MG-Cars.info

Welcome to our Site for MG, Triumph and Austin-Healey Car Information.

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 GT V8 Factory Originals Technical - Porous block

I was describing to a well-known engine builder problems I was having with my factory V8 cooling system pumping up with air. I described how I had tested the coolant for combustion gases as I suspected head gasket but that came up negative. He suggested the block might have become porous. Two things strike me here - I had already said there was no combustion gases in the coolant, and doesn't the allow block have steel sleeves in the bores? Am I missing something or is he?
Paul Hunt

Unless someone replaced your engine with a Vega engine you have iron liners, and I don't think your block, however porous, could act as a semi-permeable membrane, allowing air in but no water to escape. The most logical place for air to get in is the coolant recovery circuit, as I'm sure you are aware. I suppose it would be possible for a failing water pump to cavitate and suck air without leaking. It also seems that many have had problems on the 3.5 with the thermostat bypass and/or heater circuits.
George B.

I had a problem like this on another car once. The coolant recovery tank had two outlets. One went to the rad and had a long tube that goes to the bottom of the tank. The other is an overflow for the tank. If they get switched, your rad pumps out coolant when hot but sucks air on the cool down. I took a couple of times in the shop to figure out something that sounds so simple.
Something to check.

Kelly
Kelly

George - I only posted the very short version of what is becoming a very long tale. The pressurisation occurs progressively while underway and air is building in the top of the rad forcing first air, then coolant out of the cap on the expansion tank.

Kelly - my system is as you describe but the pipes are definitely in the correct places.

Thanks gents, any other ideas? I really need to get to the bottom of this before a rebuild.
Paul Hunt

Paul, when does the coolant/air get blown out - during operation or when the engine is shut down?

If during engine running, will it do it while idling in the driveway?

Should we infer that the engine temperature doesn't go up anytime?

Wayne
Wayne Pearson

Could you be ovrfilling the system?
Could you have an airlock when filling, eg heater?
Personally I would suspect head gasket, as you obviously do.
The MGF has what they call "jiggle" valves which let air out but no water in or out. There are similar items available for domestic heating systems.
T Green

It's been my experience that a failing head gasket can pressurize the system under a load, and not during idle or cruise situations. And it will not necessarily put coolant in a cylinder because it can still hold cooling system pressure out yet it will let the much higher combustion pressure leak by. I would try a leakdown test on it, it's possible you might find a bad cylinder.The block porosity problem is more a diesel thing, very common in the 6.9 Int'l for example.I have seen some Rover blocks crack behind the sleeve that we were able to weld and repair but hopefully this isn't the case here.
Dale

Gents:

Pressure builds very gradually during use, I can see this happening on a pressure gauge I have installed by T'eeing into the pipe that leads from the rad to the expansion tank. After 60 or so miles when starting from cold the pressure will reach that of the cap then stabilise as the cap starts venting, this is with a 20lb cap as well as the correct 15lb. The 20lb cap does seem to allow me to go further before bleeding the rad, but eventually blew the bottom hose 3 miles after I got back from Le Mans! I have also fitted a coolant level sensor in the rad and as the pressure rises I can eventually see the level dropping. I also have a catch tank on the expansion overflow. At some point I decide the coolant level has dropped enough so stop. I can remove the pressure cap and/or rad plug within a few seconds of stopping, with much hissing of air but no steam i.e. it isn't boiling. The expansion tank is completely full, the catch tank usually has some in it, and the top of the rad is full of air. I transfer everything back to the rad and continue on my way. The pressure starts rising again but very slowly, and this time I can usually go more than 100 miles before having to stop again, presumably because the pressure increase due to warm-up has already occured so I am only seeing the air build up.

It hasn't done it idling in the driveway, possibly because I haven't left it for the hour or two that it would require! That is a possibility, but there are no air bubbles when the system is fully up to temp and the fans have cycled on and off a couple of times, even when revving the engine hard.

The only time the temp goes up is if I let the coolant level drop too low, which is to be expected.

Definitely not overfilled, when cold there is perhaps only an inch in the bottom of the expansion tank, which rises to about mid-way as the engine warms up.

