Friday, September 29, 2006

Fuel Gauge fault traced

The full story is that the gauge worked fine for years. I took the tank out and welded in a new bigger bore outlet for the PI, washing the tank and generally having it at all sorts of angles in the process. I think the gauge worked fine after that but immediately I'd done the welding I replaced the tacho cable and disturbed the instruments in the process. The gauge then read zero so I pulled it out and found a dodgy earth, I put a new terminal on the earth (I think this is the earth for the light and nothing to do with gauge operation). So then the gauge read full (wrongly but I thought it was OK).
Last night my brother in law came round (I'm recovering some lost data from his laptop for him) and so I pressed him into fault finding service - he used to race a Vitesse in the early 90's and is a lifelong Porsche technician so he can be very useful :-) Testing voltages and resistances, pulling wires off the sender and joining them together all indicated that there was an intermittent fault with the sender. As the tank was quite empty I could pull the sender out and sure enough, the fine wire of the variable resistor was broken.
There was nothing at all wrong with the gauge!
We tried to solder a repair but the old wire refused to take any solder so in a bodge I managed to weave some fuse wire into the right place and got the sender working. We reassembled it all and it seems to be OK - I'll get another sender ASAP and hopefully replace it before the RBRR!
The computer fixing then swallowed the rest of the evening so my metering unit modification will happen today!

No comments: