Jump to content

Dihedral in flying wing design?


Recommended Posts

I am not convinced at all that dihedral and particularly washout is necessary with Zaggi type plan forms.

In the case of my swept back wing it has had washout, rebuilt without it. Fianally built in the present form, which I latterly came to see is the same idea that many of the WW2 wings proposed were to incorporate. That is a conventional wing section inboard, to a section break, where a reflexed section was to be used.

With respect to the reflexed section, I pinched the method from the designer of the Windfreak used (probably the best plank thermal glider to date, certainly back in the late 80s). The idea was to take a known good performing wing section, then by a method I do not quite remember draw a line to the camber line to produce a smooth tangential intersection. I used the method on E205, then I suddenly saw that if I choose my aileron width carefully, I could use a solid aileron, shaped on one side and Robert is pretty much your uncle. In my case a built up reflex, would probably be less accurate.

My experiments on the swept wing included a central fin, using rudder to turn. This needed both a big rudder and dihedral to work. Inboard at that version were the elevators. The present version with the winglets has flown without them (they plug in). In this guise there can be visible side slipping. Plus the ailerons do not work well.

As you can imagine all this work took place over a few seasons. In the end the wing now would have been competitive at 100s size, 40 years ago. Probably not today, although with electric flight, the need to tow well and take the stresses is not necessary. Plus glass wings (built by those who can) win every day at thermalling.

If I knew then, what I discovered over the years, I would have built a Windfreak. As my model is no better and I have seen them thermalling in competent hands, thermaling when rotating like a Frisbee, plus having an incredible ability to penetrate at speed. From memory it used a modified Aquila wing section, that was used by Lee Renuad (I think he developed the section), who also designed the Grand Esprit and the Sagitta.

My own thoughts are that for everyday flying, not much matters that much. It is only if competing, that real efficiency is required in all parts of the flight envelope, from towing, loitering, penetrating and hopefully thermaling on nothing.

Just build what you think you want, then trim and modify as necessary. It is all part of the fun.wink

Edited By Erfolg on 01/12/2017 10:38:44

Link to comment
Share on other sites

Posted by Engine Doctor on 01/12/2017 09:48:29:

The Zagi by Avicraft had a small amount of dihedral built in as wings were joined on a flat surface/table with top of wing flat on table. The taper of the wing section gave it the dihedral on the bottom surface.

Edited By Engine Doctor on 01/12/2017 09:49:15

Yes, that is so, which is as I said above, normal practice for flying wings and a fair number of low wing aerobatic planes.

Link to comment
Share on other sites

Mark, your experiments with non liniear washout is very interesting. In simple terms it seems that Bowers' wing is built with a small amount of washout to midspan and the balance of the total washout, beyond that to the tip. My understanding is that the tips produce upwash (negative lift) in normal flight, the down force provides longitudinal stability with the forward C of G. In cruising level flight both tips must produce some drag as they are doing work (albeit negative lift).

You wish to turn. A down-going aileron (for example) will no longer be producing a down force (negative lift) as the effective angle of attack of the tip is reduced (the wing rises). Therefore drag is reduced on that side too, causing that wingtip to advance (positive yaw in the direction of roll). Conversely the up going aileron on the other wing tip increases the down force at the tip, increasing drag, retarding the tip for the down going wing. Positive yaw in the direction of roll again.

The wing is more efficient because of the non-liniear lift distribution and because it rolls in the same direction as yaw. That is what seems to be the explanation at any rate but I may be wide of the mark!

Regarding dihedral, the Bowers wing seems to have dihedral at the root that reduces across the span of the wing. Or is that just the effect of the wing twist?

Edited By Piers Bowlan on 01/12/2017 12:56:43

Link to comment
Share on other sites

Ah, yes, that answers my question, it seems that the wing simply has wash out built in, as I suspected.

I am not sure that the concept is that revolutionary. Perhaps it would be better to say I am not sure that building wash out in is revolutionary. The vast majority of my models have wash out built in. In my case to control tip stalling.

