Skip to main content

Finally got around to looking at #PEH578L's petrol gauge. Multimeter showed that the fault was at the sender, which is a simple float on an arm operating a variable resistor.

Opened it up and was slightly surprised to discover a wire-wound rheostat. Also that three of the turns are broken which would explain why it doesn't work. So unless I feel like re-winding it, I probably need a new one.

The inside of a variable resistor. A very fine wire is wound hundreds of times around a flat insulating former. There's a wiper that can move over the turns of the wire and it's worn a bright track. Three of the wires have broken where the wiper runs over them and are sticking up away from the rest.
#peh578l

Alexandra Lanes reshared this.

in reply to Ben Harris

Ben Harris
Ben Harris
• •
Also, I have a new camera! Well, I say "new" but when I factory-reset it it defaulted to a date in 2007, so not very new.
in reply to Ben Harris

Alexandra Lanes
Alexandra Lanes
•
@Ben Harris And I think the coils are different lengths to match the shape of the tank?
@Ben Harris
in reply to Alexandra Lanes

Ben Harris
Ben Harris
• •

@ajlanes The tank is basically a cuboid, so I don't think that's what it's for.

It might be compensating for the fact that the float is on a pivoting arm, so the rate of rotation changes with depth.

But now that I look at it, I think the variance may be the wrong way around for that, so maybe it's compensating for non-linearity in the dashboard gauge instead.

Maybe I should do some experiments...

@Alexandra Lanes
in reply to Ben Harris

kæt
kæt
• •
@ajlanes If I was designing this, I'd fill it up by 10%'s and draw a graph and not worry about which part was causing it! 😄 That doesn't stop someone downstream being interested in its origin, of course!
@Alexandra Lanes
⇧