John,

In your first solution, I think you have your circuit backwards. If I read
the issue correctly, the new sender has a resistance of 0 - 30 ohms and
what the gauge expects is 0 - 90 ohms. I will be honest though, I don't
have a clue of how a gas gauge and sender work.

-joe


On Sun, Nov 17, 2013 at 9:19 PM, John Rehwinkel <[email protected]> wrote:

> > Electronic design is not my strong suit so I thought I ask if someone
> here could offer some advice.. I'm not even sure if this is possible.
> >
> >  I need to make a fuel tank sender and it's gauge compatible. The gauge
> is expecting to see a sender (basically a variable resistor) that has a
> value 0 to 90 ohms. The current sender in the tank has a value of 0 to 30
> ohms. The original gauge, changed long ago is NLA and while an appropriate
> sender is available, changing it is going to be an issue.
> >
> >  I was hoping for a simple circuit that would accomplish this. It would
> have to be slightly adjustable and not affected by the supply voltage that
> vary by several volts depending on the battery load and charge condition.
>
> There are a few ways to approach this.  I'll address them in order of
> increasing complexity.
>
> The easiest approach is to make the existing gauge look like the gauge the
> sender expects.  To do this, parallel it with a resistor.  In effect, you
> want a resistor that's half the resistance of the existing gauge, so it in
> parallel with the existing gauge, will be 1/3 the resistance.  Then the
> (approximately 3x) current from the sender will be divided unevenly, with
> about 1/3 the current (equivalent to a sender with 3x the resistance) going
> through the gauge.  Making part of this resistance adjustable will give you
> the adjustment you're after.  Note that this is the approach I would use if
> it were mine.  Note that there might be some odd effects on linearity, but
> fuel gauges aren't famous for linearity anyway (I had a VW where full to
> half barely moved the pointer, but it had lots of resolution near empty,
> where it mattered more).
>
> The next most complicated approach is a lashup known as a "current
> mirror".  It uses transistors to make the current through one transistor
> control the current through another (or several more).  By playing with
> transistor betas and paralleling transistors, you can make it assymetrical,
> so the transistor driving the gauge tracks the one driven by the sender
> proportionally.
>
> After that, you get to relay type circuits, where the sender operates
> something like an optocoupler (ideally one of the linearized kind), which
> then drives an amplifier that in turn drives the gauge.  This can also be
> done mechanically, with the sender driving a galvanometer hooked to a
> potentiometer that acts as a surrogate sender, with the desired resistance
> range.
>
> After that, you get to microcontroller-based arrangements, where the
> sender drives an analogue-to-digital converter, and then the microprocessor
> uses a lookup table to drive a digital-to-analogue converter which then
> drives a transistor controlling the gauge (PWM works just fine for
> automotive gauges, as they don't respond quickly).
>
> Once you have a microcontroller and a lookup table, you could also drive a
> nixie display or a bargraph indicator if it'll fit (just to offer something
> on-topic).
>
> A side note - some cars (like my old Volvo) actually have a voltage
> regulator in the dash so the gauges can run on a more stable voltage supply
> (I think it was 8 volts).
>
> - John
>
> --
> You received this message because you are subscribed to the Google Groups
> "neonixie-l" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> To post to this group, send an email to [email protected].
> To view this discussion on the web, visit
> https://groups.google.com/d/msgid/neonixie-l/D0DBC2B4-F25D-414E-A652-9D2B52D1C688%40mac.com
> .
> For more options, visit https://groups.google.com/groups/opt_out.
>

-- 
You received this message because you are subscribed to the Google Groups 
"neonixie-l" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send an email to [email protected].
To view this discussion on the web, visit 
https://groups.google.com/d/msgid/neonixie-l/CAE%2BVk6NdExiFWBYMkgu6jsHWJGPS%2B5oSp%3DAc3B9NQf7o9%2BYS9A%40mail.gmail.com.
For more options, visit https://groups.google.com/groups/opt_out.

Reply via email to