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 - fluctuating Temperature

The temp on my 72 GT fluctuates quite a lot . It never goes into the H zone but seems to rise and fall at different revs (though not necessarily getting hotter with high revs or vice versa). The engine periodically runs-on, too. The temp guage and sender are new, but I noticed that I have a 'fast road ' distributor fitted (43D) ie no vacuum advance. The car runs well and does not pink but could this type of distributor cuase this problem? I'd like to fit electronic ignition but don't want to buy it yet if I should change the distributor first. Any ideas?

Paul Flush
K Flush

Oh, forgot to mention: The car has the usual electric fan, the old fan blades are definitely the right way round (now!) and I fitted a new thermostat.

Thanks in advance for any help

Paul Flush
K Flush

Usual electric fan? The factory didn't start fitting them to 4-cylinder cars till 77. Whilst aftermarket fans (depending on adjustment) and most modern cars have a closely controlled fan which seems to cut in and out without the temp gauge varying much the factory system allowed for quite noticeable temp gauge movements. So in itself movements of the gauge while the fan cuts in and out is not neccessarily a problem. It's more a case of load versus speed that will cause the needle to alter - a high load and low speed will cause the temp to rise and the fan to cut in, lower loads or higher speeds less so.

Running-on and pinking can be a symptom of the engine design (my 73 is very prone) and petrol octane. Pinking is controlled by the amount of timing advance, running-on by mods (see http://www.mgb-stuff.org.uk/wn_fuelframe.htm and click on 'Running-on') or stalling it with the clutch before you switch off (yuk). The distributor type has no effect on running-on as there is no spark at the time it is happening. However if the timing is such that it is causing excessive heat in the cylinders when running you are more likely to get running-on. A high idle speed also exacerbates it.

Forgive me, but have you flushed your cooling system?
Paul Hunt

I once had this problem with a 72 B. Turned out to be a poorly grounded voltage stabilizer. FWIW.
Jim
Jim

Could be. North America had the electric temp gauge from 67, Canada from 68, Germany and Sweden from 70, and Norway later than that. Other markets had the dual, mechanical gauge until 76.
Paul Hunt

The fan is a Kenlowe and the gauge is the dual mechanical type. I can't hear when the fan cuts in or out on the road to know if this would account for the fluctuation but maybe I can fit a warning light temporarily to check. I have an anti run-on device fitted but it still runs on some of the time. I use 98 octane unleaded. The water in the system seems very clean and has been draind a couple of times recently. I could try flushing the system properly though. What about the radiator cap? Would that contribute if it was a bit weak ,say? What sort of gauge movement is normal, by the way? And finally, is there any advantage to using the non-vacuum advance distributor?

Thanks

Paul Flush
K Flush

Paul-
If the temperature tends to fluctuate while cruising on the highway, the most likely suspect is a faulty thermostat. A centrigugal-advance-only distribuor will give better throttle response because the engine will be making full power for any given RPM, but more heat goes with more power, so fuel economy will lessen and the valves will have a harder life, too. The engine will be more prone to running-on, especially if there's carbon build-up in the combustion chambers.
Steve S.

What kind of AR valve have you got? I tried one that dumped air into the manifold and it did absolutely nothing for the running-on even though I knew it was opening. I eventually developed my own system - see http://www.mgb-stuff.org.uk/wn_fuelframe.htm and click on 'Running-on'. The only time a bad rad cap will cause the problem is if it so bad it allows the coolant to boil and so you lose it, but that would be obvious. Generally if it doesn't chuck any water out a few seconds after switch off after a good run i.e. heat soak causing the temp to rise even further, it won't happen on the road. I have an illuminated switch connected to my electric fans, so I can see when they are on as well as switch them on manually. A non-vac advance will do nothing for your problem and you will lose cruising economy into the 'bargain'. The temp needle should not fluctuate during steady running in clear air, and neither should the fan cut in for that matter, only when idling or slow running behind a bus or truck or something i.e. you have little ram-effect of air through the rad. The only time I have see a temp gauge fluctuate running in clear air is in a V8 with a blocked steam pipe, I have never heard of this happening on a 4-cylinder which is completely different. As Steve says, a sticking stat could cause it.
Paul Hunt

Well, I'm not sure where this leaves me - I already fitted a new thermostat and nothing changed. Would it make any sense to run without it temporarily to see what difference it made? The non-vacuum distributor was already fitted when I bought the car but it sounds as if I'd might be better changing it anyway. The AR valve is the type available on the MGOC website but Paul Hunt's mods look interesting. I'm not sure if it's more annoying that it doesn't do it all the time or not! Maybe this inconsistent temp affects that too. I'll keep trying!

