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 - Temp Gauge Troubles

Hi All,

I've got a few questions here, all related to the temperature gauge on my '63 B.

The temperature gauge always reads high - for instance, at 65 degrees F ambient, it's reading 100. At 160 - 170, it's reading 220. Sometimes, the needle keeps climbing (when the temperature isn't) and even gets into the the oil pressure portion on the gauge (I have an infrared non-contact thermometer, so I can see what the actual temperature is). It drives me crazy because I've had the car less than three months and I can't convince myself that it's not really overheating.

Does this sound like a sensor problem, or a gauge problem? Is it possible to replace the sensor without replacing the gauge, too?

Can the gauge or sensor be repaired? Is there anybody out there that does the repairs?

Until I have the time and money to replace/repair the gauge, I'm thinking of getting a gauge from JC Whitney or somebody, and just tucking it up under the dash. Could someone please tell me what type of thread is on the temperature sensor that screws into an 18G block?

When the time does come for removing the gauge, I need to get the speaker panel off so I can reach up to the gauge. I tried to remove the panel when I replaced the light and wiper switch, but it seems like it's totally wedged between the bottom of the dash and the gearbox tunnel. Is there a way to remove it without destroying it?

OK, that's all for now.....

Cheers,
Greg
Greg Van Hook

The "Safety Gauge" encompasses two seperate gauges in one. The temp gauge, which uses a Bourdon tube, is faulty. It can be rebuilt, but new ones are available for about $150 for a complete unit. RAY
RAY

As Ray said this is a mechanical gauge both for temperature and oil pressure. Oil tubing and capillary tubing from the bulb in the cylinder head go through the bulkhead and under the dash.

Generally what happens is the capillary gets broken and you get nothing as far as temperature. You could maybe open it up and try to calibrate.

I think it's hard to find aftermarket temperature senders or bulbs to fit our cylinder head without leaking. I believe the issue is the taper seat where the adapter screws in. You go to an elecrical gauge and sender for a later MGB temperature. But, then you have to find another place for an oil gauge.

I'd be tempted to replace with original dual purpose gauge. But, it's pricey at $175 in Moss. An alternative would be to have Nisonger (www.nisonger.com) rebuild your gauge. Check ebay also. Years ago I replaced mine that had a broken capillary with a used one that has worked fine and is accurate.
Robert McCoy

Greg - Be aware that the infrared thermometers don't read the same on different materials or even the finish difference on the same material. See the article at: http://autospeed.com/cms/A_110883/article.html that talks to this issue and what the emissivity is of different materials.

Taking the speaker panel out will not gain you a whole lot of extra room to get at the temperature/oil pressure gauge as there is a fixed panel behind the plastic one. I have always found the best way to get to items in that area of the dash is to remove the radio first.

The sensor thread in the head is 5/8 - 18 SAE.
Cheers - Dave
David DuBois

OK, I just emailed Nisonger for details. Looks like they do nice work.

I don't have a radio. Just the blanking plate. I tried half-heartedly taking it off when I replaced the switches, but was afraid I'd break that, too!

I did come up with a way to get the switches in, since my hands don't fit up behind the dash. I taped a string to a piece of coathanger wire, and fed it through the switch hole from the front of the dash to the footwell, leaving some string hanging in front of the dash. Then took the string off of the rod, and taped it to the switch lever. Then pulled the string back out through the whole, and the lever along with it. Then I also had something to hold onto while threading on the bezel nut thingy.

Greg
Greg Van Hook

I'll second David's comment on emissivity of materials, since use of thermal cameras in a lab environment has been part of my job description. Metal tends to transport heat away, rather than emitting it as infrared energy, so it will look cooler (to the infrared instrument)than it actually is. This is a low-emissivity material. Coating the metal with a low thermally conductive material (high emissivity) will yield more accurate results with an infrared thermometer.

Here's a solution to calibrate your infrared device. Place a thermocouple on the surface to be measured, and use the thermocouple measurement to determine the infrared thermometer's accuracy in measuring similar surfaces. My infrared instruments have always accommodated a way to change the emissivity of the surface measured, so the infrared device will calculate the correct temperature based on the surface's emitted energy.

Having written the above, I read the article Dave notes. It's a good article, but use the thermocouple technique to determine the accuracy of your instrument, and also to help determine the emissivity of the surface you want to measure.

Human skin has an emissivity of 0.98, simply because we generate our own heat energy.
Fred Doyen

Thanks Fred,

I know where I can borrow Fluke meter with a thermocouple.

Greg
Greg Van Hook

Actually, I guess I'll find out, but how much of a difference could there be between an infrared reading and a thermocouple reading as a result of emissivity? Like 10s of degrees? If it's just a few degrees, it doesn't really matter in this case.

