1999 EV 11 Fuel Pump/ECU Problem

[missinglink]

Okay, here goes my tale of woe. 
The fuel pump died on my 1999 EV11. Clicked on the key.....no fuel pump buzz.  Checked the fuses and relays, all good. Pump was getting power but wouldn't run.  Pulled the pump checked it out and it was dead. Installed a new pump, the bike cranked up and ran just fine... I put everything back together and called it a day.

Next day, clicked on the key....no fuel pump buzz.... bummer.  Went thru the exercise again, now the pump wasn't getting power... I finally figured out that the pump relay was not getting it's ground thru the computer (pin 28).  I jumper-ed the relay to ground (bypassing the computer pin 28), pump ran, bike started, everything was good.  Took off my jumper, clicked the key on and everything still worked...WTH???? Next day, same thing all over again. It seems that if I get it to work once then it's fine until the next day.  I checked the ECU connector, cleaned all the contacts, tested for a good ground on pins 1 & 19 thru the cable, I peeked into the ECU - nothing was obviously fried.  Everything looks good but obviously something is not right.  I hate intermittent problems cause you know its gonna fail at the worst possible time.

Now, I can rig my bypass jumper to make sure the bike will run and not leave me walking someplace but I am really curious what signal the computer needs to make that ground on the initial key on.  Once the bike starts and is running the pump runs continuously using that pin 28 ground.  The coils are powered thru that same relay so it needs to stay energized to power the coils.  Does anyone have a schematic of the internals of the ECU and/or knows what it needs for that initial power up of the fuel pump relay?  It almost seems like that relay logic is not really needed and is only a point of failure waiting to happen. 

Anyone have any info on this?

Thanks.

[LowRyter]

mine did that but it was fuse, not the relay.  The fuse had minor corrosion.  I replaced the fuse and cleaned the  connector.    Also the fuse block itself had come loose. 

Happened again just last week and it was the relay that needed cleaning and again the fuseblock had come loose. 

[oldbike54]

 Just to confuse things , the wire that powers the pump on my Jackal doesn't show power , but the pump runs just fine . Been trying to figure that one out for weeks after replacing the pump 

 Dusty

[fotoguzzi]

Original Siemens relays? start by replacing them.. get some at DPGuzzi.com

how many miles on your bike?

[missinglink]

The relays are original but they seem fine.  As soon as I put a ground on the pump relay, as the ECU should be doing, it energizes and sends power to the fuel pump.  I have 65k on the clock and have never had a problem before. I was just wondering what tells the ECU to ground that pin 28 when you turn the key to the on position.

The bike was running okay when I finished working on it earlier today.  Lets see what it does tomorrow.

Regards

[Dharma Bum]

What Fotoguzzi said!  Siemens relays, especially  with that kind of mileage, are inhabited by gremlins. Get the DPguzzi replacements.

[JoeB]

Suggest as advised, relays from Dan. You'll start to see a pattern here.

[fotoguzzi]

has the battery been disconnected lately and did ALL the ground wires get re-connected?

ignition switch clean?

[twhitaker]

I remember there being reports of the relay sockets being loose. Shuffling the relays sometimes helps.

[LaMojo]

 I think the P8 CPU were easily damaged by voltage spikes caused by jump starting, wires shorting, etc.  A diode in the start up circuit would crap out but that wouldn't explain why after you initially provided a ground to the relay it would start and run normally for a day.  Are you making a ground test directly on pin 28 with everything in the circuit connected? (pull fuse to relay for test).

[oldbike54]

 Jack , aren't the '99 models equipped with the P15 ECU ?

 Dusty

[fotoguzzi]

Probably a 98 that was called a 99, don't think they made any '99's.
Pin 28 would be on the big (98) cpu, no?

[LaMojo]

Dusty,
98 & 99 EVs were virtually identical -equipped with P8s.  Changeover to P15 in 2000.  James mentions pin 28 as the grounding pin for the pump relay which is correct for a P8.  P15 doesn't have a pin 28.  Electrical diagram for 98 EV as shown on This Old Tractor site is incorrect as it shows a P15.   The diagram for a 1996 1100i should be used for the 98 & 99 EV.

Time flies.  I bought my EV new in 1998 - It's hard to believe it will by 20 years old in a couple of years from now.  Obsolete by today's standards.

[oldbike54]

Probably a 98 that was called a 99, don't think they made any '99's.
Pin 28 would be on the big (98) cpu, no?

