HobbyKing.com New Products Flash Sale
Reply
Thread Tools
Old Mar 25, 2013, 06:03 PM
Registered User
United States, MA, Shrewsbury
Joined Mar 2013
86 Posts
Question
stabilizer (kk2 board) for training?

using a flight stabilizer with fixed wing firmware loaded, and getting any old aircraft without aileron servos attached (to force it to be RET, at least at first)...
couldn't almost any aircraft be converted to a trainer? with the exception of stuff like edf jets and warbirds.
the effect of gyrostabs is like a virtual dihedral wing, automatically leveling and preventing weird pitching
gyrostabs also reduce the effect of wind

http://www.hobbyking.com/hobbyking/s...rol_Board.html
joekitch is offline Find More Posts by joekitch
Reply With Quote
Sign up now
to remove ads between posts
Old Mar 25, 2013, 10:56 PM
Rocket Programmer
jasmine2501's Avatar
United States, AZ, Mesa
Joined Jul 2007
25,356 Posts
I don't know if rudder only would be quick enough for a gyro to stabilize the craft, but it's worth a try. With the KK you could certainly increase the stability, but it might not be like the AS3X planes. They make good trainers because as you said, they are better in the wind and they don't go off line and need to be corrected as much, allowing the student to focus on their flight path. It's certainly worth a try - the KK is cheap enough, just protect it well and isolate it from vibrations and give it a try. The worst that can happen is it crashes your plane or it doesn't work and you have to build a quad copter
jasmine2501 is offline Find More Posts by jasmine2501
Reply With Quote
Old Mar 26, 2013, 10:31 PM
Registered User
United States, MA, Shrewsbury
Joined Mar 2013
86 Posts
bump, because this might be a real boon to new pilots. you buy one model that's normally more advanced than you can handle, but use a KK2 board with open aero to dramatically stabilize it. make the model RET at first (but keep the ailerons plugged into the kk2 board, just not the receiver), then slowly get good enough to put the ailerons into the loop
joekitch is offline Find More Posts by joekitch
Reply With Quote
Old Mar 27, 2013, 04:32 AM
An itch?. Scratch build.
eflightray's Avatar
South Wales U.K.
Joined Mar 2003
12,895 Posts
Wont you end up with a model that wont bank, as the stabilizer will continually try to keep the wings level ? And then rely on just rudder yaw to turn ?

I would suggest you try it out on one of your four channel no dihedral models first. Then let us know what happened.
eflightray is offline Find More Posts by eflightray
Reply With Quote
Old Mar 27, 2013, 07:52 AM
buyer of the farm
United States, FL, DeLand
Joined Mar 2009
3,947 Posts
You have to remember that 3-channel planes are purposely designed for cross control of roll and yaw with the rudder. That isn't true with most 4-channel planes. If you have a 4-channel plane where there is no or very little cross coupling with the rudder and all you get is yaw, the plane will not be flyable without the ailerons, except in a straight line.

Any virtual dihedral your stabilizer can obtain on such a plane would be through its control of the ailerons. The rudder could do anything it wanted, it would do nothing to correct an adverse roll condition. Since adverse roll in a non self-stabilizing plane is a certainty, a crash would also be a certainty.

So no, your idea wouldn't work with the vast majority of 4-channel planes.

I'd suggest that it certainly would work if you just hooked it up as a 4-channel plane and simply skipped 3-channel RET operation altogether. Then the stabilizer could easily be used to give virtual dihedral. The plane would answer the controls just fine. Instead of having to input opposite aileron to unbank the plane at the end of a turn, the stabilizer would relevel the plane, just like a RET 3-channel would do. Also if you found yourself disoriented, with enough altitude you could just throttle back to half throttle or so, let go of the controls and the plane would recover to straight and level just like a Champ.

