Need help with Fuel Pump

[Pandafish]

Hello all!  I'm having a horrendous time troubleshooting a fuel pump problem on a 2011 DT300i, perhaps someone can help.

When I turn the key, instead of hearing the happy whine/whirring of the fuel pump, I hear a buzzing or series of clicks from the fuel pump relay for 5-10 seconds (priming).  It continues this noise when trying to start the engine.

- CELP is showing Error codes 41 and 31
- Battery is 3 months new and is at 12.6v
- Starter motor sounds strong
- Replaced the fuel pump, old one tested at 2.5 ohms (bad), New one tests at 1.7 (good) - Normal is 1.9+/-0.3
- Switched out relay with others - All make the buzzing or clicking
- Relays test good when 12v applied, clicks once and I get continuity
- When I disconnect the fuel pump the relay opens and closes normally
- Fuel pump makes no noise, pumps no gas

I'm really at my wit's end here.  If anyone has any ideas of what to test or troubleshoot - I'll be happy to do so and report my findings - The scoot has her clothes off already, in the garage, from the previous work. 

Many thanks in advance.

[eamartin]

Can you bench test the fuel pump by removing it from the bike and applying power directly from the battery?

[Pandafish]

That may be difficult without accidentally shorting the battery (not enjoyable).  The harness on the pump is male, and the pins are small and tightly packed together.  I may have some R/C connectors I could repurpose.

I've been tried jumping the relay harness, and it really didn't do anything, much to my surprise.  I assumed that would make the pump run - It's possible that I am misunderstanding what the relay is for, but under my current paradigm, it doesn't make much sense.

[CROSSBOLT]

The relay is switching high current while the ECU is supplying very low current to turn on the relay. I apologize, you asked what time it was and I told ya how to build a clock! Gimme a minute and I will look up those codes on my other computer.

Karl

[Pandafish]

The relay is switching high current while the ECU is supplying very low current to turn on the relay. I apologize, you asked what time it was and I told ya how to build a clock! Gimme a minute and I will look up those codes on my other computer.

Karl

Exactly Karl, that's what I thought, because the fuel pump would take a decent bit of current you wouldn't want running through the ECU.  But jumping the 'hot' side of the relay should cause the fuel pump to run - But that doesn't happen.  I tried the old and the new fuel pump.  Nada.  Made me think maybe I didn't think that relay is serving the purpose I thought it should be.

The relay is confirmed good and functions normally when the fuel pump is disconnected.  The rapid clicking is a mystery - It makes me think the voltage to close it is intermittent from the ECU, or just too low.  The voltage measures 12v all the time across both sides of the relay harness.  I don't get it.  The signal side should only have voltage when the relay needs to be closed to run the pump.

I wish this were a mechanical issue.  I'm much better with mechanics than electricity...

[CROSSBOLT]

No code 31 listed for DT300i. Code 41 is faulty fuel pump but then list ohm values for E &F suggesting fuel level sensor. Well, both pump and sensor are in the same assembly. There are four wires out of the fuel pump, two for fuel level and two for the pump. Since there seems to be no difference in fuel pumps, old and new (except for ohms) the problem is elsewhere. Look for shorts or opens in the harness. So you have to look under the front cover where the relays and the ECU are all the way down to the fuel pump/sender on top of the fuel tank.

Karl

[Pandafish]

The ohms on the fuel sensor check out, and it is reading the fuel level sensor accurately.  I also played around with the float when I had the old one hooked up for testing.  I did find there is a reporting delay of 5-10 seconds from the float to the gauge itself, it's probably so the readings aren't all over the place when fuel sloshes around.

I can trace the wires - It's actually not a long run from the front to the tank, and reseat any connectors I find along the way.  I'll also do a continuity check from the fuel pump relay harness to the fuel pump harness - I'll let you know what I find.  I think I may also do the ECM continuity tests listed in the service manual, in the same section.  As far as I can tell, the ECM is working just dandy, but it's quick, easy, and rules out another cause.

[CROSSBOLT]

Well, as usual, I got part of it wrong! There are four wires but not all together as I implied (that gets me sorta off the hook!). There are two to the fuel pump, red with black tracer or stripe and green for ground. Two for the sensor or fuel unit as they call it, yellow with green tracer and lavender with white tracer. Blue with white? Anyway, there are two separate plugs but it seems you have all that figured out! Yes, the delay cuts the fluctuations of sloshing fuel. While you are re-seating connectors you may want to smear a little dielectric grease on the pins and holes to prevent any further corrosion. Ringing out the ECU is a good idea. You will find it since you have a good handle on eliminating what the problem isn't!

