MG-Cars.net

Welcome to our resource for MG Car Information.

Recommendations

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 - Temperature gauge sending unit

Hello everyone,
I had my cylinder head converted for unleaded fuel awhile ago, and ever since then my temperature gauge has been reading high. With the engine cool/cold and the ignition off, the gauge reads cold as expected. When you turn on the ignition before starting the engine, the gauge rises to very close to normal - where it used to read once the engine had warmed up. It goes up from there after the engine is started and warms up.
I noticed the machine shop installed the temperature sending unit using teflon tape to seal the threads. Has this changed the grounding of the unit? Could this explain why the high readings?
Glen 71 roadster
G Nicholas

'I noticed the machine shop installed the temperature sending unit using teflon tape to seal the threads. Has this changed the grounding of the unit? Could this explain why the high readings?"

I suppose that it could be the disparate reading. There is no reason to use any kind of thread sealant as the seal is the tapered end of the sender. The use of tefoln tape could be causing the problem you are seeing. try applying a ground lead to the edge of the sender and see if the reading straightens out. Cheers - Dave
DW DuBois

"When you turn on the ignition before starting the engine, the gauge rises to very close to normal"

As a North American 71 I'm assuming you have an electric gauge, therefore the problem is more likely to be electrical than anything else.

The system works by the resistance of the sender decreasing as it gets hot, therby increasing the current through the gauge. If the tape was insulating the sender from earth/ground you would get a low or no reading, not a high reading.

It sounds like they fitted an incompatible sender when they did the head job, perhaps breaking the original. It's not totally clear but the senders and the gauges seem to have changed twice. The first change occurred at different times for sender and gauge (Aug 71 for the gauge and Dec 74 for the sender), which should mean they are compatible. The second change for both was for the 77 model year. The sender type is usually denoted by the colour of the insulator, it seems that one if not both of the earlier senders was red, the later one was black.
paulh4

As Paul stated, the sender itself is most likely at fault. Once you start to tighten the sender into the cylinder head, any Teflon on the threads will be displaced by the pressure being placed on the threads. Unless the sender is loose enough to fall out, it should be grounding properly to operate the gauge as it did before the machine shop worked on the head. RAY
rjm RAY

As Paul has said this sounds like an incompatible sender. The gauge will be fine. The car has probably got a voltage stabiliser so will need a sender compatible with that. Sounds like it's broadly working OK, but just needs a little more resistance. The sender will have a part no on the spanner flats, which may give you a clue.
P A Allen

If the sender was *faulty*, the fuel gauge would read high. If that's OK then you can assume the sender is OK, and they put out an average of 10v to both gauges.

With one gauge reading high it *could* be the gauge but unlikely after a head job, so much more likely to be the sender not being compatible with the gauge (rather than not being compatible with the stabiliser). However if you can't get a compatible sender, or the one you have seems to be correct, you could tray recalibrating the gauge to the sender. They are very crude and a small adjustment can make a huge variation at the needle - this http://www.mgb-stuff.org.uk/electricstext1.htm#adjust is for the fuel gauge but they are exactly the same. However afterwards you would need to make sure that it moves into the H zone when the coolant temperature reaches 100C.
paulh4

Paul, of course you are quite correct that the sender doesn't have to be compatible with the voltage stabiliser, as such. What I meant was that there are some senders (now fairly rare), that look superficially similar, but are for use with gauges that don't require a voltage stabiliser. I've got one of these on my UK spec '72 MGB used in conjunction with a pre '64 Mini temp gauge. They were original fitment on other stuff at about that time, before everybody switched to voltage stabilisers. These were never original fitment on an MGB. That was a useful link, by the way, which I'll save to my favourites.
P A Allen

"for use with gauges that don't require a voltage stabiliser."

Interestingly the North American electric oil gauge used on Mk2 cars up to 1971 (capillary after that) was just such a beast. However it is complicated by the schematics showing it connected to the voltage stabiliser for the first year of use, direct to fused ignition thereafter. Complicated still further for that first year by the wire changing from green (fused ignition) to light-green/green (stabilised 12v) then back to green again as it moved between stabiliser and the gauges. How much of that was just an error in the schematics I don't know.

The sender contained its own stabilisation, and was the same as used in the 1989 Toyota Celica I had at one time.
paulh4

Thank you everyone for your insights. I guess I am in the market for a new sender. I did try something though, just to rule out a bad connection. I disconnected the green lead that connects to the sender, and inserted my simple circuit tester probe into the end of the green lead. It has a probe, a grounding clamp and a light in between. When I turned on the ignition, the light came on for 2 seconds, went off for a second, came back on for two seconds, etc. Not on steady, but pulsing rhythmically. I don't know anything about wiring but when I connect my tester across a battery, it lights up steadily.
No other wiring issues on the car.
G Nicholas

Just on last thing. Based on what I read here, I grounded out the green lead to the block, and the gauge went directly to hot. I think this is as it should be, and the only thing not working is the amount of grounding being permitted by the sending unit (too much). I'll be ordering a new sender. Thanks again.
G Nicholas

Should be green/blue, not green, but easy to confuse.

The pulsing on and off with the ignition on shows you have the original thermal (as opposed to modern electronic replacement) instrument stabiliser, which although counter-intuitive ironically does result in a stabilised gauge reading with these thermal gauges.

Yes shorting the green/blue to earth should result in full-scale deflection.
paulh4

I have replaced the sender in my B with the correct unit, only to have the temperature increase. Then I replaced the voltage stabilizer with one of the MOSS electronic units only to have both the fuel and temperature gauges become unstable (still reads hot). The car doesn't overheat and using a mercury thermometer at the radiator filler cap, I note 190 F temperatures on a 100 day with the A/C on. So I ordered a Smiths mechanical temperature gauge so that I know what is really going on. The gauge in the midget has always been dead accurate.
Glenn Mallory

"So I ordered a Smiths mechanical temperature gauge so that I know what is really going on."

The last time I replaced a capillary unit it read significantly lower than before, barely gets more than half-way from C to N whereas before it was always close to N. Test it in boiling water before fitting, and if it is a C-N-H gauge it should reach the edge of the H zone.
paulh4

This thread was discussed between 13/07/2016 and 29/07/2016

MG MGB Technical index

This thread is from the archive. The Live MG MGB Technical BBS is active now.