Is this low flow indicate bad fuel pump ?

[AC439]

I think I finally have to admit something that I hate to admit - fuel pump.

Today I did the fuel pump flow test as per FSM 12-50. I did it exactly as stated on the procedure and I did it twice just to be sure.

Measured flow rate is about 120-130 c.c in 10 seconds. I remember reading another member posted result of 60 cc/10 seconds. I know the spec calls for 230 c.c in 10 seconds or more.

The fuel rail is clean and I just did manual injector cleaning. The fuel filter was replaced 3 years ago and I drive less than 5k miles/year on this car. I did open the access to the fuel pump top, unplugged the electrical connector and sprayed electrical contact cleaner both ends and put it back before I did the flow rate test.

Anyway, the car drives ok except the little 2000 rpm shutter. I think I'm closing into the problem but figure I would post and ask if the 130 cc /10 seconds flow is really bad and I should get the pump changed.

Next I would do is to rent fuel pressure gauge from autozone and see what pressure I'm getting. If I have to change the fuel pump, I would. Doesn't seem its too difficult but its just dirty over that top of the pump (externally).

I checked my spark plugs many times and never see a rich fuel condition....

[InAccordance]

Pumps are dirt cheap so may as well swap one out.
13 bucks +ship on the ol rockauto

I know I had an electrical issue a while back that was burning up pumps, went through 4 before I figured out the problem. The one in it now is an el cheapo from rockauto and it's supporting DCOEs so the quality seems decent.

[AC439]

Yeah, I'm leaning towards to just change it. Now I'm tempted to flush fuel tank but am worrying about "stir up the dust". Since there must be sediment at the bottom, I'm afraid this may clog the new " sock" filter. Or I better just change the fuel pump and forget about flush or clean the tank...

[AC439]

I just realized on rockauto the pumps and the sock filter are separate items. I'm looking into the Kemso 340 LPH model as Shane86 had installed. This comes as a kit with everything I needed for $30.

[2oodoor]

Pumps are dirt cheap so may as well swap one out.
13 bucks +ship on the ol rockauto

I know I had an electrical issue a while back that was burning up pumps, went through 4 before I figured out the problem. The one in it now is an el cheapo from rockauto and it's supporting DCOEs so the quality seems decent.
What was the issue and how did you correct it?

[InAccordance]

There was something in the hazard circuit that was causing power to run to other things. Like, the car could be sitting without the key in it and you could randomly hear things kick on. Fuel pump was one of them, it would run dry and burn up.
Haven't gone into it in depth to figure out where the fault is but I just pulled the hazard fuse and everything is fine. Only thing I dont have is the hazard function (that switch on top of the steering column).
It would cause things like the door dinger to go nuts and dash lights would flash while driving. It was amusing at times tbh.

[2oodoor]

That is strange, key off circuit back feeding ign switch or fuse blocks somehow.
Anyhew Im about to go with pump #4 myself. I think Im going to just run new wires and use a relay/ toggle switch. I need to confirm what wires at the pump are + & -
I am carbed btw believe it or not, weber 38 b20a.

[InAccordance]

I was close to rewiring the pump too but figured out the problem for the most part so became a moot point.
Carb brother highfive!
a20a1 twin DCOE

[2oodoor]

Ahh yeah!
I have a pierce manifold and a pair of dcoe in a box, they may wind up on the car one day...

[Shane86]

I just realized on rockauto the pumps and the sock filter are separate items. I'm looking into the Kemso 340 LPH model as Shane86 had installed. This comes as a kit with everything I needed for $30.

It's a good pump for the price It can handle 6-8 PSI boost pressure no problem

[AC439]

Thanks and good to know. I won't run boost though. I'm planning to take the fuel pump out tomorrow to take a look. I would expect the sock filter super dirty or may even clogged. The fuel pump is original as well as the sock and never been removed from the tank.

[Dr_Snooz]

Just look in the bottom of the tank when you have the pump out and see what's in there. My guess is you'll be underwhelmed. If it looks real bad, you can clean it out then.

That hazard circuit seems to have lots of gremlins. About 10 years ago, if I'd use mine, it would disable the blinkers for a day. Now the hazards don't work at all, but everything else does. :huh:

[AC439]

You are right Snooz, it is definitely underwhelming. I was expecting to see a very dirty sock filter and lots or sediment at the bottom of the tank. But I saw none of those.

