The MGA With An Attitude
What Happens When It Boils? - YIKES ! - CO-200C

A pressure cap allows cooling system operation at higher temperature without boiling. Using a 7 psi pressure cap and plain water you can peg the 230dF temperature gauge without boiling. Using a mix of ethylene-glycol antifreeze and water it can go even higher without boiling (although glycol carries less heat than water).

But there is a limit to this magic. At some point the temperature may go high enough to make it boil even with the moderate pressure applied (somewhere beyond the end of the gauge scale). As the coolant temperature approaches the boiling point you can have some strange things going on inside the cooling system, and the engine might self destruct with hardly a notice.

There are substantial temperature differences at various points in the cooling system, which is what drives the heat out of the engine. The hottest points will be in the cylinder head near the exhaust valves. The coolest points will be at the bottom of the radiator. When the system gets hot enough to boil, it will start boiling first in very small hot spots around the exhaust ports in the head. Then the strange fluid dynamics will take over.

As the fluid begins to boil at the surface (inside of the cooling jacket of the head) it forms a thin layer of steam close to the hot surface. By laws of thermodynamics, it takes a large amount of heat to change a small amount of water into steam (the latent heat of evaporation). As a result you get only a small amount of steam at first. Very close by on the other side of that thin film, the steam, being hotter than the water, is conducting heat into the cooler water, and the steam is condensing. In this manner the steam is carrying heat away from the head and depositing it in the liquid nearby. But this only works in a very narrow range of temperature. Unfortunately steam is not a very good conductor for heat.

For this to continue with only slightly higher heat flow, some of the liquid water must be able to move around to come into intimate contact with the hot metal surface. The end result then is a great deal of turbulence. Water explodes into steam bubbles at the surface and moves suddenly away from the surface due to a sudden local increase in the steam pressure. Some nearby liquid then moves in to replace the escaping steam bubbles. The steam bubbles come into contact with slightly cooler liquid nearby and lose some heat energy, then suddenly turning back into liquid. The resulting collapse of the steam bubble promotes more turbulence in the water. This results in a violent local boiling action which has the potential to make a lot of noise, but at first may not expel very much coolant from the system.

When you see (or smell) the first wisp of water vapor from beneath the bonnet, get off the throttle, and find the first possible place to get off the road and stop. If the vapor does not stop immediately, shut the engine off, take a deep breath, and have a LOT of patience.

Please do NOT remove the pressure cap at that time, as that would result in a violent steam explosion. As the pressure drops suddenly a substantial amount of water must instantly turn into to steam to reduce the remaining water temperature to the threshold boiling temperature at normal atmospheric pressure.(212dF). This instant conversion of water to steam happens throughout the water volume as the pressure is released. The resulting appearance of high pressure steam throughout the entire volume will expel large amounts of boiling water (and equal temperature steam), tending to scald anyone in the vicinity (namely the unfortunate person who just removed the pressure cap). The steam burns will be worse than a burn from scalding hot water, because as the steam hits your skin and condenses it gives up huge amounts of latent heat, and your skin cooks rapidly.

Now assuming you do not remove the pressure cap at this time, most of the cooling system will still be operating normally. Most of the water will still be liquid and still flowing normally (except possibly carrying some small steam bubbles along with it). The radiator may still be dumping most of the heat as required to prevent boiling. Only the small amount of excess heat in the head which the radiator cannot quite get rid of will be causing the boiling, and only a small amount of water will be involved in the boiling action. As a result only a small amount of water converted to steam will remain in the steam state.

The small volume of fluid which remains in the steam state will be pushing fluid out of the overflow at the neck of the radiator. At this stage the overflow fluid will be almost completely liquid with only a small amount of escaping steam, but the overflow can increase in a very short time. If you have the good sense to spot a little escaping steam, get off the road, and shut the engine down, then you have a good chance of getting out of this unscathed. While the coolant is boiling gently under moderate pressure it may be hovering around 250dF. This temperature alone is actually not detrimental to the cylinder head. Plenty of newer cars run with higher pressure radiator cap and run all day long at 270dF with no problem.