 Oops , for some reason I read Bassa 

 

Dusty,
98 & 99 EVs were virtually identical -equipped with P8s.  Changeover to P15 in 2000.  James mentions pin 28 as the grounding pin for the pump relay which is correct for a P8.  P15 doesn't have a pin 28.  Electrical diagram for 98 EV as shown on This Old Tractor site is incorrect as it shows a P15.   The diagram for a 1996 1100i should be used for the 98 & 99 EV.

Time flies.  I bought my EV new in 1998 - It's hard to believe it will by 20 years old in a couple of years from now.  Obsolete by today's standards.

 Yeah , but it's still a pretty bike 

  Dusty

[missinglink]

I definitely have a P8 computer, big square thing.  I had it apart to see if anything in it was visibly fried.  I tried starting it yesterday but no luck. - No fuel pump buzz when I switched on the key.  I didn't have time to work on it but I will today.  I'll try the suggestion to check for the ground at the relay socket. Sprayed the key switch with some cleaner, moved wires around, moved the throttle to exercise the TPS - nothing. I did find that the check LED has a broken wire on it but I don't think that has anything to do with it.  I'll fix that today too.

I am still curious what tells the CPU to put a ground on that pin 28.  If it is just the key being on then it seems a little useless.  Why not just ground it and be done with it.  I did read somewhere (can't remember where since I have more miles on me than the bike  ) that when you close the throttle the fuel pump stops to help with the deceleration.  Seems odd to me since that same circuit powers the coils but maybe that is why they have that connection.  Come to think of it I guess no power to the coils would help with the deceleration too.   

Anyway, Thanks for all the replies.  I'll post back with today's efforts.

[missinglink]

Okay, I'm back. Been away for a while but got back on this today.  I fixed the code LED, set up the test mode and tried to start the bike.  No start but no codes either.  Just for laughs I pulled the tank and replaced the fuel filter and air filter. Changing out that fuel filter is a nightmare.  I went thru the whole troubleshooting thing again and I have it narrowed down to the fuel pump again.  The power is getting to the pump but the pump won't run.  This was a brand new pump that only worked once.....  I pulled the pump, hooked it up to a 12 battery and.....nothing.... . It checks with an ohm meter at about 3 ohms (the original one had 0 continuity) but it won't run.  I contacted the vendor for a warranty replacement, let's see how that goes.  Does anyone know if these pumps have to be full of fuel to run?  It had fuel in it when I took it back off.  Hopefully I will get the replacement without too much hassle.

Regards.

[guzzisteve]

Yes, needs fuel in it for lubrication.

[oldbike54]

Yes, needs fuel in it for lubrication.

 Yes , but it will run dry , doesn't need gasoline to kick on .

 Dusty

[pehayes]

The relays are original but they seem fine. 

Wish I had a coin for every EV owner who said that.  Get new relays from Dan.  Cheap insurance.

Patrick Hayes
Fremont CA

[Muley]

Wow, I just saw this.   This problem is similar to what I was seeing last weekend.  Of course, I too have the P8 computer.  I'll be following up on the final solution to this.  James, please keep us posted.

I wonder if Guzzi ownership can help prevent dementia, I mean, you have to use your noggin to keep one runnin', right?

[Kiwi_Roy]

I am still curious what tells the CPU to put a ground on that pin 28.  If it is just the key being on then it seems a little useless.  Why not just ground it and be done with it.  I did read somewhere (can't remember where since I have more miles on me than the bike  ) that when you close the throttle the fuel pump stops to help with the deceleration.  Seems odd to me since that same circuit powers the coils but maybe that is why they have that connection.  Come to think of it I guess no power to the coils would help with the deceleration too.   

The ECU grounds pin 28 for a few seconds when you turn on the key, that primes the injectors then it turns off.
When you crank the bike the ECU picks up the signal from one of the sensors (I think the flywheel one) and starts the pump again.
No, it doesn't turn off the pump or coils when you close the throttle.
I suspect you have blown a track off or perhaps burnt out a component.
If the pump runs when you first turn on the key (timer controlled) but not while cranking (sensor controlled) it could be the flywheel sensor. I have heard that these can pick up steel shavings from the ring gear which stops them working.

It's probably ok to run it by grounding the pin but be aware the ignition coils and pump are sucking power as long as the key is On

Do a Google search for efiman.pdf it gives a great description on the fuel injection particularly the P8 but most of it applies to the later models as well.