While I had the pump out, I soaked the sock filter in a cup of gasoline, then sprayed brake cleaner over it, followed by carb cleaner. The first rinse was kind of light brown. I did not get much dirt out from the sock filter, it was relatively clean and in good shape consider the pump and filter had not seen any daylight since the car left Japan. I let it dry and put it all back. Car drives the same, no change.

I have the engine bay fuel filter on order. I will replace it and do the flow test again. With fuel pressure test (have to go to autozone for tool) and flow test, I think it will confirm the weak fuel pump.....

[AC439]

Ok, I got the fuel filter delivered right after my last post so I went and changed it out. Test drive shows absolutely no change. Next will try to get the autozone tool to measure fuel pressure and do the fuel pressure regulator test. I think I am pretty certain its the fuel pump if the problem is not with the pressure regulator.

BTW, I measure the voltage at connector on top of the fuel pump and its 12 volts (engine off) so main relay and wiring is good to the pump.

[InAccordance]

Ahh yeah!
I have a pierce manifold and a pair of dcoe in a box, they may wind up on the car one day...

I tried getting ahold of them but didnt get anywhere. Seems like they arent interested in making that one anymore.
Guy I ended up going with is in the UK... DanST Engineering. Only cost 312$ after dollar to pounds conversion. He has the pattern now and can make more.

That last pic AC439 posted, look like a little tub looking thing inside the gas tank... mine doesn't have that, strange...

[Dr_Snooz]

it is definitely underwhelming.

The pump on my Suburban crapped out a couple years ago. It's another '89 and was someone's hardcore wheeling rig before it came to me. They had monkeyed around with the tank before, repaired a rock hole, didn't do it well enough to make CA smog rules happy, it leaked vapors, stunk like gas horrible if I topped it off, etc. I was expecting a real nightmare when I looked inside. I didn't find much of anything though.

Keep us posted on your progress. I'm curious to find out what the problem is.

[AC439]

I decided to get a fuel pump from rockauto (picked the first one on the list for LXi). Received and installed today. I reuse the original strainer since it is not clogged and seems in good shape. Initial test drive (cold engine) was strong but once it warmed up just a little, the shutter at 2k stills shows but seems less. Now I notice a little miss fire around 45 MPH too. I am disappointed that it is not the fuel pump. Will go to autozone or O'rileys to borrow a fuel pressure tester and see what kind of pressure on the car. Pretty much all ruled out except fuel pressure regulator. I'm still at a loss with this....

[AC439]

Today I went to HF and got the $20 Fuel injection pressure gauge set. It works fine on the service bolt and no mod needed to get it to work. Just put teflon tape on all fittings.....

Here's the result -

Engine running, vacuum hose connected on FPR - 32 PSI
Engine tunning, vacuum hose disconnected and plugged - 41 PSI

Then I taped the gauge onto the windshield and have my son read the numbers while driving:

Lowest measured - 31 or 32 PSI
WOT - 39 PSI

FPR working and adjust fuel pressure in proportion to throttle position. It will regulate fuel pressure in different throttle positions such as light throttle around 34 PSI and moderate throttle gives 37 PSI.

When the stumble happened at 2k, it was about 37 PSI.

So, I did not get it fix and fuel pressure is good with new pump, filter, cleaned strainer etc. I thought I found the problem with the low measured volume test but it wasn't the cultpit.

I think I'm back to focus on the electrical side....

p.s. Also checked the fuel injector resistors and they are all within spec.

[Shane86]

Did you say you replace a ignition coil already

[AC439]

Did you say you replace a ignition coil already

Yup Shane. I thought I was closing in to the cause of the problem but with new fuel pump and fuel pressure test, it doesn't seem to be caused by staving fuel supply.

Also out of desperation, just did a quick seafoam via brake booster. No noticable change afterwards. The car has no problems rev up to 5-6 k but just at light throttle gives the little shaking / misfiring around 2k....

[Shane86]

Sense everything's basically been replaced it almost sounds like the TPS sensor. throttle position sensor on the throttle body

[AC439]

Yeah I can certainly check the TPS. I had worked on my 2000 Accord TPS before and able to cut a slot on the two bolts to replace with M5 screws for adjustment. Who knows, it may be off after 32 years of engine vibration and needs a calibration. Its basically a variable resistor and I'm hoping the carbon track inside is still smooth and not giving me hiccups. When I have it out, I'm going to compare it to my 2000 accord TPS and see if they can be interchangeable physically (still would need some connector and wire cutting if it will work).

