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 MGB Technical - Core plugs - the best way to seal and secure?

Eventually it happened! I popped a core (welch) plug at a sprint meeting last weekend (middle one along the side of the block). Over the years I have seen so many of these fail at competition events where the engine is given a prolonged workout, especially with larger bore engines it must be more than faulty installation. I suspect the factory recognised this as a weak point in the engine as far back as the MGAs as the Works Sebring cars had retaining straps fitted.
What is the general opinion on how best to prevent these failures in competition engines?

(1) Dorman style expanding plugs (quality is a bit suspect on the ones I have seen)
(2) Original part, properly fitted with a good sealant (Stag?)
(3)Fit retaining straps similar to works cars and also described in Peter Burgess' book " How to power ..." (P60) (Requires the engine to be out of the car to drill and tap block)
(4) An alternative suggested to me is fit a floating tapping plate inside the block just slightly larger than the hole, then drill a hole through the centre of the welch plug and affix with bolt and sealing washer to the tapping plate (sounds a great idea).
(5) other options?

Mike Ellsmore

Dished of cupped?

I have a Gold Seal engine and I understand one of the things done with those was to fit cupped. I did have one corrode through many years ago, but the replacement and the others have been fine so far.

I also understand the block has to be machined making the recesses deeper to fit cupped.

I've had good results from stainless steel plugs in hard woking B series engines
They are mighty tough to put in and you tend to feel a bit guilty whacking them so hard to flatten them out but they stay there
I wipe a smear of Loctite flange sealant round the hole ,not on the plug as it gets wiped off as it goes in but if it's in the hole it gets squooshed up as the plug goes in, then fit the plug in, carefully with a large socket that fits right out on the edge--being carefull to get it started in nicely as the stainless plugs are very easy to damage--Once it's seated use a decent sized drift to work on it till almost but not quite flat to get the most expansion out of it
I've seen plenty of plugs not fitted properly even by reputable builders, with just a dent in the middle--they need to be just short of flat to do the job properly

Pro Torque are the distributor for them, but don't sell to the public, you have to contact them to find out who's the retailer in your area


just for info's sake, we used to have heaps of plugs blow out of our chev racer in the early days and found the only fix was to grind every second blade off the water pump impeller to slow the water down
Lotus did a similar thing with Elans by using a smaller diameter crank pulley to slow things down
Might be a project for you to find a smaller pulley ,but then if this is your first then probably an overkill for you--?
William Revit

I've always thought these would be a good idea if you could get them in the correct size and at reasonable price.
R.A Davis

I always have the block machined for cup type core plugs.

And on race engines also use a steel strap across.

Colin Parkinson

Paul, dished not cupped.

Willie, thanks for comments, hard to change crank pulley or water pump pulley. Iíll check out ss welch plugs. Using an air chisel with a planishing bit tool may be the way to fit core plug in situ.
Will keep you posted on progress!

Mike Ellsmore

Air planisher--that's a bit flash
William Revit

I use copper dish plugs and built a concave tool to fit them. The concave surface is just slighly less than that of the plug. It pushes the plug near flat.

I also use permatex core plug sealer.

Here is a picture of the tool

Bruce Cunha

My issue is my engine is in an MGA so not a lot of room to hammer.
Mike Ellsmore

Betson bolt
Dominic Clancy

FWIW, a mechanic buddy of mine used to put the plug in the block, followed by a second plug that fit tightly inside the first plug. I cannot recall any failures on his method.


79 MGB
gary hansen

That might work with cup type but these are dish
It's a side plug that's blown out
William Revit

May be look to the reason for the plug coming out? On B engines, especially large bore ones, over advance can lead to detonation which knocks em clean out of the block.
Some years ago we ran up a sprint B with a 1950 engine on our old Clayton water brake which holds the full load and makes the engine work hard. We noticed a pink but, as the car had run ok all season and won it's class we thought we would get a baseline for tuning. As soon as I put the load on and we had dragged down from 6500 to 5000 the centre core plug flew out! Hammered a new one in, backed off the timing and away we went!
As a matter of course we always strap the dished core plugs in place a la MGA Factory Comp Dept style.
Cup ones stay in very well, we were told ( I think by Chris Betson) it was common practice on factory Gold Seal blocks to machine for cup plugs.
Peter Burgess Tuning

I have now removed the welch plug and cleaned the few layers of paint off it. Appears to me the failure was due to poor installation originally (I'm innocent, as on my rebuilds I thought why disturb them, they have never given any trouble). If you look at the attached photo you may just be able to see (4) dent marks like from a ball peen on the shinny side. On the water side there is little corrosion (Amsoil Coolant Boost) but there is 2.5 mm gap under the ruler.