A way to give virtual RET would be to use a computer radio and program in a rudder/aileron mix so you don't have to use the aileron stick to begin with. Same deal--the stabilizer would provide the virtual dihedral. But what good is that when you learn to fly RET with primary roll control, the rudder in that case, on the "aileron" stick? Why even try to learn to fly the plane with the left stick? I guess you could hook rudder to the right stick, aileron to the left (which you would not use) and program the mix, but I don't quite understand what the advantage is. After all, the pilot doesn't really know or care what control surface the stick is connected to, he's only concerned what it makes the plane do. So I don't see this approach being too useful in making learning to fly any easier.

The problem with all these approaches is this guy is learning to fly and knows NOTHING. How's he going to program the stabilizer properly? How's he going to program the computer radio properly? So the limitation of using the stabilizer to help someone learn to fly is that he has to have an instructor to set everything up so the plane behaves properly. For the many, maybe they're the majority, who learn completely unassisted, this rabbit trail gives them no hope at all.

However, for RTF producers, this could be a whole new world and a whole new market.
Rockin Robbins is offline Find More Posts by Rockin Robbins
Last edited by Rockin Robbins; Mar 27, 2013 at 08:06 AM.
Reply With Quote
Old Mar 27, 2013, 10:19 AM
Rocket Programmer
jasmine2501's Avatar
United States, AZ, Mesa
Joined Jul 2007
25,356 Posts
RR, are you not aware of the AS3X planes, which are standard four channel planes stabilized by a gyro. The KK board can do airplane stabilization, so I'm afraid I'll have to disagree. I think this would work, just wouldn't work very well because gyros generally need to be able to cause quick movements in order to be effective. But, i don't think the slow response would cause any errors, you just wouldn't be able to have complete stabilization because the rudder would always be hunting and overshooting.
jasmine2501 is offline Find More Posts by jasmine2501
Reply With Quote
Old Mar 27, 2013, 10:21 AM
Rocket Programmer
jasmine2501's Avatar
United States, AZ, Mesa
Joined Jul 2007
25,356 Posts
Quote:
Originally Posted by eflightray View Post
Wont you end up with a model that wont bank, as the stabilizer will continually try to keep the wings level ? And then rely on just rudder yaw to turn ?

I would suggest you try it out on one of your four channel no dihedral models first. Then let us know what happened.
No, the stabilizer doesn't just stabilize for level flight, or yeah it would be fairly useless. Instead, it's part of the control loop, receiving commands from the pilot, and it stabilizes desired motion as well as level flight. Same as a tail gyro on a helicopter.
jasmine2501 is offline Find More Posts by jasmine2501
Reply With Quote
Old Mar 27, 2013, 11:59 AM
buyer of the farm
United States, FL, DeLand
Joined Mar 2009
3,947 Posts
Quote:
Originally Posted by jasmine2501 View Post
RR, are you not aware of the AS3X planes, which are standard four channel planes stabilized by a gyro. The KK board can do airplane stabilization, so I'm afraid I'll have to disagree. I think this would work, just wouldn't work very well because gyros generally need to be able to cause quick movements in order to be effective. But, i don't think the slow response would cause any errors, you just wouldn't be able to have complete stabilization because the rudder would always be hunting and overshooting.
Sure, I'm aware of the AS3X planes, but they aren't made for the beginner. They're all sport fliers.

I'm just saying that you can't turn a lot, probably most, 4-channel planes with the rudder. The plane has to bank and then the turn is actually accomplished by the elevator, according to the AAOPA. No bank, no turn (except for 3D planes with insanely effective rudders which can do an awesome flat turn).

Same thing if the plane will bank just a little with rudder. Since the response would be sluggish and well after any rudder input, overcontrol or undercontrol would be the response of the gyro stabilizer. Chances are it could never get ahead of the plane and would be unable to avoid catastrophe. Banking effects from turbulence could well be magnitudes more important than that resulting from control deflection.