I suspect what the video shows, is an academic, in the same vain as our BEB. They obtained their Phds by understanding and quantifying some minute aspect of a system etc. That is increasing the level fundamental understanding. It could be that this is his objective, it is to potentially help engineers design and build more efficient wings (in this case), by introducing another parameter, which has validated information, which improves predictability of the wing performance.

For simple people such as myself, I know washout works in taming to a large extent stalling characteristics. It also perhaps explains why many of my models do not seem to suffer from adverse yaw, needing no rudder when turning.

What really rings true with me is that most, engineering formulas are based on a simplification of the real world. I always am reminded of the assumptions in bending theory, of lamella free to move independently, the uniform (homogenious) properties of the material and so on. Where a real beam is nothing like this. The same was true on columns, where it was assumed that the column could be hinged in the middle, the load was not truly axial. Or the column could fail in shear if short. In neither case was the fact that stresses would not be tri--axail on the surface, be factored in and so on. All of which re-enforces my impression that this is a guy, who understands his topic in a depth that ordinary engineers are only vaguely aware, when reminded, needing to reach for a text book to remind themselves.

At the end of the day most relationships have to be simply modelled, accepting and hopefully understanding the limitations.

Link to comment
Share on other sites

And of course, if you want equally good performance inverted, build the wing straight and make sure that most of the elevon area is out towards the tip so that the trimmed reflex achieves a similar washout effect. Or investigate ‘drooperons’ as used on the Zupair Zulu for example.

Link to comment
Share on other sites

Just looked up what a Res Dart is.

Far less complicated than my creation. On that basis i think it will work better and build easier.

The problem with the way i went, by steps, is that I have a fence by the ailerons to try and limit cross wise flow.

The winglets were an experiment, as I had a 120" thermal glider which had them. It certainly held a thermal turn well and seemed to glide better with the tips in a turn. I had always been sceptical about end plates, in that they created drag, probably as much as a ordinary tip, although I guessed more. The winglets seem to overcome this. They do require a bit of work, in that my tips are a flat bottom, clark "Y" (but thin) section. They are also class covered on the flat face for strength.

In essence I would choose a span, select a airfoil, a few spruce spars, within a week you will have something very similar to a Res Dart. Once you have decided where the CG should be, a few calcs (moments) with respect to motor and Lipo and the power pod has designed itself.

Link to comment
Share on other sites

That canard is almost a dead ringer for my O/D Jetex job except my foreplane was a full delta and had just an underfin.Very stable and would hover during first phase of thrust in a stiff breeze. I always use washout and reflex on all my wings and tailless jobs. Even on rectangular models such as ARABIAN KNIGHT .A novelty plan in Aeromodeller in late 50s .A prime example. I built the original F/F and intend doing a R/C job in future .Plan available from plans sites like AeroFred etc.

Kiwikid I love your stable lovely jobs

Link to comment
Share on other sites

Again, my canard, in this version was an experiment in what I could do with Depron. I know a lot of people use it as a skin, my thoughts was along the lines of a solid wing, with spruce spars, balsa separator and a glassed wing using water based varnish. The wing section chosen is a HLG section, very thin, low camber.

They were all part of my Depron, glass cloth and varnish experiments.

This model flies pretty much every week, what ever the weather.

For me a Depron model is not just for still days, they have to stand the rough and tumble of everyday, what ever the weather flying.

wp_20160502_15_11_36_pro.jpg

To the extent, that I hit a tree in a near gale, and it repairs well, still flying, the finished repair is hard to tell from new.

The wing section was an experiment with a section I have now forgotten the name off. For depron or just sheet works OK. Although I would not use it on a thermal machine. Apparently the idea is that stagnant air fills in the back and reasonable streamline results at most AOA.

Plus the wing uses a so called living hinge, again, works extremely well.

I am now thinking there was a time when I experimented a lot.

Link to comment
Share on other sites

Posted by Erfolg on 01/12/2017 14:47:37:

Ah, yes, that answers my question, it seems that the wing simply has wash out built in, as I suspected.

