Fuel pump won't stop priming?
 

Car is an NB1 with a DW200 which replaced the stock unit when the motor went in over the spring.

Weird thing just started happening. When I put the key to the 'On' position, I hear the fuel pump whine to prime... except it doesn't stop. This seems to mostly happen when the car is cold. Testing after a drive leads to the expected 1-2 second prime noise and then things turning off. If anything, the car seems to be running a bit richer than usual during warmup.

I'm a little unsure how to diagnose this since it happens only when cold. I'm pretty sure it's coming from the tank because I can hear it pretty well if I pop off the gas cap and listen through there. Can't really hear it once the car is on though. The whine just stops then, so it's almost like the priming pulse won't turn itself off when the key is just in the 'on' position and the car isn't running.

I don't think it's MS doing anything because I haven't touched anything on there related to the fuel pump, but I'm attaching the log just in case. Any help figuring this out?

Light green wire at the fuel pump relay is the trigger from the ecu. Can also be found at the diagnostic connector.

Measure for continuity to ground on that wire. The pump should only run when that wire shows continuity to ground. If it's running when it's not the relay is sticking.

Thanks, that's what I was thinking. I might pull the relay out to see if I can make it click or just check it for continuity with a multimeter first. That should make it easier to figure out what's up.
I'll check the fuse box for continuity as well, which is about the same thing as far as I can tell. Appreciate it!

You said it was intermittent. If it is then you can't test it outside of the car because you won't know if it's exhibiting the issue or not.

Don't make this harder than it is. When the pump is running when it shouldn't be (key on engine off) set your meter to continuity and put one lead on ground and the other lead on the fuel pump pin in the diagnostic connector under the hood.

Ahh,that makes sense.

So just to test, i pulled out the plug to the relay and put the key to the on position and got no noise. Okay, so I confirmed the fuel pump relay location. I had pulled this (and the injector fuse) out when doing a compression test a few weeks ago. Put it back in and ensured it was tight.

I connected in the diag box with the car off and it reads a high number, no noise.
With the key on engine off, reads a negative number similar to above except negative and it has the noise for continuity. However, the fuel pump has been priming and then stopping when I did this test. I'll report back with what I find when it's acting up, i.e. not turning off. I'll leave the MM in my car for now. Because science.

Any meter I've ever used will show resistance when on the continuity setting. You can't have negative resistance......

What meter are you using?

Its just a basic MM. I confirmed that it's on continuity setting. Connecting fuel pump and ground when the key is on but engine is off leads to the beeping you'd expect from a completed circuit. It doesn't make the noise with the key out. I'll see if it's any different if / when the problem returns.

Polarity shouldn't matter right because it's just testing for a complete circuit? It reads at one when the leads aren't touching anything and beeps when they're contacting one another.

Correct.

To clarify, what we're testing is whether the relay is running without the trigger from the ecu or not.

If you turn the key on and the pump continues after the initial prime then test for continuity. If you don't have continuity to ground on that wire and the pump continues to run then that would point to the relay contacts sticking in the on position.

Okay, so right now I basically confirmed the relay isn't sticking because:
a) the priming noise stops (duh)
b) when the noise isn't happening, there IS continuity using the diagnostic box.

If I start seeing a lack of continuity WITH the noise, then I can pinpoint it to the relay.

I might just order a couple of spare relays for the motor, like the main relay etc. Some necessary ones to keep in the glovebox in case something fails.

Thanks for your help man!

So the issue is happening again. When I check continuity with the pump continuously priming, it still shows that there's continuity.

What gives?

Intermittent short to ground on that wire or it's the megasquirt. Either the tune or the hardware.

I'd unplug the megasquirt when you're seeing the issue then check for continuity.

I drove the car around for a few minutes and it was back to normal when I got home. I didn't get a chance to check continuity again, but I thought I was supposed to see continuity when i wasn't hearing continuous priming?

I can order a new relay and try that. Ill also follow up and disconnect the megasquirt to test if, err when, it happens again. Other than that is there anything else I should try? I'll put the new relay in as well when I get it Any reason to pull the pump? I don't think that's where it would be having issues.

I can look up some wiring diagrams, but where's the ground for the pump? I haven't done anything with wiring lately, but couldn't hurt to try. It feels that the issue only happens when things are cold if that makes sense

Alright I'll take another crack at explaining this cause you still don't understand it.

The lt green wire at the fuel pump relay we're talking about is the wiring coming from the fuel pump output on the megasquirt. When the megasquirt grounds that wire the relay is triggered and power is provided to the fuel pump. If that wire is grounded for any other reason the same thing will happen. We are trying to determine if the megasquirt is grounding that wire of if the wire is shorting to the chassis somewhere.


You should ONLY have continuity to ground on the lt green wire when the pump is running. If the pump is running when the lt green wire shows no continuity to ground see post #2.

