:? Had a rather confusing experience last weekend. (99 Bonnie, 92k miles, VIN "K") Seemed to me like fuel starvation. Here are the details: Casual, easy driving on 2-lane road in South Dakota Black Hills. -speeds anywhere from about 20 to 60 mph. -very winding road, but not doing any "performance driving." -engine temp normal, 195. -ambient air temp probably upper 80'*, maybe lower 90'* -A/C on. -fuel tank level about 1/3 to 1/2 tank. Started out with one case of slight power drop for about 1 or 2 seconds. Then fine for a few more miles. Then the serious power drop-outs started, just like engine died, but it never did. -just dropped down to idle speed, -no matter where the throttle was. "Feathering the throttle" didn't help. Wide throttle opening didn't help. (only difference, naturally, was strong acceleration during the 1 or 2 second spurts of power) Tried lower gears, increasing rpms, still no change. Stopped, removed the big plastic trim cover, reinstalled the oil filler tube & cap, did the typical under-hood survey, found nothing visible, audible, or otherwise abnormal, reseated numerous sensor connectors. Even at idle, the engine would act like it was going to die, but aways struggled back to life. -pulled gas cap, had some vacuum, but didn't change the problem. Limped back down the road (down hill) a couple of miles to a small piece of civilization. (C-Store, a few cabins to rent, etc) Let it set for about 45 minutes with hood open. Restarted, still same problems. With the Gas Filler Cap off, and my ear to the filler tube, tried to listen to the normal pump running sound, thought it sounded quieter than normal. -also seemed quieter than normal in the car too, when doing the test of listening for the couple of seconds of pump run when just turning on the key, without starting the engine. -relay always ran the pump for the typical couple of seconds.

So, we spent the night there, and had the car hauled in to nearby town the next morning. Got new pump, strainer, and filter. Tested fuel pressure before changing anything, and read 40psi, with only a slight occasional fluctation, before, during, and after normal operating temps. (of course ran fine, and seemed normal at that time) Anyway, changed pump, strainer, & filter. (Filter was just fine, and only the slightest dirt on the bottom of the strainer) Next day drove the 170 miles back home, and ran perfectly.

So, what happened ?? (-I've had in-tank pumps fail on three other vehicles through the years, and each of them was a totally dead, hard failure. -not like this). Was it really the pump ? :?: Your thought please Thanks !!

There'* a diagnostic connector for your fuel pump directly behind your battery. Connect a jumper from the green connector to the positive post of your battery. The fuel pump should run.

Read resistance from the connector to ground with the ignition off. Should be a straight short through the pump motor, but a higher resistance could show corrosion on a connector. Turn the ignition on, and it should go from short to open. This shows the fuel pump relay is working properly.

This is really interesting to me as I'm having lean effects and 40psi at the rail also. I've always thought pressure was the direct effect of volume on a given space. To me that would mean that as pressure drops volume would drop an equal amount....is it possible to have 40psi (not THAT far from spec) and get drastically less flow? Maybe I'm just tired and not thinking about it clearly...

I once had a similar problem and started out by replacing the fuel filter. Although it sure seemed fine by inspection, I went ahead and changed it as it was already out of the car, I had a new one, yadda, yadda. Fixed the problem. I was a little perplexed, but pleased, as the filter sure didn't look dirty or full of crud and I could easily "blow through it". Perhaps the problem was simply a need to replace the filter?

i had a similar experience in a 96 grand am. turned out being the MAP sensor. the pcm couldn't decide if it had a ton of air, or no air at all to work with. as a result, no matter where my foot was, it would lag. if i floored it, it would lag. then every now and then, it would send the proper reading to the pcm, and the car would jump like crazy at the sudden added power.

very crazy. went to replace it, but it just started working normally again, so I didn't have to bother with it. I'd check that out...

p.*. if the MAP did die, then what you were explaining would make perfect sense - the fuel pump would still work just fine, but only putting out as much gas as the PCM calls for. so you'd have it working at a pace that seems way to low, but its doing just what it'* supposed to...

Thanks to everyone for the replies

To Dr Jay, you bring up a good point to ponder. The relationship between pressure and volume can be kind of a "sneaky one." But when analyzing the basic laws of science at work (hydraulics, of course) it really becomes quite logical. I'd really like to dive off into the full explanation ,,,, BUT I'm always hesitant,,, -thinking that a lot of people would rather be spared of all the details. (You know how it is when we encounter those whose motto should be: "Don't confuse me with the true facts. My mind is already made up.")

ANYWAY, in our rather simple hydraulic picture here, we could have a condition where the fuel rail pressure is right where it should be, but the volume is too low. I'll TRY to explain what I mean. For the purpose of explaination, I'll use the example of a nearly plugged filter, or a severely pinched off fuel line. This could reduce the volume capability (ie flow) very severely, but under conditions of low demand, (engine not under load, ie. idle, low cruise, etc) the flow is still sufficient to produce normal fuel rail pressure. -and that is also assuming the fuel pressure regulator is working correctly too. HOWEVER, just as soon as the system demand exceeds the capability of the system to supply the necessary volume (flow rate) then the pressure is gone, (-or seriously reduced) -hence immediate lean condition. (fuel starvation).

