psi.

If the pump is being supplied with appropriate voltage/amperage including a proper ground circuit, THAT PUMP IS DONE. Or at least the tank needs to be opened-up for inspection of the pump, the connecting hose, and the fuel filter sock on the end of the pump.




The guys at NAPA are totally wrong about pipe size differences increasing fuel pressure.
 
Top