"Gene Kearns" wrote in message
news
On Sun, 22 Feb 2004 17:31:16 GMT, "Tony Cox" wrote:

It's not clear to me that the stresses in shock heating should be
the same as for shock cooling. Just because people don't
(apparently) suffer from the effects of shock heating doesn't
by itself preclude the possibility of shock cooling.


Agreed, but I think the point not to lose sight of is that it is the
rapid *change* in temperature that causes the problem...
high-low...low-high... does it really matter?

The problem is induced stress, caused by different temperatures
at different locations in the metal of the engine. Temperature rate
of change observed on (say) the cylinder head is just one factor
which might predict the temperature variations throughout the engine;
my point is that there are other factors too. These factors are
different on take-off and landing. So even if the temperature rate
of change in one régime is higher than in another, this doesn't
necessarily indicate that the maximum stress in the engine is
higher. So no, I don't agree that rapid change (as measured by
some CHT) is the cause of the problem.

High-low and low-high *do* matter if the stress limit of metal
is a function of temperature, even if the internal stress fields are
similar for shock heating and cooling. Assuming that maximum
stress occurs at the instant that take-off power is applied (or
when power is reduced during descent) one would expect the
engine materials to be subject to this stress when they were
relatively cool (or hot, in descent). Stress limits of metals at
different temperatures are probably different.

{snip}
Whether this makes metals more or less susceptible to
cracking, I've no idea - but if shock cooling is a fact and shock
heating isn't, we should expect metal to be more fragile at higher
temperatures.

The word "fragile" is not used much in reference to metallurgy and
doesn't convey a lot of meaning to me. Remember, we are talking about
the forces involved in incidents akin to pouring a cool liquid into a
warm punch bowl.... not a pretty sight.... and is common to all
crystalline substances (amorphous or not).

OK, more likely to crack then. (and if you're keen to be pedantic,
crystals are never amorphous).


Also, since the initial airspeed is higher, changes in
cooling (speed-up during descent, closing cowl flaps) will alter
temperatures (and introduce differentials) on the 'cooling' side which
are far less uniform than during the initial climbout.


No, remember here, this is about transfer of (heat) energy. There is
more capacity of rushing (relatively cooler) air at descent, thus the
greater energy is cooling..... just the opposite is true at climb,
where the combustion process holds the upper hand and can cause
temperatures to soar in non-uniform ways.

You'd really have to do a finite element analysis to verify this.
But you can reasonably assume that if the airspeed is higher, then
the *variation* of the airspeed over different parts of the engine
is higher too. This means that substantially different amounts of heat
are being transferred away from different parts of the engine, which
(other things being equal), will mean that there'll be a temperature
variation over the surface which could itself lead to cracks.

The point is not to quantify this one way or the other - as I've said,
you need to do some complex modelling to be sure - rather, it is
to show that you can't predict the stresses in different modes of
flight by simply looking at the CHT rate of change.