An air lock due to blockages in the coolant passages is a possibility, a system flush with clean water came out perfectly clean. I have warmed the system up with the car in a 'nose up' position until all air bubbles ceased, but it still happens when I get back on the road.

Head gasket was my first thought but the coolant and air in the system has been tested several times under varying conditions and is showing no contamination.

See what I mean about a long story?

Keep the ideas coming, chaps.
Paul Hunt

Paul,
Facinating problem. How has the problem developed? Suddenly, progressively over a period of time or is this a "new" engine or car? This may be significant and will help the thought process of those racking their brains. I was told some time ago that 3.9 blocks were susceptible to porosity, but I have no evidence to support this (a typical "bloke down the pub" story?!!). Logic would suggest that block porosity would become a problem progressively and not not suddenly? Good luck. Pete
Pete Green

Going back to the original question how would porosity become a problem on a sleeved block, which is a question I ask in all innocence? The problem started suddenly earlier this year, on a factory-spec (as far as I know, although a mate with an uprated factory car did say he had trouble keeping up with me on the way back from Le Mans) car I have been running as a daily driver for eight years and more than 60k miles. I will say that it has tended to vomit after switching off for as long as I have had it, but fitting a new rad cap has always cured it for a year or two. It is not vomiting now, just pumping out whilst underway.
Roger Parker, a contributor to this BBS, has said since I first posted this that in his experience of a fleet of V8s he has known of a couple that vomited or just ran hotter than the others, or both, until they replaced the short engine i.e. block, crank, con rods and pistons. I feel the end is nigh ...
Paul Hunt

The only thing that I can think of now is that there is a localized hotspot in your engine that is causing the coolant to boil, but not raise the overall temperature of the cooling system. That could explain the non-combustion gas(es) in the coolant. I'm not sure, but I don't believe a failing water pump could suck air in through the seal on a pressurized system - although sluggish circulation could account for hot spots.

Grasping at straws,
Wayne
Wayne Pearson

On a cast iron block porosity can eat right into the cylinder and coolant can enter the base or get into the chamber.Mostly when this happens it was an air pocket in the casting that was bound to fail.I have even seen this on a brand new Chevy Bowtie thickwall block.I have always thought that this was the reason a lot of British motors have dry sleeves from new,to eliminate the possibility of a bad casting ruining a block.I don't think porosity is your problem here. Another thing you might want to try is to pull the spark plugs and pressurize the cooling system and let it sit overnight.If you have a bad head gasket or crack it might leak coolant into a cylinder and will spray out of the plug hole when you crank it over with the key.
Dale

Dale - as I say several coolant checks have come up negative for contamination from combustion gases, but pressurizing the system when cold and seeing what happens (to the pressure at least) overnight. However if 15psi (or 20 if I use the other cap) is enough to cause leakage from the coolant to a cylinder would it be reasonable to expect massive leakage from the cylinder to the coolant under cylinder pressures?

Wayne - that is one of my straws too, a very localised boiling at one spot causing a small but steady stream of bubbles like in the bottom of a saucepan. Twice when I have drained and refilled the system I could only get back in about half the stated capacity. Either that's a massive air-lock, or the block and heater retain more coolant when 'drained' than I would expect. For a big hot-spot I would expect to see a big rise in temperature after switch off but it never goes above half-way to the red zone, i.e. about the standard switch-on point for the fans, i.e. about 90C/194F. Neither are there any bubbles in the expansion tank once the system is fully up to temp.

I'm picking up the component parts of a short engine from the aforementioned Roger Parker this week - very generously for free as he needs the space. I'll get the bits evaluated for a rebuild, and as long as the block is OK at least I'll have a 'second string' if my block turns out to be bad when finally stripped. But I would still like to get to the bottom of the cause ...