Peter, I am pretty sure detonation wasn't the cause as the engine was tuned on a dyno after recent rebuild - knock sensor probe used, total timing set at 27 deg BTDC (123 distributor, 11:1 CR)

Pressure surges might have had a minor influence as the radiator cap is remote from the radiator on the overflow tank connected with a small pipe but primary cause I reckon is down to bad fitting.


Mike Ellsmore

Not a happy looking plug Mike
I'm sure you'll be able to improve on that
Once the centre has been turned inside out like that the 'expansion' effect is lost, it's no different to being curved the other way,
-almost, but not quite flat is the target for maximum expansion
William Revit

It hasn't been turned inside out - the left image is the dry side of the plug and is still convex. The side with the ruler is the wet side showing it could still go another 2.5 mm before it is flat. It was hit 4 or 5 times with a ball peen hammer in different spots rather than finished with a dolly about half the diameter.
Mike Ellsmore

Not fitted properly!

Use some loctite thread locker in the cleaned recess.

Don't use a ball peen OR a drift to set the plug. It should also not be flat!

Use an ordinary hammer's flat head to strike the centre of the plug several times so that a dimple is formed in the centre of the plug - the centre of the dimple should be level with the edge of the plug - no more, no less.

Pictures of the process at - item 7.
Chris at Octarine Services

Thank you Chris. Good advice. Mike
Mike Ellsmore

, a question
Why a dimple in the middle,I don't understand the reasoning behind that, doesn't sound right to me, -If it's curved inwards it would be no different to leaving it curved out wards---I've always fitted them almost flat, still convex a tiddle but almost flat, and using a decent drift

Have a look at this-In my opinion this is how they should go in, but don't use two hard hammers together if you like your eyes the way they are---
William Revit

They are not curved in - they assume an "W" shape in cross section and that is the strongest configuration - just as the factory fitted them.

But then the factory fitted steel plugs - not brass - maybe the weaker brass plugs need more spreading - I would not use them.

I have never had a steel plug blow out and I have fitted 100s of them.
Chris at Octarine Services

Hmm - don't know - a W shape would be concave in the middle, which is the bit that seems wrong to me
Anyway, each to their own I guess
William Revit

Hi Mike
I just wonder why it came out under 'anger' conditions? Did you run same fuel/booster as when dynoing?
What was the weather like when it popped?
I have seen pressure surges from high rpm pump cavitation but never on a B. Surge enough to blow rad cap! Are you running a larger water pump pulley than a chrome bumper one?
Peter Burgess Tuning

Still thinking -
Although the plug appears not to have been fitted over well there is one other maybe cause
I have seen some B's fitted with a thermostat and also the bypass sleeve
In a case where the thermostat isn't open far and a heap of revs are used there isn't really anywhere for the coolant to go and the plugs would be under extra pressure
You don't happen to be running both a sleeve and thermostat by any chance--?
William Revit

I don't subscribe to the pump pressure theory - the pump is just a paddle, not a positive displacement pump.

It does not create any significant pressure, it simply moves the water under centrifugal force and should there be a blockage in the flow, the pressure would soon cause the water in the pump to "stall".
Chris at Octarine Services

Have a read of this Chris
Note where it says the shockwaves from cavitation....These shock waves can be intense enough to damage a water pump, crack a block or a cylinder head.
I have seen this effect on the rolling road on high rpm engines, it is as if the head gasket has failed....water blows out from the rad cap or overflow.
I have also seen cavitation caused by steel impellers on a B rather than a cast one. Blew the Payen head gasket! We also fit a rubber bumper water pump pulley to slow the pump down.
Peter Burgess Tuning

Still not convinced!

The article says

"It is often incorrectly believed that a pump [or a boat propeller] that cavitates creates air bubbles."

That is wrong - boat propellers certainly do not form steam! - see

The article is confused over cavitation and nucleate boiling.

