M6 Fuel Pump not Running - No Power / No Fuel / No Start

[PaulL]

I have a 2022 Response TXi. Has been running well, went for 50 hour service last week. We've been out twice since with no issues.

Today on our first run the engine cut out. Cranked but wouldn't start. After a short while, cranked and started. Seemed to run fine. Skied a run. Switched skier, on idle ran rough then cut out. Cranked but no start, occasional kick or fire but run rough and immediately stop. Towed home.

Talked to Malibu dealer, went through a variety of options. Pulled fuel filter, no water (good news), everything seemed clean. But after that the fuel filter didn't refill, which pointed to fuel pump. Engine diagnostics also reported basically no fuel pressure.

Pulled fuel pump(s), both pumps run when given 12v directly.

Fuses to both fuel pumps look fine, and have 12v on them.

Swapped relays, no difference.

Presumption is:

1. Fuel pumps are fine

2. Relays are probably fine on a near new boat with 50 hours, unlikely both would be broken

3. Something is leading the computer to not turn on the fuel pump, which in turn means it has no fuel and no fire.

Coincidentally, I also have a bilge pump problem today (was running fine yesterday). The screen reports that the auto bilge is running, but it isn't. When I manually turn it on, it also doesn't run. However, if I turn off the battery cutoff, and use the knob to lift the float, it then runs. When the cut-out is turned on, lifting the float does not cause the bilge to run. My presumption is that when the bilge is just using the float it runs fine (I can hear it pump). But when the ECM is turned on, it's somehow intercepting the float signal and/or not providing power to the bilge, for reasons of it's own.

This makes me suspicious that there's a fuse or a control circuit that isn't activating, and that is preventing both the fuel pump and the bilge pump from running. Having said that, I don't see a bilge fuse anywhere, and the fuel pump fuses are clearly on top of the engine....I can't see that there'd be another fuel pump fuse somewhere else. And I have no idea why something would break in a way that means the fuel pump and bilge pump don't go properly, but things like cranking the engine and the fuel rail pump is all working.

Questions:

1. Has anyone experienced anything like this, and have any advice?

2. What is the trigger that would normally cause the fuel pump in the tank to be activated? I would have assumed having no fuel pressure would cause it to prime (and the engine diagnostics reports no fuel pressure), but perhaps there's something more complicated needed? 

3. A thought that occurred to me in writing this - perhaps I should manually fill the fuel filter. That might give enough for the fuel rail pump to pick it up and pull through - although it was full when I took it off to check the filter, so if that was enough to get it running it would have been running then. Any thoughts on whether this would/could be a useful step?

Meantime, I've entirely disconnected the battery, on the logic that it might cause the computer to reset overnight (and I have the battery on charge, as I've been cranking a bit trying to get it to start). 

[tvano]

you said "Fuses to both fuel pumps look fine, and have 12v on them."

petty point: look fine or did you meter them?  

recommend getting the meter out if you didn't.

[justgary]

The ECM should prime the fuel for a few seconds at key on.  If that is not happening, you most likely do have an electrical problem.  I would use a meter or test lamp as @tvano suggests.  Resist the urge to probe through wire insulation or to deform connector sockets by using too much force.

I would also check the engine harness connector and the engine ground lugs at the rear of the heads.  By check I mean remove, clean, inspect, and replace.  The problems you mention can be caused by poor grounding.

[rennis]

I have nothing constructive to add here, but holy s*** that boat must be a rocket with the M6!  bravo! 

[PaulL]

3 hours ago, tvano said:

you said "Fuses to both fuel pumps look fine, and have 12v on them."

petty point: look fine or did you meter them?  

recommend getting the meter out if you didn't.

Fair point. I'm pretty sure they're fine but I will do that today. It's a near new boat, I don't really see why it'd blow both fuses.

2 hours ago, justgary said:

The ECM should prime the fuel for a few seconds at key on.  If that is not happening, you most likely do have an electrical problem.  I would use a meter or test lamp as @tvano suggests.  Resist the urge to probe through wire insulation or to deform connector sockets by using too much force.

I would also check the engine harness connector and the engine ground lugs at the rear of the heads.  By check I mean remove, clean, inspect, and replace.  The problems you mention can be caused by poor grounding.

It primes the fuel rails at key on, I can hear a slight whine from that pump. The fuel pump in the tank though, nothing.

