|
|
This is an interesting debate. I go away to work for a day and everyone
seems to have sided with Tracy... wonder why :-) Again I preface my
comment by knowing that NASCAR prefers thick rads (of a sort), In
reality, there is a happy medium, i.e. we all agree that the "ultimate"
thick rad wouldn't work... 4 square inches of surface area in a tube
extending back through the fuselage. I could fit in cars and planes -
but each packet of air can only get as hot as the hottest coolant. The
ultimate thin rad wont work either - just too damn big (and maybe too
much drag, and maybe not).
Nonetheless, all effective radiators are wider than they are thick, even
in the aircraft that have used them. This is different from
counter-flow heat exchangers - but that is a different topic.
> Posted for "Tracy Crook" <lors01@msn.com>:
>
> I'll embed my comments in the appropriate places below.
As will I.
> > Posted for "David Leonard" <Daveleonard@cox.net>:
> >
> > I think the key to the argument for a thin rad lies in the fact that
> > parasitic drag is a function of the velocity SQUARED. Implying to
me
> > that passing air quickly through a rad pays a disproportionately
higher
> > drag penalty than passing a slightly larger amount of air slowly.
> > Comparing 2 rads of the same volume; one thick, one thin: if we are
> > going to, say, pass 10 cubic feet of air through in 5 seconds, it
will
> > need to be going faster in the thicker rad (with proportionately
less
> > surface area)- as already mentioned.
>
> Keep in mind we are not flying a radiator through the air, it's the
> AIRPLANE. Unless you can show why more air through the system causes
less
> drag, you have to conceed the point that less air means less drag.
>
More air CAN cause less drag if it is going slower. Again the argument
lies in the fact that drag is a function of velocity CUBED (not squared
as I previously stated - I think maybe laminar flow is velocity squared)
but mass of air to only the first power. A quick search found this
classic discussion of parasitic vs. induced drag which illustrates the
velocity cubed point:
http://www.paraborne.com/how_airplanes_fly_3.html
Drop the velocity of the air, the drag drops by a 3-fold factor.
Increase the mass, and drag only increases by a one-fold factor.
More air has the added benefit of not requiring each packet to gain as
much heat, allowing an overall higher temperature gradient.
> >
> > As you point out above the thicker rad will allow each packet to get
> > hotter because it will be spending more time crossing the thicker
rad
> > (so less air is needed). But this is offset somewhat by the fact
that
> > the slower speed in the thin rad allows each packet to gain more
heat
> > per inch and spend almost as much time in the rad.
>
> As soon as you use the "slower air" through the thin rad argument, you
> loose
> the vaunted higher efficiency of the thin rad. Paul often makes the
point
> that the rear part of the thick rad is not working as well due to the
> pre-heated air. If you slow the air through the thin rad, the same
thing
> will happen in the thin rad.
I think that the vaunted higher efficiency comes mostly from lower drag,
thermal efficiency has a much smaller effect on cooling drag. I made
the above point merely to point out that those two effects somewhat
cancel each other out. More air means greater thermal efficiency
(higher average temperature gradient - so slightly reduced needed
surface area and weight) balanced by more air creating more drag. But
it is the speed of the air that is cubed in the production of drag.
>
> >
> > Paul once illustrated it best for me when he said the ideal would be
to
> > take a 'small' packet of fast air and use an efficient diffuser to
> > expand the cross section and slow it down (now higher pressure),
pass it
> > slowly and efficiently through the rad. Yes, most energy is now
lost,
> > but obviously not ALL of it (otherwise it would never leave the
cowl),
> > and use an efficient diffuser to go back to small cross section and
high
> > speed.
>
> Major and widespread error here. Please explain why a *high pressure*
in
> front of a thin rad will result in *slower flow* through the thin rad
> core.
> The lower velocity in the aft end of the diffuser is being confused
with
> the
> velocity through the core of the radiator. They are NOT the same
thing.
> The rad (even a thin one) is a restriction and the air is rushing
FASTER
> through it due to the pressure recovered by the diffuser, not slower.
>
High pressure in front of the rad in its self is not helpful. The key
is of course is minimal pressure drop across the rad - it is important
that there be high pressure on the back side of the rad as well. This
high pressure is then used to accelerate the air back up to exit
velocity. The important thing again (put a different way) is to
minimize pressure drop across the rad per cubic inch of rad. This is
how the Mustang supposedly got some thrust out of its cooling system -
it sped up the exiting air to faster than the speed of the aircraft
(though I don't believe there was actually any thrust produced from the
cooling system). True that the mustang had a thick rad, but it was even
wider. Probably as wide as they could make it and still have 7 degree
ducts.
>From flow dynamics we were always taught that it is the higher airspeed
across the top of the wing that causes low pressure and lift etc.. (not
really completely accurate, but close enough). The point is that with
the fast airplane type outside air we need to slow it down to car type
outside air for flow through the rad AND speed it back up again before
we put it outside the cowl.
> >
> > The other key point then becomes "efficient diffuser." IMHO, the
happy
> > medium is to use the broadest rad that will fit without disturbing
the
> > profile of the aircraft - and still leaves room for efficient
diffusers
> > (whatever that is) (7 degree rule and all I suppose).
> >
>
> Compare a 7 deg. difuser for an 18" x 18" thin rad and compare it to
one
> for
> a 9.5" x 9.5" thick rad. Do you have a prayer of fitting the big one
> under
> the cowl?
> Another reason for using thick rads.
True. That's why I say use the largest surface area rad you can fit
WITHOUT deforming the airplane.
>
> Thanks for the opportunity to debate this issue in a non-censored
> environment David. No offense is taken or meant
Ditto.
David Leonard
|
|