MG-Cars.info

Welcome to our Site for MG, Triumph and Austin-Healey Car Information.

Parts

MG parts spares and accessories are available for MG T Series (TA, MG TB, MG TC, MG TD, MG TF), Magnette, MGA, Twin cam, MGB, MGBGT, MGC, MGC GT, MG Midget, Sprite and other MG models from British car spares company LBCarCo.

MG MGB Technical - Temp gauge

The temperature gauge in my 71 B reads too high. It starts out at "N" and rises from there. My mechanic suggested a grounding problem so I poked around and found the voltage stabilizer on the firewall. I could barely contort myself just enough to see it, and feel it. The two spaded wires were both tight, but the unit itself was mounted a little wobbly. There seemed to be a place for two securing screws, but only one was fitted. Could this wobbly voltage regulator mount result in high temp readings?
Thanks
G Nicholas

The voltage stabilizer needs to be tightly grounded.
Glenn Mallory

The voltage stabilizer is position (horizontal) sensitive as well as needing a good earth. It also controls the voltage to the fuel gauge as well. Hence if it's faulty it will affect the accuracy of that unit too. Am I right in assuming that the gauge is reading "N" with a cold engine as soon as you turn the ignition on? Try disconnecting the wire from the sender unit. It should then go to zero. If you then ground it, see if it shows "H". If so it's not the gauge but probably the sender unit. I have to say I'm not familiar with the electrical temp gauge, so this is how I assume they work.
Allan Reeling

If it's a stabiliser earthing problem (as seems likely) the fuel gauge would also read high. The gauge readings will also vary with system voltage i.e. be higher with the engine running and alternator charging compared to just turning the ignition on. The attached shows the effects, from left to right, of a working stabiliser, one with a bad earth with just the ignition switch on, and one with a bad earth with the engine running. As you can see the difference between first and last is almost half a gauge, which ties in with you seeing N with a cold engine and rising from there.

As far as I'm aware only one screw is fitted.

paulh4

Thank you everybody for your answers. At Allan's suggestion today I disconnected the green wire at the sensor, and with the ignition on, and a cold engine, the gauge registered "C". I then connected a continuity tester to the green wire and to ground. The tester has a small bulb in it. With the ignition on, the bulb pulsed, close to one blip per second, and the gauge read half way between "N" and "H". I then grounded the green wire directly to the block and the gauge went right to "H".

As per Paul's suggestion I started the car and watched the gas gauge. It recorded the same as when the ignition was on, and the car not running. I recently ran the gas much closer to empty than I normally do, maybe 1/8 of a tank, and had no unexpected out of gas issue. I had previously asked my mechanic to order a temp gauge sending unit, and will let him sort this further.

It might take a few days, but I'll keep you posted.
Glen
G Nicholas

Glen
If you are fitting a new temp sensor--
Don't over tighten it as this can damage it and change it's output
And no threadtape as this 'can' insulate it from earth

willy
William Revit

The electric temp and fuel gauge systems are identical in how they function. The pulsing does show that the stabiliser is switching on and off as it should, which it wouldn't do if the earth was bad. Neither that nor the visual test would show whether it was outputting higher than it should, but the fact that the temp gauged showed 'normal' with a cold engine when the fuel gauge had read has little as 1/8th without running out does indicate it is the temp gauge system that is the problem.

There were three gauges and three senders over the years, and one of the senders is not compatible with the other two gauges, only its own gauge. Getting the wrong mix will read either high or low. The senders had different coloured insulators - red for early and black for later, which I think dates from the start of the 77 model year.

http://www.mgb-stuff.org.uk/electricstext1.htm#tempgauge has more info, including a chart showing the relationship between resistance and gauge reading, which is from an 'early' gauge and sender.
paulh4

The best way to solve the electric temp gauge inaccuracies is to fit a mechanical gauge with a capillary tube. I tried various senders and stabilizers and one combination read higher than the next. Changing stabilizers also affected fuel gauge readings.

When adapting the capillary gauge, be sure to check the accuracy with a mercury cooking thermometer on the stove prior to fitting. I had to pull the needle off and make a minor correction at boiling to make sure the mechanical gauge was accurate at the high temperature end.

I got the Smiths mechanical gauge on Ebay. It does read in °C instead of °F but what was Fahrenheit thinking with 32° and 212° anyway?

Regarding stabilizing the voltage for proper fuel gauge readings, I used an adjustable stabilizer (generic from Ebay) and set it to read "F" with the tank topped off. It reads exactly "E" when it runs dry to I was at least able to get the electric fuel gauge to read accurately, albeit with a bit of work.
Glenn Mallory

"what was Fahrenheit thinking with 32° and 212° anyway?"

http://mathforum.org/library/drmath/view/52561.html

Which begs the question of why he chose 96 for the temperature of the human body.
paulh4

Thank you everybody for the excellent comments. I promised an update so here it is. Doing some of the testing suggested, I managed to convince myself the sender unit was at fault. I installed a new one and things changed, but not perfectly. The gauge now starts at cold, rises to normal quickly, and after a few minutes settles about 2/3 of the way towards hot.

I purchased a digital cooking thermometer with a probe that I inserted into the top rad tank. On a cold motor I got a reading of 71 (garage temp I guess). I held onto the top rad hose with the engine running and the temp steady at 71. After a couple of minutes, the hose warmed and the temperature began to climb quickly to 170. After a run on the road on a hot day the gauge peaked out at 177. I think this shows the cooling system is just about perfect. So I think the bottom line is my previous sender was bad. The current one is working but mismatched to my gauge which is also working. I guess my new normal in 2/3 towards hot. Thanks again.
G Nicholas

"I guess my new normal in 2/3 towards hot."

Should be OK, but for confirmation you could blank off the rad to get the coolant temp up towards 212, and check the gauge reaches H by then.
paulh4

This thread was discussed between 01/05/2017 and 12/06/2017

MG MGB Technical index

This thread is from the archives. Join the live MG MGB Technical BBS now