On Fri, 13 Jun 1997, Michael Fisher wrote:
> Sorry Ray but I think you are missing some physics of the problem. I do fluid
> dynamics for a living and have studied this problem in great detail (along with
> several engineer and physicist friends around the country). I did not believe
> this myself until I looked at the problem and tried it on several different
> cars.
I said that the thermostat might prevent the water pump from cavitating
(so flow would be faster with an open thermostat than with no thermostat),
and might direct flow so as to prevent hot spots, again providing better
cooling with the thermostat in place. What I find hard to believe is that
the thermostat improves cooling (at any ambient temperature) *because it
slows coolant flow*. The argument that the thermostat slows flow, thereby
keeping coolant in the radiator longer so it has more time to lose heat
makes no sense to me. Slowing flow results in a temperature gradient in
the radiator, which means the cool side of the radiator is not
transferring as much heat as the hot side, and it also means that the
engine passes more heat to each aliquot of coolant in the block.
If I understand you correctly, you say that you and several physicist
friends studied this extensively and you did not believe it yourself
until you tried it and it worked. Did all your studies lead you to
believe it was not true? I've thought about it a lot, too, and I
conclude that faster coolant flow is always better. If removing the
thermostat makes the engine run hotter, then either the thermostat helps
direct flow to give an even temperature (certainly possible) or the flow
actually slowed down when the thermostat was removed.
If, on the other hand, your analysis led you to a sound theoretical reason
why slowing flow would improve cooling, then I'd appreciate your
explaining the physics in an understandable way.
W. R. Gibbons Dept. of Molecular Physiology & Biophysics
Univ. of Vermont College of Medicine, Burlington, VT
gibbons(at)northpole.med.uvm.edu (802) 656-8629
This archive was generated by hypermail 2b30 : Tue Sep 05 2000 - 09:44:16 CDT