Replaced the fuel pump for $60

[jtraceS2R]

Hi guys (and gals). N00b here. Could use a lil' help with some electrical gremlin extermination!


So the bike is a 2000 M900i.e.. It's new to me. Ran well when I picked it up from the PO. Sat for three years after a scary high-side that left some bent levers and tank rash, but no structural damage. It's got a brand new Shorai $200 battery. Here's the train of troubleshooting:
The bike cranks like a champ, but shows no sign of turning over.
Narrowed it down to fuel delivery, as both plugs spark against the engine head.
Sure enough, the pump doesn't prime when the key is switched on. (Didn't pick up on this immediately, as I only cranked it 2-3 times before things went south.)
The wiring for the pump routes through the fuel level sensor, and the low fuel light on the panel lights up, despite the full tank.
The pump worked when I just connected it to the battery.

"AHA!" I thought, "I'll replace the sensor!". And so began my search for a fuel level sensor on the ebays. As it would turn out, there was one that some parts place in CA had up for sale for $60. It came inside a somewhat beat up steel S2R tank. The $60 tank looked better than my current tank, so I though this was a surefire 2 birds, one $60 stone. Sadly however, the same problems persist even with the new tank, as I did not think to trace back down into the connector before swapping the level sensor. The ol' multimeter says one or two of the pins are dead, and my official wiring diagram di Italia doesn't show that part of the system very well. Both the new pump and the old pump spin when I just connect them to the battery. Does the pump have some failsafe that keeps it from running when the level sensor registers "dry" so that the motor wouldn't burn up without liquid in it? I traced one of the wires back to a yellow box under the seat with "Guillera" written on it. The connections on it looked a little toasted, so I cleaned them up and dialectric greased them. Also to no avail. It was at this point that I tore off my headlamp, cut the lights and stormed out of the garage.

Any thoughts? I'm at my wit's end here and am one kidney shy of being able to afford taking it to the Duc dealer to have them troubleshoot. All words of wisdom would of course be greatly appreciated!

[Speeddog]

That 'yellow box' is the relay that controls the power to the ECU and fuel pump.

See if your local old-school auto parts store can cross reference a replacement.

That should take care of it.

[Buckethead]

It was at this point that I tore off my headlamp, cut the lights and stormed out of the garage.

Well there's your mistake.

Everyone knows you have to throw at least one wrench if you want to actually fix something.

[brad black]

the fuel level sensor has nothing to do with the pump or power thereof.  it's a completely separate circuit.  terminals 1 and 2 at the tank connector are the pump.  3 and 4 the sensor.  no connection between them.

the pump is powered by the ecu via a relay, which should be under the tank, just in front of the battery on the lh side.  a large, hooded, sealed relay.  check for power into and out of that.  the same relay powers the coils and injectors, so if you have spark, it must be ok.

there isn't anything between the relay and fuel tank component wise.  so i'd be checking for power through the circuit and finding out where it stops.  a test light is best for this.  is there voltage in the red wire (terminal 1) in the 4 pin connector at the tank?  if so, there's every chance there's a break in the wire were it goes through the fuel level sensor.

you just need to follow the circuit.  don't buy anything else until you've found a problem.

[jtraceS2R]

Much thanks to all! This is my second monster. I've had the other (carburated) one for ten years come August, and it's been the best bike I could ask for. Tires. Oil change. Tires. Oil change. Valve check, etc.. Nothing major. So, despite thousands of miles in the saddle, I have almost no experience with the nebulous multimeter. Is there any resource the DMF might recommend, online or analog, that would dispel its voodoo magic? 

[oldndumb]

http://lmgtfy.com/?q=how+to+use+a+multimeter

[jtraceS2R]

Big thanks to all for the direction. Been wicked busy. The first time I've had to be in the garage at all will be tomorrow night. I'll give it all a whirl. And thanks for the intro to lmgtfy.com, oldndumb! I'll be using that quite a lot as well.

[jtraceS2R]

Wow. What a crazy couple weeks. So tough to stifle the urge to get in the garage with weather like this, but sometimes life gets in the way!

Progress-
So I looked into the hooded relay next to the ECU. The original unit looked fine but I tried to troubleshoot with it removed per suggestion. I have the fuel pump out of the tank and hooked up to the correct pins through the connector near the rear shock via some alligator clips. With the relay disconnected and the switch on, I jumped around from pin to pin. I found that I could indeed cross two pins there and get the pump to run. So I thought... relay, right? There's the continuity break. After installing a new (thankfully $30) relay, still same symptoms. Since I can get the pump to work from as far back as the relay, I suppose the ECU isn't supplying the correct voltage through the correct pins at said relay, and that it's not a break in the wiring between the relay and pump. Thoughts? The multimeter reads -.06 when I hook everything up like it should be. Anyone know how to troubleshoot an ECU unit? They are not cheap. A quick Google search found one for $1000 on the ebays. It was new OEM, but I can't imagine I could get one for sub $500. I also tried to jump around in the pins of the yellow Guillera relay under the seat. No dice.

The only other bit of troubleshooting intel is that there was this crazy gas overflow from the filler cap. A broken vent line went completely awry into total shit-show one night. There was a 2' radius of gas under the bike when I came into the garage the next day. Gas was covering the whole motor and had gotten into the airbox a bit (which drained through the appropriate holes). Though smelly, gas is not known to entice electrical gremlins. I feel this is pretty extraneous, but it happened between the last time the bike ran and the first time it didn't.

Any help would of course be elementally appreciated. The local shop is $100/hr!

[Howie]

You cannot trouble shoot the computer, but you can eliminate it as a cause.  The ignition/fuel relay is powered by the computer, purple/black wire from pin 15 on the computer triggers the relay.  The relay is a standard relay, terminal 30 is power from the fuse box, 87 supplies power to ignition and the fuel pump, 86 is ground and 85 is the trigger.  85 is energized, when the key is turned on.  Power goes through a coil to ground creating a magnetic field..  The magnetic field closes a set of contact points allowing power to to go to ignition and the fuel pump  Here is a diagram  http://www.bing.com/images/search?q=bosch+relay+wiring+diagram+5+pole&id=A3A5478FBDCE6EF41C82C57CB18D3C85EF5462A8&FORM=IQFRBA#view=detail&id=B6FDD2DAC9A1478143F314EAF4C286F6EEF29F90&selectedIndex=1   Substitute fuel pump for horns and computer output for switch.  You could have removed the relay, applied battery positive to 85, 86 to battery ground.  Then used your ohmmeter to see if you have continuity between 30 and 87.  No continuity?  Bad relay.  $30 saved!  Randomly probing on circuit controlled by computer without knowledge is also not a good idea.

No bike for 3+ weeks, $60 for a sender, $30 for a relay?  Maybe $100 an hour ain't so bad?

[jtraceS2R]

Thanks Howie. I'll poke around again this Sunday and still try to keep from throwing things. Can't wait to get this thing up and running. Thankfully I have another couple machines to tool around on but I really want to get this one in particular out on the track this summer, which is 1/3 over already!