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 C-N-H calibration

Several recent threads related to real or preceived overheating problems have mentioned temperature gauge readings in hot weather to the right side of the N. I have a similar "problem" when running at sustained higher RPMs when the ambient temperature is above 80 deg. F. In my case this occurs on the way home in the heat of the day. In the morning, on the same route, the gauge needle sits just the to right of N, which for my car is normal.
I have searched the archives for specific information on the the corresponding temperature readings for the C-N-H markings and could not find any definitive data.
The best info to be found is on this forum at www.mgcars.org.uk/electrical/temp/html
Here are given values for the sender resistence in ohms vs. temperature.
Using my $10 Radio Shack multimeter and a pirated meat thermometer I obtine dreadings that were quite close top those given, except for the very highest.
Background: 69B, original engine, 76K miles. Ignition and carbs as standard. 50/50 mix of Prestone antifreeze, 180 deg. thermostat, 13 psi cap.
I obtained my temp readings at idle (or fast idle) from the open radiator, using the meat thermometer.
I obtained the resistance readings from the flat blade of the sending unit, with the sending wire disconnected.
Procedure: As the engine was warming up and with the thermometer in place I quickly note the thermometer reading, disconnect the sending unit wire and take a resistance reading. This has to be done quickly as the readings change rapidly. Having the multimeter negative probe clamped securely to a good ground first is big help. One also has to be very careful around the spinning metal fan blades. I also repeated this over several days after running hot and obtained the following:
Water temp: 80F, 531 ohms (engine off, no gauge reading); 177F, 96.3 ohms, needle just to the right of N; 180F, 76.4 ohms; 195F, 68.9 ohms.
Although I did not get a resistance reading at N, the water temp was 167 deg. F, exactly as given the data online. I did not get any higher readings since at idle the engine temps would begin to drop. I have not yet tried blocking the radiator off to force a higher temp reading since my results appear to be consistent with the published data.
From this I conclude that a reading just to the right of N (needle over the upper right vertical stroke of "N") is normal, around 180 degrees runnig temp, and a reading between N and H corresponds to about 195 degrees.
According to the online data H corresponds to 212 degrees, the boiling point of plain water. My original factory handbook recommeds using "Blucol Antifreeze" and has information about the correct mixture to prevent frost damage, but says nothing about Blucol's ability (if any) to modify the boiling point. The gauge appears to be designed for the use of plain water only.
I submit therefore that when using modern coolants that a gauge reading of H is no longer a cause of real concern, although it is troubling to see. A Prestone mix of 50/50 is good to about 270 deg. F.
Soory for a long post, but I would be interested in the opinions and experiences of others regarding to this favorite of topics.
Regards, Andy
Andrew Blackley

Andy,

Good question.

A engine may have max bhp at say 80c but the protective oil needs 100c so a compromise of 90c may be reached for a road car.

To lower temp requires more water at additional weight, so to increase bhp may reduce power to weight another comprises.

The water temp therm controls this and should give a N reading. A oil cooler therm set at 80c controls oil temp.

Over 30mph the airflow through rad should give a N reading.

If stuck in traffic the fans take over to maintain N reading(pref to electric rather than mech).

Low temps mean oil is not working, high temps means reduced bhp, very high temps means turn heater on and roast feet to let engine keep cool.

Cars are designed to stay at N, otherwise it should be a known reason to deviate.

Paul

Paul

Paul: The point of my rambling post was to offer the possibility that the temp gauge was designed to register information based on operating procedures that were common before the widespread adoption of modern closed cooling systems, i.e. the use of plain water (preferably softened) in temperate conditions, and adding anti freeze (not "coolant") during winter months or as needed. Hence a Hot reading at 212F.
A gauge calibrated to read H at 212 degrees in a modern car would be in the red zone too often for most owners peace of mind. A case in point: I have a 98 Jeep Wrangler and the temperature guage reads 100-210-260, with 210 in the straight up position. This past week I drove my 98 Jeep Wrangler to the North Carolina shore over the mountains in 95-100 deg. F air temp. The gauge stayed near 210 on long steep climbs at 70+mph, wereas is normally sits just to the left at about 180F. It dropped into normal as on the down hill grades, but was running a bit hotter than usual.This was due to the conditions at hand. The stock cooling system can only shed so much heat when the air temp is so high (and the temp at road level is much higher than the air temp too). It seems that the designers of the Jeep took this into account and labeled the gauge accordingly.
I have though about using a modern aftermarket gauge installed in tandem with the stock unit, as a test. Browsing at the local parts supply house I noticed that these also were labeled with 210 at normal , and Hot at 260.
Regards, Andy
Andrew Blackley

Andrew,

Living in UK air temp is not usually a problem.

