View Full Version : TR4/4A Fuel Gauge TR4A

05-10-2012, 03:08 AM
I have current to the left side of the gauge via the green/black wire with the ignition turned on.

I have no current to the right side of the gauge from which another green/black wire no doubt heads off to the sender in the tank.

At the sender in the tank I have good ground and zero current at the green/black wire.

I can only conclude that my fuel gauge is shot.

Sound correct?

05-10-2012, 06:57 AM
Sounds right. Here's a schematic (https://www.advanceautowire.com/tr24a.pdf) that shows current coming from the voltage stabilizer, through the gauge to the sender in the tank. Sounds like you're losing something through the gauge.

05-10-2012, 09:32 AM
The wiring schematic only shows a green/ black wire between the gauge and the sending unit. Power is via a light green/green wire coming from the voltage stabilizer. You say that you have each side connected to a green/black wire so check that. Then be certain that you have continuity between both ends of the green/black wire (check at the gauge and the sending unit for continuity)just to be sure that it's the correct wire between the sender and the gauge.

Connect a temp ground wire to the other gauge post and see if you get movement of the gauge while having a light green/green power from the voltage stabilizer to check the gauge. It should show full tank. That would prove the gauge good or bad. If you have the wires reversed it will move towards empty so be sure you have the wires on the correct posts. If the gauge works then focus on the sending unit as the problem.

The sending unit controls the ground to the gauge and as it varies the gauge shows different readings.

Hope this helps.

05-10-2012, 09:47 AM
Are you actually measuring current, or voltage?

My next step would be to measure the resistance across the two gauge terminals, with the wires disconnected. If it is open or more than about 200 ohms, the gauge is bad.

<div class="ubbcode-block"><div class="ubbcode-header">Quote:]If you have the wires reversed it will move towards empty [/QUOTE]
I disagree. The TR4-6 gauges do not care which wire is which. They work by heating a bimetallic spring, and the heater puts out the same amount of heat regardless of polarity. No current should read empty, maximum current reads full.

05-10-2012, 10:10 AM
Thanks. I just used a test light. I can check voltage. How much would a expect at that point? The feed off the Voltage Stabilizer is pulsing. On off on off on off. But the current at the gauge is constant, no pulsing. I assume that's normal, maybe how it "stabilizes."

I'll also check resistance across the posts.

Of course another question. While testing all this I let my lights stay on for 15 minutes. The amp gauge sure got hot. Normal, too?

05-10-2012, 11:16 AM
It's normal for the original type "stabilizer" to flash on and off; in fact that is a good test for whether it is functioning properly or not. (Aftermarket "electronic" ones don't flash.)

But you should see that same flashing at the gauge, unless the sender is reading empty or the wire is shorted to ground. With the tank near empty, the voltage on that side may be too low to light your test lamp. Mine isn't visible below about 5 volts (which would be around 1/2 tank).

Nope, the ammeter shouldn't be getting that hot for just the lights. Sounds like probably an iffy connection at the back of it. Try pulling the quick connects off and polishing up the tabs with a Scotch-Brite pad or similar mild abrasive; plus crushing the quick connects a bit to tighten them up. If that doesn't help, you may need to replace the quick connects.

I've been test-driving a product that is supposed to help clean up contacts and keep them clean https://www.caig.com/ I can't really say if it works or not though, haven't been using it long enough.

05-10-2012, 08:27 PM
The pulsing light at the Voltage Stabilizer is nice and bright.

The test light to the left terminal of the gauge is also bright and, yes, your are correct Randall, it also pulses. That wire is a green/black shielded connector.

The opposite side is also Green/Black but the connector is not shielded, and that one shows "zero" life using test light.

On a voltage meter the left green/black wire, with the pulsing current, shows fluctuating voltage of 5 to 11 volts.

The resistance between the poles of the gauge, measured by my multimeter, is .062.

I'm in awe about all I don't know. But I love the education. So what does all this mean?

05-10-2012, 10:30 PM
With the tank near empty, the voltage on that side may be too low to light your test lamp. Mine isn't visible below about 5 volts (which would be around 1/2 tank).

I'm lost on the theory I guess. I had no voltage at the sender, and I assumed that meant no voltage traveled through the gauge to the sender side. When current goes from the gauge to the sender, what is happening? How does it work?

You guys here all taught me to be too finicky. So, I've already pulled my gauge apart to see inside. I see some awfully weak and frayed wires. It would take some real skill to solder them back.

Please take a look, and share your thoughts please. You can see the little frayed wires:

05-10-2012, 10:33 PM
Here's another:

05-10-2012, 11:58 PM
Hmm, I've never been very good at this part, but let's try a quick lesson in electricity:

Voltage is a potential, it doesn't really "flow". The usual comparison is water pressure. Imagine for a moment, a dam with water behind it, and a closed valve at the bottom. The pressure at the valve is kind of like voltage, it is just a potential for flow.