BTW I have fixed a couple of the P8 computers where the safety diode (pin 10) was blown, that is there to prevent the ECU trying to start if you accidentally put the battery in backwards, happens quite often.

Update: On re-reading your post you say it worked for a while after you had grounded the pin 28, the ECU cannot ground the relay as well as your jumper, perhaps you have a bad connection or weak relay, grounding the pin gives it a little more voltage to pull in. I know it's hard to measure the Voltage across the relay pins but that might show something
The coil pins are the 2 outside small ones 85 & 86
----- 30

----- 87

|  |  |    85, 87a, 86

[screamday]

I wonder if Guzzi ownership can help prevent dementia, I mean, you have to use your noggin to keep one runnin', right?

Or....live fairly close to Wayne. 

[rodekyll]

The fuel pump doesn't stop at idle (the engine needs a constant, pressurized supply), but there is definitely a change in the pump's spin rate as you move the throttle off or onto idle.  I don't know if the ecu is speeding it up (can't see a control circuit for that in the diagram) or if the increase/decrease in flow at the same pressure is making the impeller change speed.  But it does happen.

[CalVin2007]

The fuel pump doesn't stop at idle (the engine needs a constant, pressurized supply), but there is definitely a change in the pump's spin rate as you move the throttle off or onto idle.  I don't know if the ecu is speeding it up (can't see a control circuit for that in the diagram) or if the increase/decrease in flow at the same pressure is making the impeller change speed.  But it does happen.

   Correct,RK... Voltage increase when the engine speeds up causes the fuel pump to speed up. Watching the voltmeter on mine shows it quite clearly. 

  Terry

[oldbike54]

   Correct,RK... Voltage increase when the engine speeds up causes the fuel pump to speed up. Watching the voltmeter on mine shows it quite clearly. 

  Terry

 Yep , took me a bit to figure out why the wires to the Jackal pump showed no juice when the pump was not hooked up , but when installed the pump runs fine . There is a mechanical check ball in the outlet side of the pump that pressures up and shuts down the pump after priming , after that , it is controlled by current .

 Dusty

 Dusty

[rodekyll]

Not quite . . .

The fuel pump relay is controlled by the ecu.  The ecu sends it a signal when the key is turned on for that 3sec prime.  Then it shuts the pump off until it senses the engine turning at some speed.  Once it thinks the engine is running it returns power to the pump.  If the engine stops turning the ecu shuts the pump off again.  If you want it to reprime you need to toggle the kill button.

[Kiwi_Roy]

   Correct,RK... Voltage increase when the engine speeds up causes the fuel pump to speed up. Watching the voltmeter on mine shows it quite clearly. 

  Terry

The ECU isn't wired to the pump, all it controls is the relay, perhaps it may have some control but only by not firing the injectors.

From the Efiman document
Electric fuel pump
This is a roller positive displacement pump with the motor submerged in the fuel. The motor is a
permanent magnet carbon brush type. When the impeller is turned by the motor, the fuel is moved from
the intake orifice to the delivery port. The amount of fuel is determined by the rollers which adhere to the
outer ring while the motor is turning. The pump has a check valve to prevent the fuel circuit emptying
when the pump is not running. It is also equipped with an overpressure valve which short-circuits the
delivery when the pressure is over 74 psi to prevent the pump motor from overheating.
Delivery: 26.5 gal/hour at 44 psi with 12V power supply. The pump draws about 4.5 amps.

[oldbike54]

Not quite . . .

The fuel pump relay is controlled by the ecu.  The ecu sends it a signal when the key is turned on for that 3sec prime.  Then it shuts the pump off until it senses the engine turning at some speed.  Once it thinks the engine is running it returns power to the pump.  If the engine stops turning the ecu shuts the pump off again.  If you want it to reprime you need to toggle the kill button.

 OK , but if my Jackal starts for a second and then dies in cold weather , the pump primes again W/O any input from me . Just figured it was because the pressure had lessened .

 Dusty

 

 

[oldbike54]

The ECU isn't wired to the pump, all it controls is the relay, perhaps it may have some control but only by not firing the injectors.

 I thought the pump runs at a constant speed and the regulator controls flow . Help me Roy .

 Dusty

[Kiwi_Roy]

OK , but if my Jackal starts for a second and then dies in cold weather , the pump primes again W/O any input from me . Just figured it was because the pressure had lessened .

 Dusty

I think you have a fault, perhaps losing the signal to the ECU relay for a second causing it to reset the cycle.
May be as simple as a switch that needs a good clean.