Rear means rear on a v-drive install, rear on a direct drive install, or rear meaning near the transmission? I'll have a look this morning and see if I can work it out - the internet didn't have any information to offer me on this. :-) 

1 hour ago, rennis said:

I have nothing constructive to add here, but holy s*** that boat must be a rocket with the M6!  bravo! 

Yes and no. I was debating which engine. My brother finally said "when you have a choice of engines you always get the big one." That sounded like good advice. We probably would have been fine with the M5. We put the speed prop on, but even so we run into the rev limiter on about 2/3 throttle. We had a 2003 Response LX previously with the Mercruiser 330HP and a 4 bladed stainless prop, I think that had about the same hole shot we have now. Having said that, it's pretty effortless and water sport speeds, figure we're not working the motor very hard. Should go for decades.

[PaulL]

4 hours ago, tvano said:

you said "Fuses to both fuel pumps look fine, and have 12v on them."

petty point: look fine or did you meter them?  

recommend getting the meter out if you didn't.

OK, metered both. No issues. I think the relays are the issue - either the relay isn't turning on, or the relay broken (much less likely - I swapped them, there's two of them etc etc).

3 hours ago, justgary said:

The ECM should prime the fuel for a few seconds at key on.  If that is not happening, you most likely do have an electrical problem.  I would use a meter or test lamp as @tvano suggests.  Resist the urge to probe through wire insulation or to deform connector sockets by using too much force.

I would also check the engine harness connector and the engine ground lugs at the rear of the heads.  By check I mean remove, clean, inspect, and replace.  The problems you mention can be caused by poor grounding.

I'm not sure where to trace the electricity too. I know it must come to the relay, and I can't really see power on there. I guess I need a wiring diagram for the relay to know which side should be live. If I pull off the surround I think I can get a meter on to most of the sockets without forcing. If I know which is the on that's supposed to be live / the main circuit, I can use wire to jump across it and see if the pump runs then. That would localise the problem to being upstream of the relay.

My gut's telling me it's the ECM misbehaving. But my gut won't get Malibu to replace it. :-) So I need some diagnostic that tells me that. I guess that means a wiring diagram for the output of the ECM, and I could test there? I imagined it would be on a bus these days, but perhaps there's a packet of wires coming out I can test somehow.

[justgary]

The diagram for a relay should be stamped on the case of the relay.  It might be underneath, but it should be on it somewhere.  Also, it is very common for relays to have a plastic cap that has small tabs holding it on.  You might be able to remove the cap and probe directly inside on the terminals.

In the case of the relay I show below, pins 85 and 86 are the relay control coil.  The ECM puts 12v across those to close the relay contacts.  Pin 30 will probably have 12v on it, most likely through a fuse or breaker.  Pin 87a is the normally open connection, and should be unused for a fuel pump.  Pin 87 should power the pump (via pin 30) when the relay closes.

[PaulL]

OK. A lot more data but no more information. So I'm mostly writing this to help me form my thoughts - I think this is going to the shop.

1. The relays are a Panasonic/NAiS ACM13221, and of the form CMI-R-12V. The ACM13221 means 1=Form C, 3=Sealed, 2=Plug-in, 2=with resistor inside, 1=12V. The data sheet is here: https://www3.panasonic.biz/ac/e_download/control/relay/vehicle/catalog/mech_eng_cm.pdf, an equivalent diagram is here:

2. When the ignition is on, there is 12v on both socket 87 and socket 30 (with the relay removed). I have pictures of this, because it's hard to believe, but I can't see how to actually post the pictures here. :-(. I pulled both fuel pump relays in case one pump relay was somehow energising the other, but looks to not be the case.

3. When I pull the fuses for the fuel pump, that should result in there being no power on socket 30. Sure enough, pulling the fuses means no power on pin 30, but still power on socket 87. Pictures of this also....

4. This must mean the power is coming from the fuel pump somewhere - i.e. the fuel pump has some sort of short or is receiving power. So I unplugged the fuel pump plug at the fuel tank. At that point the 12v on pin 87 disappears - so that 12v is associated with the pump being plugged in. This probably means that the pump negative wire has the 12v on it, but arguably the ECM could be noticing the pump is there and then putting 12v on it somehow...no idea how.

5. Then I looked for 12v anywhere in the fuel pump plug, checking each of the sockets. Nothing is coming from the wiring. It's hard to check the pump side, the angle of the socket means I'd need to take the pump out (again), and that wasn't a fun job.  (Edit, come back to this at point 9. My problem was using the pump earth as my negative)

6. I checked for voltage from the pump ground (labelled as such on the wire) to the nearest metal element - I have 12v. That's not really good - that means I have 12v somewhere inside the fuel tank I think.

