jim beam

hi

can't figure this one out:

have an '89 civic dx hatchback & the fuel gauge "sticks".

have replaced the gauge - same problem.
have replaced the sender - same problem.
have checked the earth in the trunk - the trunk light now works, but the
above problems still persist.

have checked the wiring - 7.6v at the sender, the book says should be
battery. but if i disconnect, the gauge goes below zero, if i short, it
goes to full, so behavior indicates that's ok. and if i disconnect,
then reconnect, the gauge moves to register the fuel level present at
that time, then it stays at that level and won't move again.

i've also double checked the sender and it's output is within the
resistance ranges quoted in the helm manual.

anyone got any ideas?

Chip Stein

has it ever been wrecked? i had an accord drive me nuts like this,
turned out it had been wrecked and the tank bent so the float touched
the baffle.
otherwise start with the obvious. the book says battery voltage, you
have 7.6 volts. find out why that is first off. if there's a known
problem like this and you keep throwing parts at it, you'll never fix
it.
Chip

jim beam

the wreck theory might explain everything! i'll check it out this weekend.

thanks chip. i'll keep you posted.

j

Chip Stein wrote:
> has it ever been wrecked? i had an accord drive me nuts like this,
> turned out it had been wrecked and the tank bent so the float touched
> the baffle.
> otherwise start with the obvious. the book says battery voltage, you
> have 7.6 volts. find out why that is first off. if there's a known
> problem like this and you keep throwing parts at it, you'll never fix
> it.
> Chip

jim beam

chip, you got it! yes, the float is sticking on the baffle. awsome
diagnosis! thank you a thousand times.

jb



Chip Stein wrote:
> has it ever been wrecked? i had an accord drive me nuts like this,
> turned out it had been wrecked and the tank bent so the float touched
> the baffle.
> otherwise start with the obvious. the book says battery voltage, you
> have 7.6 volts. find out why that is first off. if there's a known
> problem like this and you keep throwing parts at it, you'll never fix
> it.
> Chip