Basic (stupid?)Electrical Question

Bernard Littau bernardl at
Sat Nov 2 11:18:50 EST 2002

> I must be missing some basic concept, but here goes:
> I'm trying to locate an electrical "leak": new battery goes dead after a
> week or so of sitting.
> With ignition off and all interior light switches in the off position, I
> disconnected the battery at the ground and found -after charging- 13 volts
> between the ground terminal of the battery and the ground strap.
> So I figure
> there must be a completed circuit someplace.  I then removed and replaced
> all the fuses, one by one, expecting that when I pulled the one on the
> guilty circuit, the multimeter would drop to zero. It never did -including
> the one fuse on the knee bolster aux relay board. It just
> continued to read
> 13 volts.
> Can anyone point me in the right direction?

If I am reading this correctly, then you are taking the fuses out one at a
time, and replacing them after checking the multimeter.

If you had hypothetical car with two fuses, each with a current draw, and
you did your test taking one at a time out, you would always see 13V on your
meter. If you removed one, and then the other, when both were out your meter
would read 0V.  Now add them back one at a time.  13V as the first one goes
in, and still 13V as the other goes in.  You still don't know if the second
one has a current draw or not.  See where I am going?

I think you need to take all the fuses out, and see what you read.  If you
still get 13V, then it is a non-fused item which is drawing power.  If you
get 0V, then add the fuses back one at a time until one shows 13V.  The last
one you added caused the circuit to complete.  Now pull that fuse out again
and keep adding back the other fuses to see if there are multiple offenders.

Sensing voltage is also not a very good test, as a very small amount of
current, like the keep alive for the radio presets, or the clock, can cause
a meter to read 13V.  You need to measure the current drawn for each fuse,
and decide if it is too much.

Measuring the current with a typical meter means running that current
through the meter.  You must be careful not to run more current through the
meter than it can handle.  Most expensive modern digital meters are well
protected, and you won't hurt them too easily.  A cheap meter, or an analog
meter, can be hurt by hooking it up as you describe above, and trying to
start the car.  That's a worst case example.

How to decide if a draw is too much?  I use this: most car batteries are
around 60Ah (Amp hours).  Such a battery can theoretically deliver 60 Amps
for one hour, or one Amp for 60 hours (in practice this is untrue, but it
will do for our needs :-)  So, if you left your car standing with a 1 amp
current draw, in 60 hours it would be very very dead.  I routinely leave my
Audi for 2-4 weeks and have no problem starting the car.  2 weeks is 336
hours.  A draw that would drain the battery completely in two weeks would be
60Ah/336h = 0.18A or 180 mA.  That is not much current, about 2W (one small
instrument cluster light).

You are looking for a very small but consistent draw of current.  A trunk
light would drain your battery in days.

Or, you are looking for an intermittent higher draw of current.  Your
radiator fan, for instance, comes on from time-to-time when the car is off.
While I have no expectation that this is the case here, it serves as an
example:  If the fan did come on secretly, late at night while you were
asleep, it could draw down your battery.  The vacuum pump for the power door
locks is a more likely candidate.  It replenishes the vacuum if it leaks
down.  Maybe it runs for a few seconds once an hour if there is a leak.
Bottom line here is you may need to find a higher current but intermittent
draw.  You will likely not find this using the pull the fuses and measure

Start by pulling all the fuses and measuring current, then adding one fuse
at a time and measuring current.  Report back on what you find and I'm sure
you'll get more help on what to do or look for next.


Bernard Littau
Woodinville, WA
'88 5ktq

More information about the 200q20v mailing list