7. Unplug the tank ground wire. The 12v isn't coming from the pump, it's coming from the pump ground wire.

8. Unplug the tank ground wire. Check pin 87 on the relay again. Still has 12v on it. So the 12v is on the pump ground wire, but it's also presumably on the negative terminal of the pump plug.

9. Unplug the pump plug, check for 12v anywhere, but this time checking to a known -ve/earth on the engine. Remember the relays are removed. There should be no power on the pump. Two of the 4 terminals still have 12v on them - presumably the negative terminals. This didn't show up last time because I used the pump earth for my -ve, and I've already worked out that had 12v on it.

Upshot of all that. I think my fuel pumps aren't working because both the +ve and -ve terminals have 12v on them, so there's no voltage differential. The pump ground also has 12v, and although I haven't pulled the bilge pump, I'd hypothesise that it probably also has 12v on the negative terminal which is why the computer thinks it's running, and why the float doesn't make it run. It also explains why the bilge pump runs when the battery cutout is disconnected - it now has 12v from the float and the hard wired battery lead, and the -ve no longer has 12v on it from whatever short/switch is causing that.

In turn, I'd expect that to be a short somewhere on the negative circuit, somewhere that's in common between those two elements. Weirdly, they're probably the two that aren't just earthed to the motor like everything else is. The motor earth (-ve) can't be charged because nothing would go, and because all these measurements were earthed off the motor. So, my diagnosis would be that I have a short on the negative terminal of things that:

a) aren't connected to motor earth (so all the other engine related stuff seems fine)

b) aren't part of the house wiring - all the lights, dash, seat heater, internal heater etc are working

c) for the fuel pumps specifically, it's impacting both the fuel pump earth as well as the fuel pump -ve, presumably those both run to the same or a similar place

I was wondering if the ECM could be somehow doing this, but thinking about it I don't think it can - all those negatives couldn't be running through the ECM, that's why the fuel pumps have relays, to avoid that much current running through the computer.

Does anyone have any insight into where those -ves would run to? They can't go directly to the battery, so they must go to a circuit somewhere.

It's also very weird to me that a short circuit of that type wouldn't either short the battery and drain it, or spark and carry on. I guess it may have just worked out that way because of dumb luck, but that particular element doesn't quite add up for me.

A different option here (given the boat's recently been in for service) is that someone has unplugged something and plugged it back in backwards, rather than a short. That could explain 12v on the negative lead without the battery going flat. Wouldn't explain why the boat ran for 2 days, and I think the bilge pump was working OK. I have a vague theory that the rail fuel pump may have enough pull to suck all the way from the tank, and then something happened to break the suction so it was sucking air - so in that case it perhaps could have been running without the tank fuel pump for a couple days, but that does require a bit of magic to be true.

 

Edited February 18, 2023 by PaulL

[justgary]

Back up a few steps and tell us about pins 85 and 86 on the relays.

[PaulL]

7 hours ago, justgary said:

The ECM should prime the fuel for a few seconds at key on.  If that is not happening, you most likely do have an electrical problem.  I would use a meter or test lamp as @tvano suggests.  Resist the urge to probe through wire insulation or to deform connector sockets by using too much force.

I would also check the engine harness connector and the engine ground lugs at the rear of the heads.  By check I mean remove, clean, inspect, and replace.  The problems you mention can be caused by poor grounding.

I went through everything I know, and despite having checked continuity on the ground, I decided a poor ground could be causing a bunch of elements to lift to the battery voltage - I don't really see how that happens when the relay is disconnected, but maybe some other element that is poorly earthed is lifting the whole engine to closer to 12v. I checked a few other relays, and the rail pump had about 4v on pin 87, one of the others had about 3v. Seems a lot like a poor earth.

Looking at the earth more closely, the bolt head was scuffed. On a near new boat, that means someone has had it off or at least tightened it (every other bolt is pristine and painted). I took it off, light sand with wet and dry sandpaper on the housing, then each of the 3 wire blades that connect (including the solid copper one that is the engine earth). Reassembled, no difference. I still feel like it's an earth problem of some sort, but not sure. 