Paul Flush
K Flush

I have similar symptoms on my GT. The needle comes up at high speeds and under load but never gets anywhere near the red. Over-retarded timing definitely exacerbates it. However, I am pretty sure it only started doing it with the introduction of a Kenlowe, followed quickly by the addition of spotlights that must take a good chunk of frontal area from the grille. The blurb with the Kenlowe specifically points out the the car is going to run warmer without the mechanical fan, which helps the efficiency of the engine (just use decent oil). If it helps, I've had new rad, stat, head gasket, manifold gasket and flushed the waterways in the block in the time I've had the car and nothing has made a difference. As I was once told, if it's not in red, it's not a propblem!
Steve Postins

Steve-
If the problem appears at higher engine speeds, check the intake hose to the coolant pump by squeezing it. Some of the hoses on the market are so poorly made that they collapse at higher intake speeds, restricting the coolant flow. An adequate quality hose should refuse to collapse no matter how hard you squeeze it.
Steve S.

I have never yet come across a top or bottom hose that you can't compress by squeezing. There may be some out there, but the fact you can compress a standard hose by hand is not, repeat not, neccessarily a problem. I'm also very sceptical about the abilty of a standard MGB engine to suck a hose flat by the action of the pump in the absence of other problems. High speed (how high?) almost certainly does equate to high load given that the effort required to push the car through the air increases at the square of the increase in speed. Electric fans are usually fitted for their greater cooling ability at low speeds or idling, the mechanically driven fan is usually doing far more work (consuming power and making noise) than is required once underway in clear air and the ram-effect should be more than enough. But anything blocking the airway will reduce the effectiveness of the cooling system. I have both mechanically (CB roadster) and electrically (RB V8) cooled MGBs and neither shows any tendency for the needle to rise at motorway speeds or beyond, more the reverse. None of which helps solve the problem.

You *could* try running without a stat just to see what happened. In theory warmup should take longer, and it should run too cool when underway in clear air. There should be proportionate rises and falls in cooling system pressure and temperature, and as part of an investigation into a cooling system problem on my V8 I plumbed in a pressure gauge. This is easy on cars with a separate, pressurised header tank - you just put a Tee in the small pipe that connects the two - less so where the pressure cap is on the rad. At least that allows you to confirm that the gauge movements are accompanied by pressure changes i.e. it is not simply a dodgy gauge. It could be a head gasket problem leaking combustion gasses into the coolant at larger throttle openings and loads.
Paul Hunt

Well the hose squeezes with the resistance of a paper cup on mine so it certainly looks a little suspect. The needle moves upwards once you are running at 4000rpm or more, so 80+ in top. Just to complicate things its 'normal' position is well to the left of centre and it just about hits the middle when run hard, so I really should get a thermometer and find out what the actual temperature is in these conditions. I'll report back!

Steve
Steve Postins

My temp fluctuations are at much less speeds than 80 or 4000 rpm. It varies between a bit left of centre on the gauge and a bit right of centre. It hasn't, so far, gone as far as the 'H' zone. It's a bit hard to pin down exactly what gear/revs sees it moving up and down so I'm going to try a few of the suggestions that have been made already and report back. I noticed that as I neared home yesterday (ie getting slower) the temp rose. When I stopped the car the Kenlowe wasn't running but switched on about 30 seconds later. It's set around the 'N' position but maybe that's not the best place. I have a feeling that the suggestion about the head gasket may be the right one (though the car is running pretty well) but I'm still hoping for an easier solution!

Paul Flush
K Flush

Update: On my GT the normal running temperature (needle left of centre) is about 84C/183F*, N equates to 90C/194F*, and mid-way between N and H 100C/212F*. My Kenlowe cuts in at a little over 90C/194F. Hard running therefore takes me from about 84* to near 90*.

