Matching fuel guage to sender

[donsayers]

Re: Matching fuel guage to sender

May I add one more thing: No attempt to modify the circuit will work until you replace the voltage regulator. The voltage regulator supplies the gauge with a time weighted average of 10 volts. it does this with a bimetallic strip that heats up in response to the battery voltage and turns the output on and off to give an average of 10 volts over a period of some seconds. Th regulator is usually to be found in a small rectangular package bolted to the back of the speedo.

There are solid state regulators available that give a true 10 volts that cost £12 or so, available from car wiring suppliers. I fitted a regulator based on a 50p reg from Rapid electronics.

This one modification has made the response of the fuel gauge more repeatable.

[macdonald.dm]

Re: Matching fuel guage to sender

Hi Don.

I was assuming that the clocks used by Keith DIDN'T actually have a voltage regulator installed.

I can't claim credit for the circuit, as it was given to Keith by a clever friend of his! However, its function IS based on the gauge being a current-measuring device.

The whole idea (I think) is that the gauge was calculated as requiring a certain current (around 50mA) when the resistance of a MATCHED sender was about 240 ohms whilst indicating an 'Empty' tank. However, the actual Ford sender used has only a resistance of 147 ohms when 'Empty'. The guy's idea was to tweak the voltage to provide 50mA with this resistance, arriving at around 7V.

As you have rightly said, there are two problems: (1) it's not linear, and (2) with a 'Full' tank, and low sender resistance, this WILL affect the voltage outputted from the Pot, making its operation even more unpredictable!

I've acknowledged both these points in my posts above, but feel it's still worth a punt as it would be a cheap and simple solution, and should at the very least provide an accurate indication of a tank that's approaching 'Empty'! I kinda hope that it'll also read pretty close to 'Full' when full, although it might be a case of it just going a little above 3/4's!

Or it might not really work properly at all...

[b_caswell]

Re: Matching fuel guage to sender

Bit late now but the resistor I put in the circuit was 16 ohms (not 160 hms). Im totally lost with the thread of this and need a couple of asprins. Easiest way is to draw a line on the gauge glass with a crayon!!

[robert]

Re: Matching fuel guage to sender

Wow, this has taken off since I last looked! Some great ideas here - just shows there are often many different ways to solve a problem.
Going back to Paul's post of 13 Sept, no problem using my circuit in your article. Also, the alternative components you list are all ok - nothing here is critical except that the op-amp should have an input range including ground, which the LM324 does so that's fine.
Finally, I now have a simple spreadsheet for calculating the values - you just feed in the min and max gauge current and sender resistance and it calculates the resistor values (no pots required). Let me know if want a copy.

[b_caswell]

Re: Matching fuel guage to sender

AH thats what the blank pics in misc were about. I can lend anyone a chinagraph pencil in black.(It writes on glass)

[b_caswell]

Re: Matching fuel guage to sender

See Berli forum (Dane 20 sept) he has just extracted the ballast resistor wire from a loom.. Is this any good?

[Dane_Rescueman]

Re: Matching fuel guage to sender

it is nearly one metre long, buried in the main loom, and decrepit. A proper resitor would be better for you. The ballast wire is designed to allow 7 volts at the coil under operating current which varies but is about 6 amps to my old coil. The wire is probably only about an Ohm or so. (in my case ∞ Ω !)

[Dane_Rescueman]

Re: Matching fuel guage to sender

OK, so I can't send ASCII coded symbols.
should read- - - - -
(In my case Infinity Ohms !)

[robert]

Re: Matching fuel guage to sender

Just to tie two loose ends together, the voltage regulator Don mentions (23 sep) is doing the same job as the 30 ohm pot, if I've understood correctly. So if you use a 7.2V regulator (or an LM317 adjustable regulator set for 7.2V), you should get the same result without the resistance wire, broom handle and sticky back plastic! Also, you can adjust the voltage with the '317 using a low power (cheap and won't get hot) pot if you want it adjustable.

