Build your own HO229 WW2 German Jet Fighter!!

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

Aerowerx

Well-Known Member
Joined
Dec 1, 2011
Messages
5,543
Location
Marion, Ohio
Yes, I have a copy. It's a good primer. I went a lot further with the maths than the book does. If you want your tailless aircraft to work over the full CL range, you need to variable twist or things get very, very out of whack away from your design point. For instance, lots of drag at cruise if you wash out the tips enough for a safe stall. Have you got to the bit where Nickel says that elliptical distributions are the best way forward?

VB, I have no illusions of being able to teach Al anything, but I'd be very interested in hearing how he may have overcome some of the obstacles I have come up against. specifically, getting it to work fairly well over the whole CL range without delving into the wonders of wing warping.
Yes, I have read what he said about elliptical. In fact, this is my 2nd read-through.

But keep in mind that there has been research done since this book was written. Even Prandtl himself, that developed the lifting line theory, later on stated than bell-shaped was better than elliptical. And the Hortens also used bell-shaped, at least in their later designs. This, for some reason, has been practically ignored.

And Nickel himself states several times in the book that you will never get perfect elliptical, only "good enough" (whatever you want that to mean).

As for morphing, I remember one idea where the wing tips were sliced like a loaf of bread and then covered with a flexible membrane. The slices are mounted on a tube running span-wise. Only the very tip slice is fixed to the tube, and the tube is then rotated to get different twists.
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
12,849
Location
Memphis, TN
Rudders are a quick fix. Personally I have a hard time building the same model twice. I have done it a couple of times, but usually it has to be years apart. How many iterations do you want to build to go have fun flying an RC? Slap some rudders on and go fly. Scientific building would be different, the goal is not fun but the object. Really not the same.

I’m really sad that the N9M crashed on multiple levels. Clearly as an experimental aircraft never to reach public hands, it flew quite well to regular experienced pilots. One of the airplanes magazines like FlyPast or Aeroplane had a great article on it. The pilot flying it said if you can fly a T-6 you would have no problem. It is WW2 fighter sized requiring experience. It’s not for student solos like a Cub or 150. Engines catching fire and flight controls jamming on an airplane that has had lots of success goes in the bad luck category not helped with shoestring budget. Clearly When it took off the plane was not right like hundreds of times before. I got a kick of seeing it’s brother at the Smithsonian, but I really wanted to see the N9M.
 

Red Jensen

Well-Known Member
Joined
Jan 1, 2018
Messages
71
Location
Palmdale, CA
So he is going to be there this year? I'd developed the impression that after retiring that he would be abandoning further development of his ideas, leaving that up to his students and the rest of us? If he is going to be there this year, and I actually get to attend, it will remove some of the sting of not being able to attend his forum last year.
Yes he is running it again this year.
 

Aerowerx

Well-Known Member
Joined
Dec 1, 2011
Messages
5,543
Location
Marion, Ohio
...The big downside to these distributions is that the optimum twist changes with CL....
Question for you pictsidhe.

What is your mission profile? What I envision for my (future/maybe/possible) design is spending 90% of the time in cruise, so I optimize for that. So long as the wing tips do not stall in the other flight modes, does it really matter?

My "mission" is to have a touring aircraft for short or medium cross country flights for sight seeing and long weekend trips. Cruise at 100 mph. Pilot and passenger. 30 lb luggage. 10 or 20 gallons fuel. Besides the challenge of just designing such an aircraft!
 

Red Jensen

Well-Known Member
Joined
Jan 1, 2018
Messages
71
Location
Palmdale, CA
I think so, but this is a practical aeroplane model..it can be fitted with LG etc.
Juke,

It suffers from yaw instability because of the E spanload coupled with all of the area ahead of the CG which is destabilizing. Again, P spanload doesn’t require vertical surfaces for stabilization, and yaw control can be accomplished by other means, landing gear or not. How do I know? I’ve built several dozen of these from 12” to 25’ span.
 

Speedboat100

Well-Known Member
Joined
Nov 8, 2018
Messages
1,011
Location
Europe
Juke,

It suffers from yaw instability because of the E spanload coupled with all of the area ahead of the CG which is destabilizing. Again, P spanload doesn’t require vertical surfaces for stabilization, and yaw control can be accomplished by other means, landing gear or not. How do I know? I’ve built several dozen of these from 12” to 25’ span.
I have seen flying wings before, but show us yours please.
 

Red Jensen

Well-Known Member
Joined
Jan 1, 2018
Messages
71
Location
Palmdale, CA
As requested, not a vertical tail in site. Now you explain to me the difference between a E and P spanload please.
 

Attachments

Last edited:

Aerowerx

Well-Known Member
Joined
Dec 1, 2011
Messages
5,543
Location
Marion, Ohio
As requested, not a vertical tail in site. Now you explain to me the difference between a E and P spanload please.
E is elliptical, right? But what is P anyway? Prandtl bell shaped? Remember Prandtl himself later recanted on the elliptical in favor of the bell bell shaped being best.
 

Red Jensen

Well-Known Member
Joined
Jan 1, 2018
Messages
71
Location
Palmdale, CA
E is elliptical, right? But what is P anyway? Prandtl bell shaped? Remember Prandtl himself later recanted on the elliptical in favor of the bell bell shaped being best.
Yes P is Prandtl. If span is not constrained, for the same weight of materiel, P gives lower induced drag for the same root bending moment. P is not the answer to everything of course as noted above and elsewhere, but it does solve the all wing issue.
 

Aerowerx

Well-Known Member
Joined
Dec 1, 2011
Messages
5,543
Location
Marion, Ohio
Yes P is Prandtl. If span is not constrained, for the same weight of materiel, P gives lower induced drag for the same root bending moment. P is not the answer to everything of course as noted above and elsewhere, but it does solve the all wing issue.
Ok, thanks.

