resistance mismatch

[walnutron]

Hi,
I'm having a problem with a new fuel level gauge in my 55 Chevy. I have changed the instrument cluster and because of size constraints, I could not use the OEM fuel level gauge. The new gauge would like to "see" 0 to 90 ohms from the sender(0=empty and 90=full). The OEM sender's resistance ranges from 0 to 35 ohms(0=empty and 35=full). The gauge operates on 12 VDC and shows a resistance of 196.4 ohms between the positive and ground terminals. Any help would be greatly appreciated. Thank you, Ron

[pebe]

Hi walnutron,

Can you provide a link to the specification details of your new gauge.

[walnutron]

Hi,
The gauge is an Equus fuel level gauge part # 7362 and the link is http://www.equus.com/Product/7362/2"-Chrome-Fuel-Gauge-GM . If you need any more information, I'll try to find it. Thank you for any help that you can give. Thanks again, Ron

[pebe]

Hi Ron,
I've looked at the data sheet and searched elsewhere for how it works. It looks like the inside may be like this.

http://www.google.co.uk/imgres?imgurl=h ... Q&dur=1235

If so, a resistive shunt across the left hand coil may do the trick. You measured the resistance from + to - (both coils in series) as 196.4ohms. Can you measure across just the left hand one (+ to S)?

[walnutron]

Hi Pebe,
I metered 97.8 ohms from the positive terminal to the sender terminal. Thank you, Ron

[pebe]

Hi Ron,
It looks like the two coils have about equal resistance values. So as far as the sensing resistor is concerned it is being fed, in effect, by two resistors of approx 97.8 ohms each in parallel, from a potential of 6V. You can check that by connecting up your 30ohm sensor. With the resistor at max value you should have a voltage of about 2.5V across it and the gauge should be reading about 1/3 of full scale.

If that is correct, then you need to shunt each of the coils in the gauge with a resistor of about 47ohms. The one from sensor to ground only needs to be 1/4W but the one from sensor to +12V needs to be 2W rating.

I hope that works out alright for you.

[walnutron]

Hi Pebe,
Thanks, I'll give it a try and get back with you. Thanks again for your time and knowledge. Thank you, Ron

[walnutron]

Hi Pebe,
I'm sorry, but I made a mistake and didn't isolate the + side of the fuel level gauge when taking the meter readings. The fuel level gauge is being fed from the volt gauge and flawed my readings. With the fuel level gauge terminals now totally isolated, these are the ohm readings: from + to sender terminal (114 ohms), from + to ground terminal (332 ohms) and from ground to sender terminal (218 ohms). Thank you again for your time and patience. Ron

[pebe]

Hi Ron,
No problem. It just needs a change in the resistor values.

Check that the voltage on the sender terminal reads about 2.25V (with 12V battery) when your 30ohm sensor is wired in and set to maximum resistance.

If OK, then two resistors are needed, as before. A 57ohm resistor between Sender and +, and a 108ohm between sender and ground. The nearest in ‘preferred’ values are 56ohms and 100ohms and these will make very little difference to the accuracy.

The battery voltage can go up to 14V when it is charging, so the 56ohm needs to be 3.5W minimum, so use a 5W wirewound one (keep it away from wiring etc. because of the heat dissipated). The 100ohm one dissipates 0.19W so you can use a normal ¼W or 1/2W one.

There is an alternative way of doing this using pulse width modulation. These extra resistors would then not be used but an IC and a transistor and a few other bits would be needed. Let me know if you want details.

[walnutron]

Hi Pebe,
If the voltage at the sender checks out, it may take several days for me to get the resistors values we're are looking for. I'll keep you posted. Thanks again! Ron

[walnutron]

Hi Pebe,
You were right on with voltage reading at the sender terminal. I metered 2.25V with the battery at 12V and the sender at max. resistance. How much current will the sender see with our 57 and 108 ohm resistor set up? I did a check with a 12V battery and the OEM gauge and the OEM sender. When the sender is at max. resistance, I metered 1.7V at the sender terminal and .06MA going to the sender. In the last post you had mentioned an alternative method. Would pulse width modulation send less current to the sender? Thanks again for being so helpful. Ron

[pebe]

Hi Ron,
The current through the sender would vary between 500mA at zero resistance and 250mA at maximum resistance. Rather a lot!

I have discarded the PWM method I originally mentioned. It relies on switching on the sender for only one third of the time, so effectively the sender appears to be 90 ohms, rather than 30ohms. To eliminate needle vibration in the gauge the switching frequency needs to be several hundred Hertz. But at that speed the (unknown) inductance of the coils in the gauge comes into play and would affect the reading.

So I have thought of a circuit that would make the 30ohm sensor emulate a 90ohm one, without those high currents. It will take me a few days to finalize the detail.

Pebe.

[pebe]

Ron,

Check your PM box.

pebe

[pebe]

Hi Ron,

Here is the circuit. Basically, it passes a small current through the 30ohm sensor, P1. The resulting voltage across it is fed to the input, pin2, of IC1. The output of the op-amp switches on the transistor which shunts R4 and R5, effectively varying the current drawn from ‘S’ pin of the gauge. The figures in blue indicate the voltages as P1 is varied from zero to its maximum resistance.

R1 and D1 & D2 form a stable 1.4V supply which is fed to R2 and P1. Varying the sensor from min to max resistance gives a voltage of from zero to 0.68V into pin2 of the op-amp IC1. The low value of 1.4V for the supply means that only 40mA max flows through the sensor.

The output of IC1 controls TR1 so that its current shunts R4 and R5 until the voltage that is tapped off P2 and fed into pin3, equals the voltage fed in from P1. With P1 at max, the voltage at pin ‘S’ of the gauge is 4.3V, and that gives the same result as connecting a 90ohm resistor between ‘S’ and ‘-‘ of the gauge.

D3 is fitted in series with the negative supply to the gauge. That is because it is not possible to get zero volts between collector and emitter of TR1, even when it is switched hard on. It will not effect the operation of the gauge because it is the difference in current through its two internal coils – not the supply voltage – which operates the needle.

To set it up, adjust P1 for max resistance and then adjust P2 until the voltage between ‘S’ and ‘-‘ of the gauge is 4.3V.

[walnutron]

Hi Pebe,

Thank you for the circuit and now I'm anxious to get started, but I have a few questions.
- Can R2, R4 and R5 be 1/4 watt resistors?
- Is .75 watts sufficient for P2?
- Is R3 a 1K-2watt resistor?
Thanks again,

Ron