Paul Hunt

Paul, you say you have an expansion tank and a catch basin. The layout could be a cause of the problem. A true expansion tank should be pressurized, directly connected to the system at its highest point, and be preferrably higher than that point. In this case the catch basin would be unpressurized and connected via siphon tube to the expansion tank overflow. Since you say the top of the radiator empties out while the expansion tank is full it's clear your system is different from that so maybe that will help some. Also, some BOP blocks had core shift when cast, resulting in the iron sleeves being offset in the alloy cylinders. This created a thick wall on one side and a thin one on the other, about as good a recipie for hot spots as I can think of. No way to know without pulling the heads though, whether your block suffers from that problem.
Jim Blackwood

Jim - factory setup of tube from almost (minus about 1/2") the top of the system (rad) right beside the top hose inlet to a pressurized expansion tank that is slightly below the highest point. This has an overflow tube that originally dumped on the ground, now routed to an unpressurised bottle lower than the tank. I have given considerable thought to why coolant is being pumped into the expansion tank, and from there to the catch tank, leaving air in the rad and have come to the conclusion that it is due to the relative locations of the inlet from the top hose and the outlet to the expansion tank. While coolant is flowing there is a large flow from the top hose inlet past the end of the expansion outlet, and if there is anything going to flow down the expansion pipe due to there being a higher pressure in the rad than the expansion tank there is a fair chance it will be coolant. To test this out I have fiddled a metal pipe in through the rad filler hole and out through the expansion pipe and soldered to it. Now the open end of the expansion pipe really is right at the top of the rad, and well away from the top hose inlet, both factors ensuring it is in any air that is present and not coolant. Early days yet but since doing that I still have pressure building, but when it exceeds that of the cap it seems to be the built-up air that is being transferred into the expansion tank and from there out of the cap rather than coolant, and the coolant level in the rad is remaining at a satisfactory level. Not a true fix, but if it continues to work at least I won't have to stop every 100 miles or so and give it a 'blow job' to push the coolant back from the expansion tank into the rad. Only until a rebuild with or without a replacement block.
Paul Hunt

Paul
Back in the mist of time, I had an 1800 with a warped head - the gas test showed nothing and the local garage owner removed the rad cap - filled the rad to over-flowing before starting her. She then blew bubbles.

You have propbably tried this - but visible circulation also showed that the water pump was working.

My money is also on a hot spot but I would just check the thermostat is not jammed shut. Notoriously unreliable things (Just take it out for a while)!

Good luck
Roger

Paul,
Sounds like an excellent dodge. Another good option would be a smaller pressurized vent line from the top of the expansion tank back into a high point in the radiator, or some other high point in the system. The idea being to have a higher air pressure at the new vent line than the inlet to the expansion line in at least some point in normal engine operation. In this mode the expansion line does not necessarily have to be at a high point in the radiator,(your original point close to the hot return would be good, as the expansion tank can provide extra cooling) as the vent line will allow liquid level equalization through the expansion line. This allows the system to equalize much better during steady state operation.
Jim Blackwood

Roger - have tried that, until the fan cuts in for the first time there have been some bubbles, particularly when the engine drops back to an idle after revving. But after the fan has cut in and out for the first time I have seen none. Interestingly, because revving the engine causes a pressure drop of a couple of pounds at the top of the rad this has the effect of sucking coolant back from the expansion tank then pumping it back in as the revs drop, I can see the level rise and fall. Anyone else see this?

Similarly, any factory V8s like to report the level of the coolant in the expansion tank when cold, when the fans cut in, and when they cut out? Mine is about 1", about mid-way, and very near the top. Cap off all the time.

Jim - I may be misunderstanding but wouldn't a pipe from the top of the expansion tank to the top of the rad tend to prevent the rad drawing *coolant* back from the expansion tank as the system cools, drawing air back instead?
Paul Hunt

Well your waterpump is working - and if there is no blockage either at the thermostat or internally on account of an old blanking plug or the like - could the suction effect of water movement be sucking in air through a split component - say the fan switch ?

Paul - have ypu got any tell-tale damp joints/engine plugs ?
Roger

It looks to me as if there is some confusion on how the remote tank works. The tank with the pressure cap on it as found on original MGBGT V8s and later MGBs has one primary purpose, to remotely locate the radiator cap so an MGC like bonnet (hood) would not be necessary with the taller radiator positioned forward.