Finally, having had another look, that water pump hose is truly floppy and I can see the exterior wall contracting somewhat at higher engine speeds. I realise that it's never going to be rock hard but I think Steve S. has a point with this one.
Steve Postins

If the normal position is well to the left of centre and it only gets up to N when run hard then that says to me that the stat is either missing or stuck open. Feel the top of the rad as the engine is warming up from cold. It should stay cool until the needle gets near N, then get hot in a rush. If the top of the rad warms gradually and constantly as the engine warms the stat is missing or open. This will cause the fluctuating temp gauge readings you describe, as the stats job is to continually vary the flow of water through the engine and rad to maintain a relatively constant engine outlet temperature. Without the stat the rad will be cooling the water the maximum amount all the time i.e. way too much unless the engine is working really hard.
Paul Hunt

Thanks Paul,

The stat seems OK in that I can see it open on the gauge and as you describe in a whoosh just past "N" or at about 90* according to the measurement scale I made (although it's an 82* stat?). The needle at "N" on my Roadster only equates to 80* so I guess the calibration must be out on the GT. It's the guage with the wire wrapped pipe so I don't think there is much to be done about it? Checking the archives there seems to be a good range of interpretations of the actual temperature at "N" so I wouldn't really know what's correct anyway.
Steve Postins

A small development! This time after a run I switched off and shortly after water was coming out of the radiator overflow pipe. Does this mean the cap is faulty after all, or that there is too much pressure in the system due to a possible head gasket leak (as suggested by Paul Hunt)? The temp rises more at low speed in 4th gear but holds steadyish at higher speeds. The car is running very well but I have the impression that it is running a bit too hot generally. I suppose being too hot would also cause the running on sometimes?

Paul Flush
K Flush

I think Jim of Ohio had an excellent suggestion. Make sure the voltage regulator is OK before getting into mechanical issues. It could be that the temperatures are not fluctuating at all, only the voltage to the gauge is fluctuating. Amps drawn by the electric fan could be affecting the voltage levels at the gauge if the voltage regulator isn't up to snuff.
D Maples

In my case the gauge is mechanical, but the regulator is new anyway. The Kenlowe wasn't running when it was overheating, in fact, so that adjustment I have already made. Still, I'm a bit concerned about water coming from the overflow.

Paul Flush
K Flush

Paul,

My trawls of the archive suggest that a little overflow is quite normal and some people go to the trouble of fitting a bottle to catch it. As for the head gasket idea, a compression test is easy to do and will give you a good indication. If gas is getting into the coolant and heating it you should see bubbles in the water as you rev the engine. If you can get sophisticated, a gastester used in the air space at the top of the radiator is definitive if it detects exhaust gas.

By the way, sorry for hijacking your thread!
Steve Postins

IMHO there shouldn't be any overflow with the exception of during the first warm-up after overfilling. If there is water coming out when the engine is still running it is either boiling or a blown head gasket is allowing compression gases to push the coolant out. If it suddenly chucks out a bit of coolant a few seconds after switch-off again it is boiling, but this time because of heat-soak. The first thing to do is to replace the radiator cap which is a cheap and easy fix. Check the new cap against the old by placing the tops in your palms and pressing the internal ends together against the spring. If one moves before the other it is weaker, they should both move at the same time. I had problems with a new cap and was able to demonstrate it to the sceptical counter staff by getting half-a-dozen caps and showing them that the one I had bought always moved first despite them all being marked the same. If it still happens with a new cap then the coolant is getting hotter, and hence higher pressure, than the cap can cope with. This could be due to localised boiling, or the engine generally producing too much heat.
Paul Hunt

This is turning into some thread! Just had another run and this time no spill from the overflow. This is hardly a definitive survey so I'll try a few more runs. I adjusted the Kenlowe to come on sooner so the fluctuations are still there but slightly less movement. I'll get a compression test done and maybe but a new radiator cap anyway since that's cheap and easy. What readings am I looking for in a comp. test. What are normal variations? Don't worry about the 'hijacking' - it's turning into quite an interesting thread!

Paul Flush
K Flush

Paul Flush, Perhaps you had the cooling system overfiled as Paul Hunt suggested. Compression should run around 150 psi, but it's more important that all cylinders are within 10 or 15 psi of each other. Good luck, Clifton
Clifton Gordon

Hells bells. After all this playing around I've tracked down tell-tale bubbles in the coolant. My excuse is that it only happens at high revs, so any idiot could have missed them. Really.

Steve Postins

This thread was discussed between 02/05/2003 and 11/05/2003

MG MGB Technical index

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