Wind tunnel

Help Support HomeBuiltAirplanes.com:

rtfm

Well-Known Member
Hi,
Well, I have a lot of Q-Cell goop to put on the Razorback this weekend, and a lot of time just sitting around waiting for the resin to set. So I think I'm going to give the wind tunnel a go. My plan is (roughly) as follows:
1. Acquire the factory's big stage fan (LOTS of blow power there...)
2. Build a tunnel out of some heavy duty black plastic I have in the shop and attach it at one end to the fan cage, and at the other end to the anti-turbulance box (a box stuffed with straws)
3. Construct the wind tunnel test area as follows: (1) Foam base, with grooves cut to accept perspex sides (2) insert two perspex sides (3) attach top (perspex if I have sufficient, or if not, plywood) (4) Secure the test area with duct tape, and ensure relatively robust.
4. Secure the test area to the anti-turbulance box
5. Sides of the test box will have two grooves into which the rollers attached to a tubular wing "spar" will fit. This is to enable the wing section to roll back and forth in the grooves as drag increases/decreases. The 'spar' will be anchored at the beginning of the grooves with a strain guage ($2-Shop scales - absolutes or actual measurements are not important. I want to measure change, and relative movement under different conditions) 6. Once the wing section is fitted, I'll attach a$2-Shop strain guage to the TE to measure relative deflection

The wing section will have (1) a fowler flap (good practise and trial/error to get the rails sorted out for the real thing) (2) a speed brake. The idea is to see:
• How the wing section deflects (both vertically and horisontally) as the flap is extended
• How the dropping of the speed brake affects the drag and the lift
Benefits of spending my time thus employed:
1. If deploying the speed brake destroys the lift significantly, now would be a good time to discover this.
2. It will allow me to experiment with the shape of the flap rails
3. It will also give me a good way to raise my skill levels before tackling the actual wing with its rails, flaps and (possibly) speed brake
4. I will use the actual linear actuator and rig up the control linkages as they will appear in the plane. Again, an excellent test/experience bed for the real thing.

So there it is. The project I'll be working on between sanding sessions. Got to keep the enthusiasm levels high during interminable sanding sessions!

I promise to take lots of pictures.
Regards,
Duncan

Autodidact

Well-Known Member
Hey, Duncan. How will you measure lift? I too, would love to have a wind tunnel.

DeepStall

Well-Known Member
Sounds like a fun project! Can you throw out some dimensions/sketches for all of this? Some things you may want to think about:

• Is your flowpath fan - antiturbulence box - test section or antiturbulence box - test section - fan? (The latter with the fan downstream is standard practice -- mass is conserved, the air going into the fan has to come from somewhere! And this gets you much better flow quality). Putting a bellmouth on the intake (preferably with a contraction going into the test section) also helps.
• I've seen a g rad student build an antiturbulence box using straws -- it took him and helper a week and an enormous number of straws to get the job done! He eventually bought a piece of aluminum honeycomb, and put screens at both sides.
• Have you done any math/measurements to try to estimate the Reynolds number your model wing will be operating at? Your force measurements may be off if the Reynolds is too different from the actual flight value.
• How will you react the lift force generated by the model, while keeping all parts of the tunnel aligned?
• If I understand correctly, your model wing will go from wall to wall? Be careful when applying any "2D" data you collect to your real "3D" wing.
• What sorts of deflections are you expecting to see with flap deflection? If your wing is anchored at both ends, the deflections will be very different than in-flight, with the wingtip free. If the model is built at 100% scale, the deflections should be pretty small, no? If you're building a sub-scale model, how will you scale your structure to get representative measurements?
• For the kinds of low-tech load tests you're describing, a car top or whirling arm rig might be a simpler/cheaper answer. Before you go to the trouble to build a full wind tunnel, think about the kind and quality of data you're ultimately interested in, and confirm that building the tunnel really is your preferred route.

I really encourage risk-reduction testing like this. But beyond "let's build something cool and have fun," be sure you're doing something that will truly answer the questions you have.

rtfm

Well-Known Member
FROM TOPAZ: Dang. RTFM, I apologize - I went to answer this post of yours and "edited" it instead! And now I can't find an "undo". Again, my profound apologies for disrupting your post.

What I was attempting to add to the discussion was, "One assumes there will be a "glue them all together" step in there? Lest the windtunnel blow the contents of the boxes - the straws - through the boxes and into the tunnel." Prompted by the part in your original post, below:

... 1. Anti-turbulance straws - I fully intend to make this a three minute build. It took them a week? How big was their tunnel? My method: Buy a dozen boxes of straws. Or 20. Chop off the top and bottom ends of the boxes. Stack the boxes on top of each other. Done.

