Posted for Gary Casey <glcasey@adelphia.net>:

 Johannes,
 
 You are thinking correctly.  I agree with Chris in that some of your problem
might be in the sealing and shape at the inlet - it is  critical that there be
minimal leaks there.  Using the airspeed  indicator as a pressure sensor is a
good idea, but you have to be a  little careful regarding the reference
(static) pressure.  You have  referenced it to the cabin, which is probably
okay, but there will be  some error in the measurement.  If you have an
alternate static  source that references the cabin it would be best to turn
that on  before making measurements and then compare the "airspeed" indication
 to that.  Also, remember that your cooling pressure ("air speed")  gauge is
indicating speed, but the pressure it is measuring is  proportional to the
square of the indication.  Example: if you had an  indicated airspeed of 71%
of the aircraft indicated speed, the  pressure is only 50% of the total
dynamic pressure.  I think you have  a good location for the measurement with
the hose looking at the fuel  distributor, which will block any dynamic
contribution.  I've found  that the airspeed inside the cowl is low enough to
pretty much  eliminate the sensitivity to the orientation of the pressure hose
-  anywhere in the cowl looking in any direction will pretty much read  the
same thing.  I've doubled the hose back on itself and then  wrapped a cloth
around it or you can get a scrap of open-cell foam  and stuff the end into it,
but none of that will make much of a  difference.  Bottom line is that you
would like at least 75% pressure  recovery in the upper cowl, which will give
an "air speed" reading of  87% of the aircraft indicated airspeed (the
airspeed indicator  measures pressure so, yes, you do want it to read HIGH,
unlike what  was suggested in one post).  You are getting 56% recovery, not
 terrible, but could be better.
 
 As for the lower plenum pressure, you would like it to be as high  (not low)
as possible GIVEN adequate cooling.  If you are running too  hot then you will
probably want to lower the lower cowl pressure.   Your lower cowl pressure is
only 12%  of the dynamic pressure  (airspeed indicates 35% of the aircraft
speed), actually on the low  side.  I think you should improve the efficiency
of the nozzle that  accelerates the air back up to speed at the outlet.  I
believe from  your pictures the lower edge of the firewall ends in a sharp
edge,  causing turbulence at the outlet.  This should be a smooth converging
 shape, converting the lower cowl pressure into velocity, reducing drag.
 
 The real question is, what is your objective?  Do you have to improve  the
cooling or are you trying to reduce drag?  In either case the  shape and
sealing of the inlet is very important.  In either case the  shape of the air
outlet is important.  After those are optimized then  reduce the area of the
outlet until the cooling is adequate.  From  what I can tell the area of the
inlet is not the critical factor - an  oversize inlet, if correctly shaped, is
not a big factor in drag  reduction.
 
 Hope this discussion helps
 Gary Casey