Because this remote radiator cap has a reservoir it can perform the same function of maintain a full radiator as the coolant level fluctuates some just as a non-pressurized overflow tank would. Both systems tap air from the top of the radiator and route it to the bottom of a reservoir so it bubbles up and is expelled first. As either system cools, coolant is drawn from the bottom of the reservoir. The non-pressurized system keeps the pressured system full and the system in the MGB will always have air in it. Cooling performance will only be hampered if the coolant level drops enough to allow air in the radiator. Adding an overflow tank to the MGB remote system ensures the coolant level does drop in the radiator and is easy check without opening the system.

Neither of these systems requires the coolant reservoir to be mounted above the radiator because the coolant is drawn from the bottom of the tank.

An alternate system is the surge tank. A surge tank also has a reservoir, but with two lines; a large one connected to the bottom of the radiator and a smaller one connected to the top of the radiator. As a surge tank is filled coolant flows the bottom and air in the radiator moves through the top line and out during fill up. A surge tank must be higher than the radiator or air won’t exit the radiator.

The suggestion to add a second line between the top of the radiator and the stock tank sound like an attempt to convert to the stock system to be a surge tank. There are a couple of problems with that; first the stock tank is mounted no higher than the radiator and the original line would not function efficiently because it also connects to the top of the radiator and routes up and down.
George Champion

Correct George. The surge tank also has the ability to radiate a small amount of heat, supplimenting the radiator if designed so that it has a positive flow through the tank as was done on some early sixties Ford sedans. However, use of the second line with the MGB tank should work *if* coolant in the tank is kept above the level of the top of the radiator, *and* the lower line is not higher than that at any point.

Paul, the system works primarily on gravity, but its action can be enhanced by small pressure differences in the system caused by the operation of the water pump which will vary according to running conditions. I stand corrected, a connection at the water pump inlet rather than hot water return would tend to suck water from the surge tank, and draw either air or water from the top line (by the hot water return) into the surge tank. I had it backwards as George points out. Use of a top line the same size as the bottom line should allow some level of fluid circulation (depending on size)through the surge tank in a topped up system, then use of a non pressurized recovery tank should allow the system to stay full.

Jim
Jim Blackwood

Roger - I have noticed a little dampness round the thermostat housing - even after flatting, a new gasket and sealant - and the front cover gasket but that was for a very long time before this pumping-up started. One of the core plugs weeped very slightly when I first tried using a 20lb cap *after* the problem started, but stopped after a few days.

Even after short heating/cooling cycles, once the system has had air *added* to it, it remains above atmospheric pressure even when fully cold, and I believe that is stopping the normal process which purges any air from the rad i.e. after draining and refilling. On a long enough run the steady build-up of pressure from air being introduced into the cooling circuit eventually vents through the cap, and in an ideal world this would only be the introduced air after it had bubbled through the expansion tank. But because of the proximity of the expansion pipe to the top hose inlet, the excess pressure is taking coolant with it, so the rad gradully empties and the expansion tank gradually fills, at which point it starts venting coolant. I don't want to do any permanent major replumbing to get round what after all is a fault elsewhere that has recently arisen, the small mod I have done to the expansion pipe inside the rad is easily reversible.

Picked up the short engine from Roger Parker this morning, everything looks in pretty good nick. So it is off to Peter Burgess shortly for a report on its potential for a rebuild, then I have to make the decision as to which way to go.

But I would still love to get to the bottom of what is wrong.
Paul Hunt

Check your head gasket, block and head surfaces very carefully when you tear it down. Also look for very small cracks in the area of the valve seats or below the exhaust seats. Pressure from the combustion chamber has to be getting into the water jacket somehow, but in minute enough quantities to not show up in the test. The combustion chamber is the only part of the engine that is capable of producing an air pressure of 20 psi over ambient. If it were vaporized coolant, condensation upon cooling would produce the expected pressure drop.

jim
Jim Blackwood

This thread was discussed between 10/09/2002 and 19/09/2002

MG MGB GT V8 Factory Originals Technical index

This thread is from the archives. Join the live MG MGB GT V8 Factory Originals Technical BBS now