Current is kind of like the water flowing somewhere. So if you open the valve, you get some water flow. How much flow depends on how high the water is behind the dam (the voltage), plus how far you open the valve (which is comparable to resistance). If you measure the pressure before the valve, it will remain nearly the same, regardless of how much water is flowing. If you measure after the valve, you'll get nearly zero, again regardless of how much water is flowing.

After that, the comparison kind of breaks down, so we'll let it go at that. The important bit is that current and voltage measure different aspects of electricity.

The voltage stabilizer doesn't stabilize the voltage, what it stabilizes is the gauge reading. Since the gauge mechanism responds relatively slowly, the VS can work by switching between full battery voltage, and zero volts. (You didn't measure zero because your voltmeter doesn't respond quickly enough.) The VS is needed because the battery voltage is not a constant 12.0 volts but can rise to as much as 15 volts under some circumstances (like driving down the road on a cold day) and possibly drop below 12 volts (idling for a long time with the head lights, heater, and brake lights on).

Inside, it has a coil of resistance wire, effectively an electric heater, wrapped around a bimetallic strip. The bimetallic strip consists of two different kinds of metal bonded together, with different rates of expansion with temperature. So when it gets hot, one side expands more than the other, and forces the strip into a curve. The end of the strip in the VS has a contact that closes when the strip cools, and applies ignition voltage to the heater (and to the gauges). The heater heats the strip up, the contacts open and the cycle repeats.

This sounds very complicated and Rube Goldberg, but for 1960 technology it is actually very simple, rugged, and most importantly, cheap. There are a surprising number of original voltage stabilizers that still work today! (Quite a contrast to my 1970 Audi, which used a crude electronic regulator that seemed to fail every 3 or 4 years.)

Inside the gauge is another heater and bimetal strip (on the left in your photos above). But this time, the end of the bimetal strip is connected to the indicator needle, through a linkage that magnifies the movement of the strip. How hot the heater gets depends on how much power it dissipates.

Since one side is supplied with (essentially) a constant 10 volts, the resistance of the sender is what controls the current through the gauge (and the voltage across it). Power is voltage times current, so the heater temperature (and hence needle position) depends on the resistance of the sending unit.

Clear as mud? There will be a quiz on Friday!


05-11-2012, 12:27 AM
Back to an earlier question: When the fuel tank is nearly empty, the sender resistance should be fairly high. So if you measure the voltage at the sender side of the gauge, it would be nearly the same as the output from the VS. The heater gets very little power (small voltage times small current), the strip doesn't bend and the gauge reads empty.

As the level in the tank rises, the sender resistance goes down, so at full, the sender resistance is nearly zero. If you measure the voltage at the sender side of the gauge, it will be nearly zero. The heater gets a lot of power (large voltage times large current), the strip bends to maximum and the gauge should read full.

What I'm not sure about is the .062 measurement you reported. Very few ohmmeters read as low as .001 ohms (the leads have much more resistance than that), so I'm guessing that it was actually .062 Kiloohms (62 ohms) or .062 Megohms (62,000 ohms). 62 ohms would be fairly close I think, while 62,000 is much too high.

But either way, you saw a definite voltage drop across the gauge with (I assume) the gauge reading empty. So something is wrong with it.

05-11-2012, 04:26 AM
Where on earth did you learn all that?

So, can I apply voltage and test that gauge, or will I blow it without the VS in between?

The sender side is that side of the gauge that I said was dead, correct?

I'm not sure I'm understanding the direction of the current.

It goes from the VS, to the gauge, and from the gauge to the sender?

Then the resistance at the sender is what causes the needle to move or behave the way it does.

But there must be measurable current or voltage at the sender, or something is wrong at the gauge.

05-11-2012, 07:51 AM

Be sure to check that ammeter's connections too. I had a problem on mine. About 3 years ago I mistakenly left my lights on for 30 minutes while car was parked in garage. By the time I discovered it, a little smoke was coming from the dash. Checked everything and appears no harm done. So I carried on.

Then later on I noticed each time I took the car out, the ammeter face steams up in the first few minutes of each drive. No other issues so, I carried on.

Then, about a year after I am driving the car, switch on the headlights and the engine flat dies... What the...? With Randall's help, I test the voltage at the battery and get 12.4 volts. I then test the voltage at the ignition coil and I'm getting only about 6 volts. Hmmm. Traced it back to corroded/burned connectors on the ammeter.

I guess when I left those lights on, the copper connectors were just corroded enough to cause resistance, which causes heat, which causes more corrosion....endless circle. It was so bad the rubber covers on the spade connectors had melted.

So, pulled ammeter, replaced blade and spade connectors with new and new rubber boots. Added some dilectric grease for good measure.

Result is a full 12.0 volts at the coil now. And I cannot tell you how much better the car starts. Fires straight-up the instant I turn the key!

Who'd have known?! So, a warm ammeter is a warning of things to come. I'd start with a test of your volts at the coil.


