I thought I had a bad fuel pump, but now I have no idea what is wrong.

S

Scottwax

New member
Since Tuesday night, occaisionally when driving, my car would feel real sluggish, sometimes when lightly accelerating or at a steady speed and I would have to step harder on the gas to maintain my speed, step hard enough and it would downshift and accelerate again.



Took it into my Honda mechanic this morning, figuring the fuel pump was getting weak. Nope, passed all testing, idle, cruise speed and it didn't lose pressure when you turn off the car. EGR valve passed, TPS passed, MAP Sensor passed, distributer (coil, ignitor, wires, etc) passed, PA (atmospheric sensor) passed. Even changed the ECU (I had a spare one at home) and still it was not running right or throwing a CEL code. Vacuum gauge needle was rock steady at idle and accelerating, so no apparent valve problems and at 2500 rpm, the vacuum didn't drop off, so we could rule out an exhaust blockage.



On the test drive after changing the ECU, I finally had to get gas, since I had let it get low in case the fuel pump had to be changed (tank has to come out), and after putting in about 5 gallons, I was driving back to the shop and after driving maybe 1/2 a mile, my car wouldn't accelerate past about 45 mph, either in 4th or 3rd, finally after about a mile, the CEL came on and the engine problem cleared up. Ran fine all the way to the shop and then at 15 mph in the parking lot, same thing.



Pulled the code, 43 which is fuel delivery. Pulled the (2 month old) fuel filter, dumped it out, couldn't see any debris, could blow through it. Put it back on, ran decarbonizer and fuel injector cleaner through the engine. Still felt sluggish at times on the test drive. My mechanic took it out and noted a few times acceleration seemed sluggish. Checked timing and it was right on.



I went with him on a long test drive and a few times it was sluggish, but at least now it would accelerate instead of just maintaining speed, and it was only sluggish for a few seconds, same on my drive home....so the problem doesn't seem as bad.



I filled up on gas last Thursday and the problem began on Tuesday, so I figure if it was bad gas, the problem would have shown up sooner, right?



Now, today, the car runs fine on the freeway, drove all the way to my brother's in Rockwall (about 42 miles freeway) averaging 70-75 mph. The problem occurs under light acceleration, I can hit the gas harder, the car will downshift and accelerate normally. Seems to mostly occur when accelerating lightly between about 1500-3000 rpm. Comes and goes too. I added a bottle of fuel injector cleaner and more gas today.



I think extended freeway use, including a full throttle burst from 70-85 mph with no problems, bogging, etc rule out the fuel pump.



BTW, 1990 Accord, automatic.
 
Did you check the O2 sensor? If it's contaminated, or getting lazy, the A/F ratio will be out of whack. This happened to me.....I installed ceramic coated long tube headers, and when I first started it after the install with brand new AC Delco O2 sensors, the coating smoked for a few minutes. I didn't think anything of it, then starting 2 weeks later, whenever the car would go into closed loop (using the O2 sensors for the A/F ratio instead of just the Mass Airflow Sensor), it would stumble and give the same symptoms you described, barely accelerating. It finally got to the point where the car was running so rich when it went into closed loop that it would nearly die at idle, then when the service engine soon like came on, it would run perfectly smooth......this is because the PCM detected both O2 sensors getting slow response and put it into open loop, ignoring them, and using pre-programmed fuel maps (14.7:1 A/F ratio). It's definitely worth a try....shouldn't be hard at all to change either since the manifold is right in front of you once you open the hood.



Hope you get it worked out...keep us posted!



-Matt
 
Hey Scott... had a similar problem on a Saturn I just sold. I had replaced the fuel filter as I thought that was most obvious. Then I did the air filter also just to be safe. No improvement to speak of.



Then, all at once, I replaced the PVC valve, spark plugs and wires. Something in that combination did the trick. It ran like a charm for the week we drove it before we sold it.
 
Thanks, Matt. My mechanic says normally a bad O2 sensor on my car won't affect how the car runs too much, but I do know mine is pretty old, I changed it when I got a header and that was about 160,000 miles ago.