I am not sure that the concept is that revolutionary. Perhaps it would be better to say I am not sure that building wash out in is revolutionary. The vast majority of my models have wash out built in. In my case to control tip stalling.

I think it is not just the concept of using washout Erfolg, as the Horten's wings all had washout but Bowers claims that it is the non-linear distribution of the washout across the span that is new. However I know that the Hortens did use different sections from root to tip and vary the washout span-wise in their research. They were in communication with Lugwick Prantle and knew all about 'bell shaped lift distribution' as per Prantle's 1933 paper on the subject. The Hortens built 63 wings during their lifetime (half during WW2). They all featured drag rudders and suffered from adverse yaw, so like you, am not too sure where Bowers is coming from. I wonder if he will achieve the so called 'proverse yaw' at larger scales? In another lecture he mentioned that he was doing this research as a private venture.

Getting back to dihedral, from Karl Nichol and Michael Wolfart's book on Tailless aircraft, it seems that dihedral in swept flying wings is not necessary or desirable, as the wing sweep provides sufficient lateral stability. Bowers wings seem to feature some dihedral but the dihedral appears to be greatest at the wing root and decreases to the tip. Bower's wing is not only twisted but appears bent across the span (as viewed from the front elevation). Some decent three views would be helpful.

It seems that you have had a lot of success using the KF wing sections Erfolg. They have done some wind tunnel research into these sections but unfortunately they only seems to work at low reynolds numbers, - model sizes.

Edited By Piers Bowlan on 03/12/2017 09:12:21

Link to comment
Share on other sites

I found the presentation interesting, not just for its content. Perhaps more about what it represented.

Perhaps the first thing that people like me tend not to recognise, is that a awful lot of we think of as fact, is often just accepted logic. Perhaps the person who I have had the deepest conversation with respect to Phds is the son-in-law, his is a medical Phd, as he now says, he is a proper doctor. In his case he was assigned a area to investigate where his sponsor was seeking to verify the accepted science as true or not. As such the Phd would come from the finding new facts etc. The reality was that much that was seen as logical, did not match up to a scientific investigation.

I would expect that the good BEB was a similar tale. It is surprising how much has not been thoroughly investigated. The company I worked for Phds were ten a penny, people like myself with Hons degrees were as common as grass in a lawn. In our case, we worked with the convention, the Phd were supposed to spot opportunities etc.

I can easily see how it is or was logical to assume that the lift profile (if that is the correct term) has been seen as elliptical. My logic would start with each unit area will generate the same amount of lift (negative pressure, call it what you will), the form would be a retangle. Some one clever than me comes along, and then says, but you visually can see (in some circumstances) that there are tip losses, the profile is not a rectangle, it is elliptical. Nobody at this stage has measured or is able to instrument a wing. Later some one does and concludes that there are some variation from the predicted values, giving a whole range of reasons as to why, lthough concludes it is elliptical ( as this avoids arguments).

Perhaps strangely, a vet in my old club made an observation along the same lines as the NASA Phd or more probably lots people have had similar thoughts, although that is as far as it went. The majority, such as myself, will say, hmm, very interesting. Let me get on with my Horner tips.

For our NASA Phd, he thinks, perhaps wings can be designed better if we really understand what is going on at the tip. As we were made aware, he noticed the discrepancies between the assumptions and what can be observed with such things as bird flight.

Being an active, ongoing Phd, involved in fundamental research, he was not letting go. His problem is a day job. The day job depends on funding from Government and Industry (private enterprise). Private Enterprise tends only to fund things that help them make money or save it.

Now I think what we were witnessing, a guy who believes (and I suspect is correct) that the lift profile is Bell shaped. He is looking to get funding to do some work in the area. The bait is, you could build a more efficient commercial aircraft by having access to knowledge of more efficient structures, a improvement in energy usage and so on.

The presentation could all have been preparing a presentation to those with cheque books, that would enable a Phd, do what he loves, investigating fundamental aspects of science in a limited area.

