The Flying Box - DEVELOPMENT

[glydr]

Hi there,

Cool concept... makes me want to cut one myself.

I saw once in a magazine (that means I have absolutely no credit for this idea) that a guy made a flying cube much like the box concept. Instead of flying flat side down (like the current flite test box prototype) the cube flew corner side down, diamond style. The motor had a good deal of down thrust and dragged the cube around high alpha. I think the diamond style provides a dihedral effect which the square style doesn't enjoy.

The ingenious part was turning; the motor (located in a similar position to the flite test prototype) was attached to a servo which provided turn control. To turn left the motor would angle to left and the cube followed suit. Brilliant! Elevator control was left off this version with climb and descent controlled by motor thrust.

I'll have a dig around and see if I can find this magazine and if there are any other pearls of wisdom given by it.

G

PS - Awesome show!

 

[Rixx]

Hi guys. I'd try moving motor braces forward, mount motor as pusher, and stuck lighter battery in the center, in front of the motor. Tapering elevons (narrow at center, wide at edge) could improve handling, as center part is not contributing to roll much, just brakes airflow.

 

[glydr]

I found the mag: QEFI April 2002, 'Over Here' column, page 34.

In any case the aircraft in question has some information found on RC Groups: http://www.rcgroups.com/forums/showthread.php?t=210475

Here is a photo from the RC Groups review.



Reading the magazine article I found that I had the thrust angle wrong in my previous post. It's actually "18 degrees of upthrust, and some 20 degrees of left and right". Not sure if upthrust will help with vertical take off.

Seems to have a very forward CoG (like your's, Chad) with the battery in the bottom corner, quite far forward. Alas no details as to the exact CG location.

I'm thinking of doing one with a servo controlling up and down thrust, as well as left and right. Or simply an elevator vane somewhere in the box. Hmmmm.

Looking forward to the next proto! It would be great to make one of these look like a postal package!

Cheers,

G

 

[mrfourtysevenman]

CHAD
put rocket fins on the outside that way it doesnt roll like a barrel when it tries to fly. the prop wash inside the airframe is rotating the whole thing like a cow stuck in a tornado. the stabilization must come from outside the airframe.

 

[colorex]

CHAD
put rocket fins on the outside that way it doesnt roll like a barrel when it tries to fly. the prop wash inside the airframe is rotating the whole thing like a cow stuck in a tornado. the stabilization must come from outside the airframe.

However, stabs on the outside might ruin the look of the box...

 

[Gundy]

I once did a "cylinder plane" and had the same issues you are having here.
The answer I came up with is now called a "vector control system". a vector system moves the motor to change the direction of the plane. no control surfaces. just point the motor where you want to go.
To keep it stable, keep all the weight at the bottom. gravity will keep it heavy side down.
Your issue with it being to twitchy to control is a size issue. Make it larger and everything slows down.
My cylinder was 36" in diameter.
It flew great in calm conductions.

Later,
Gundy

 

[ExAir]

Flying Box

Very cool, can't wait to see it work!

One consideration is that the roll moment exerted by the elevons is only about 8cm but the pitch moment is about 30cm. That would certainly make the pitch more active for a given degree of deflection. Putting a vertical rudder with opposite moving "rudders" would double at least the wetted surface area for roll inputs. May help, maybe not.

I've seen this one implementation of that: http://www.youtube.com/watch?v=2ssdQ5qoH0I&list=FLzV3CxJH2WOmtL8djZ7MS2A&index=24

Well done brother!

 

[ananas1301]

That video is cool. I like how he makes all those rolls but still still had the thing under control. My respects to him for that.

The good thing you could see on this is that he only used rudder to steer from left to right and vice versa. The elevon function was kinda unnecessary unless you want the ability to roll, because in fact you have all control over the plane only by rudder and elevator.

Rudder or vectored thrust seems to be the way to go for the flying box I think.

 

[Rcplaneguy]

Hey Chad, have you thought of going with some wings on the side and trimming them down and testing them one length at a time? May help your problem of figure out the problem, and also, a gyro on the roll would be great.
Thanks!
~Rcplaneguy1

 

[nibnobsam]

The first time I read "Flying Box" I thought more of a helicopter than a plane.

If you make it a heli, or hovering-box, it would instead require much tail-heaviness to hold steady in the air. It would also need trimming to prevent torque roll.