Oh good, you noticed that I'm a TOTAL noob at wiring haha.

I only checked continuity in the diagnostic box! Where would you ground the lt green wire? Can I pull the wire off the relay and use the MM on the prongs of the relay itself?

Sorry if this is vague as I'm operating off memory but I had a similar issue. My brain-built ms3xwas shipped with many issues and it was either Q16, Q2 or D4 in the fuel driver circuit that had failed causing this. If you already checked everything else, that's where I'd start, or figure a way to put a switch to stop the fuel pump for safety purposes.

The wire at the relay is spliced to the one at the diagnostic connector so they're electrically the same.

You need pictures and I can't make one until Wednesday. Hang tight or hope someone else has more time than I do.

I still would like to know the model of the meter you're using. It makes the spooning easier.

If the FP wire in the diag box has constant continuity to ground, the relay, and therefore the fuel pump, is going to stay constantly on. This is either because the MS for some reason is commanding the FP stay on, or there is a short to your chassis somewhere.

If the FP wire in the diag box has continuity to ground for only 3 seconds when you KOEO, then your relay is most likely sticking.

If you want to check at the relay socket, you won't hear the fuel pump obviously, but yes, you can stick your MM probe in the ground side of the relay's coil (light green) and test to see if it's staying grounded or not.

Spartan, here's the MM.

https://cimg6.ibsrv.net/gimg/www.mia...ed49c22de3.jpg


If it's MS related wouldn't it be a constant issue and not intermittent? Can I do the test you're recommending with the relay in the car?

Also, is there another compatible relay on the NB that I can swap with to test? I think the antenna or defroster relay was the same for the NA, but I don't think that's the case for the NB since I can't find part numbers that are interchangeable.

If your meter shows resistance with the probes isolated from everything, throw it away lol. I use a fluke, it shows 0 ohms when the probes are isolated and providing an open circuit.

On continuity it reads a straight 1 with nothing touching it. If I touch the two probes together, it beeps showing continuity. It reads zero pretty much on all the other settings.

In any event, if this is the relay:

https://cimg0.ibsrv.net/gimg/www.mia...b007b39402.jpg

If I have it set to continuity and I have it on the two ones across from each other on the right in the image (with the holes), I get the MM beeping, i.e. showing continuity. This confirms that it's the relay right? Since it's not supposed to show continuity with the car off (no ground?)?
After I took the plug in and out under the dash, the relay worked as it was supposed to. Pulling the plug out and retesting the prongs as above showed the same thing. Honestly, it's like anytime it acts up, it starts acting fine if I pull the plug and put it back in... If it was the MS and not the relay, wouldn't pulling the plug out do nothing?

With continuity, the number was reading in the 60's or something.

Edit: NVM. This probably actually tested nothing because the relay wasn't getting power lol. But hey, it's kinda weird that unplugging and plugging back in seems to fix it.... I should be getting a relay shortly. Will test that and see if it makes the intermittent issue go away.

Wires for the MM?

No that is the relay that powers the ecu and a whole ton of other stuff. The facturly fuel pump relay is next to the steering column. I think even the factory one of these is solid state (not certain because mine melted itself pretty shortly after upgrading the fuel pump) so it cant stick intermittently. It can fail in either energized or de-energized state though.

I've had an NA fp relay apart. It's not solid state. I don't see any reason why they would have gone solid state on a NB.

There isn't any reason, because they didn't. It's a normal relay.

Usually manufactures go solid state so they can use a PWM controlled fuel pump, which I don't think is the case on an NB.

I can take a picture of the innards of the relay. I actually got a spare from a local guy that parts cars. I couldn't get the old one out from the housing so I took of the entire plate with all the relays and ended up swapping the innards of the relay while leaving the housing where it was so I wouldn't break it. It acts like it should on a cursory test. I'll find out if it too has an issue in the next few drives I suppose. I have the old one in the glove in case something happens. Picked up a spare main relay as well because reasons.

Hm. I upgraded to a dw200. That shouldn't change anything enough to require a different relay, no?

TLDR; On this episode of monkey's writing shakespeare, I just swapped the relay. Lets see if that 'fixes' the issue, if said issue is an old relay sticking intermittently.

You wouldn't think a dw200 would be an issue considering its draws well under 10 amps and I think it's a 30 amp relay but track guys are killing the relays with stock pumps.

Definitely in for results as I chose to skip dedicated fuel pump wiring when I did my dw200.

I keep a spare relay in my glove box though.

It takes an additional 30 mins or so to wire in a dedicated relay to power the fuel pump. It will allow the pump to see more stable voltage especially after everything starts heating up. I went the lazy route and bought the DW fuel pump hardwire kit. Which honestly is a ebay/amazon relay pigtail with 12v+ lead being like 10ft long with a chinese fuse holder, and a decent looking relay. $40. Id rather not worry or question how the DW200 pump is doing so I just opt to upgrade and be done with it.