Cooling the engine is essential to prevent component failure through thermal fatigue, I'm not sure a B engine is designed for Hot to be at 260.

The combustion chamber area must be cooled sufficiently to prevent pre-ignition and
detonation. The higher the combustion chamber temperatures are, the more the ignition has to be
retarded to avoid the onset of the aforementioned
pre-ignition and detonation. This causes a reduction in engine output; particularly torque. Further torque
losses are caused when an engine is running too hot by increased inlet temperatures, creating a less dense fuel/air mixture. Again the engine may need tuning to suit these higher temps.

At what stage temperature wise is it necessary to hit panic button?

Paul



Paul

Andrew,

I think that if you try to run two water temperature gauges, neither one will read accurately. You would have to wire them in parallel. In this case, the original gauge would see a resistance that would be lower than the actual resistance of the sending unit. When you wire two components in parallel, the resistance is lower than that or either component. The original gauge would be reading the resistance of the sender and the new gauge, while the new gauge would see the resistance of the sender and the old gauge.

I don't know what the resistance of the gauges are, but I do know that the resistance of a standard d'arsenval movement is not very high. The lower the resistance of the gauge, the more it will affect the circuit.

Perhaps a more accurate test would be to plan out a road course that would include some hills and so on. Run this course several times to establish a good baseline with the original gauge. Then disconnect the old gauge and connect the new, temporary gauge. Run the course again and compare the results. The trouble with this is that you can't control things like ambient temperature and traffic conditions.

Probably the most accurate method would be to put a tee pipe fitting in and use two senders, one for each gauge. Make several runs, recording the results (including ambient conditions). Then switch the senders, so that the sender that was connected to the original gauge is now connected to the new gauge, and repeat the test. The switch will help factor out any slight differences that there might be between the senders.
Paul Noble

Paul N: My test, if I ever do it, would indeed use a new gauge and the new sending unit together. I hadnt given too much thought to the installation, but the tee sounds like a good idea.
The points of my "thesis" were a) that the original factory gauges are not calibrated to reflect the use of modern coolant, and that H on the gauge is 212 deg.F, and that N at 167 deg. F is probably too cold for engine efficiency;
b)a running temperature of 195 deg. F with high ambient air temperature is not a cause for concern.
Indeed, John Twist in his technical book posists that a normal running temp. of 190 deg. F is desirable.
Andrew Blackley

Andrew,

A quote from Redline as I use Water Wetter

Since 50/50 glycol solution has about 4 times the viscosity and only 70% of the thermal conductivity of water, the thermal convection coefficient for a 50/50 glycol solution is approximately 50% of the coefficient for water. Water in the cooling system is capable of transferring twice as much heat out of the same system as compared to a 50/50 glycol coolant and water solution. In order for a 50/50 glycol mixture to reject as much heat as water (amount of heat rejected is independent of the coolant), the temperature differentials at the heat transfer surface must be twice as great, which means higher cylinder head temperatures.

I already said that a temp between 80 -90 would be ideal which ties in with John Twist at 190f

The coolant temp at bottom end is governed by Thermostat and say opens at 82deg C so if running normally N should be 85deg C or 185f, so your reading may be low or you are using lower thermostat.

With high ambient normal running may be higher at expense of bhp but not a problem and as I mentioned above may be preferable than having to carry more water.

However a problem arises when stuck in traffic (UK prob) for over an hour with high ambient and thermal runaway occurs and then its nice to have a safety margin rather than a car with a high temp under normal operating conditions then getting caught in traffic and watching needle climb off scale.

Was H 212deg at bottom of white danger zone giving a full guage reading of say 230?

