Incorrect fuel gauge

[ericbushby]

Good evening Gentlemen,
I am looking for a little help with the old fuel gauge problem. The car is a 1967 S3.
My gauge always reads high and has never gone below 1/4 reading even when the tank is empty. On investigating I believe I have found the problem.
I have the correct 0 to 90 ohms sender which is reading 40 ohms with 5 gallons in the tank.
I now believe that I have the later type gauge. It is definitely a bi-metal strip type and I read in the archives that it should be a moving coil meter type for this sender arrangement.
I suspect it was all they could get at the time.
I have also found that the voltage regulator is giving 8 volts output instead of 10v. I wonder if that was adjusted to try to reduce the meter reading.
Now, it would be helpful if someone could measure the resistance of the fuel gauge instrument (not the sender) as extra confirmation. this one is 62 ohms.
I am thinking I need to buy a new regulator and gauge.
Any comments, corrections or suggestions will be most welcome.
Thankyou
Eric in Burnley
1967 S3SE DHC

[Craven]

Hi Eric,
As you have said your cars gauge should have a ? moving coil? type gauge, to match your 0 ? 90 ohm tank unit. In this arrangement low resistance = low reading on gauge, but with a bi metal gauge however low resistance = high gauge reading.
A bi metal tank unit is typically 230 ohms on empty and around 30 ohms when full, so your tank at 40 ohms would show almost full on the gauge but the indications will move in opposite direction as the fuel level changes.
Ron.
PS
Possible, PO may have fitted your tank unit up-side-down

[ericbushby]

Ron,
Thankyou for your comments.
curiously, the sender was upside down when I bought the car, but the gauge worked in reverse and I put the sender the correct way up so that the resistance increased as the fuel level fell.
But as I said before the gauge now reads far too high.

I have now bench tested it with a 10 volt supply and a variable resistor in series and got the following results:-
It is a simple thing and just gives more deflection with more current through it.

230 ohms 37mA Pointer moves off rest and to a fraction (half a pointer) above zero

66ohms 80mA Half Full

24 ohms 210mA Full

It is almost as if I had a 230 ohm sender, it would work quite well set up as in an early 90 ohm arrangement.
I wonder if this gauge is not for a Lotus at all. The label is missing.
Of course, I am trying to correct faults and mistakes and want to use the correct, original method.
Does anyone know if the correct gauge for the 90 ohm sender is available. I have not tried any possible sources yet.
Any suggestions are welcome,
Eric in Burnley
1967 S3SE DHC

[Craven]

Hello Eric,
From what you have written it would seem you have done a double take on the fitment of the tank sender unit. Originally the unit would provide 0 ohm when the tank is Empty and 90 ohm when Full.
This is correct for the original gauge, called by some as a moving coil, I think it?s more akin to a moving Iron type gauge but that?s neither here nor there, other than a second hand unit I don?t think any repro?s are made of this type.
Your experiment has shown the requirement for the bi metal gauge to indicate correctly, by inverting the tank unit you have presented the gauge with a range of resistance which make sense giving you the gauge readings you see, your Empty 90 ohm is too low ( 230 required ) to give an Empty reading but producing an upscale reading, your Full 0 ohm is too low ( 30 required ) and the gauge reads well over the full mark pushing the needle to its mechanical stop.
The pic is of what I believe is the original fuel gauge in my own S3.
May help.
Ron.

[ericbushby]

Ron,
Thankyou for the picture. So that is the meter I need then.
There seems to be something wrong in your explanation in your first paragraph.
With a higher resistance in circuit the meter would move very little and read empty. As fuel is added then the resistance reduces and the current increases causing the meter deflection to move towards full.
That is how this bi-metal gauge behaves but a moving coil or iron type would do the same.