Last edited by a moderator:

DeepStall

Well-Known Member
Re straws: the guys were cutting them all to ~7cm length to minimize the pressure loss. The straws were just inside the inlet, roughly 60cm by 100cm which was then contracted down to a 20cm by 15cm test section. So, lots of straws, lots of cutting, lots of gluing them all together. Your method sounds very straightforward, but I would expect full-length drinking straws to cause some pretty significant losses.

Building the wing prototype will be valuable. I'm still not 100% understanding how you're going to use any force/moment data you collect (absolute or relative to some defined state), but as long as you're clear, go for it!

rtfm

Well-Known Member
Hi,
Pressure loss... Mmmm I'll give this a bit of a think. Thanks.

The problem I face with using Fowlers and a speed brake is, I have no idea (and can't find any information on this either) how deploying the speed brake will affect pitching moment or lift. It could be that when I deploy the speed brake, it will disrupt the airflow over the flap and I'll fall out of the sky. I suspect, however, that the speed brake will have little or no effect on either the pitching moment or the lift. If so, I'll be a happy camper, and will build my wing like this, using speed brakes alone to alter the glidepath.

I promise to post lots of photos. And maybe a video... Basically, a bit of fun in the midst of all this interminable sanding!

Cheers,
Duncan

Well-Known Member
Using a very fine wire mesh also helps a lot. A good tip for cabin air inlets (straws work even better) if they're too loud.

planecrazzzy

Well-Known Member
Straws ???

Sounds like something mounted in the back of a Pick-up truck...

High into the wind stream.... Driving at 70 mph

should give plenty of data...

and less gluing...
.
Gotta Fly...
.
.

Aircar

Banned
Duncan--I am troubled. Firstly why you are having to wait around for your micro to harden (usually you have to work as fast as possible just to get it on before it hardens and tailor the hardener to match the temperature and quantity involed ( it is polyester based -yes?)

But, secondly, what you are proposing -and only slightly on the 'wind tunnel' aspect but that too.

You do know that the design of fowler flaps is a well established bit of aeronautical practice --that there are thousands (yes) of technical reports dealing with the loads and lift drag etc of all permutations of Fowler flaps (and slotted,double slotted,triple slotted , plain etc etc --this is not an area needing original research even if that was feasible )

Also you should possibly spend some time looking into the design work of Peter Garrison on his Melmoth 1 and 2 --they are almost unique in being equipped with Fowler flaps (for a homebuilt) and represent the most complex set of wings on almost anything ever built by an amateur . Peter is a technical journalist by profession and in the late 60s to 70 wrote a series of articles describing how he researched the design aspects of his intended long range aircraft by spendingtime with reference books and the research papers in the Royal Aeronautical Society library in London --he wrote up his design 'evolution' and personal learing in UK Pilot magazine and did the preliminary design of an aircraft based on his work called Sprite --he returned to the US and continued to develop his ideas and ended up with the Melmoth (others did build the Sprite in the UK) and detailed the construction of that over the -many -years it took to design and build.

I think you would benefit a lot from reading his building log and the tribulations that accompany any sophisticated development program --and to understand the neccesary background knowledge and how to apply it.

Just as an aside the Fowler flap was invented by one Harlan Fowler who was not an engineer and did it from inspired intuition as much as anything --having a big slot between the upper and lower wing surfaces would have been expected to ruin lift at first glance rather than almost double it .

You don't need to do any 'wind tunnel' testing but do need to study the ways to design a flap and use existing accurate data and you would find that your proposed tunnel would just collapse due to the lowered pressure in the contraction if there was any sort of air velocity --the old wind tunnel at DSTO had 1/2' thick steel walls for very subsonic speeds and the pressure forces are not small if viable ReNo testing is involved --also you can only do a small amount of flow turning in a straight tunnel before 'ground effect' and tunnel blocking take over.

Just well intentioned advice taken in that vein I hope.

Maybe someone can post a link to Peter Garrison's site.--?

Hot Wings

Grumpy Cynic
HBA Supporter
Log Member
You do know that the design of fowler flaps is a well established bit of aeronautical practice

Aircar, I'd like to pick your brain for a moment. While there may be plenty of data available for this kind of aerodynamic configuration, I've had very little success finding any information of modern airfoils suitable for plank style flying wings, particularly something as unique as Norman has posted using a kind of Junkers flap attached to a flap. Can you point my to any papers on testing a configuration similar to the BOK-5?

It's this kind of unique configuration that might justify a series of car top wind tunnel tests?

rtfm

Well-Known Member
Hi Aircar,
If there is so much information out there as it applies to my configuration, then answer two simple questions, and I'll re-think my intention to have fun with my wind tunnel:

Questions:
Fowler flaps deployed - 30 degrees. On approach.
Deploy speed brake (hinge line = main spar)
What happens to pitch? What happens to lift?