Takes a little longer than 30min if you're taking your time to make sure the wiring is run properly.

Sure it could. While doing the fuel pump, OEM harness is fully accessible and it really can't be taking much time unless you are really unorganized with your tools and supplies and count the time it takes looking for all that lol. My point was it just doesn't take much to wire one in, then you know it's good!

I just did it a few months ago on my '95. It took me longer than 30 to do it making sure all the wiring was tidied up and not going to rub. I even installed the relay right next to the ECU (which is behind the passenger seat on the 94-97) which meant I didn't have run wiring all the way up to the dash.

https://images2.imgbox.com/e2/60/wQVVFp4c_o.jpg
https://images2.imgbox.com/25/93/g2H1DALg_o.jpg
https://images2.imgbox.com/22/d1/vhNsXonB_o.jpg

Swapping the relay didn't fix it for me. It seems to 'reset' everytime the car warms up or if I unplug it and plug it back in.

Plz poast when you get a chance haha. Everything seems to be working fine, but I can definitely hear things priming indefinitely when I first turn the key. After things warm up, it clicks off audibly.

can you "force it" off by unplugging relay?

So put key to 'on' position and then unplug the relay? I'll give this a go tonight.

Is your thought that, if it keeps priming then it's elsewhere in the circuit? i.e. a short somewhere.

Interesting approach!

The pump can't get power any other way than through the relay. Pulling the relay will definitely shut off the pump but that won't tell us anything we don't already know.

That wouldn't make the pump run when it shouldn't.

The relay is just a switch that gets flipped by an electromagnet. The electromagnet is a just a coil of wire. When the coil has electricity flowing through it you get a magnetic field that physically moves an electrical contact mounted on an arm.

Wht/blu in the diagram is the 12+ feed for the coil and for the actual power feed to the fuel pump. We have power on this wire only when the key is on.

The LT GRN wire is the ground for the coil supplied by the ECU. So the pump will only run when there is power on wht/blu and ground on lt grn.

We know wht/blu is doing it's job. Since the pump is running when it's shouldn't it means LT GRN is being grounded when it shouldn't. Either the MS3 is grounding the wire when we don't want it to or that wire is making contact with ground somewhere along the wiring run.

The same LT GRN wire is also split and runs to the diagnostic connector underhood. This means we have easy access to the wire underhood for troubleshooting.

The original testing I told you to do would rule out a bad relay. Since we now know it's not the relay we need to figure out if it's a megasquirt issue or a vehicle wiring issue. Easiest way to do that would be to unplug the ecu with the key on engine off when the pump seems to be continuously running.


https://cimg2.ibsrv.net/gimg/www.mia...322b8eff2d.png

No. I've never messed with an NB but there's nothing in the diagram to support that. How long the pump primes for looks to be entirely controlled by the ECU and has to be time based.

My Hardwire Relay sits right next to the fuel pump harness grommet. No wires go up to the dash. Get's signaled from the OEM relay and draws direct from the battery. All wiring is within the parcel shelf area tied up to the factory harness. 2004 MSM

Which is connected to the LT GRN fuel pump output off the ecu. Which is triggered for a set time to prime the system.


Highlighted is your regulator. When pressure reaches the point the regulator is set for the extra fuel is dumped back in to the tank while the pump continues to run.



https://cimg3.ibsrv.net/gimg/www.mia...92b9b40e29.png

its time based. so either the relay or the ecu.
by that logic, he shouldn't be having this problem, at all, but here we are, so......




in typical fashion, this is being over-theorized and will go on for many pages lol
if you pull relay and it dies then problem is either relay or ecu. or you can pull relay and check if trigger voltage is being supplied continuously.

or just not care and hardwire it bypassing all of this

I has hardwire kit I'll sell you. :naughty:

Tell me how this fixes his problem? See my reply to brainzata. See the diagram I posted.

When you HW the pump you're still using the same LT GRN wire from the ecu and the same output from the ecu. This is where we already know he has a problem.

Just to clarify how simple this is to test. Turn the key on. If the pump continues to run with the engine off, unplug the ecu. Tell us if the pump stops when you unplug it.

That's too easy. Way too easy. We need to discuss this for a few more pages before he tries that

how should he feel when doing it? what emotions should he expect and how will the car feel?

:likecat: for you.

I'm going with timid and apprehensive for $300, Alex.

I was actually responding to midtenn but forgot to quote. I understand the pump is turned off after a few seconds of priming. Thanks for the info though

Oh crap that was my bad. I thought you were boileralum.

Pump does absolutely nothing when the ECU is unplugged. No priming, period.

Plugged everything back in and it decided to act normal... for now. I'm sure it'll start sticking again the next time I turn it on.