To me, the only way (-or at least quickest, simple way) to determine this would be to be able to monitor the fuel rail pressure while driving under conditions of various engine loads, from light to fully loaded (hard acceleration, up steep hill, wheel dyno, -whatever)

Oh, also, thinking of the pressure/volume relationship, it is possible to have the "reversed condition." -that is adequate volume, but low pressure. -such as would be caused be a bad regulator, or possibly pump. -but having dissected several "in-tank" pumps, I'm thinking that their design is not real likely to produce full volume without full pressure capability too.

Anyway, "keep the thoughts flowing !!" -and Thanks to ALL !!

(I think if I had a MAP failing, it should set a code, -maybe not? -anyway it didn't.)
-and yes, it surely could be the filter, although the "quick and dirty" test showed it to be okay. Thanks again.

Awesome explanation!
I get the feeling thats whats going on with mine. I get 40psi but have classic symptoms of a lean condition...On that note, I got ahold of the fuel pump people earlier. They were supposed to have my pump but the tech guy (real nice guy) said he was getting ahold of Walbro to check with their data sheets to ensure it will fit...pretty good service I'd say


Oh and for future reference, I never have too many details I love learning as much as I can.

Interesting you should mention your MAP sensor, I just replaced mine today. In hunting down my lean condition (not accepting JUST the fuel pump) I decided to test my MAP. Its supposed to have 1-2v at idle and since it gets 5v in I'd assume somewhere near that at WOT. I pulled 0.2v at idle and 0.3v at WOT (close to wot). Wow huh? Interesting my computer must have picked up and re-learned it because the boost gauge would still get to half of where it should be. By replacing it I can now get it to read where it should be. The idle is MUCH smoother and so is throttle response. Oh and I didn't even get a code for it! I think my computer just ignores everything, although I have done a few things to set it off.

Testing is easy. One way to do it is to disconnect the plug and use another set of wires to connect the left and right side of the plug to the proper connections on the MAP. Then test the voltage at the middle plug on the MAP.

I didn't have connectors so I did it the cheapie way. I just took a sewing needle and lightly punched it into the middle wire and tested it that way. Much quicker and easier I think.

Anyway try that out, you might be surprised at what you see .

DrJay, WOW Thanks for the great info -especially pertaining to the MAP sensor info. (-and thanks also to everyone else too, mkaake for mentioning the MAP, and willwren for the pump testing info. DANG, wish I'd known about that one, this weekend, while it was still in failure mode,,,)

I will try the MAP testing. -soon, I hope.

Hey I had to laugh about your comment on your computer "ignoring everything." Man-o-man, that is how I felt last weekend, as the stupid system just got sicker and sicker, until it was down to first gear, and barely pulling itself down the road. Now I must ask, for as danged smart as that stupid computer is supposed to be, HOW IN THE HECK, can it not figure out that "we have a serious problem here, Huston."

(-I have to think that my computer was doing some funky computing, -maybe it was so confused that it just figured that everything it was seeing was FUBB ! -(that is, "Fouled Up Beyond Belief")

Gotta Run. -more later.

Regarding fuel pumps, I have seen them die both ways from slow and agonizing to sudden and fatal.

Regarding fuel pressure, it is not only pressure but more importantly as mentioned, pressure drop. Ideally, if your gauge hose is long enough, tape the guage to the windsheild and see if there is a pressure drop when you accelerate. The same thing can be accomplished with a short hose by brake torqueing it. The pressure should hold relatively steady. If there is a pressure drop it is a sure sign of a failing pump.

Ranger, excellent point. And I sure did wish that I could have done exactly that: monitor the rail pressure under load, at the time of failure. (the next day, before replacing parts, I did the brake/torque thing, but at that time everything was running fine anyway). Which brings me to the point of where I'm going to check on the availability of a fuel pressure gage w/remote transducer (fire safety) that I can add to the car interior indefinitely (-can always remove it, and use it on another car -will check Auto Meter, and others).

Say everyone, I was just reading through the 1999 H Platform Service Manual, sections related to this problem. AND, found something VERY INTERESTING Get this: (quoting) "The PCM alters fuel pump speed by varying the duty cycle on the pump speed control PWM circuit to the fuel pump control module. Under normal conditions, the fuel pump speed control PWM circuit duty cycle is set at 33%. When higher fuel volume is required due to increased engine load (MAP sensor value over 90kPa) the PCM switches the fuel pump speed control PWM circuit duty cycle to 100%. At the higher duty cycle, the fuel pump control module increases the voltage to the in-tank fuel pump, allowing a higher volume of fuel to be delivered to the fuel rail."

To that I say, "good grief " Talk about unnecessary complexity.

AND DrJay, there you have the exact reason that changing the MAP fixed your lean problem. I'm guessing that your pump was always running at 33% duty cycle. Specifically, the pressure was fine (at low demand) but you were into the throttle, the volume delivered by the pump was insufficient.