Subaru Fuel Pump Ground Wire?

Mike: The fuel pump wiring is visible under the rear seat, but there are also a few wires for the fuel level sender in the tank. The later AWD cars have 2 fuel senders, not sure if this is the case on yours (is your car FWD or AWD?). The pump is under the cover that has hoses coming out of it. Chck for voltage with a multi meter (with the ignition in the on position), and you can tell which wire is positive and negative with the meter. To ensure you are checking the pump wiring, pull the pump fuse and test again, if no voltage, you have the right wires. Connect some wire to the ground wire(solder it to ensure a good connection), and ground it with a self tapping screw anywhere convenient on the metal under the seat.

Is the fuel pump in the tank? Do you often run until the gauge is on empty? If both are 'yes', that could be the problem.

I've read elsewhere that in-tank pumps depend on being covered by fuel to keep them cool, and thus to prevent them from burning out, you need to be sure to never go under 1/4 tank.

Steve

You gotta love a design that takes an overheating electrical device and submerges it in an explosive liquid and at the same time allows a rupture in the fuel line to spray gasoline all over a nice hot engine.

I mean, you GOTTA love that.

-DanD

LOL! Have you thought about the fact that the brush commutation (i.e., lots of sparks) also takes place in the fuel (I know - the third component - air/oxygen - is missing - but I always try to keep myself more than one failure mechanism away from death whenever I can). 8^)

Bill Putney (to reply by e-mail, replace the last letter of the alphabet in my address with "x")

Do you have a reference for this "fact"? Considering the number of people who run ther cars with less than 1/4 tank of fuel, I have a hard time believing this is true.

-- Vic Roberts

Though there are some here who I respect but who would (no doubt politely) disagree with me, I also question this accepted "fact".

As a veteran fuel pump components designer, my 2 cents is this: Cool fuel flows thru the pump (all around the armature and inside the case/magnet housing). Though the fuel surrounding the "can" certainly would contribute something to the overall cooling, my guess is that it has, at most, a second-order effect to local peak pump temperatures relative to the internal flow (primarily because the external flow is not in direct contact with the primary heat-generating components, such as armature and brush/commutator interface). Because of this, I believe that the most that can be said about running with the tank low in fuel is that it reduces the design safety factor a little. I'm only a sample of one, but I sold my '86 turbo wagon last year with 275k miles on it and the original fuel pump, and I never paid much attention to keeping the fuel level up.

Bill Putney (to reply by e-mail, replace the last letter of the alphabet in my address with "x")

I remember having a fuel pump discussion where it was brought up that most (if not all) FI pumps were of the roller cell variety - a type of positive displacement pump. This, combined with the fact that the motor was a DC series wound type (will over-rev with no load) would mean that these pumps should never be allowed to suck air. In addition to over-revving, the rollers can overheat very quickly if they aren't immersed in fuel.

Air is only missing from the fuel tank when full. When the level drops, the missing gasoline is replaced with air... air that's chock full of vapors. When you get real low on fuel and it sloshes around when you speed up, slow down, or turn, you can even suck air into the pump.

I prefer not to think about it.

Hey, ever been in a car that had the fuel line spring a leak? With a tank-mounted fuel pump it just keeps right on pumping and spraying that gas all over your nice hot engine. It happened to a friend of mine driving down the autobahn.

The car was totalled.

Another friend had a leak in his fuel line but since the pump wasn't mounted in his tank it resulted in air in the fuel line.

The car stalled.

I don't about you, but I don't care for in-tank fuel pumps. It makes working on them much more dangerous if you have to remove the fuel tank.

-DanD

I hear you on that one. When I used to work on design teams as a supplier on GM/Delphi and Ford fuel pumps, I used to say to their engineers that if I had been the first engineer to think of running fuel through the commutation area of an electric pump, I would have kept my mouth shut. I also am convinced that, if that were not already being done today, in today's legal climate, *no* manufacturer would ever be the first to try it after the company lawyers looked at the idea.

Bill Putney (to reply by e-mail, replace the last letter of the alphabet in my address with "x")

Don't let lawyers make design decisions, there are more quantitative ways to make such decisions. Given that cars with nearly empty gas tanks are not blowing up everyday (sample of millions) I gather this design has proved to be quite effective.

-- Dominic Richens | snipped-for-privacy@alumni.uottawa.ca "If you're not *outraged*, you're not paying attention!"

You didn't indicate whether yours was a turbo or not, but on turbo models, the fuel pump connector can become burnt, causing a no start condition. Just to eliminate the potential, pull up the rear seat and have a look at the connector. Pull it apart and examine the connectors. If you see any black or brown marks, you may have located a potential problem.

WHAT IF ALL FUSES HAVE POWER *EXCEPT* #13 fuel pump fuse regardless of key position. Yest light shows power to ALL sockets *EXCEPT* #13 fuel pump socket? WHAT keeps power from that ONE socket ?

No thanks. Discussions in Usenet stay in Usenet. Wonder how the folks over there might react to telling them to move their web-based discussion to Usenet. Get the point?

Besides, you decided to join a *19-year old* in a web forum, and only linked to it here because you replied decades late. You want folks here to see you very disconnected reply. Instead just start your discussion (and likely start a new thread to do so).

Did it ever occur to you that he might not have noticed it was ancient? That fairly frequently occurs...