Usually though, it throws a code pretty regularly and other than that one time, the CEL hasn't been on. Mileage has been good also, but maybe the O2 sensor got contaminated by bad gas?
 
zzyyzx said:
Hey Scott... had a similar problem on a Saturn I just sold. I had replaced the fuel filter as I thought that was most obvious. Then I did the air filter also just to be safe. No improvement to speak of.



Then, all at once, I replaced the PVC valve, spark plugs and wires. Something in that combination did the trick. It ran like a charm for the week we drove it before we sold it.
Click to expand...



I know the plugs are good, they are only about 6 months old (NGKs), the plug wires are a few years old (ACCELs) but seem fine and I replaced the PCV valve about a year ago. It just seems if I had bad plugs or wires, the problem would be worse at full throttle or at freeway speeds. The PCV is cheap though, I can just replace that and see what happens.
 
It's possible, I've seen it happen to some people who fill up at gas stations that use certain additives in the gas.



Also, check the ECU grounds......I've also seen stuff like this happen on the Camaro/Firebird forums from something as simple as corroded battery terminals :shocked
 
Okay, I can check engine grounds. About 6 months ago, I cleaned every single ground connection and the spot each ground was attached too. Maybe one of them worked itself slightly loose. Battery terminals look real good though.



I swapped out ECUs with no change, seemed properly grounded.



Anyway, I really won't be doing anything to my car for a week or so. My parents told me I could take my Dad's car to get my kids next week in Kentucky (my parents are on a cruise to Rome), so I just picked up his car a few days early. I don't think my Accord is going to strand me, but I have 4 fully scheduled days between now and when I leave Wednesday for a 930 mile drive and I don't want to take a chance.
 
Im having the same issue on my ranger, it happens so sporatically, I really am at a loss of what it could be. I still need to change my fuel filter, not sure it that will do anything, keep us updated, I want to know what to look for!
 
Sure sounds like a fuel filter. Question: when you changed it 2 months ago, what did the old one look like? If it was REAL bad, you could have a clog elsewhere - is there another filter ahead of your fuel pump, perhaps? (I know nothing of Hondas, but some cars have as many as 3 fuel filters, some only replaceable with major units like the fuel pump). Sounds like Honda did a pressure test, did they do a flow test as well? Just some ideas.
 
prowler said:
Sure sounds like a fuel filter. Question: when you changed it 2 months ago, what did the old one look like? If it was REAL bad, you could have a clog elsewhere - is there another filter ahead of your fuel pump, perhaps? (I know nothing of Hondas, but some cars have as many as 3 fuel filters, some only replaceable with major units like the fuel pump). Sounds like Honda did a pressure test, did they do a flow test as well? Just some ideas.
Click to expand...



The only other filter is the sock on the fuel pump...and the gas tank has to be removed to get to it. With 277,900 miles on the car, if I pull the tank, a new pump is going in even if the current one (which is the original) tests fine.



It was flow tested too. Passed. Problem is only under light load, and yes, the fuel pump/filter were the first things I thought of and he checked.
 
MattZ28 said:
Hey Scott, I just realized we have the same birthday! June 7th....except mine's 26 years after yours :D
Click to expand...



Which would make you 16, almost 17? I have a bowling ball older than you. Seriously!
 
My brother is going to drop my car off at his mechanic's while I am in Kentucky. My mechanic is very good at r & r but not as good as diagnosing as my brother's mechanic....but he is 40 miles from me so I only use him when my guy is stumped on something. His best guess over the phone is the O2 sensor, especially since the last time it was changed was when I installed a header, 160,000 miles ago.



Once it is fixed, I will revisit this to let y'all know what was wrong. Thanks for all the suggestions!
 
Hope you can figure the problem out, but with 278k miles on the odometer, I think you've gotten good service from your Accord. How about an engine swap? Just kidding. I've been looking for an excuse to do an engine swap on my Civic. Good luck.
 