I think I will have to try to find a second hand gauge and have it overhauled if necessary.
Thankyou for your help

Has anyone out there got one spare please.
Eric in Burnley
S3SE DHC

[Craven]

Hi Eric,
My comment that this type of gauge is NOT a conventional Moving Coil movement but more akin to a Moving Iron gauge was because of how it operates. The pointer shaft, mounted in the familiar fashion on hair springs carriers a small permanent magnet, NOT a coil, there is NO moving coil in this instrument. The pointer move is controlled by interaction between fields produced by two electro magnets fitted in the case and the fixed magnet on the pointer. This instrument is a 3 terminal device and is NOT simply in series with the tank unit, a cleaver design which makes it essentially independent of applied voltage, this is why there is no voltage stabilizer fitted to early cars.
This instrument reads Empty when there is zero resistance between the connection ? T = Tank ? on the gauge and chassis.
When or if you find the correct gauge will have to turn your tank unit back up the correct way.
Ron.

[ericbushby]

Thankyou Ron,
There is still more to learn then.
I have not got a proper wiring diagram of the S3 and I have been using a combination of S2 and S4 drawings up to now. I will sort that out a.s.a.p.
I am not too bothered about having to turn the sender over as it comes out fairly easily.
If I cannot get the correct gauge I may have to buy a 230 ohm sender and then I can use this Bi-metal gauge on 10 volts.
I do not want to do that but it would work much better than the system I have now and is easily reversible.
Thanks again for your help,
I will update when I have a result,
Eric in Burnley
1967 S3SE DHC

[billwill]

You should find more information on all Smiths instruments in the books linked from this topic:
lotus-electrical-f38/smiths-instruments-documentation-from-the-horses-mouth-t29442.html?hilit=instruments

[Craven]

Hello Eric,
If you were to add say a 100 ohm resistor in series with your present gauge and tank unit it would bring your indicated fuel level within the scale of your Bimetal gauge.
Empty would become 190 and Full = 100 not ideal but you may have a better working indicator than you have now. Anyway may give you an idea to play with.
Mini Van uses a 230 tank unit 6 screw fixing although at an angle. ?19 I think.
Ron.

[ericbushby]

Ron,
I can see that you are thinking about me and I appreciate that.
I have tried several mods to the circuit with no satisfaction. An extra resistor just flattens the response and as you can see from my bench testing just 66 ohms causes a half full reading. It needs the full range to vary with the fuel level.
I have now decided what to do, that is until I change my mind.
The nice looking nearly new gauge has gone back in, having improved the lighting which is why I took it out in the first place. I will continue to use it like this for a while. I may fit a new sender when it is off the road for something else. That might not be long because the diff is leaking!!
It seems unlikely that I can recreate the original system, but I now know that it will work well with a 230 ohm sender arranged so that it gives high resistance at empty.
Thankyou for the minivan tip. I traded in a TR2 for a Minivan and used it to carry diving and camping gear to Spain a few times in the early 60`s.
I got ?200.00 for that TR !!


Bill,
Thanks for the Smiths info. That is very useful
Eric in Burnley
1967 S3SE DHC

[Craven]

Hi Eric,
The idea of adding in a fixed resistance was to bring both FULL & empty are on scale, the bimetal gauge requires a span of 200 ohms your tank unit provides a span of just 90 ohms so clearly it will only ever produce approx. a ? scale deflection, it?s just where on the gauge it fits.
Perhaps using your variable resistor, pick a position on the scale that would be suitable for empty, I think you are most interested in this point, adjust to give this, measure its resistance subtract 90 and use this value to pad out system but it must be more than 30 ohms. If you add in just 66 ? 45 = 11 ohms the halfway point, your full tank reading will still be off scale.
Your current tank unit can only give a span of ? the gauges scale, however SPYIDA make a system basically an amplifier that will, supposedly, match any gauge to any tank/sender unit. https://www.spiyda.com/magento/index.ph ... d-mk2.html
Must be some psychology between Minivan and Elan ownership, a couple of mates also have owned one before an Elan, mine was owned by a chimney sweep, probably why it was cheap but beggars can?t be choosers back in those days, never sure if it was just the smoky engine I could smell.
Ron.

[ericbushby]

I have a satisfactory result.
While the diff. was away for repairs I bought a new Minivan sender locally for ?18.00, (thankyou Ron.)
When checked it went from 20 ohms to 250 ohms.
I re-drilled the sender plate to put it upright and extended the arm by 2" to give the required 9" swing for the Elan tank.
The gauge now gives an approximate indication of the tank contents throughout the range and I am quite pleased with it.
Thankyou for help and comments.
Eric in Burnley
1967 S3SE DHC