I think I'm pretty close, because when my 180 degree thermostat opens, it still registers on the gauge (albeit at about 220 degrees or more), while my infrared readings are all in the mid 170s on the block. So I know I'm within 10 degrees anyway.

Thanks,
Greg
Greg Van Hook

The difference in calculated temperature for given emissivity increment will be much more pronounced for high emissivity than for low emissivity surfaces. I suggest, if you are able to input emissivity to your instrument, that you keep changing the emissivity until the infrared thermometer agrees with the thermocouple.

What's important, though, is that the thermocouple be carefully affixed to the surface to avoid inaccuracy. Make sure your measurement doesn't use an area with a large thermal gradient, and point the infrared "eye" as close to the thermocouple as possible.

Oh...don't forget that the infrared device is much more quick than the thermocouple in producing a measurement. I'm not sure about these thermometers, but the thermocouple has a tendency to produce an average over however long it takes to react to a temp change (depending on thermocouple type). My instruments are instantaneous, but if I designed a thermometer, I'd put a several second average into it for these applications. It's best to let your measured surface reach a steady state, if possible, to accomodate whatever averaging the two measurement devices incorporate.
Fred Doyen

Stick the bulb in a pan of boiling water. You'd have to drain some anti-freeze and get the bulb out of the head.

Robert McCoy

Robert,

OK. Clueless here in Upper Black Eddy. What happens when I put the bulb in boiling water?

Greg
Greg Van Hook

You should read 212 degrees at the gauge or a few less if its off the boil. Just a way to check calibration.
Robert McCoy

Oh, darn. I was hoping that maybe that was a way to "reset" the gauge or something....

My poor old gauge, though, would be beyond max'd out at 212, probably pointing up at about 90 psi in the oil pressure range!

Thanks,
Greg
Greg Van Hook

Wow, something is really wonked in that gauge.
Robert McCoy

Greg - If your gauge is really reading that high, you might try putting the gauge in a pan of water, bring the water to a boil and reset the gauge needle to 212°. Then start with cold water and heat it up with a candy thermometer in it and compare readings every 10 degrees from 160° up to 212°. That may be all you need to correct the gauge. That said, if it doesn't track within a few degrees, then it is time to send it to Nisonger or APT. Cheers - Dave
David DuBois

I'll try that after the show on Sunday Dave, but I suspect it'll be on it's way to Nisonger!

Regarding temp readings, FWIW here are some comparisons between my Smiths gauge (G), a Fluke thermocouple (F) and my $24 infrared non-contacting thermometer (I) from Harbor Freight (all readings in degrees F):

Thermostat Housing (painted aluminum)
G = >230 F = 180 I = 182

Temp. sender boss on head (painted steel)
G = >230 F = 175.3 I = 182.8

Head at #4 cylinder
G = >230 F = 175.0 I = 179.4

Radiator tank (cooling down after a few minutes - painted brass - it is brass, right?)
G = 220 F = 133.7 I = 140.6

Obviously, the Smiths gauge is history. The differences between the Fluke thermocouple and the IR thermometer were fairly small as to be insignificant in this situation. Looks like aluminum has the highest emissivity (it had the smallest gap between the thermocouple and the IR thermometer readings).

Cheers,
Greg




Greg Van Hook

I made a temperature gauge read high like that by shorting the metal capillary pipe across the back of the alternator with a big flash. After that the gauge read constantly about 30 degrees F high. I had it repaired by Speedy Cables in Wales (cost £50 GB pounds) - not very convenient for Greg.
Mike Howlett

Greg,

As Mike eluded to above, disconnect the battery when threading the capillary tube through the dash and into the head. It probably would not matter if you are taking your old gauge out and replacing it or sending it out for repair but when you are putting your new one in, definitely disconnect your batter terminal leads. For safety, both of them. I reined a good gauge when the capillary tube hit the hot lead on the solenoid on the inner fender while threading it up to the head one time. Hard lesson!!!

Also, I think your thermostat housing should be cast iron on your car. Don’t think that would mater a whole lot in this thread. Just a comment.

HTH

Fred
Fred Wright

Hi Fred,

Thanks for the tip. I've been known to zap things 'cause I didn't take the time to disconnect the battery.

My car has the '69 - '71 head. I sure thought it was aluminum - it just doesn't weigh much.

Cheers,
Greg
Greg Van Hook

MGB factory heads are cast iron. Aluminium heads are aftermarket - and I would have thought, of relatively recent origin.

My 69 is certainly cast iron.
Regards
Roger
Roger T

Oops, I'm only referring to the thermostat housing.

Greg
Greg Van Hook

This thread was discussed between 05/06/2009 and 12/06/2009

MG MGB Technical index

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