More on V8 engine cooling
Tony Lake (Flamenco Red 2489) from Northamptonshire sent in a useful note as a follow up to Bill McCullough's painstaking research into the cooling system used on the Factory MGBGTV8s. (Apr 08)

I read Bill McCullough's report on V8 cooling with interest. It took me back to the mid 60's when I ran some cooling tests on a V8 diesel engine. There was a lot of doubt about the development data because of overheating reports from service and pretty clear evidence that truck radiator capacity was too small. Initial data was very variable, mainly because of difficulty in measuring coolant temperature rise accurately and repeatably. Finally NPL certified 0.1degC resolution mercury in glass thermometers were used and then it was it was possible to discern the effect of changing ambient temperature on heat rejected to coolant. Up until then it was acceptable to do a heat balance knowing fuel consumption, power output, heat to exhaust gas, alleged heat to coolant with a rider that the gap was due to unmeasured radiated heat from the engine and other measuring errors. One test allowed the ambient temperature to increase over time whilst I continued to measure coolant temp rise, there was a linear relationship between increasing heat rejected to coolant and reducing heat transfer radiating from the engine as the ambient temp increased, I quit at about 60degC when I got a chill from running in and out of the hot test cell.

My experience of overheating with a Factory MGV8 goes back to its purchase in 2001. It had leaks, a sticky thermostat, a broken Kenlowe controller and a dud block coolant sensor that operated the cooling fan. Even after I removed the Kenlowe package and fixed all the other bits it still took a long time with the car at rest for the cooling fans to do their job and drag the ccolant temp down. If the car was moving and generating a bit of ram air then equilibrium was restored pretty quickly.

I realised my knees were getting warm and then then I realised the cooling fans were sucking air from the engine bay

Whilst investigating fan operation I realised my knees were getting warm, and then noticed the cooling fans were sucking hot air from the engine bay which explained why the cooling down process took so long. With the cooling fans fitted in the right direction as pushers and the car at rest in traffic top tank temperature rises and falls, as the fans do their job, the engine gets hot but it is under control and never in the red, at least in the UK. I wonder how many V8s still have this problem? I also came across it on a TF21 Alvis.

I believe the root cause of high coolant temperature can be laid at the door of inadequate underbonnet ventilation

I believe the root cause of high coolant temperature can be laid at the door of inadequate underbonnet ventilation. At low forward speed and high engine power output there is a clear relationship between heat soak from the exhaust manifolds and low airflow which reduces the capacity of the slow moving underbonnet air to absorb radiated heat, the result is an increase in coolant outlet temperature, the heat has to go somewhere. The heat rejected to coolant increases as the air temperature around the engine increases. The high pressure pick off point for heater supply and its return to suction side of the water pump is good practice generating a sensible pressure drop to encourage flow. The problem of overheating is exacerbated by opening the heater valve when the coolant is already approaching a critical point and in effect doubling the amount of very hot coolant that bypasses the radiator. The heater core and its fan are not very effective at extracting heat so the temperature drop back to the suction side of the water pump is negligible. It is likely that the temperature drop from top tank to bottom hose is less than 10degC, so the cycle once started is vicious and
Back to Contents listing

self sustaining, unless power demand is reduced and engine speed is reduced by pulling as high a gear as possible or the heater is switched off!

The thermostat is a vital part of cooling system control, it has to vent air and also help provide a stable engine operating temperature so I am loath to modify its function. I have driven in freezing conditions this winter and at motorway speeds noted how the gauge remains on the cool side of normal, the thermostat is clearly not fully open so in those circumstances allowing low temperature coolant into the top tank via extra holes in the thermostat is of no benefit to engine or heater operation, that coolant is replaced by very cold liquid from the bottom hose of the radiator. At lower road speeds in town the engine runs a little warmer, as the ram air effect reduces. My heater is adequate and although it is adjustable I can switch off the fan if it gets too hot. The best operating conditions for the engine are when oil and coolant temperatures are as close to each other as possible, in the range 85/90degC. Every 6degC above 93degC mineral oil oxidation rate doubles, accelerating additive depletion rate. Low temperatures encourage condensation and sludging as well as increasing oil shearing resistance. I suspect that is why 82degC is such a popular thermostat range in older cars. In winter there may be a case for blanking off part of the radiator to modulate the ram air effect, but then one would have to be quick off the mark to remove it in standing traffic.

The best operating conditions for the engine are when oil and coolant temperatures are as close to each other as possible, in the range 85/90degC

The pressure cap is another key component, it deals with high ambient temperature by raising the boiling point of the mixture, it also suppresses afterboil in a full cooling system, but it can't cope with a leaky one. Changing the pressure cap for a lower pressure type will not alter the temperature at which the coolant reaches equilibrium, that is determined by the engine heat rejection rate, ambient temperature and flow rate of the cooling ram air or the electric fan air supply and subsequent coolant temperature drop through the radiator. A 15psi cap will raise the boiling point of water to 120degC at sea level. For every 1000ft above sea level the boiling point of water reduces by approx 1degC. If the effect of antifreeze is factored in then the boiling point of the mixture under the same conditions is raised further. This compound effect is very beneficial. There are more benefits, antifreeze in the form of a glycol, usually propylene rather than ethylene which is toxic, contains corrosion inhibitors which are vital to protect the water side of a wet liner engine. Before the cooling system warms up and the thermostat is still closed the suction induced at the water pump can result in cavitation which will reduce coolant flow and damage the water pump impellor. Once the thermostat is open the eye of the water pump sees a positive pressure which stops cavitation. This extra pressure is also seen at the cylinder liner which can be excited by piston slap on the power stroke. This instantaneous vibration can cause cavitation which might eventually be terminal, however the inhibitors and high coolant pressure combine to reduce the likelihood. The very hot areas in the cylinder head adjacent to the exhaust valves depend on good coolant flow, local nucleate boiling of the coolant will occur in this area which a good pressure cap and correct glycol/water mixture will mitigate. Engineers like this scenario, plenty of virtue out of basic necessity.

(Updated 18.4.08)

You can refer back to Bill McCullough's V8NOTE374 by following the link below. As this article runs to eleven pages with photos and diagrams, it can be downloaded and viewed as a PDF document. It is included in Volume 11 of the V8 Workshop Notes series available on CD. V8NOTE374
Copyright reserved by the V8 Register