Thread: Error in Fuel Gauge

I'm going to have a crack at solving this, I'm not an auto electrics pro, but I dabble and I like a challenge, so here goes:

First thing to check is the wire that runs from the fuel guage to the sender unit atop the tank, because it sounds like you have a short circuit. Make sure this wire is connected to the fuel sender unit ONLY and is not short circuiting back to the chassis (earth). It could be pinched somewhere out of sight under the bulkhead. Try disconnecting the original wire off both the sender unit and the gauge and run a temporary new wire to see if a proper reading is now made. If it seems to be working now, then replace this wire properly.

If it still reads FULL when a new wire is connected, then it'll have to be a process of elimination, seeing which parts in the chain are working properly, in order, to fully rule things out as you go.

So, start at the voltage regulator. Turn the engine on and check the voltage coming in is in the normal range, like around 14V. Read the output from the regulator for a constant 10V. Remove the wires from the regulator to the gauges and with the engine running, read the voltage from the regulator to earth (chassis or engine, etc). It should be an unwavering 10V, even with changes in engine revs. Old regulators wear out and stop delivering the steady, lowered voltage that the guages use as a reference point for a correct reading.

If you have a steady 10V, then reconnect gauges and move on. If not, replace. You can replace with an inexpensive modern solid state regulator for better performance and durability.

Next, check the guage itself.
Maybe try this (I think it would work), wire up from the 10V feed to the fuel gauge, then from the guage to a known working water temp sensor (it's a varible resistor like the fuel level sender). Earth the sensor back to the chassis and acquire some cold, warm and boiling hot water. You should be able to get stable, varied readings on the fuel guage as the tip of the sensor is placed into the different temperature waters.

Alternatively, you could wire up to the water temp sensor in your engine when the coolant is warm to at least see if the guage will give a different reading to the normal FULL it is displaying (disconnect the wire to the water temp gauge from the sensor first).

If, in all the different temps, the guage still reads FULL!! like it normally does, then it's faulty and will need replacing.

If the guage seems to be behaving correctly now and shows differing, but stable and consistent readings from the different temperatures the sensor is immersed in, then the problem is likely to be from the varible resistor in the fuel sender unit itself.

Replacing the sender unit should have the problem solved.

That's the best I can offer, I'm keen to hear how it goes.
Cheers, Nige

Originally Posted by captain_funbeard

I'm going to have a crack at solving this, I'm not an auto electrics pro, but I dabble and I like a challenge, so here goes:

Originally Posted by captain_funbeard

First thing to check is the wire that runs from the fuel guage to the sender unit atop the tank, because it sounds like you have a short circuit. Make sure this wire is connected to the fuel sender unit ONLY and is not short circuiting back to the chassis (earth). It could be pinched somewhere out of sight under the bulkhead. Try disconnecting the original wire off both the sender unit and the gauge and run a temporary new wire to see if a proper reading is now made. If it seems to be working now, then replace this wire properly.

If it still reads FULL when a new wire is connected, then it'll have to be a process of elimination, seeing which parts in the chain are working properly, in order, to fully rule things out as you go.

So, start at the voltage regulator. Turn the engine on and check the voltage coming in is in the normal range, like around 14V. Read the output from the regulator for a constant 10V. Remove the wires from the regulator to the gauges and with the engine running, read the voltage from the regulator to earth (chassis or engine, etc). It should be an unwavering 10V, even with changes in engine revs. Old regulators wear out and stop delivering the steady, lowered voltage that the guages use as a reference point for a correct reading.

If you have a steady 10V, then reconnect gauges and move on. If not, replace. You can replace with an inexpensive modern solid state regulator for better performance and durability.

Next, check the guage itself.
Maybe try this (I think it would work), wire up from the 10V feed to the fuel gauge, then from the guage to a known working water temp sensor (it's a varible resistor like the fuel level sender). Earth the sensor back to the chassis and acquire some cold, warm and boiling hot water. You should be able to get stable, varied readings on the fuel guage as the tip of the sensor is placed into the different temperature waters.

Alternatively, you could wire up to the water temp sensor in your engine when the coolant is warm to at least see if the guage will give a different reading to the normal FULL it is displaying (disconnect the wire to the water temp gauge from the sensor first).

If, in all the different temps, the guage still reads FULL!! like it normally does, then it's faulty and will need replacing.

If the guage seems to be behaving correctly now and shows differing, but stable and consistent readings from the different temperatures the sensor is immersed in, then the problem is likely to be from the varible resistor in the fuel sender unit itself.

Replacing the sender unit should have the problem solved.

That's the best I can offer, I'm keen to hear how it goes.
Cheers, Nige

G'day Nige,

Thank you for your reply.

So would it be the voltage regulator for the instruments - Part Number 579468 which screws in behind the speedometer? Or do you mean a voltage regulator fitted on to some alternators? Some sources also call this part the instrument voltage stabiliser too don't they?