The pressure reduction on the backside of the pump vanes will cause water at over 100 degrees to suddenly boil - nucleate boiling, with its attendant shock waves - so I think the issue is not cavitation at all!

The primary function of the water pump is to produce flow to assist the thermosyphon effect and to reduce the incidence of nucleate boiling by removing the localised hot water.

When running freely there IS a pressure difference across the pump - it is that difference that produces the flow. But block the outlet and the pump stalls - no flow, no pressure across the pump.

Yes, slowing the pump speed will result in less pressure reduction across the pump vanes and avoid or reduce the boiling in the pump but it will also reduce the flow through the head which may lead to boiling there.

Increasing the rad cap pressure will increase the boiling temp of the water, reducing the incidence of boiling in the head and the pump.
Chris at Octarine Services

We're probably getting in a bit deeper than Mike's problem really but

Not on an MG but a little story for you-
On our old methanol burning Chev the water pump was driven off the front of the crank and when we taught it to rev properly it was always blowing welch plugs out (cup type)
Just by luck I happened to see one of the other teams' pumps apart and it had less vanes on the impeller than ours- pulled ours apart and ground every second vane off and never blew another even on the longer tracks where the revs were right up there for longer--------------water pump pressure-?

Also, with rattling bores doing damage,
It's well known with diesels getting little vibration bubbles on the sleeves and spearing pin holes straight through em
Ford actually sell an additive for the coolant in the powerstroke diesels that you poke in every 7,500klm to modify the coolant to prevent bubbles forming on the sleeves and to limit the chance of cavitation
If it's not used ,200,000klm pulls them up and the sleeves are gone and the waterpump housing near the impeller is all eaten away from cavitation

I think it would be safer for Mike to never start his engine again------------lol
William Revit

Just curious. My friend is rebuilding the engine in a late '80s BMW M6. When the block came back from the machine shop I noticed that the plugs are inserted concavely instead of convexly as in an MG I think the theory is that the water pressure will press on the dome pointing in and make it expand as opposed to the water pressing against the dome pointing out making it contract. Does this make any sense? I can't imagine there is enough water pressure in the block to make a difference but I really don't know.

Just wondering.


J. K. Chapin

I would imagine that the BMW has cup type plugs which as you say look concave ,but they have a support edge around them and are already an interference fit as they come and have to be forced in the hole by hammering with a suitable sized punch or socket that sits on the outer edge of the plug
The dish type plugs (convex) are same size as the hole and slide in ,up against a shoulder, then have to be expanded by reducing the amount of dish in them, (hammered flatter)

Hope that explains it for you
William Revit

Perfectly. Thanks Willy. Jud
J. K. Chapin

I thought I would bring you up to date with my core plug replacement (I have been away for awhile touring Norway and surrounds in a modern mgtf!).
I decided in the end to go with the Dorman repair plug described here
for a couple of reasons, it is hard to hammer a disk plug in an MGA and I have a good store of Dorman plugs (I ordered (5) a few years ago and they sent me 5 packets of 5 for the price of 5 - you have to love Amazon) so I thought why not test one on a 1950 cc engine that gets a bit of a thrashing. So far so good , about 100 kms and a round of the Interclub hillclimb at Rob Roy and another hillclimb this weekend.
The trick to holding them while tightening the acorn nut is using a 7/8" offset ring spanner. I used Wurth Flange Sealant to ensure seal around the edges (I use this as head stud sealant also, on the advice of my engine machinist).
I will send in another post my first attempt where I over tightened the nut and snapped the bolt - it shows how they deform uniformly on the coolant side.

Mike Ellsmore

See attached pics.
Left hand side shows before and after tightening the acorn nut. RH side shows symmetrical deformation on the cooling side and results of my aggressive tightening of a 5/16" UNC bolt (I had to drill a couple holes through it so I could lever it out).
If you don't see another post here you will know it is still holding together!

Mike Ellsmore

Very Edwardian - reminds me of the time machine in the film of the same name :o)

Reminds me of Chitty Chitty Bang Bang ...
Chris at Octarine Services

Ha ha ha ha---
---it reminds me of a Lancia Fanalone

William Revit

This thread was discussed between 11/06/2019 and 08/11/2019

MG MGB Technical index

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