I also followed the leads from the fuel pump to source - they join into a large loom that's going into the ECM on top of the engine. I removed that to look at it, but there's no user serviceable parts I can see - just a big loom plugged into a computer, and a big heat sink. I'm guessing the negative leads and the ground all flow into that unit, and are managed within there. Perhaps replacement of the ECM is the answer, or at least something to try.

There is a connector in the fuel pump harness, where it joins onto the larger loom. I've detached that and checked there, again with no fuses in the fuel pump I get 12v on two of the lines (presumably the negative lines).

Again, I think I'm at the limit of what I can do with it - it's pointing to a short or a wiring problem in the ECM, and that requires tools and parts I don't have to work on.

[justgary]

The ECM does not power the fuel pumps directly; that's what the relays are for.  The ECM most likely provides a ground for the relay coils via an open collector circuit.

Did you verify that pin 30 goes to the fuel pump?  Malibu is very good about using standard colors in their harnesses.

[PaulL]

1 minute ago, justgary said:

Back up a few steps and tell us about pins 85 and 86 on the relays.

One shows continuity to ground. The other (I think 86) I just rechecked, and gets voltage for a few seconds on power up, then the voltage goes off. I'm assuming the ECM is turning on the pump for a few seconds, then waiting for engine start before turning it on again. So far as I can tell this doesn't result in the pump running - I can't hear it running, and there's no fuel coming into the fuel filter, and no pressure when I push the Schrader valve, nor does the engine diagnostics report any pressure.

[PaulL]

10 minutes ago, justgary said:

Did you verify that pin 30 goes to the fuel pump?  Malibu is very good about using standard colors in their harnesses.

I haven't directly verified it, but when I unplug the fuel pump lead at the tank the voltage disappears. That's not exactly proof, I'm not sure my multimeter leads are long enough to test continuity from the pin 30 all the way to the fuel pump lead. I could test on the harness where there's a connector (near the fuel filter), but I'm pretty confident that the fact the voltage goes away when I unplug it means it's running to that plug.

The wires on the M6Di are all wrapped with black mesh - I can't see the colour on any of the wires anywhere. Which is nice and tidy, but not ideal for checking what goes where. :-)  Maybe I could get the missus to hold one end of the multimeter on the relay, and I'll see if I can stretch the other to the fuel tank.

Edited February 18, 2023 by PaulL
Add more info.

[PaulL]

20 minutes ago, justgary said:

The ECM does not power the fuel pumps directly; that's what the relays are for.  The ECM most likely provides a ground for the relay coils via an open collector circuit.

That's my expectation, and yes, that's why there's a relay. Having said that, the wires all seem to run to that same unit, and it has a set of monitoring that can detect open circuits on the relays (know this because when I pulled the fuel rail pump relay the computer complained of an open circuit), which may mean it's doing more than just terminating all the negatives on a common. Probably above my pay grade, and either way, the negative seems to run to the ECM, and therefore whether it's a common collector or something different, the problem is probably inside that unit.

Although, thinking a bit harder, when I cleaned the earth on the engine, there were three wires/blades that went to there. One was clearly from the battery. Presumably one of the others is the ground from the ECM. One had a little bit of rubber or silicone on it where the blade connected to the wire, but I don't think enough to stop a connection. I sanded the body of the blade itself, and tightened it well.

Anyway, into the shop for proper diagnostics tomorrow. I just feel like it may stay there for a week or more, which annoys me. Peak water ski season here.

Edited February 18, 2023 by PaulL

[justgary]

So, with the relays in place one at a time, can you hear/feel them click when the key is turned on?  If so, the ECM is trying to run the pumps.

I would next try to jump pin 30 to +12v with a wire to the starter B+ lug (with the relay removed).  That should run the pump motor.  Try both of them.

[PaulL]

1 hour ago, justgary said:

So, with the relays in place one at a time, can you hear/feel them click when the key is turned on?  If so, the ECM is trying to run the pumps.

I would next try to jump pin 30 to +12v with a wire to the starter B+ lug (with the relay removed).  That should run the pump motor.  Try both of them.

The relay appears to click. I wasn't sure, so I had to remove all the others, and it's definitely that one clicking (along with others) - I can feel it. The pump doesn't run.

Jumping across from pin 30 to pin 87 also doesn't run the pump. I don't have a wire suitable to jump to the starter, but I have 12v in the relay, it's not exactly the same but is something I can do with the tools I have. It makes sense to me that it doesn't run if there's already 12v on the negative, although that could be a "weak" 12v if the ground is poor. I guess bridging pin 30 and 87 is just the same as the relay turning on, and we can hear it click so we already knew that didn't run the pump.

