On 21 Jul 2011 17:51:43 GMT, John Carter wrote:
> After a lot of testing, I found my meter was giving bad results - too
> high on voltage. Problem was the 9V battery in the meter was low, so
> I replaced it and got good readings to work from.
>
> What I found was a "leaking" fuse. During the process of tracing the
> path of power from the battery, I was getting ready to do some fuse
> panel checks, and for some onkbown reason, decided to change the 20A
> cig lighter/aux power fuse. While out of the circuit, I held it up
> to the light and could clearly see the filament between the two
> blades was broken. I checked it with a meter set for continuity and
> when I applied the test leads, a small LED on the top of the fuse lit
> up. The fuse (labelled Smart Glow) is supposed to light up if it
> blows so that you can see which one blew. BUt how does it get its
> power if the circuit is broken ? I don't know, but apparently it
> will allow a reduced amount of voltage to come through the circuit,
> which is why I was getting (after meter battery replaced) 10.1 VDC
> reading at the lighter/aux sockets.
>
> Now I don't know about you, but I have always understood a fuse,
> when working properly, to completely shut down the circuit it was
> supposed to be protecting, not allow some power to get through. Maybe
> they think the circuit is protected with reduced voltage?
>
> I talked with a technician who verified that these types of fuses
> sometines leak like this one - but said in his experience, they did
> so all the time, inspite of the manuf's cautions. He said that this
> type of fuse causes more problems than it fixes, and besides, if you
> know what in the car is failing (wipers, radio,etc.) you just look at
> the fuse map found on the fuse box lid and pull and test the
> appropriate fuse. He advised me to throw all my spares of that type
> in the trash, and go get some "real" fuses, and in addition check to
> see if there were any like that in te panel and replace them. I'm
> headed out this PM to do this.
>
> Mystery solved, one of those things we used to call a "funny", which
> we experienced all the time back in the days of computer mainframes
> with discrete component circuit cards.
Does this relate to an earlier post of yours? It would've been helpful
if you had replied in thread.
It makes sense. Clearly an LED circuit must have a finite resistance, so
as to allow the current to flow that lights the LED. Thus a
high-resistance device would end up powered through the fuse, as long as
there was no low-resistance load also in series with the fuse. For a
hypothetical example, a 100 kilohm LED in series with a 1 megohm load
would reduce the voltage by ~10% instead of cutting it off.
--
Gene E. Bloch (Stumbling Bloch)
|