I have read Al's paper. Just wanted to make sure P was Prandtl. As Nickel likes to say, in "Tailless Aircraft" book", there are an infinite number of Bell shapes. Prandtl is only one of them.

From what I read both Prandtl and the Hortens came to the same conclusion without knowing it.
 

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
7,427
Location
North Carolina
Question for you pictsidhe.

What is your mission profile? What I envision for my (future/maybe/possible) design is spending 90% of the time in cruise, so I optimize for that. So long as the wing tips do not stall in the other flight modes, does it really matter?

My "mission" is to have a touring aircraft for short or medium cross country flights for sight seeing and long weekend trips. Cruise at 100 mph. Pilot and passenger. 30 lb luggage. 10 or 20 gallons fuel. Besides the challenge of just designing such an aircraft!
I was originally thinking of a 103 floater. I'd like to build a very long distance motorglider. Next year, with this still only in spreadsheets, it could be something else.

BSLD are currently best designed for a single Cl point. Drift away from that, you lose its advantages. Once the structure and stability have to be designed around a range of distributions such as you'd get from a rigid wing, the advantages diminish.

Lets take a wing twisted for cruise. Strength is often air tested at maximum CL. That requires maximum twist. With a cruise twist and stall CL, the lift moves outwards, meaning you need a stronger, heavier wing than you would if it had a stall twist. The tips have a higher loading, you may need to fatten them up to prevent stall. That's more tip and root weight and more twist needed for BSLD. That means the twist will go even further out of whack away from your design point... With the CL moving outwards on a swept wing, it also moves backwards. Now you are nose heavy and having a harder time maintaining pitch. Accomodating the edge cases needs a fair amount of compromise. So much so, BSLD looks hard to implement and still get a noticeable gain in improvement. This is almost certainly why there have been so few. Notice how the people saying that BSLD is optimum, were theoreticians...

I'd like to get all the aero calculations into a spreadsheet then run a genetic algorithm, I think that is the way forward. Maybe I'll work out wing warping, maybe I'll work out how to do it without warping...

I just looked at the last spreadsheet I did. Changing the CL by 1.0 alters the optimum twist by 15 degrees. The distribution changes quite a bit if you don't accomodate for that somehow. My spreadsheet didn't account for sweep very well, it really needs a different model than my straight wing + kludge one.

FWIW, my spreadsheet planform evolved into something like the Horten ones. They were definitely onto something.

The conventional rectangular wing aircraft is a benign but low performance design. They aren't just easy to build, they are easy to design, you don't have to get things so right. They are still perfectly capable of killing many pilots, though. Once you start to optimise the wing and reduce the margins, you have to get more things nearer ideal to avoid nasty things happening.
 

Speedboat100

Well-Known Member
Joined
Nov 8, 2018
Messages
1,011
Location
Europe
Last edited:

Red Jensen

Well-Known Member
Joined
Jan 1, 2018
Messages
71
Location
Palmdale, CA
Ok, thanks.

I have read Al's paper. Just wanted to make sure P was Prandtl. As Nickel likes to say, in "Tailless Aircraft" book", there are an infinite number of Bell shapes. Prandtl is only one of them.

From what I read both Prandtl and the Hortens came to the same conclusion without knowing it.

Horten and Prandtl did arrive at similar solutions. There are pictures of them together, but no evidence that they ever collaborated. We have flown both spanloads, currently the P is working better.
 

Red Jensen

Well-Known Member
Joined
Jan 1, 2018
Messages
71
Location
Palmdale, CA
Yes...now explain the E spanload.

About spanload: http://www.dept.aoe.vt.edu/~mason/Mason_f/WHMAOESemOct.pdf

Red since you seem to have some formal education on the subject...isn't a wing with wider chord always more effective than a vary small chord...when the proportions are the same ?

Its not my job to educate you. You are the one who professed all flying wings needs rudders, and I showed you that isn't true. Now you are mixing terms, talking about chord. Please try to stick to the subject.
 

Attachments

pictsidhe

Well-Known Member
Joined
Jul 15, 2014
Messages
7,427
Location
North Carolina
P spanload meaning Prandtl? IIRC, Prandtl favoured the sin^3 distribution, though he wrote it differently. R.T. Jones independantly showed that root bending moment decreases at even higher exponents, though at a diminishing rate. I've been playing with exponents between 2 and 3 in my mathematical dabblings. That appears to be the aerodynamic sweetspot for stability and efficiency criteria as well as reducing the root moment. I have only tested with pencils and electrons.

I'd be very interested in hearing if the Bowers team has a way to minimise the lift distribution variation with CL on a rigid wing?
 

Red Jensen

Well-Known Member
Joined
Jan 1, 2018
Messages
71
Location
Palmdale, CA
P spanload meaning Prandtl? IIRC, Prandtl favoured the sin^3 distribution, though he wrote it differently. R.T. Jones independantly showed that root bending moment decreases at even higher exponents, though at a diminishing rate. I've been playing with exponents between 2 and 3 in my mathematical dabblings. That appears to be the aerodynamic sweetspot for stability and efficiency criteria as well as reducing the root moment. I have only tested with pencils and electrons.

I'd be very interested in hearing if the Bowers team has a way to minimise the lift distribution variation with CL on a rigid wing?
Currently we haven't delved into trying to optimize off nominal CL. We have discussed what you proposed (ie flexible skin to tailor for other CL) but that is it so far. With Al now being retired, I think the innovation may be over, at least for this iteration. There is another paper in the works howwever. We're working on the Mars wings mostly now.
 
2
Top