Maybe a quad copter in a cardboard box

 

[sfactor1]

how about adding 2 servos for a "rudder" where they can move like a rudder in 1 mix and in another mix for aileron control, make them all move clockwise or counter clockwise so when u input a roll, all 4 turn in a windmill (like a swastika) pattern so it should roll like a drill bit! the added servos should allow u to mount a smaller batt, preferably as close as possible to the axis of roll

 

[Non Action Man]

Hey Chad, great to see something a little different!

My thoughts: It seems you are after two things here, vtol and forward flight, two very hard things to mix together on simple project, as you would have seen by the box falling over as soon as it got off the ground. But of course its only hard, not impossible!

If you do want to have vtol as one of the goals then you will have to centralise the GC to the "vertical" axis (the axis the motor is on). But as you are aware, this will introduce many other problems with torque roll.

Have a look at this video of a vtol "uav" it is a circular aerofoil and the trailing edges are the control surfaces. The mechanism of lift is very different to what you are doing, but perhaps the control method could be the same, just "inside out".

Instead of the control surfaces deflecting outwards, you could could keep a more intact box appearance by having your surfaces deflecting inwards, hinged off the inner surface of the outer box - not explained very well sorry! You would need to use a gyro to control the torque roll, with some small surfaces that counter act the roll. As long as they are in the prop wash, they should be able to stop the roll - In the video you can see a ring of small static deflection surfaces, and a few active ones.

This would need a CG further back than you have it now, to reduce the balancing act to keep it vertical. But under forward flight it wont really level out.

On the other hand, if you don't really want vtol, then I'd go with the diamond "mode" corner down approach, then you have 4 wings with a large dihedral instead of 2 wings with none. Thrust vectoring might be the way to go too. I think its more about getting a battery, motor and speed control to fly and then make it look like a box.

anyway, just some ramblings,

NAM

 

[flyingbiscuit]

NAM is right. VTOL and forward flight are hard to mix. The military has been trying a long time to do it efficiently. The best way to go with this is to run corner down (diamond) by moving battery, RX, and ESC to the corner. Hinge all four outer back edges in a v-tail mix using ailerons and elevator, and since the center post is already hinged use it for rudder. After that the best vertical take off you're going to get is what you already have, unless you make it a helicopter.

 

[znurtdog]

For Box Sake! Paint that thing so we know which ways what! Oh, and how about some leds?

 

[StoneKap]

Hi Guys,
There is a lot of really great advice here. The challenge that I was going for was to maintain the integrity of the box. That means I can't cut up the walls of the box. The other is to have minimal markings on it. I'd also like it to fly flat side down. With this criteria in mind, I think I'll try the rudder next (most popular suggestion) and adjust the CG (2nd most popular).

 

[Manfet]

Hi Guys,
I think I'll try the rudder next (most popular suggestion) and adjust the CG (2nd most popular).

Well many made the suggestions but if I remember correctly most just suggested one of them and not both. In my opinion both are overkill for the start. But adjusting the CG has the benefit of better VTOL abilities. But the better flight in terms of stability is probably reached through the rudder. At first I would have advised you to go with the rudder but CG in the middle is kind of cool because I would like to see the box banking in the air.

Good Luck with your enhancements

 

[SunShine]

Hi Chad!
be careful with the cg, if you center it (put it behind the prop) then you will have insane amounts of roll since you have no wings, also, it will be very instable, like a heli without a gyro...
what i would do: leave the cg, and ad rudders, you don't really need this plane (box) to roll so you can avoid adding weight by using one servo for the elevator and one for the rudder. if you want it to VTOL, than that's another story, than go 4 channels and add a 3 axis gyro, also center the cg, and push it faaar back.
just my 2 cents
Also, as a video maker i strongly recommend to ad some paint too the outside of the box
-A

 

[F16V1PER]

I do not think the bank-and-yank method will work with this type of aircraft with this current setup. So if you want more turning I would suggest either putting a rudder on it or actually putting small delta wings on it. And of course making the control surfaces bigger.

 

[sfactor1]

what im suggesting is to basically build something like the upcoming eflight hypertaxi encased in a box.....that thing flys insane!

 

[gKadonoff]

What I would try is making it as similar as possible to a box kite. If you look at pictures of box kites you'll see that they basically have two separate boxes that are connected. This might give you more stability.

And for the control surfaces, I would put them on the sides of the box, not on the inside, when the control surfaces are on the inside like that they are only really deflecting the prop wash which is very turbulent. If the controls are on the sides, they're going to be deflecting steady air flow around the sides, possibly giving you better control.