If it is the instrument regulator affecting the fuel gauge then this could explain why the temperature gauge wasn't reading either. It was staying with the needle sitting down on the cold after the engine was running for 20 minutes. I also noticed that the temperature gauge needle was fluttering too.

Your message might be killing two birds with one stone

I will let you know this weekend when I get to play ... ooops work on the Land Rover


Kind Regards
Lionel

Things I have read on this forum:
1. Put a decent earth wire from the sender mount to the chassis; don't rely on the tank being earthed.
2. There is an electrical thingy in the instrument cluster, which controls the current to the gauges. JDNSW knows what it is called; in the meantime search "fuel gauges" on this forum.

Check that the sender unit does actually work. I once had a Morris Minor which had a stuffed one, because someone had connected the battery backwards and stuffed the sender (wrong polarity),

Cheers Charlie

Hello Charlie, Nigel & All,

The first part number was off my old 1975 printed Land Rover Catalogue (579468). My new Brooklands Catalogue 1988 has what looks like the same part listed as 148876. Seems to be more hits on the internet with the newer parts number. See attached photo.

Kind Regards
Lionel

Three likely causes for the symptom.

1. (more or less covered above) Voltage stabiliser malfunctioning. This can be the device itself playing up, or simply that it is incorrectly earthed. On the Series 3, it is mounted on the back of the Speedo (Series 2a negative earth on the firewall behind the instruments and relying on the mounting for an earth. This gives an average of 10v, by switching on and off with a period of a second or two - if functioning correctly, a meter at the sender will show this fluctuating voltage. There may be aftermarket IC versions that show a steady 10v, but I have not encountered one.

2. With the change from positive to negative earth in 1967, the type of gauge changed. If the gauge and sender of the two types are combined you will get this effect - the tank and/or gauge may have been swapped - it is unlikely the gauge will have been, so probably the sender. The change in resistance is in the opposite direction, and the total resistance is different as well. (My guess is this is most likely)

3. Shorted wire between gauge and sender.


A poor tank earth would not cause this symptom, although it will cause incorrect readings.

Hope this helps,

John

Originally Posted by Lionelgee

Hello Charlie, Nigel & All,

Seems to be more hits on the internet with the newer parts number. See attached photo.

Kind Regards
Lionel

That's the regulator I was referring to. Mounted on the back of the speedo. I've only taken multimeter readings off a new electronic one I bought last year. It outputs 10V steadily. Wasn't aware that the original ones achieved this voltage by switching on and off. That's probably why they wear out. The gauges have a bimetal strip that heats on one side when current passes through, this current is affected by the variable resistor in whatever sender unit it is connected to. This in turn affects the needle. The gauges are slow to respond, so an on/ off current would still show a steady reading on the gauge.
I think a flickering needle might be evidence of a faulty regulator, not sure. Sounds like you've got some investigating to do, good luck.

Here's a link to the new solid state regulator/ stabilizer I'm using:

SOLID STATE VOLTAGE STABILIZER Land rover series 2,2a,3 | eBay

It's made up of modern electronics sealed in a box with silicon so very moisture proof. Seems good quality, the seller has them listed constantly so obviously makes a lot of them.

Originally Posted by captain_funbeard

That's the regulator I was referring to. Mounted on the back of the speedo. I've only taken multimeter readings off a new electronic one I bought last year. It outputs 10V steadily. Wasn't aware that the original ones achieved this voltage by switching on and off. That's probably why they wear out. The gauges have a bimetal strip that heats on one side when current passes through, this current is affected by the variable resistor in whatever sender unit it is connected to. This in turn affects the needle. The gauges are slow to respond, so an on/ off current would still show a steady reading on the gauge.
I think a flickering needle might be evidence of a faulty regulator, not sure. Sounds like you've got some investigating to do, good luck.

Here's a link to the new solid state regulator/ stabilizer I'm using:

SOLID STATE VOLTAGE STABILIZER Land rover series 2,2a,3 | eBay

It's made up of modern electronics sealed in a box with silicon so very moisture proof. Seems good quality, the seller has them listed constantly so obviously makes a lot of them.

Hello Nigel AKA Captain Funbeard,

Thank you for the reply and the link to the electronic version of the device. I do have a couple of spare original bi-metal type ones sitting happily on a shelf ... somewhere, which were bought NOS. I just need to find them . I will let everyone know who have been kind enough to reply to my posts how I go with fitting a new one over the weekend.

By the way - in the near future I intend to fit some more classic Smiths gauges to replace the mix-match extra gauges that have been fitted to Baldrick by previous owners since 1976. I read on one Smiths gauges supplier's site that a number of gauges require additional instrument voltage regulators to be fitted. How many gauges can the electronic version handle?

Kind Regards
Lionel

Hmm... Couldn't say. Maybe query the guy who builds them directly. Good question, I'll have to check that for myself too.
My original bi-metal regulator was powering a smiths and a VDO gauge in addition to the stock gauges and seemed to be performing fine, but unsure how the extra load would affect a solid state regulator. I'll contact him and get back to you.