Regards,
Duncan

Last edited:

Matt G.

Well-Known Member
This isn't quite the same configuration that you'll have (I also doubt you're using the same airfoil, too) but it might be interesting reading, nonetheless.

Aircar

Banned
'Dear Abby......" I'll do my best ;- Offhand the BOK 5 looks like a servo flap mounted on the plain trailing edge flap --compare it to the "sparrow strainers" on Quickies and Dragonflies --fixed down lifting airfoils that counter the up floating tendency of the flap .

It may not be -- given that the trailing edge should need to stay in a negative deflection to trim a tailless plank type -unswept- wing .

If Norman reads this thread I think he is the most likely to have an idea on what the idea is ( I have a couple of books on Russian aircraft and some with flying wing stuff that might have a better view of it --if I find something I'll get back here.) --can you link to his posting on it ?

Thanks to Matt G. for the link to Peter Garrison's (new) site -- Peter is pretty self deprecating in regard to his work --I took dozens of slides of the Melmoth at Oshkosh in 1974 as it was the only really sophisticated homebuilt anywhere near sailplane levels and showing the result of real design and research --Rutan had the Vari Viggen there but it was pretty boxy and the only other comparable design in novelty

Duncan --you can look up the NASA Langley NTIS database and plug in all the key words --the design of Fowler flaps usually centres around getting the optimum location of the wing TE and flap nose (the slot setting geometry)

Regardless of what sort of flap you contemplate having airbrakes in front will reduce the lift and moments by drecreasing the dynamic pressure behind them --and therefore also the downwash over the tail all of which will change the trim of the whole aircraft . You could possiblt reduce the effect on the horizontal tail by raising it to the top of the fin (T tail ) thus sort of reinventing the Alvarez Polliwagen --an aircraft which was 'cloned' and modified with little understanding by a rather unscrupulous 'designer' who touted the result as his own design and "evidence of the design genius of XXX" (himself) on his website --without writing the book on the misadventures with this thing I understand now that the RAA will not register them.

Duncan-I backtracked your thread some time ago just to get a feel for things and it seems that you have blithely gone from a "freewing" through various changes of mind to the current Fowler flapped arrangement without ever precisely describing what you are trying to acheive --the "Fowler flap tail" iteration and other 'passing fancies' would themselves represent major design tasks to even analyze for feasibility -- and need a bit more than even a basic understanding of flight mechanics . _The Freewing concept is in the 'advanced' and highly sophisticated category and so far has only been demonstrated with one piece wings like the Spratt or Flying Fleas which have relatively low speeds and power and big damping from the fusealge and second wing/tail or the UAV type like the scorpion built by SCALED that had the split tail and boom 'flying' each wing half (being in effect two 'half' conventional aircraft split down the centreline pivotted about the motor )

Why do you want to have airbrakes mounted in front of the flaps anyway ? Why wouldn't you just fit a belly board like the VariEzes ? Even wing tip drag producers like the spli ailerons on the Grumman A6 Intruder (see some photos) --to avoid ant effect on the tail or to destroy lift ? --also makes for directional stiffness . These are all the sort of fundamental design decisions that are made by a designer very early on --your talking about building the fuselage or something else 'without knowing what sort of wing or where it is going to be ' on one posting seems a little bit --no a lot - cavalier and worrisome if you have a real intention to build something this way.

You could get a 'second opinion' from Bill Whitney --the most respected consulting engineer in your part of the world (or Australia) --I think he is still alive and will give advice on amateur built --or Daffyd Lllewelyn as another up there.

You do need some advice from an independent and qualified source though.

Aircar

Banned
To be more specific Duncan your "two simple questions" are nothing like simple at all ( as you must know) and often the asking of a question itself betrays the real problem . Hopefully you can get something from reading Peter Garrison's builder log at least --he fitted speed brakes to both his aircraft to allow rapid descent without chilling his engine and because he has good span loading and gliding performance (low sink rate ) in his second design -- with such short wings as you show on your avatar you won't be needing to increase sink rate especially with "30 degrees of fowler flap" out --it will start sinking like a brick as soon as you back off power and like 99% of other powered light aircraft without airbrakes .

