Beyond Streamlining (part deux)
Hi all,
Since I don't see it yet ... please use this topic to continue the amazing "Beyond Streamlining" topic that John McGinnis started a couple of years ago and now exists only in the archives.
John's demonstrator aircraft is amazing (what we've seen of it.) Mayyyybe the Synergy discussion belongs in a new thread now. :-) The issues and discussion of aerodynamics, thermodynamics, synergy, syntropy, etc. remains relevant and deserves discussion. Maybe I'll even have something to add besides eyeballs this year.
Dynamic Pressure, Induced Drag, and ... Restarting
First, I agree with Hoghead. Many hands (or pocketbooks) make light work.
Now, back to the real purpose of this discussion -- next-generation aerodynamics. Here are some thoughts brought about by post #192 in the archived thread "Beyond Streamlining...."
Here is John responding to a question about dynamic pressure. I shuffled the order around so it makes a bit more sense for my points.
In 1752 (!) Jean d' Alembert proved that an object in steady motion through a
perfect fluid (inviscid, which means free of viscosity) would in fact have no
drag at all. Modern experiments since the 1950s have shown this to be a fact,
as when we (correctly) use power to take away the influence of viscosity, drag
drops to essentially zero. This is what Goldschmied shouted from the rooftops
in the 80's.
This makes perfect sense to a physicist. It is only when we must do work against resistance that we need to apply power. We do similar things in the lab all the time with air tables (think air hockey) for less-friction experiments. On the other hand we only want to remove the viscosity right next to our skin. We need to air to drag other air along with it so that our wings work and throw big bunches of it at the ground. (That's induced drag though, not dynamic pressure.
Yet because of the 'impossibility' that total drag could be less
than the form drag revealed by the 'dynamic pressure', it was quite literally
not understood for a long, long time.
It's easier to just "do the math" than to actually understand a problem and the system where it exists.
As it turns out, the problem is found in the concept of dynamic pressure
itself. But taking only one-half of the circular kinetic energy of mass flow is
a mathematical sleight-of-hand that allows us to 'lock down' air movement long
enough to assign it a number. In the other branches of physics this is a no-
no. You can't know both the position and the the velocity of a particle at the
same time.
ummm ... when that particle is extremely tiny, extremely fast, or extremely energetic. We have a long way to go in terms of personal flight before we need to worry about the speed of light. (More's the pity, actually.) :-) Yes, dynamic pressure is sleight-of-hand. It got out of hand. A standing joke in the physics community goes like, "Today we're going to study milk. For our first-order approximation, we'll assume the cow is a sphere." John's point is valid. For a century we have split the "airplane in flight" system into tiny, bite-sized chunks we (thought we) could work on mathematically. We pretended that what happens in the airstream five feet from the wing no longer matters. It matters, but it's going to take some work to prove it other than empirically.
OK, with respect to your post #190, we need to separate subject matter.
Let's say we're flying in a neutral bouyancy airship so that there will be no induced
drag. Classically what we are taught is that the dynamic pressure (1/2 the
product of the air density and the velocity squared) times the area and drag
coefficient will yield the drag. Under this paradigm there will always be a drag
value that is approximately double what it would be if we used power to
eliminate boundary layer viscosity, instead of simply trying always to blast
our way through it. ... No physical mechanism corresponds to the '1/2' in the
equation (!) It is an assumption, long buried,
Whenever I see an equation of the form y = 1/2 aX2 and that equation is modeling a physical system, then we're most likely looking at one of the many forms of Newton's second law, F = ma (force = mass times acceleration.) The right-hand side of that equation is most (but not all) of Newton's law written with respect to velocity instead of acceleration. The whole equation of course being f = 1/2 mv2 + v0. The 1/2 comes from taking the instantaneous value of "change of position" (the first derivative, if you're math inclined.) On the other hand, we appear to have been waving our hands at and ignoring a meaningful value. None of this has anything to do with viscosity but only change of position. The whole "we can treat viscosity as if it were pressure" assumption has got to go though.
Even the 'neglible' low viscosity of our atmosphere can and does create major
havoc when circumstances allow persistence to develop. Just as a bullwhip
translates 'negligible' forces and velocities at the handle into supersonic
energies at the tip because it's a supple and transmissive medium of adequate
length....
Great John, introduce chaos theory. Wasn't this deep enough? :P
What is low induced drag? For our purposes it is low wake vortex.
So induced drag is not the energy we put into making lift, but something else? We certainly need to provide the impetus to do both. (I'm not--yet--schooled in aerospace engineering, but I am a physicist and systems scientist. The fun parts of John's ideas come with the systems more than the physics. If we can deal with boundary layer viscosity then we should see a change in drag. I'm curious if the measured changes are density-dependent.) (I wonder if we could beg our web provider for a Greek alphabet too ... ) :-)
. . .
Anyone want to talk about systems theory? Feedback, feed-forward, signals, control theory, etc. and digging into how Synergy sizes up with other aircraft?
New data on Static Pressure Thurst - Large Discrepancies with Goldschmied Pubs
This entry is for anyone who has been wondering why Goldschmied Propulsion hasn't taken over the world. I have collected many papers on the topic, and one was recently published by AIAA that may shed light on why nothing of note has appeared in literature in the intervening years.
I would point those who are curious at AIAA 2012-0866 "Computational Study of the Embedded Engine Static Pressure Thrust Propulsion System" which attempts to reproduce Goldschmied's wind-tunnel results in CFD (ie the Goldschmied body). The bad news is that the only discrepancy found is the pressure upstream of the embedded fan - CFD predicts far greater delta-P than the measured data. This directly relates to the amount of power required by the embedded fan to achieve static pressure thrust - and this power consumption is far higher than Goldschmied calculates, in fact so high that it is higher than that necessary for a non-embedded propulsor (ie conventional external propellor). It is unfortunate that the power used to drive the fan was not directly measured (or perhaps not reported since it would have been inconvenient) by Goldschmied - he back-calculated based on velocity and delta-P (ref. AIAA 1987-2935-856). I suspect that this one pressure port was somehow measuring some dynamic pressure instead of the static pressure it should have.
I can say fairly confidently now that with this datapoint as well as some independent simulation on my own largely confirms that the data presented by Goldschmied is erroneous. I suspect industry came to the same conclusion long ago.
However, there is one ray of hope on this topic. In the formal report to the Navy (DTNSRDC-ASED-CR-02-86 Wind Tunnel Test of the Modified Goldschmied Model with Propulsion and Empennage Analysis of Test Results - avail on DTIC), Goldschmied points to another body shape that should perform substantially better. This shape is called out as the R8 body from Smith, Stokes & Lee's "Optimum Tail Shapes for Bodies of Revolution", AIAA Journal of Hydronautics in 1981 (search AIAA.org). The bad news is the ordinates of this shape are not called out correctly in Goldschmied's report - the X and Y coordinates are the same for the first 26 datapoints, which is perhaps more example of his less-than-meticulous nature.
Best,
John