The pump does run when outside the fuel tank and directly powered with 12v. So I know it's not the pump broken. The last test would be to supply the pump with a new -ve, but the only way I have to do that is to unplug the harness again - and then I'm basically doing the same thing as when the pump was out of the tank. I guess the improvement is I can see the fuel coming into the filter, and see the fuel pressure on the diagnostics. I'll think on whether I have a way to do that that won't wreck something - I'm powering pins then instead of a socket, so high chance of managing to "cross the streams" and blow a fuse somewhere. I'll see what wiring I have that I could use perhaps. I think I have a cable with spring clips on either end I could press into service.

[Steve B.]

Sure sounds fuel isn't making it to the 2nd pump for whatever reason. Please have a good fire extinguisher close by and hopefully a 2nd person while dealing with fuel delivery.

Guess I'd start at the beginning and confirm fuel is coming out of that tank pump. However you have to rig it up. Then confirm it makes it to the 2nd pump.

Keep us posted.

Steve B.

[PaulL]

Yeah, it's going in for warranty. I got to the point where I was about to jump the fuel pump directly off the starter motor live, and then thought "hmm, no fuse, fiddling with two wires poked onto the pins of a plug I just disconnected, known 12v on the fuel tank ground.....how many things can go wrong here? I'm pretty sure whatever information I got would just confirm it needs to go into the shop. It can go into the shop tomorrow.

[Steve B.]

Good call ! Let the pro's do their thing. Some tasks are well worth it and this is one of them. Good job diagnosing.

[PaulL]

Further thinking to add (but I'm not touching the boat any more). I think the lack of a proper ground on some things is the right diagnosis.
1. The continuity tester showed connection from the battery negative to the engine block, and zero voltage differential. I think the engine block is properly grounded

2. There was a voltage differential between the engine block and the "fuel pump ground" cable, and no continuity. That means that somewhere between the fuel pump connection to ground, and the actual ground on the engine block, there's a break

3. If this was some sort of common negative/common ground, used for all the things that aren't actually bolted to the engine block, and if that common ground had maybe a break in the harness or something unplugged, then if any of the things using that "ground" had power to them, that "ground" would actually be lifted to 12v. No current flow, but voltage would be lifted

4. That would explain pretty much all my symptoms, whereas a short wouldn't - if there was a short of some sort that was putting power on the ground then it would drain the battery / blow fuses etc. The fact that's not happening says it's more a floating/disconnected ground for those components, rather than a short

So I think that's what I'm telling the boat shop - looks like a ground problem on at least the bilge pump and the fuel pump, perhaps more things as well. And explain to them the diagnostics that suggest that. Hopefully it's as simple as a cracked wire in the harness, or a poorly seated connection on the ECM. It could be the lead on the engine earth - there's three leads on there, one of which just looked a bit different than the others, it may have been poorly terminated.

Edited February 19, 2023 by PaulL

[PaulL]

Good resolution today. Took it to the shop, and despite no booking, they did it within the day. Which is good, since it looks flat tomorrow morning.

There's a second earth on the other side of the motor, low down, with only a couple of things earthed there. It was fully loose - about 3-4mm of thread, the nut not tightened at all on the stud. Not sure what else that nut was holding together, only had a quick look when I picked it up. The guy doing the work only found it by accident, he was working through the wiring checking every connector was tight, and suddenly the fuel pump ran. He'd checked a connector near that earth, and bumped it. 

Weird that they'd have two separate earth locations on the same engine, when all the wires are running through a wiring harness anyway - wouldn't you just run them to the same place? Anyway, seems they don't.

[Steve B.]

WOW !

Glad you got it. You thought it through and knew the correct diagnosis, it was only the location that you lacked. Wonder where the other ground leads go to? For now, just ski !

Steve B.

[BuFootin]

Can you give more of a description on the location of the second earth ground please?  I have a 2022 TXi and wanna check it out.

great diagnosis btw👍

[PaulL]

1 hour ago, BuFootin said:

Can you give more of a description on the location of the second earth ground please?  I have a 2022 TXi and wanna check it out.

great diagnosis btw👍

The main earth lead is sort of behind the fuel filter - starboard side to the stern.

This one was basically in the same place on the port side. Maybe a little further towards the bow. I'll look tomorrow and give a better description though.

[BuFootin]

Great!  Thanks 🙏🏻