[donsayers]

Re: Matching fuel guage to sender

No! The voltage regulator gives the dials in your dashboard a regulated 10 volts. The dynamo or later alternator in conjunction with the battery does not give a steady 12 volts. It can drop quite a bit when all the loads are on and you are idling, or go as high as 14 v when revving hard. Without the voltage reg the dials would go up and down in response to speed. Whatever circuit you are using you will have to have some sort of voltage regulator. A simple resistance wont work.

[robert]

Re: Matching fuel guage to sender

I was suggesting using a 7.2V voltage regulator INSTEAD of the resistor(s)! If you have a 10V regulator and then a (7.2v) potential divider, you'd be better off with a lower voltage (7.2V) regulator and no potential divider.

Anyway, given that we have a stable reduced voltage source, something else is wrong. The Smiths gauge I measured originally required 120mA full and 20mA empty and had a resistance of around 60 ohms. So, for the 10/180 sender for example, the empty condition would require a voltage of 20mA x (180+60)R = 4.8V. Now at full we have a current of 4.8V/(10+60)R = 68mA. This is just over half scale - just like fitting a series resistor! If you use a pot instead of a regulator it will be even worse.
Perhaps I have misunderstood how this solution works but it doesn't look right to me.

[macdonald.dm]

Re: Matching fuel guage to sender

Hi Don and Rob, and everyone else!

You are absolutely right in all you say about the limitations of this 30 ohm potential divider circuit.

The only situation I can see it POSSIBLY working is with the values given by Keith's equipment, and it's pure chance that his gauge and sender might JUST make it suitable for this simple set-up.

And certainly a 7.2V voltage regulator would be a much better option than the pot-div as it is, indeed, regulated! (Even without it, tho', the effect of voltages variations of <12V to >14V will be reduced by around half, but a regulator would certainly be better.)

Given Keith's values of 22 - 147 ohms for his Sierra sender, and the 33 - 240 ohms required by his gauge, the two sets of figures JUST HAPPEN to correlate well together, and the currents provided by both sets should be pretty close if the voltage driving the bigger set is 12V and the lower set 7V.

This would ONLY apply to Keith's set-up, and anyone else with a similar clock and sender.

I can see it being pretty accurate almost across the scale, with perhaps the gauge not reaching 'full' when full.

On the other hand, knowing the vagaries of electrickery, it might just end up being too unpredictable! However, it is such a simple and cheap solution that I can't help thinking it's worth a bash!

[robert]

Re: Matching fuel guage to sender

Ah, ok I get it.
However, it's only right if the voltage is stable and the gauge resistance is negligable - see image.

[macdonald.dm]

Re: Matching fuel guage to sender

Hi Rob.

What is 'Supply R'?

I see what you mean. Setting up would be a case of tweaking the output voltage to get a good 'Empty' reading. I suspect - and your figures would seem to confirm - that the gauge would then NOT every read 'Full', but I think it could come pretty close.

The effect of voltage fluctuations on the gauge would be reduced by nearly half, since the voltage used would be similarly reduced.

But, clearly an experiment!

How easy is it to set up and use the LM317 adjustable voltage regulator you mentioned? How much does it cost?

[robert]

Re: Matching fuel guage to sender

Hi Donnie and all,

Supply R is an approximation of the source resistance of the pot. As you mentioned earlier, when current is drawn from the pot's tapping the voltage will drop according to this resistance. It is calculated from the parallel combination of the two parts of the pot so if it was exactly in the middle it would be 15 in parallel with 15 = 7.5 ohms. You could work it out exactly from the position of the pot but it's quite tedius so I just guessed. It's known as the Thevenin equivalent (I've probably spelt that wrong). Eliminating this resistance is the main benefit if the regulator (as well as imunity from voltage variations and reduction in wasted current and thus heat).

I don't mean to sound negative, by the way, this IS a good solution and I think it WILL work.

Finally, the LM317 costs about £1 for the TO220 version and is set up as shown (the voltage is set by the resistors, the caps are necessary for stability and the diode is just for protection).