Karl

[Pandafish]

I traced the wires from the relay all the way down to the pump, and really every other component/harness I could get my hands on.  This scoot is very pampered, so unsurprisingly the harnesses all looked clean when I disconnected them.  I did use a dab of dielectric grease, because why not, while I am already in there. 

I did find the small trunk of wires splits off to the fuel pump.  Wired into that is the harness for the side-stand engine cutoff.  The was one open harness (Male with two dark green wires) next to it that had no mate.  I did find that if I disconnected the harness for the side stand sensor (female) and mated them up, it effectively bypassed the side stand sensor.  I'm not 100% sure that is the purpose of that harness, but it works like that.

I checked all the fuses, they were good.  ECM checks out with the pin continuity tests listed in the manual.

I checked continuity between the relay harness on the hot side (double red), all the way down to the fuel pimp harness (Red), and it good.  So if the relay would close, we'd be gin.

Now, the relay closes and opens normal, If I disconnect the fuel pump. It 'chatters' when connected, instead.  Why would this happen?  The same power is being applied on the signal side of the relay regardless.  And the hot side has nothing to do with the relay opening or closing.  Argh!

If the battery has enough power to turn the starter, it should have zero problem with a piddly fuel pump motor. 

EDIT:  Also, took voltage across the fuel pump power part of the fuel pump harness.  It is getting 11.9v - When the relay is closed for 5-10 seconds after key turn (No relay chatter). 

I also jumped the hot side of the fuel pump relay harness - Nothing happened.

Any ideas from here?

[Pandafish]

Finally broke down and took it to the dealer - I'll let you know the outcome.

[CROSSBOLT]

I will be pleasantly surprised if the dealer's techs can figure it out!

Karl

[Pandafish]

I got a call from the dealer yesterday (really solid, no BS bike shop).  No pleasant surprises for us, the dealer's techs are absolutely befuddled by this.  He said they continuity tested tested every pinout, checked all the voltages, did everything in the service manual, and everything checks out 100% perfect.

In a fit of frustration they removed the fuel pump, and reinstalled it.  Now it works...  sometimes.     He had me on the phone and it would start randomly on various key turns.  Sometimes 3 or 6 times in a row, sometimes it just got stubborn for a spell.

My dealer said his techs are out of runway on reasons and solutions - The next step would be to replace the entire wiring, hoping to isolate and eliminate a grounding issue that may or may not be present.  That's where we start running into some serious $$$$$.

I'm going to give him a call on Saturday, see if the DT300 is having a 'good day', put on her clothes, roll the dice and ride her back home.  I'll probably stay off the highway, as having the fuel pump cut out there wouldn't be a super fun time. 

While I'm kinda happy it wasn't some obvious issue where I'd have to say 'derp' and swallow my pride, I'm now stuck with a ride with questionable reliability.  Getting randomly stranded is no fun at all. 

So the saga continues - If anyone has any ideas, please let me know.

[CROSSBOLT]

Not the fuel pump, Panda. Possible culprits: (in my opinion) contacts in FP relay funky, intermittent; contact funky at ECU; funky contact in any terminal crimped on the wires for this circuit. I spout all this because it is intermittent and is still there. Sounds like a two-man job:one operating the key and one wiggling wires!

The ECU should be good since it does the self-test and does its job as long as the pump runs. This is about all I can think of. Where are you located?

Karl

[Pandafish]

I FIXED IT!!!!!   

Karl, that was solid advice and right on the money.  After towing the poor thing back home, I finally got a free weekend, I got mad, and I got down to wire wigglin'.  It was a two-man job, luckily I had sired a son many years ago.   

At first, I thought it was the harness that went into the fuel pump, as wiggling there seemed to improve the odds a slight amount.  I took the harness apart as much as I could...  and it looked pretty good.  Hrm...

Since it is right there, I also checked out the side stand interlock harness.  I had no reason to suspect it, but it looked less impressive than the fuel pump one.  I used a TINY screw driver to make the female harness a little tighter (insert joke here) AMIRITE GUYS!?!   

It worked properly 80% of the time!   PROGRESS!!!

Tightened up the female connectors a little more and applied large amounts of dialectic grease to the harness, and every other harness involved - Including the ECM harness.

100% RELIABLITY!  I cleared the code and rode all over the neighborhood this weekend, tuning the scoot on and off at every parking lot.  Probably into triple digits of key turns - Solid as a rock again.

Thanks, Karl - I truly appreciate your advice and help.  I hope this helps someone else someday.

[CROSSBOLT]

Ah, crap! You mean to tell me it was the sidestand switch causing all this? Intermittent switch contact? Good show finding it! Yeah I will remember this one!

Karl