[Shane86]

Yeah that could be the problem that's one of the main reasons I swap to a B16 intake manifold.

[AC439]

Wondering if O2 sensor is the problem too. Usually no studder first 45 seconds cold engine (open loop) but symptom getting more pronounce after the first minute (sensor is warmed, close loop). So I think if O2 sensor is bad, it may be sending wrong voltage to the ECU (may make the ECU think its too rich?).. Just my other thought.

Definitely I will check TPS calibration first...

[Shane86]

That could definitely be the O2 sensor I had similar situations like it

[Oldblueaccord]

Wondering if O2 sensor is the problem too. Usually no studder first 45 seconds cold engine (open loop) but symptom getting more pronounce after the first minute (sensor is warmed, cold loop).
So I think if O2 sensor is bad, it may be sending wrong voltage to the ECU (may make the ECU think its too rich?).. Just my other though.

Definitely I will check TPS calibration first...


I think all the EFI tanks have a bolt on the bottom that you can drain the tank with.

The first minute or so of closed loop the car may run pretty lean like 15 to 1. When I swiched to E 85 I noticed it more after a moderate shut down,like 20 mins the first 5 mins of running there would be some stumbling and an occasional stall out at a stop light. My wide bad would read 15 to sometimes 15.7 before it would balance out back to the correct 14.7 normal run.

[InAccordance]

My carbs do something like that. Like if I go to the grocery store or something for 30 mins or so, come back out and start it, runs like shit for a few minutes but is fine afterwards. Dont have any electronic controls at all so who knows lol.

[AC439]

The symptom on my car is usually no problem within the first 30 seconds (cold engine) but once pass this time period, I will feel that 2k misfire when gear shifts from 3 to 4 (automatic). The vibration will stay if I hold steady throttle, at 4th gear and around 2k rpm. Once the speed/rpm pick up, the vibration goes away. Thereafter, I can always reproduce the symptom when engine is warm. This makes me to think about O2 sensor, but most of time with a failing O2 sensor, it cause running rich. I never seen black smoke and my plugs are not dark.

Tomorrow, I am off and will check TPS voltage and the smoothness of the range. Too bad the car is not OBD1 or 2, otherwise I can connect to my laptop and read the data when engine is running. I'm hoping TPS is off range so I can calibrate it. I may also check the MAP as well but not sure if I can use a pressure gauge to check its calibration....

[AC439]

Ok, I just back probed and measured the resistance & voltage at the TPS.

I used both a digital and an analog meter to measure. First I measure the resistance range and for smooth sliding within the range between close and full throttle. With analog meter, it is fine and no intermittent or spike or anything. Then I back probed it and measure the voltage change within range using analog meter and it is smooth.

Then I connected my digital meter to measure close throttle voltage - 0.45 volt at the sensor plug (out of spec). I then move the ground connection from the TPS (still keeping center one connected to meter) to the negative battery terminal, I measured 0.49 volts (within spec). I have not have a chance to measure at ECU yet. Full throttle measured 4.53 volts.

It seems like TPS is fine and I now suspect grounding issue since I have a difference of 0.04 volt between ground point at TPS vs battery neg terminal. I may run a ground wire directly from the battery terminal to the ECU ground pin and see what will happen....

[Dr_Snooz]

I'll give you credit for persistence. LOL

[AC439]

I'll give you credit for persistence. LOL

LOL, I can't stand that little 2k misfire. Everything else I can live with !

[AC439]

I tried to take the TPS bolts off but after 2 hours I still can't so I'm giving up. The screw are not like the other typical honda TPS that you can cut a slit and unscrew it. I tried many different ways but still can't get it off. :mad:

Also when I bench tested it with a 5v regulated power supply, it does not behave as a potentiometer. Although I measured resistance and confirmed its smoothness, applying 5v to the ends and the voltage from the slider does not change with rotation. If I put the TPS connected back the car and do voltage test, it changes following rotation (but it is a little bot off now) I was like what the !@#$% ?

Then I pulled a known good TPS for a 2000 Accord and did the exact same bench test and got the same result, the slider voltage does not change with rotation but the resistance measuring does change smoothly.

