Welcome to our resource for MG Car Information.



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 MGF Technical - Don't lose your head [gasket]

It has been interesting to read about potential causes and solutions of HGF these last few months- and I think we are getting closer to getting the right answers (or is that blind optimism on my part? ;o) So I thought now might be a good idea to summarise the maintanence checks and preventative measures- especially as some information has come my way on the aetiology of HGF from the MGF racing fraternity.

Maintanence checks:
1. regular checks of fluid levels and function of the auto-bleed valve.
2. regular checks/replacement of the ECU coolant temperature sensor.
3. Ensuring adequate bleeding of the coolant system, with the use of the appropriate antifreeze.

Another potential cause of the HGF problem has come to light as a result of MGF related motorsports. Basically, from what I gather, the problem is as already been highlighted here, namely due to the remote nature of the radiator causing larger than 'normal' [in FWD applications] changes in coolant temperature entering the engine. This, as was rightly pointed out, causes the standard type thermostat to occasionally be fooled and close when the engine is actually hot, thus allowing localised 'micro' overheating. This phenomena, over a period of time, may weaken the perfectly good head gasket leading to failure.

The common solution, apparently, is to drill four equi-spaced 6 mm holes in the thermostat ring. This mod allows the cooling system to continue to function in a normal fashion, but crucially also some circulation when the thermostat is shut. Thus, when the thermostat is inappropriately shut, some cooling circulation continues avoiding the excess oscillations of temperature caused by a negative-feedback system such as this with a remote effector system.

There is a draw back- the warm up time is increased. It has been suggested to me that a compromise for a road car would be two 6 mm holes- faster warm-up time, but still adequate closed-thermostat circultion to prevent a temperature peaks causing head gasket failure.

This solution certainly sounds sensible to me, and I may try it out. It may also be an extremely sensible mod to consider if your car has already suffered from one failure- and perhaps almost essential if there is ' history of repeat failures?

What is the consensus view on this?

Rob Bell

Rob how do you check the ECU coolant temperature sensor?


Any chance of some sort of BBS meet, where the smarter/more experienced guy's can show some of these excellent tips 'n' tricks to the not so well initiated amoung us ?
What do people think ???


Kieren Gibson


As the heater circuit is open even when the thermostat is closed, this will still provide some form of cooling.
However, will it stop the cyclic problem that the 6mm holes are designed to cure ?

I may have missed something, but I think it'll solve the problem as well.
Creating the problem of hot feet though! :-)

Paul Nothard

>Any chance of some sort of BBS meet, where the
>smarter/more experienced guy's can show some of these
>excellent tips 'n' tricks to the not so well
>initiated amoung us ?


Come to the Essex meet... although beware it is very much a lure to the dark side ;-)

Whereabouts in herts are you?

Chris George

Thank you for this idea, Rob. As a newbie, I am especially interested in formulating a list of things we can do to reduce the chances of HGF. Should we add the following?

4. look at the gauges? Or is this too obvious?

5. Should we be checking coolant pipes for deterioration? Isn't sudden pipe failure a cause of dramatic 'unavoidable' HGF?

Dunno, what do you all think?

[Apologies to anyone out there sick of HGF's!]

Like Rob, I'd be interested to know the consensus view on drilling the thermostat ring. Also, what do we think for a recommended frequency of changing the ECU coolant sensor? Every 24000 miles? I'd gladly cough up for that.

How do you check the auto bleed valve. I'm just about to reread the 'Temperature Teaser' thread, so if it's decribed there, please ignore me!
David Bainbridge

David- not too obvious at all- do check the guages (but not always a reliable indication as we have heard over the years), but the corrosion checks on the rad return pipes are very important.

Paul, does the heater have the capacity to really make that much impact on engine temp? I admit that when the ECU temp sensor packed up I used the heater in this way- but I am not sure how effective it was- at least it was something I could do when the temp needle attempted to dive into the red!

ECU temp sensor replacement? From what Per has suggested regarding progressive deterioration, perhaps replacement should be every 3 years? Ten quid a shot seems a reasonable price to pay.