Aircar

Banned
Hot Wings --an interim reply -- I looked up one book and found an article from Aeroplane may 2008 "Professor Belyaev's BATWING" which was one I had in mind --another tailess but twin fuselage and heavily forward swept design. The BOK 5 is described as having combination flap/elevator/aileron surfaces but hard to see how it could trim with just one --such novel designs are almost mandatory to do free air aerodynamic test . Another tailless forward swept wing tested by mounting it off a car was the Cornelius (later as the XFG 1 -experimental fuel glider it was tested as a manned aircraft but intended to be towed behind a bomber to extend range and cast off when empty -stable enough to remain in tow without a pilot ) photos of the front fender mounted model were published in popular mechanics etc --the Cartercopter , Taylor's Duck , the Zuck Planemobil etc etc were all tested on rolling rigs but mostly only at very small scale -- I designed a rig to fit a VW type 3 chassis for testing the new crop of ultralights that emerged after the end of compulsory certification in 1976 in Australia -- just about every non conventional design crashed and many were fatal (the ligeti stratos being the most well known from being promoted in the US )

rtfm

Well-Known Member
Hi Aircar,
Yes, I have been evolving my idea of what to build for a long time. And yes, I sometimes change my ideas quite dramatically. But this is what is fun about designing an airplane. You sound as though you regard this as a bad thing?

Having bounced all over the place concept-wise, and got it out of my system, what I have now is a reasonably standard airplane, and I'm toying with some ideas regarding flaps. Can't see how this is cavalier.

Thanks,
Duncan

rtfm

Well-Known Member
Duncan--I am troubled. Firstly why you are having to wait around for your micro to harden (usually you have to work as fast as possible just to get it on before it hardens and tailor the hardener to match the temperature and quantity involed ( it is polyester based -yes?)
First, Jeffco epoxy, not Polyester.
Second, the epoxy get to the leather stage in about 20 minutes, and starts to thicken up long before that. Yes, I have to work pretty fast to get it all on before it sets. But it takes at least 24 hours before I can work it (sand back, etc). This is the 'wait time' I am referring to.

You don't need to do any 'wind tunnel' testing but do need to study the ways to design a flap and use existing accurate data
I know how to build a Fowler flap (well, more or less - but nothing that a bit of experimentation won't solve) That has never been my issue. And I know what the expected extra lift will be. That is also not my problem. As I've said often enough for it to be getting a little repetitive, I want to observe the relationship (ie relative changes) between plain wing, progressing extension of the flaps and the deployment of an air brake. And that information has proved very elusive on the web. If you know where I can find it, please tell me.

But one thing strikes me as a bit odd... Why are you (and others) so sanguine about my wanting to build a wind tunnel? I don't get it. It's fun. It will answer my questions above. And it will (as a by product) give me a reason to mock-up my real wing/flap/brake. Suggesting I mount my wing on a car is of no value to me. I don't have a car.

Regards,
Duncan

rtfm

Well-Known Member
To be more specific Duncan your "two simple questions" are nothing like simple at all ( as you must know)
Hey, play fair, Aircar. First you tell me to stop wasting my time with a wind tunnel, but rather to seach the rich information already on the web. But when I pose my two questions, you tell me they're not simple at all.

That's exactly my point!

IF you can tell me (from all the data already published) what my two answers are, then I'll stop wasting time on my wind tunnel. But if the answers to these questions are NOT out there, then my wind tunnel is a great idea, because it will answer my questions quite simply.

As to why I want to use Fowler flaps AND a speed brake, the answer is quite simple. According to the glider guys, speed brakes provide a far superior and precise way to adjust the glide slope on approach without having to fiddle with power settings (and their in-built delay in effecting a change to the glide slope). This sounds to me like a damned good idea, and one worth considering. Pretty instantaneous effects. But I would need to know what happens to my Fowler flap enhanced lift when I deploy the air brakes. And if this question has not already been posed and answered on the web or elsewhere, then I need to emperically answer it for myself.

But there is a second reason, and that is the pitching moment caused by the Fowlers. I'd like to know:
1. How this pitching moment dials in as the flaps are extended. Orion has suggested that the pitching moment may in fact come on almost immediately, and steadily increase as the flaps extend. If this is the case, one of my other "flights of fancy" as you call it - ie the horisontal stab Fowlers - might be able to exactly balance out the pitching moment, resulting in a plane which does not change pitch with the extension of Fowlers. And I think that would be an interesting experiment also.
2. I'd be interested to see what happens if I extend the Fowlers only (ie sort of like a natural extension of the top wing skin only) rather than deflecting them downwards. Is there a big difference in lift when I transition from extension to deflection? Or is it mainly incresing the drag? Or both? Again, this requires relative measurements, not absolute ones.

Anyway, that's some background to my thinking. Don't you also enjoy playing? Trying this, testing that. Doing something just because you feel like it? And if any of these ideas seem to work, then I may well consider incorporating them into my plane (yes, even at this advanced stage). I will then, of course, need a plan to verify my preliminary tests, get advice etc.

Regards,
Duncan

2