The solution to your immediate dilemma is to wait patiently until it stops boiling, then wait a little longer until the temperature gauge drops back to 212dF. While the temperature is dropping from close to 250dF back to 212dF there is pressurized steam in the system. Steam is a gas, so it is compressible. As such there will be a larger amount of steam squeezed into a given space at a higher temperature and pressure. As heat can dissipate the temperature can begin to drop, and some of the steam will condense. At the point where the system is back to one atmosphere of pressure, any remaining steam inside the cooling system can continue to condense, and the system pressure will drop slightly below one atmosphere.

If you do not have any remote recovery tank you will have to wait patiently while the system dissipates some more heat to condense any steam inside back into water. During this process it will be drawing in air through the radiator cap return vent to fill the volume being vacated by the condensing steam. The temperature will remain at 212dF until all of the steam has condensed. Then as the system continues to lose heat the temperature will finally drop below 212dF.

If you did have a remote recovery tank, and there is still liquid coolant in that remote tank, then the story goes a little different (and a lot faster). As soon as the temperature drops to 212dF as the steam is condensing it will draw liquid coolant back into the radiator rather than drawing in air. Liquid in the remote tank may by this time have cooled down to substantially less than 212dF. As this cooler liquid is drawn into the radiator it can promote more rapid condensation of the remaining steam, and the liquid may be sucked back in quite quickly. Most likely all of the liquid will be transferred in short order followed by sucking in of additional air) after which you still get to wait al little longer for the temperature to drop significantly below 212dF.

When it gets down to perhaps 200dF or so the steam will have long since condensed and the pressure will be gone. If you were a brave soul (or impatient) you might find a heavy rag and carefully remove the radiator cap. Do NOT immediately pour cold water into the radiator while the engine is still hot. Right after this sort of boiling and expulsion episode the system is bound to be low on liquid. If the liquid level is below the level of the engine head gasket there may be no liquid in the cylinder head. In that case the surface of the cooling jacket around the exhaust valves may still be quite hot, possibly substantially higher than 212dF.

There may have still been liquid coolant in this area while the engine was still running, as the water pump has the ability to push the coolant up from the radiator to the cylinder head. When you shut it off the coolant level in the engine might drop suddenly to match the level in the radiator. This can leave the cylinder head high and dry while it is still cooking at fairly high operating temperature around the exhaust valves. Keep in mind that there has to be a temperature difference to drive the heat out, so the exhaust valves and valve seats will be warmer than the surrounding coolant, which may itself have been around 250dF just moments earlier.

With the coolant suddenly drained away from the cylinder head, a little heat soak and stabilizing time can bring the heat out from the higher temperature internal parts to increase the temperature of the surface of the cooling jacket in that immediate area. In this case the surface temperature could become higher than the temperature of the now absent coolant (somewhat higher than 250dF). This may still not cause any permanent damage to the engine as the far internal parts of the valve gear regularly run at higher temperatures.

However, .... If you were to pour water into it at that time some of the water might flash almost instantly into steam as it hits the hot spots it the cylinder head, and the surface tries to suddenly reduce to 212dF. This sudden thermal shock (hot in one place and suddenly cooler nearby) has the potential to cause cracks in the cylinder head around the valve seats. Thus you may cause yourself a very expensive repair problem where there may have been no damage at all just a moment before.

Have patience in this situation and give it plenty of time to cool down before you put more water back into the radiator. When you do put water back into the radiator, especially if it is still warm, pour very slowly, and be prepared to stop pouring immediately if you notice it suddenly percolating and spitting at you. This is indication that there are still hot spots in the cylinder head which could cause cracking from the thermal shock. Take an additional 15 minute break first, maybe avoid a $500 to $1000 repair bill, and maybe drive away with fresh water and no problem.

Also of course you need to figure out why it overheated and make plans to prevent it from happening again.

Thank you for your comments -- Send e-mail to <Barney Gaylord>
© 2005 Barney Gaylord -- Copyright and reprint information