05-11-2012, 08:03 AM
Where on earth did you learn all that?
Oh, here and there. I like knowing how things work and I've been messing with electricity since I was in grade school. In high school, I collected dead televisions just to see if I could bring them back to life. And I even majored in EE for awhile (but eventually switched to CS when I learned that EEs don't usually get to build things).

<div class="ubbcode-block"><div class="ubbcode-header">Quote:]So, can I apply voltage and test that gauge, or will I blow it without the VS in between?[/QUOTE]
Smiths warns against applying more than 10 volts to the gauge, so it's probably not a good idea. But I have done it for just a second or two (long enough to see the needle climbing towards full) without any apparent ill effect. The thing is, you don't want to let the heater get too hot, as even if it doesn't burn out, it may permanently distort the bimetallic strip.

<div class="ubbcode-block"><div class="ubbcode-header">Quote:]The sender side is that side of the gauge that I said was dead, correct?[/QUOTE]
<div class="ubbcode-block"><div class="ubbcode-header">Quote:]I'm not sure I'm understanding the direction of the current.

It goes from the VS, to the gauge, and from the gauge to the sender?
[/QUOTE] Well, <span style="text-decoration: underline">direction </span>of current is a slippery concept. But for simplicity's sake, yes, we can look at it that way. It only flows in a complete circuit, so the circuit in this case is from the battery "hot" terminal, through the ignition switch, VS, gauge and sender, back to the battery ground connection.
<div class="ubbcode-block"><div class="ubbcode-header">Quote:]Then the resistance at the sender is what causes the needle to move or behave the way it does.

But there must be measurable current or voltage at the sender, or something is wrong at the gauge.
Exactly. But with the sender at "full", there will be very little voltage (and relatively high current), so it's important to consider which one you are actually measuring. Some specialized meters can read current from a single probe, but practically all general purpose ones require that the circuit be interrupted (so the current flows through the meter). Since (I am assuming) you are actually measuring voltage, you wouldn't necessarily see any voltage at that point.

The relevant equation is known as Ohm's law, which states that the voltage (volts) is equal to the current (amps) times the resistance (ohms). So when the sender is near the full position, it's resistance goes to nearly zero. The current is limited by the resistance of the gauge, so the voltage across the sender also goes to nearly zero.

BTW, here is a nice photo (courtesy Jim Miller via the British V8 web site) of the VS internals. In the upper LH corner you can see the heater coil still wrapped around the bimetallic strip. In the lower RH corner you can see the integrated circuit voltage regulator being used to replace the strip &amp; contacts.

05-12-2012, 01:01 PM
RJS, thanks, I'm doing that now. I remember a bad ammeter in my TR4 in high school. But things were different then. I just went and bought a new one at "Leonard's Import Parts" two miles from home.

05-12-2012, 02:03 PM
so I'm guessing that it was actually .062 Kiloohms (62 ohms) or .062 Megohms (62,000 ohms). 62 ohms would be fairly close I think, while 62,000 is much too high.

Right again. It's 62 ohms, meaning my resistance is probably OK and those slightly frayed wires I might just "let well enough alone."?

Fiddling with copper wires the thickness of shirt threads wouldn't be easy, and I can't imagine what I might ruin something with a soldering iron. I'm seeing an electrical shop today to see what they have.

Also, rechecked voltage. I do have voltage at the green/black sender wire at the sender. It fluctuates as the VS apparently tells it to, just like the direct side of the gauge. This all seems to say I've got good continuity and voltage where it needs to be.

So, does all this mean I've got a bad sender and have to drain and remove my gas tank and go through at that again? I can do it; just hate it.

BTW, the black wire at the sender shows a good negative, so I don't see that as an issue.

Finally: If the sender is bad wouldn't I see at least some heating of that bi-metallic arm in the gauge, and at least some reading?

Another pic for kicks:

05-12-2012, 04:48 PM
Quick test, jumper the sender wire to ground at the sender. If the gauge now climbs towards full, the sender is the problem.

BTW, one of those little cubical Facet fuel pumps works well as a transfer pump to empty the tank. If you get one with the right pressure rating, you can also carry it as an on-board spare for the mechanical fuel pump. (Mine is wrapped in a shop towel and in a big Ziplock bag in the center of the spare tire.)

Aircraft Spruce has about the best price on new ones, while used ones frequently go for about half that on eBay.

05-12-2012, 08:51 PM
If I jump direct to the green/black at the sender, meaning from the green black to ground, then my gauge works!!!

Those gauge wires still look frayed so what if I dab some conductive epoxy on them--no good or a good precaution?

And the question: does this mean its the sender that's bad? I guess so Randall. Thanks for the help!!!

05-12-2012, 09:40 PM
If you just filled the tank (rather than manually pulling up the lever), then it might be a bad float. Or it might just be a broken wire inside the sender. Either way, you're going to have to pull it out.

I'd leave the wires alone, personally. They have to flex to some extent as the gauge moves, and I suspect be more likely to break right at the edge of the epoxy.