Kieren, we could have a BBS met at Silverstone on the Saturday evening- over beers? Otherwise, as Chris suggest, pop along to the Essex roadsters meet (this evening is the next one).

Rob Bell


Think that the heater matrix (with fans on!) would have at least as much capacity as 2x6mm holes.

No facts, just gut feeling.

At Donnington with the heater on full the rad fan did not kick in after a thrashing that took the oil to 150degC.
Without the heater on, the rad fan did kick in very quickly.
So it's enough to tip the balance there.

I think we now need somone with experience to comment. :-)

Paul Nothard

Just searched the archives.
Spencer from London mentioned it in 1998 to the thread 'How to extra-cool my VVC'.

Carl was IMO the only one currently who modified the thermostat by changing to a 10 C cooler one.

Anyway, I'm just on drilling holes to a spare thermostat ;)
Thought two are enough. Ok I go and add two more ;-)

Rob, sent some pictures.
Here are some more. Go and get them.
Only the pictures are there. More to follow.

BTW Carl mentioned the other year that the thermostat disassmbling is a difficult job. Breaking fingers 'included'.

PS off for two more holes
Dieter Koennecke

> we could have a BBS met at Silverstone on the Saturday evening- over beers

Sounds good to me, I will be at Silverstone, probably Sunday only due to hopefully going onto the track, which reminds me, I need to pay up!

Would be nice to meet up with everyone, maybe even Dirk if he's over here ;-)

Steve Childs

Another alternative, perhaps?

Maybe there'd be the case for a second "heater matrix" sized "radiator" (complete with its own thermostatic fan) in the engine bay itself?

Probably "plumbed" into the lines that feed the "passenger" heater....


>I think we now need somone with experience to comment. :-)

Paul, I just added the old threads from the archive as PDF to the above link. Worth to read.
Some famous experianced comments inside.
Anyway, that drilled thermostat gets in as soon as any coolant works get done at my car !! Guess my official MG workshop will not do it.

Steve, Silverstone.... :( got quite expensive for continental F-ers. Lack of $$$ on my humble side.
Dieter Koennecke

Dieter now has updated the page to provide a full explaination of the thermostat mod. Also mentions the first suggestion by Spencer (old regulars will recognise the name!) to do exactly this mod in March '98! Couldn't believe my eyes... A solution to a problem we didn't even know existed back then...

Paul, I honestly do not know whether the heater could vent sufficient heat, or indeed would be capable of ameliorating the temp oscillations, but your observations are very interesting. The fact that the rad fan still comes on even with the heater on suggests that the heat dissipation is not adequate- albeit, it may reduce the rate of rise.
FWIW, during the summer I usually leave the heater on 'max' out of habbit- to keep my hands cosy on the steering wheel with the roof down. It'd be faintly amused to discover that this habbit has preserved the head gasket all these years...

Rob Bell

Deiter, thanks for those threads. Some really good info about HGF as well. And some old names too...
(Rob, that makes us old! Eeek!)

Anyways...Rob, think you misunderstood...
Heater matrix on full -> no rad fan.
Heater off -> rad fan needs to come on.
ie. It made the difference.
And this was with a hot hot car (oil @ 150degC).

IMO getting rid of heat from the engine can be no bad thing, so I keep the internal heater on as much as is tolerable when racing/'track-day'ing.
My gut feeling is that will be enough...

Having said that, the holey thermostat can do only good! :-)

Paul Nothard

Ah yes, I see what you are saying Paul... As you say, any method by which heat is disappated is 'a good thing'. Never gave it a thought before- but now will certainly leave the heater purposefully on 'hot' for track days and sprints!

Still, the perforated thermostat sounds like the best way to go for a permanent solution?

(Old? Uh-oh... certainly showing our age when we can remember characters like Spencer and Casey! ;o)
Rob Bell

intruduce the same theorie some hours ago at the german Forum.
Response is low currently, but you know.
Germans like precise measures. :)

Isn't any Mathematics student or Measurement and Controls Student out here who can enlighten the theory a bit more ?

