Overheating coolant

[etsuter@xxxxxxxxxxxx]

On 08/22/98 at 08:33 AM,

  FERDINANDO DI MATTEO <fredalfa@domain.elided>  said:

> Subject: Re: alfa-digest V7 #51RE: Overheating coolant

> I'm surprised at how many of you have no idea what the purpose of
> the coolant thermostat is.  Reasons for overheating are many and
> guessing solved nothing.  I shall try to tell you what I've learned
> about the thermostat.  It is there to regulate the flow of coolant
> through the engine and radiator.  If the coolant flows too fast
> through the radiator, it cannot lose enough heat before it gets
> heated some more with each pass.  An engine, any engine, depending
> on a radiator to cool its coolant will overheat very badly with
> thermostat removed.

I'm afraid that I have to take issue with Fred's explanation here.  I
agree completely that one normally needs a thermostat or its
functional equivalent, but I don't agree with the reasons given. 
Those who know me are aware that I tend to long-winded explanations,
particularly on technical subjects, so fair warning is hereby given to
the unwary.  You may want to skip to the next message.

There are three reasons why one should use a thermostat that come to
mind immediately, and perhaps there would be more upon reflection. 
The primary one is thermal control, which is what Fred was addressing. 
However, the belief that one function of a thermostat is to slow down
the flow through the radiator so that it is not "too fast" is not
correct.

Although there are a number of factors which drive the performance of
any heat exchanger and the system to which it is attached, *in
general* there will not be any loss of cooling capability when the
flow rate through the system is increased.  It is true that with an
increased flow rate there will be a reduction in the energy rejected
to the atmosphere *per unit mass of coolant*.  However, this will be
offset by the reduction in energy *per unit mass* rejected *to* the
coolant by the engine, so that there is normally very little impact
upon the cooling capability of the system - all other factors being
equal.  

The amount of energy rejected to the atmosphere *per unit time* will
be essentially unchanged, as it is driven by the combination of the
available temperature difference, the film transfer coefficient inside
the heat exchanger tubing, the thermal resistance of the heat
exchanger tubing and fins, and the film transfer coefficient between
the air and the heat exchanger fins.  All of these factors will be
*essentially* unchanged with increased coolant flow - in the absence
of other factors.  (For those familiar with heat transfer, I *am*
aware that the inside "h" will increase slightly with increased flow
velocity, but if you run the numbers you will find that the effect
upon the overall "UA" is very, very small.)

BTW - as a side comment - I think that if you look at all of the
pressure losses in the cooling system and at the performance curve of
the water pump you will find that there is actually only a modest
change in flow rate if the thermostat is removed.  Not enough to make
a major change in the dwell time of the coolant in the radiator.

Having said that the cooling capability of the radiator will normally
be essentially unchanged with or without a thermostat, I will now do a
180 and state that the presence of a thermostat or its functional
equivalent is potentially very important - just as Fred stated.  The
reason is that, even when wide open, the thermostat creates a pressure
drop in the cooling system which has the effect of raising the
pressure in the cooling system *in the engine*.  

Under low to moderate power output conditions this is probably of no
consequence, but under higher power conditions - particularly for
extended periods - the presence or absence of back pressure on the
cooling system can be a big factor.  Any engine has "hot spots" within
it, where a greater amount of energy per unit area is being
transferred to the coolant than in other locations within the engine. 
Under high power conditions this can be enough to cause local boiling
of the coolant, even when the average temperature of the coolant is
below the boiling point.

Normally the vapor bubbles will transfer heat to the remaining coolant
as they pass through the system and will condense back to liquid
again.  However, this is a *relatively* slow process due to the small
temperature difference, and the potential exists for some of the vapor
bubbles to carry through to the radiator.  If this happens, the
cooling capability of the system will take a big dump and you can
expect major overheating if operating under sustained high power
output.

The reason is that a gas or vapor has a *much* worse heat transfer
coefficient than a liquid, and the ability of the coolant to transfer
energy to the radiator tubing so that it can be rejected to the
atmosphere will drop sharply.  This will raise the temperature of the
coolant returning to the engine, which will increase the amount of
local boiling in the engine and the amount of vapor reaching the
radiator, which lowers the cooling capacity even more, so that we
rapidly head toward overheating

The back-pressure provided by the thermostat will reduce or prevent
the local boiling inside the engine so that this nasty cycle has less
chance of ever starting.  Since a thermostat is one more moving part
which has the potential to fail, it is not unusual to replace it with
a simple orifice (which will serve that back-pressure function) in
applications such as racing where modulation of the coolant
temperature is not an important consideration.

Even where there is not cause for concern about providing
back-pressure on the cooling system inside the engine, it is still
normally a good idea to use a thermostat.  It will help to ensure that
the engine is close to the design operating temperature regardless of
the ambient air temperature or power setting, so that the engine is
running with its components at the design clearances -  always a good
idea.  In addition, by bringing the engine up to operating temperature
even in a cold climate, condensation from blow-by can be prevented
and/or evaporated.  Even with modern oils it is still desirable to
avoid moisture in the oil.

For anyone who has plowed through all of the above, don't say that I
didn't warn you that I can be long-winded.

> Thermostats can easily be tested and if proven to function properly,
> an abnormal condition must be found elsewhere.  Fred Di Matteo  AROC
> Tech Advisor, Fort Myers, Florida

Here I am in *total* agreement with Fred.  Placing your thermostat in
a pan of water on your stove and watching to confirm that it opens
when you bring the water up to boiling is about as easy a qualitative
test as you will find.  If you go to the trouble of obtaining a
suitable thermometer it is also possible to make a quantative
measurement to confirm the exact value at which it opens, although
that is not as important as confirming that it *does* function.

With best wishes,
Tom
- -----------------------------------------------------------
etsuter@domain.elided
- -----------------------------------------------------------

------------------------------