^^^Is it solely because even though a coolant mixture in the above concentration ranges will raise the actual boiling point of the solution, a 100% water fill will actually
conduct the engine's heat away from the water jackets in the engine and into the water to be dissipated by the radiator tubes/fins and into the air, much more quickly and efficiently?
Every interface between the outside air and the engine block requires a temperature difference in order to heat across it.
Very simplified (there are multiple steps to each one)
- Radiator to air (conduction through the metal and surface effects)
- Cylinder wall to cooling jacket (conduction through the metal and surface effects)
- Coolant (in this case convection into the water from the cooling jacket, heating it up, at which point it flows to the radiator, where the radiator cools it down)
The sum of temperature changes at each interface determine how much heat is transferred. Or said differently, with a more efficient part in the system, the same amount of heat can be transferred with a lower temperature rise. The cylinder liner we can’t do much about, but adding a “better” radiator reduces the temperature rise across the interface between the air and the coolant.
In the case of the coolant, the temperature rise (basically stored heat) in the coolant is related to the heat flow. More heat, more temperature rise.
But since water holds more heat for every degree of increase, you can transfer the same heat with less temperature rise in the water than with glycol. Water holds about half again as much heat as pure glycol, blends come out somewhere in between.
But since we have a more or less fixed amount of heat to pull out of the engine for a given fuel and air input, and the exterior of the radiator side is pegged at outside air temperature, reducing the temperature rise at any step means keeping the cylinders cooler. And since what we really care about is cylinder temperature because of clearances, this is good, and the thermostat is there to make sure we don’t coo
Likewise if we turn up the boost and we need to dissipate more heat, the temperature difference between the cylinder wall and radiator rises. Unfortunately with the stock radiator, this results in overheating the cylinder walls.
So water is better in pretty much the same way a more efficient radiator is, as cuts a few degrees out of the chain.
For water and coolant below boiling, the temperature change is more or less linear per degree of heat transferred, and since our 50/50 mix is somewhere around 80% of the heat capacity of water, that means a 20% higher temperature rise. But the difference in boiling point between 50/50 and pure water is only about 10F, which while it sounds good on the surface, doesn’t matter a lot - you could make up the difference by running a radiator cap with 3-4psi higher pressure.
While the boiling point *might* save you on an extremely hot day going up a hill with a moderate load, the problem is that turning up the power and flogging the car, we’re trying to dump more heat, and by the time either water or mix coolant boils, you’re very close to fubar already, especially with modern engines designed to run hotter even under normal conditions (and already equipped with much higher pressure systems 23psi vs the 13-16psi systems of decades past.)
If anything, on a heavily stressed (track,etc) car, the lower boiling point of water is going keep your engine cooler for longer, but also pull you into the pits sooner and save your engine from even higher temperatures if something goes horribly wrong.
That was a bit of a novel, but the main reason to use glycol in your cooling system is the much lower freeze point.