# Fuel Gauge Damper Circuit

Hello all,

I have an old ford with a fuel gauge sender that has a resistance of 73ohms at empty and 10ohms at full.
I have replaced the old analog gauge with a digital one but it is extremely erratic.
I thought I might be able to make a damper circuit that would hold the variance to maybe 25 possible values in the 10 to 73 ohm range thus removing the erratic behavior.

I was thinking that I could use an Arduino to read the resistance of the fuel tank sender average it out and then using a digital pot, send out the new value. I don't know if this is the way to approach this or not.

I'd appreciate it if someone could lead me in the right direction.

Thanks in Advance!

I don't think that will help. What is needed is to smooth out the signal Arduino Smoothing Tutorial

rally7:
Hello all,

I have an old ford with a fuel gauge sender that has a resistance of 73ohms at empty and 10ohms at full.
I have replaced the old analog gauge with a digital one but it is extremely erratic.
I thought I might be able to make a damper circuit that would hold the variance to maybe 25 possible values in the 10 to 73 ohm range thus removing the erratic behavior.

I was thinking that I could use an Arduino to read the resistance of the fuel tank sender average it out and then using a digital pot, send out the new value. I don't know if this is the way to approach this or not.

I'd appreciate it if someone could lead me in the right direction.

Thanks in Advance!

How old is the Ford? My 1950 Mercury had a wire wound resistor in the sender. Sounds like yours is the same and the slider is bouncing between windings. Try placing an electrolytic capacitor, correctly polarized, across the leads from the sender.

Paul

Was it steady with the old analog gauge? - they were often thermal with a long time constant.

If so , Paul_KD7HB's idea could well help - choose a r-c time constant of a few seconds

Allan

How "erratic" is erratic?

As the tank doesn't drain extremely fast normally you could also take e.g. the values of the last minute and average those.

Hi,

rally7:
I have replaced the old analog gauge with a digital one but it is extremely erratic.

Adding a capacitor should work, but practically getting "a r-c time constant of a few seconds" when the sender is only 10 to 73 ohms would mean quite a large capacitance (20,000uF+)

If you think you could drive the digital gauge "using a digital pot" with an Arduino, then it should be possible to simply use a smaller capacitor and a resistor or two... you'd have to post details of the gauge to see if this could be done.

Yours,
TonyWilk

P.S. Time constant calculator does not work. Try to calculate the capacitor for 50 ohms, 1 second.

TonyWilk:
P.S. Time constant calculator does not work. Try to calculate the capacitor for 50 ohms, 1 second.

I noticed that. I just assumed I was doing something wrong.

The one linked to above never worked for calculating R or C from T and the other.

If your digital gauge has a high input impedance, add series R of ( say ) 10k, and a shunt capacitor of 1000uF.

Allan

allanhurst:
Was it steady with the old analog gauge? - they were often thermal with a long time constant.

If so , Paul_KD7HB's idea could well help - choose a r-c time constant of a few seconds

Allan

Exactly! the gauge had a bimetal strip with a resistance coil wound around it. The strip was connected to a gear ot lever that moved the needle. Heat applied to the strip varied based on the resistance of the tank sender.

Paul

Thank you very much for the input. The car is a 66 Ford Thunderbird. Here is a picture of the sender:

it is a coil spring type resistor I guess. I know just enough to be dangerous :o

The gauge is a Prosport Evo, but I can't find any details on it.

When I say erratic, it will fluctuate 50% or more. For example reading 23 then going up to 78 then back down to 41. It never stays at any single value but sweeps back and forth whether the car is moving or not.

I built an digital LED version of the speedometer and have replaced all the analog gauges with digital.

The old one fluctuated just the same. The thermal delay averaged it out. You need to do the same with software.

Paul

I'm not sure I understand how? I can do the programming if I understand the logic but unfortunately I'm severely lacking in my E.E. education (youtube educated)

rally7:
I'm not sure I understand how? I can do the programming if I understand the logic but unfortunately I'm severely lacking in my E.E. education (youtube educated)

Make a voltage divider using a resistor and the sender in series. Using an AD pin, measure the value across the fixed resistor perhaps once per second. Add that value to a counter. Then, after 60 seconds, divide the counter by 60. That will give you and average value for the 1 minute of time. You will need to calibrate that average against the amount of fuel in the tank.

Yes, you say, I can't see the tank level for whole minute. True. But on the old gauge, you could not see it correctly for several seconds. If this is a problem. scale back the time you are averaging the readings.

Paul

I don't think there is any point bringing out the Arduino sledgehammer to crack this nut until some simpler solution has at least been tried.

Even with an Arduino, you still have the problem of driving the 'Prosport Evo' gauge

A quick google came up with: EVO Series Digital Fuel Level Gauge
I guess it's likely there's something like a 200ohm pullup in there - you'd have to measure it.

I reckon it is well worth following Paul_KD7HB's early suggestion and fit a capacitor between Ground and Signal on the back of the gauge. I'd go for 1,000 uF, 25V

It may also be worthwhile double-checking the sender wiring; it looks like it has separate sender Signal and Ground pins so you could run Ground directly to the gauge with the signal (if it ain't already).

Yours,
TonyWilk

P.S. Cool dash! (looks like the steering wheel's on the wrong side tho )

TonyWilk:
P.S. Cool dash! (looks like the steering wheel's on the wrong side tho )

Are you looking in a mirror ?
that steering wheel is on the left hand side where it should be.