Fred,

 

There is a free lunch - or at least a cheap one.  Drag is an interesting component.  Those that fly rivet-bumped, strut-braced, fixed gear airplanes with poorly designed cowls and bad plenum seals are forever stuck in a world that just doesn't understand relative drag and its significance.  In a laminar flow airplane with very low form and induced drag, even something that used to be thought of as a small drag component soon looms as a large component.  Fixing the cooling drag in an already draggy airplane may buy little performance but in an airplane where the only drag is from cooling, any improvement will have great payback.

 

So, let's take just one example.  I saw a 6-8 knot indicated speed increase at cruise speeds of about 175 KIAS (depends on atmospheric conditions) by sealing all control surface and flap gaps.  Since 6/175 = 3.4%, the result was a conservative 3.4% speed improvement.  From your argument, that would mean a 10% reduction in drag.  Pretty impressive for just $110. 

 

Gap seals on a spam can would not result in anywhere near such a performance improvement because of all the other drag.  In other words, you are right that a small % change in drag would yield an even smaller % change in performance.  You are wrong to claim there is no free lunch.  The gap seal modification was my most expensive drag reducer.  There are so many others that are free and cooling drag is one of them.

 

Here's another on the power side.  Suppose ram air adds 2" to the MAP (its free also and this discussion will keep the speed around 200 Knots). If power is approximated by RPM x MAP, holding RPM constant and increasing MAP from say 25" to 27" (2/25 = 8%), then an 8% improvement in power is almost a 3% increase in speed by your estimation.  The 200 knot airplane now goes about 205.5 Knots, certainly enough to win a race against a similarly equipped challenger with some flow restricting filter on a leaky induction system pickup.

 

Combining power gains with drag reductions results in significantly greater performance.

 

No Free Lunch indeed.  Balderdash!

 

Grayhawk

 

In a message dated 2/15/2008 10:17:40 A.M. Central Standard Time, fredmoreno@optusnet.com.au writes:

“Increase the power, reduce the drag and the limit is raised but it is in no way linear past 200 knots . It is very exponential.”

Not linear at all, in fact.  At these speeds in the thick air we customarily fly in with aspirated airplanes and with the weight and aspect ratio of the wings, the total drag is almost entirely parasitic drag.   Only a vanishingly small amount of induced drag (drag arising from lift) occurs at the speeds discussed. 

This means that to a very good approximation, with a propeller airplane, power (shaft horsepower) goes as the cube of speed.

So it is cubic curve.

However…. One can linearize the curve for small deviations without incurring much of an error (mathematically throw away the third order and higher terms) to arrive at a simplification for constant conditions (same drag coefficient, same flight conditions) –

Which is:

For a 1% increase in speed, you need 3% more power.

For a 2% increase in speed, you need 6% more power

For n% more speed, you need 3n% more power for small n (say less than 10%)

Test - for 10% more speed, the simplification yields 30% more power.  Compare this to 1.1 cubed which is 1.331, or 33.1% more power.  So the simple approximation is not bad.

It is MUCH BETTER to reduce drag than increase power.  Unfortunately, for constant power, to get a 1% speed increase requires a 3% drag coefficient decrease, and 2% speed increase requires 6% drag coefficient decrease etc. etc.  (another linear simplification for small changes).

No free lunch.

Fred Moreno

AKA Captain Tuna, Chicken of the Skies