Paul
Paul

Andrew,

I think that would be an interesting test. I know that temperature gauges used in automobiles are not terribly accurate. It would be nice to know just how inaccurate they are. I was not intending to be critical of your test or of the idea of such a test. Its just that I wasn't sure if you were going to simply wire both gauges into the same sender. This would, as I explained, distort the reading of both gauges.

I am having a minor problem with my engine temperature. Many people come to this BBS seeking help for overheating. Mine is just the opposite. I just replaced the clutch and while I had it out, I flushed the radiator, replaced the thermostat with a 195 degree one and generally cleaned things up. My wife and I just went on a long (50 miles one way) trip yesterday. It was about 90 degrees outside, but the engine of my 1970 MGB never even got to the N mark. It stayed about 2/3 from C to N. I realize that running the engine too cold is not desirable, but how do I get it to heat up? Put a little mud in the radiator like the stuff I rinsed out?
Paul Noble

Paul N,

What temperature thermostat are you using? As Paul W points out, the thermostat controls minimum operating temp. This is interesting as I recently changed temp senders in the 71 GT, the old one was reading obviously low. It would indicate about 2/3 between C and N when at NOT (normal operating temp). The new one indicates almost exactly N under normal driving conditions. I suspect the variances in senders and gauges render the temp reading as a SWAG. I'm tempted to use a temperature gauge calibrator on my sending unit to determine exactly what temp the C, N, and H indicate.
Paul K

Paul In Surrey: I am running a 180degF thermostat, so as a result my normal gauge reading is just slightly to the right of N.
My little test was not terribly scientific, and the accuracy of a kitchen meat thermometer creates a margin of error in the experiment, but I did come close to the published resistance values.
Regarding your last question, I dont know, as I am not brave enough to test my thesis to the bitter end, i.e. block the radiator and let the temps reach H on the gauge!
I think that 230deg.F is NOT normal or desirable for any car, new or old.
I agree about not wanting to be stuck in traffic and watch you engine go "on the boil", so certainly any margin of safety you can obtain if your car is liable to be under those conditions would be wise.
Reading the info at the Electrical Help section about voltage stabilizers throws another possible glitch into the mix.
I guess the moral is that everyone may wish to test their own cars.
Regards, Andy
Andrew Blackley

Andrew,

If thermostat working correctly you should see guage reach temp and then fall as thermostat opens, this gives guide to your guage reading, it should then settle above this initial reading at N. With high ambients or traffic it will move onward and upward.

I'm sure a start to solution (keeping guage at N) is alluminium rads which according to MGOC gives a 30%increase in cooling (are any modern day cars fitted with non aly rads), together with shrouded fans should bring B cooling into 21st century, although heat does not leave engine bay very readily.

Paul N,

On basis you cannot guarantee any new part works as it should, perhaps thermostat in heated water and measure temp and then in car check when thermostat opens on guage.

The trad way of keeping temp up is cardboard in front of rad now replaced by blinds but usually reserved for winter in UK.

Paul
Paul

Andrew; Just a thought, you could use a mechanical gauge to make your checks, the big
problem would be finding some way and a place to mount the sender bulb. It would not
interfere with the normal electrnic gauge. Regards, Clifton
Clifton Gordon

One idea I thought of on the way home would be to introduce into the upper rad. hose a short piece of tubing, suitably drilled and tapped to accomodate a sender. It wouldnt show much until the thremostat opened, but we're after calibrating the gauge for the higher temp range anyways. Summit Racing has a big selection of both mechanical and electrical gauges.
Andrew Blackley

I have a multimeter that i paid way too much for, but it has an attachment that can be placed in the upper rad hose and the the hose clamped bak down . it will give temp readings.
i have never checked my 74 bgt with ac but it reads just below the n when on the open road no matter the outside temp and just above the n when in stop and go traffic. It has been nere 100 degrees here lately and i have no trouble.
I was told years ago the any guage the goes c-n-h in not going to give an exact reading, it it would then they would mark it in degrees. they change to this type of guage to stop owner fron frettin over every little change the temp guage.
I would never let my guage go all the way to h.
steve

This thread was discussed between 29/07/2002 and 30/07/2002

MG MGB Technical index

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