Hello Scott,



ill toss my shadetree thoughts out.



you have checked the obvious things i would have. based on your posts and diagnostics done, i would almost entirely rule out fuel starvation (filter, pump, regulator, etc), and look more at fuel management (lambda's, etc, as you are).



most cars will not use input from the O2's at WOT, but that does not help. also, many cars default to a limp or open loop mode when O2's are out of range (almost definately tossing a code).



a really quick thing to check: your TPS (throttle position sensor)



if you have an analog voltmeter, that is best, but digital works fine too. hook it up to the TPS's ground wire and return wire. slowly open the throttle, noting a nice smooth sweep in the needle. the sweep should be linear with the movement of the throttle). if there are any dips, breaks, etc, that is no good. if you have a repair manual (or look online), check the spec's. e.g., on EFI stangs, one usually sets the idle TPS at ~< 0.97 volts. at WOT, voltage is ~4.5 volts. the feed to the sensor is 5 volts. find similar specs for your Honda.



since the TPS tells the puter how far you have the throttle whacked open, if it is giving a bad (low) signal, it would in essence, be like you lifted your foot off the throttle until it gets past the dip.



like i said, super quick test (couple minutes total).



good luck, Scott.
 
I had a similar problem once on my '86 Accord. I had two fuel filters, one under the hood and one mounted to the fuel tank behind the left rear tire. I replaced both of those and then replaced the fuel pump. That fixed the problem for me. The fuel pump was really easy to replace. All I had to do was lift up the carpet in the trunk, pop off an access cover, and remove a few bolts and it lifted right out. Are you positive you have to remove the tank to change the pump? I've never seen a Honda that didn't have an access panel. Even my Civic has one underneath the back seat.
 
aarrgghh, this was the last thing i read/looked at last night before hitting the sack. i knew i was missing something when i posted above. then it hit me as i was drifting off - if you have not done so, check and clean your MAF wire, (if you have one - i assume you do). they can get dirty, especially if you run a cotton/gauze filter (K & N, etc) - the oil you impregnate the filtering media with gets 'sucked' downstream onto the wire. this will cause symptoms like you describe....clean it with contact cleaner. some folks have touched the wire with a cotton swab while cleaning- i dont have the guts to do that (the wires break very very easily). i just spray them with contact cleaner.



this is more likely to solve the problem than the TPS. again, this takes you 5 minutes tops.



good luck, big S.
 
GreyFixer said:
I had a similar problem once on my '86 Accord. I had two fuel filters, one under the hood and one mounted to the fuel tank behind the left rear tire. I replaced both of those and then replaced the fuel pump. That fixed the problem for me. The fuel pump was really easy to replace. All I had to do was lift up the carpet in the trunk, pop off an access cover, and remove a few bolts and it lifted right out. Are you positive you have to remove the tank to change the pump? I've never seen a Honda that didn't have an access panel. Even my Civic has one underneath the back seat.
Click to expand...



90-97 Accords only have an access panel for the fuel sending unit. The tank has to come out...believe me, if there was a way to do it without removing the tank, I am sure my mechanic (and my brothers) would much prefer using an access panel. Besides, my Chiltons, Haynes and Honda Shop manual all confirm the need to drop the tank. Goofy, huh?



HISSIN50-no MAF sensor or screen and the TPS already checked out.



Lowejackson-thought about that also and that would normally only be a starting issue or a constant drivability issue (like if the engine was thinking it was cold the whole time and ran rich).



No, it was something much easier. One of my 6 month old NGK Super Platinum (similar in design to an Iridium plug) plug's electrode was missing about 1/8th of an inch, giving that plug a massive gap to fire across. Never even thought initally to check the plugs since they were nearly new, no noticable misfire and the ECU threw a fuel code...which in retrospect, may have been because the plug wasn't igniting the fuel properly so unburned fuel was reaching the O2 sensor and making the ECU think the engine was running too rich.



Anyway, cost me $10 to fix it.
pat.gif




Most likely cause of the plug failure was detonation and (or) bad gas. I filled up a few times with plus unleaded instead of premium which is a bad thing since my timing is advanced about 3-4 degrees from stock (noticably better bottom end and throttle response) so timing stays and only premium that is not from 7-11 goes in!



Thanks for everyone's input! :wavey
 
You must log in or register to reply here.
Back
Top