Link to comment
Share on other sites

The vortex as shown is apparently nothing new. As a student some, oh dear, much longer ago than I care to remember, was taught this as as a fact, as it were a continuos process, where it could be modelled like a elongated "o" behind the TE and along the wing. To be honest i took it to be representative of a model, where the energy consumed could be calculated. By this time I was not into aeromodelling. It was just part of the fluidics curriculum for accredited mechanical engineering course. What seemed far more useful were calcs of fluid flow, in bulk systems and Reynold numbers , dimensionless analysis (which I personally have never made use of), friction on bodies and so on.

I am increasingly convinced what the NASA guy is pushing, is that rather than trying to make the wing work at the maximum right to the wing tip. It is more energy efficient, to wash out the tip, to the extent, that its drag is minimised. You will still have the vortex from the downwash and upwash where they meet. I think he is arguing that the overall system has many potential advantages. One being, that adverse yaw, can be avoided or managed.

I honestly believe what the NASA guy is saying is simple, here is the potential to obtain a small reduction in energy lost, lets investigated it, to investigate it I need some money for NASA to instigate a programme of work and suitable resources.

To me it seems like a good body of work and understanding could arise from such a project.

Perhaps the biggest gain would be the loss of a fin, he suggested that contributed perhaps 20%? of the airframe drag?

He also hinted that the gains were only optimised at specific speeds, as with winglets.

Link to comment
Share on other sites

  • 3 weeks later...

Many aspects of this thread have caused me some disquiet, mainly in that it has reminded me of some aspects of my own designed models where some aspects of performance remained unresolved although they have mainly flown successfully.

Watching the Prantl video and later some thought has possibly resolved at least one of the issues. It seems that Prantl had recognised that pressure distribution could have been Bell shaped, although plumbing for elliptical distribution.

Looking particularly at Kiwi models most had end plates or similar.

In the video, it mentions that winglets that many aircraft have now, in effect is the same as producing a bell lift wing. The Professor seems to be suggesting that the same can be achieved by washing the wing out. Suggesting that not only can the wing be made to operate more efficiently at a given airspeed, that the fin arrangement could be dispenses with, saving perhaps 10% of drag force. Not only that the adverse yaw issue is avoided.

Which brought me back to my initially wayward Canard. There was a major problem with directional control. Solved by the increased size fins. It was only after having the issue boring a hole in my mind, that perhaps the penny dropped. My wing is thin, slightly less than 8%, with very little camber. Perfectly flat, under most circumstances. I have now concluded that the wing probably suffered from adverse yaw. This will have been a first for me, in that if any of my models previously have suffered from it, it was not to much of an issue, in that under my clumsy control I have not noticed.

The washing out idea, probably works well, in that my wing, has in the past lost one of the fins, yet continued to fly well. In this case, the centre of the wing is E205, in the tip region a modified E205 reflexed, which effectively creates a washed of tip region.

v2.jpg

In the picture above you can see some tape which were all part of the development work to improve directional control. They did nothing particularly useful. I increased the plates to larger still (the first being small), that solved the directional issue. I think I had tried differential aileron movement, although the result must have been less than impressive.

Link to comment
Share on other sites

  • 3 years later...

To add to this old thread...as a matter of info..

Whilst I still plan to (someday) design my own flying wing, I have bought the RES Dart kit and have it on my workbench at the moment.

It is intended as a RES glider, so is developed to be stable. Aside from the sweep of the wing, it has plenty of washout (~5 degrees) and (relating to the point of this old thread) anhedral. Each wingtip is around 50mm lower than the root, measured at the spar.

 

Edited By David Hall 9 on 24/01/2021 11:44:37

Edited By David Hall 9 on 24/01/2021 11:45:06

Link to comment
Share on other sites

  • 4 weeks later...

I know it's an older thread but thought I'd chine in. Roy Garner did a flying wing RES model a long time ago;

Hatchet

Which appears to have a small amount of dihedral.

Peck Polymers' "Genesis" hlg also had a bit of dihedral.

I think it's possible to do, but remember that 1 degree sweepback is roughly equivalent to 3 degrees dihedral in terms of stability.

 

 

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...