Guys I need assistance with my fuel pump....it goes on as soon as I open the switch (1991 S4). I thought it would be that the PO might have jumper the fuel pump relay, so I replaced the relay with a new one...the old looked original without any modifications, and still the pump goes on when it should come on for a few seconds and then energize when the engine moves...correct? Where else should I look? BTW...I do have a new refurbished LH installed.

 

I think you mean the fuel pump runs as soon as you turn on the ignition? Your description of "opening a switch" is a bit odd - you are actually closing a switch in this case. Better to stick to the more obvious terminology.

Looks like you have an S4 - in this case the fuel pump should not turn on at all until the engine is rotating. Earlier cars had a priming pulse on initial turn on, this was dropped on the S4+.

The fuel pump relay on these cars is a simple Bosch SPST relay (no smarts) so for it to not work properly suggest the LH is faulty and always enables the fuel pump.

Alan

 

Alan; thanks...you're description is correct...sorry for my confusing explanation!

When I bought it it didn't run..I installed a new MAF and LH refurbished by JDS. I hate to think the LH is at fault, but your explanation makes sense. Unfortunately I'm very comfortable swapping a transmission, which I just did, but the electrical side of wrenching eludes me and intimidates me.

 

Given that the car would NOT RUN and stopped for the prior owner....I suspect that they may have messed with the wiring to provide constant power to the fuel pump...maybe remove the relay and see if the pump still runs ???

 

The JDS reconditioned LH should be OK so it would make sense to take a look at the wiring around the relay - maybe even the back of the panel - as Jim says it is possible someone has messed with it previously.

What year is the car?

Look for obvious non-factory wiring near the bottom of the panel for starters.

Alan

 

Thanks James...if I have an opportunity tomorrow morning before I go to work I'll test it as you advising.

 

OK you had it - on a '91 look for wiring added around plug W15 (labelled) pin 15 is the top pin on the left side of plug W.

Alan

 

Alan thank you...it's a 1991. And yes...this car was left for dead for over five years and who knows what went on. When I brought it home the board covering the panel was in the trunk along the floor mats, the LH was hanging to the side...all that area around the panel looked quite messy. I'll test it like you're suggesting and look closely around. Maybe I'll take pictures of the panel and post them....hundreds of knowledgeable eyes can only help. Thank you.

 

They may have grabbed switched power from anywhere Like Greg Brown has said He is constantly amazed at how "creative" some people have been when "fixing" a 928 Not to mention the time needed to figure out what they screwed up.

 

Thank you Alan....good homework! Will report back.

 

Absolutely correct! I once heard Mr. John Kuhns say that the main problem with 928's is the "mechanic" who previously worked on it.

 

I have seen maybe 2000 LH ECUs over the past 15 years and VERY occasionally I see one that behaves as you describe. Probably 2 like that in that time. As I don't consider it a problem I never bothered to find out the reason.

As long as the fuel pump shuts down when the engine stops with ignition still on then you have the essential safety feature. You could test this by unplugging the EZK ECU with the engine running, then when the engine is stopped with the ignition still on, listen for the pump. It should be off.

 

If someone has the ignition switch on for any reason....listen to radio, roll up windows, etc. AND happens to have one leaking injector it can easily load up a cylinder with liquid gasoline. And then on a subsequent restart where another cylinder fires first the uncompressible gasoline can literally lift the head off the engine block , split a cylinder or simply bend a connecting rod....... Sure it is unlikely but certainly possible.

 

The pump only runs for about 1 second...

 

Reading the first post it sounds like it is staying on for as long as the switch is on. Not a one second run but for as long as the switch is on. So far different than one second.