Have studied Measurement and Controls 'some' years ago myself, so I do not need to know how to setup the affiliated differential analysis formulars for open sytems any longer ;-). Dunno the english expressions though.

What have we got:
- measure of waterflow and pressure of the water pump (in the manuals IMO)
- Diameter of the pipe in front of the thermostat
- Diameter of the added ... let my name them .. 6mm bypass holes'
- lenght of the hoses to the front cooler and back to the engine
- Temperature gradient of the thermostat .. can be checked at different water temps with the one 'in my hands'
- Temperature gradient of the engine .. somebody can check at let me suggest constant 4000 revs at any known outside temperature.

Any else more or less known parameters ?
Just some ideas to mind up the scientists on board.

PS. having in mind a former colleage who meaned:
*Everything can be calculated*. His nickname was 'Zweistein' cause the famous 'Einstein' was obviously not so intelligent .. LOL
Dieter Koennecke

Ambient air temperatures?

Zweistein indeed ;oD LOL
Rob Bell

Has anyone used a water wetting agent?
That is supposed to help get rid of aeration and aid cooling.

Graham Robson

Have been following the threads on HGF and temps in anticipation of Russ's F coming to new home on Sunday :-))
Rob, one of my thoughts on reading about the localised hot-spot was 'why not remove the thermostat altogether' - being used to lower-tech cars we always did this with any engine that showed regular signs of running too hot, eg straight after a rebuild while still tight. So yes I believe you are on the right lines here, the drilled thermostat merely refines this 'sledgehammer' approach. Someone (Rog P?) mentioned using an 82 degree stat, that might also help if the normal one is higher ?

5th update online. Added some pictures with less holes. Just to know how a thermostat with one hole looks like.
Dieter Koennecke

Chris,I am based in Hitchin and live in Arlesey. How about you ?

Rob, Can't get a pass for Silverstone unfortunately - It's my daughters 1st birthday and wife is threatening divorce if I don't start paying more attention to them and less o the little red rocket. Sorry I missed the Essex meet last night, But I'll be at the next one as long as I know enough in advance.


Kieren Gibson

Had a discussion with a Rover garage this morning.

Comment on the hole in the thermostat.
'koennte man ja mal ausprobieren', which means in english a very carefull: 'Why not, this could be tried'.

On the other hand the coolant temperature sensor (brown on left top of the engine) came up. The guy mentioned about adding a resistor to the sensor and force the ECU to start the front cooler at lower temperatur.
I'll write this comment also to the 'temperature teaser' thread.

Dieter Koennecke

Graham, people have tried water wetter in the past- although many have expressed fears that it may adversely effect the performance of the antifreeze additives. I know of one anecdotal case where water wetter was present, and a HGF occured- for what that is worth.

I like Neil's idea of an additional heater matrix- perhaps controlled by an electronic valve... but it is perhaps too complex?

David, you are right, Carl is already using a cooler thermostat- 82C I think. Not sure that this would always help, but may reduce the frequency of overheat spikes?

Kieren, the next Essex MGF meet is Thursday next (the 17th). Hope to see you there (the Halfway house, A127 near Orsett).

Perhaps what we need to do is to find the best solution for a road car in terms of drill-hole number and drill hole size? Too much by pass flow could be detrimental to normal engine function, too little... (!)

Rob Bell

I don't think I'd bother with "controlling" the water flow to the additional matrix - simply put it in parallel with the heater itself.

I would, however, use an on/off thermostatic fan (completely independant of the MEMS!)

Depending on exactly where you put it, you could arrange that it didn't get much airflow unless the fan was actually running. Thus it would have little more impact than the passenger heater when the car didn't need the extra cooling (i.e. when its fan wasn't running) as in when still warming up.

You could probably even find the components from an another Austin/Morris/Rover/... to keep withing the MG tradition :-)

The drawback would probably be yet another bleed point!


This thread was discussed between 03/05/2001 and 04/05/2001

MG MGF Technical index

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