This is odd. It seems like the TPS is more than just a potentiometer. Treating it like a simple pot is overly simplified. Also, the measured resistance between slider to both sides added up MORE than the pot itself. Same with the known good one. So, it seems like there is an additional resistor between the actual slider and the slider output pin.

[AC439]

So I was unable to mechanically loosen the TPS for calibration, I resorted to electrical mod with a 47 ohms resistors added to the 0 volt side.. After the mod, I was able to bring the CT voltage to 0.49V and FT at 4.6V. However, test drive is the same, the 2k vibration still there.

I also jumped wires out for easy measurement. Until I can find a way to take the two TPS bolts off, this is for now. I'm out of ideas.

[Dr_Snooz]

Just grab your 4" grinder and grind that crap clean off.

[AC439]

May ended up doing that but afraid will break the TPS. I'm taking a break from the 3G and will update here when I can make some progress.

I have been working on my 2k accord VC gasket leak today. Its a V6 so its double the amount of work. After pulling the VC off, found a snapped head bolt on the side closer to the windshield and I didn't expect that. But there seems no leak or anything for now. Installed a flea-bay IAC and idle starts surging....:hmph: driving me crazy. I'm so tired now being working 4 hours on a laptop then another 6 hours on the 2k accord.... Have to catch some rest and putting back the OEM IAC tomorrow....

[Oldblueaccord]

the 0-5 v is the reference voltage youll never get 0 and 5 volts at the TPS. its .5 idle .45 what ever at WOT. Thats really same for any TPS even my newer 6.4 hemis read this.

Now a dead spot in the TPS sure just like the throttle linkage/carb on my old IH truck its worn in the spot you use the most...right where you need it to start off clutch it and light cruise.

like my Dad would say just "drive around it" meaning stay out of 2k range and the problem would not be there.

Also I had forgotten your an Auto that opens a whole nother can of worms with stuttering etc.

[AC439]

the 0-5 v is the reference voltage youll never get 0 and 5 volts at the TPS. its .5 idle .45 what ever at WOT. Thats really same for any TPS even my newer 6.4 hemis read this.

Now a dead spot in the TPS sure just like the throttle linkage/carb on my old IH truck its worn in the spot you use the most...right where you need it to start off clutch it and light cruise.
.

I know the voltage range already. Since I can't take the TPS bolts off, I am just modifying it electrically to put it within spec. Definitely this is not the source of the problem. Also my TPS doesn't have a dead spot. I'm an EE and have checked this already with analogue meter with bench power supply.

I just realize my water pump is going as well. So one thing after another......

[InAccordance]

That's the way it goes my man. Fix one thing and another thing breaks. Just got my speedometer working again and now the ignition switch is going bad. :uh:

[AC439]

Yesterday I changed the water pump with an Aisin WPH003 from rockauto. Took me about 3 hours. Mostly due to the very limited space and also taking time to clean up mating surface etc. I also checked the coolant intake pipe and surrounding and its dry and no leaking from the seal so I am pretty sure it was the water pump. The seal on the old pump is all flattened after many years. The bottom outer part of the old pump was a little wet compare to the rest of the pump so I assume the leak was coming from the weep hole. After putting things back all together and refilling with coolant, I started the engine and saw coolant dripping down that plastic shield against the crank pulley (to the left of the car like attaching to the splash guard of the left wheel well). I was like WTF ?! I decided to take a test drive around the neighborhood anyway. After the 2 mins test, the coolant dripping stopped. I think it was probably from the coolant spilled when I removed the water pump. I did drive the car a little more today and didn't see coolant dripping anymore so I think its fixed (as least I hope).

[Dr_Snooz]

It seems like there is always some residual coolant to give you a heart attack after you work on the cooling system.

My coolant pipe leaked pretty good after my rebuild. It was corroded and iffy when I put it in and I should have replaced it, but good luck finding one. I again should have replaced it when I saw it leaking, but I had no interest in tearing back into an engine bay I'd just reassembled so carefully. Instead, I took out the air filter box and ever so carefully smeared RTV around the pipe and inlet hole on the block. Then I prayed very hard. It's been a few years, and so far, so good.

I honestly don't know what I would do to fix it properly.

[AC439]

Yeah I was almost having a heart attack....

I changed the water pump long time ago so it was way overdue. Basically I have no memory about how to change that pump. Maybe the residual coolant was spilling out just like this time? I have no memory on it.

Anyway, I am driving the car today and see no coolant leak so I assume all is well (with the cooling system, not the 2k shake).