I'm sure there are some 4-channel planes that would do fine, but look at threads talking about 4-channel planes with dihedral and you'll see the 4-channel buzzards all over that corpse cawing "4-channel planes with dihedral fly like pigs." So RC companies don't tend to sell 4-channel planes with dihedral in spite of the fact that almost all general aviation full size planes feature both dihedral and ailerons.

That's why I think it would be best just to abandon the RET configuration. You really don't give a rip what control surfaces are hooked to each channel. All you care about is when I flick this stick left the plane does this. Since that is true both in 4-channel and RET configurations if the respective planes are designed properly, why get obsessed about RET?

Just let the gyro give the 4-channel plane some virtual dihedral to help the new pilot recover from abnormal attitudes automatically and go from there!
Rockin Robbins is offline Find More Posts by Rockin Robbins
Reply With Quote
Old Mar 27, 2013, 01:44 PM
Registered User
United States, MA, Shrewsbury
Joined Mar 2013
86 Posts
Quote:
Originally Posted by jasmine2501 View Post
No, the stabilizer doesn't just stabilize for level flight, or yeah it would be fairly useless. Instead, it's part of the control loop, receiving commands from the pilot, and it stabilizes desired motion as well as level flight. Same as a tail gyro on a helicopter.
this is correct, commands from the pilot are interspersed with commands from the KK2 board. if you want to turn, you'll turn, mostly because it's easy to tell that you're turning the rudder and to not overcorrect or ignore that input

KK2 mostly just behaves like a dihedral (preference for level flight when there's no user input) but also does lots of tiny control surface movements to counter sudden acceleration due to wind (pulling 2 gs upwards over a third of a second, when the pilot wasn't doing anything like that, is very obviously wind buffeting and can be counteracted)

ASX3 is slightly better hardware, but from what i hear, KK2 with OpenAreo is better and more configurable overall

EDIT; and what i meant by RET was that the pilot doesn't have access to the ailerons yet, but the kk2 board does. you would of course get a 4 channel plane with a kk2 board. it just turns a 4 channel plane into a virtual dihedral RET trainer temporarily.
once the pilot can RET, like with a regular trainer, he'd take control of the ailerons
joekitch is offline Find More Posts by joekitch
Reply With Quote
Old Mar 27, 2013, 02:47 PM
An itch?. Scratch build.
eflightray's Avatar
South Wales U.K.
Joined Mar 2003
12,895 Posts
So if the ailerons are not connected to the Rx, i.e. no pilot/Tx input, as in - 'and getting any old aircraft without aileron servos attached ', what will the stabilizer controlled ailerons be doing if the model tries to bank with any yaw ?
eflightray is offline Find More Posts by eflightray
Reply With Quote
Old Mar 27, 2013, 02:54 PM
Registered User
United States, MA, Shrewsbury
Joined Mar 2013
86 Posts
Quote:
Originally Posted by eflightray View Post
So if the ailerons are not connected to the Rx, i.e. no pilot/Tx input, as in - 'and getting any old aircraft without aileron servos attached ', what will the stabilizer controlled ailerons be doing if the model tries to bank with any yaw ?
for the first few miliseconds it'll try to correct it, then realize that the pilot is actually trying to bank, and law off the interruptions while the pilot completes the turn while still making minute corrections for wind buffeting. humans do things very very slowly compared to even really cheapo integrated circuits.

all of that is configurable by the way, although it can be pretty tricky
i would probably put together a guide on all this, assuming it works how it should
joekitch is offline Find More Posts by joekitch
Reply With Quote
Old Mar 27, 2013, 03:04 PM
Rocket Programmer
jasmine2501's Avatar
United States, AZ, Mesa
Joined Jul 2007
25,356 Posts
Quote:
Originally Posted by joekitch View Post
for the first few miliseconds it'll try to correct it, then realize that the pilot is actually trying to bank, and law off the interruptions while the pilot completes the turn while still making minute corrections for wind buffeting. humans do things very very slowly compared to even really cheapo integrated circuits.

all of that is configurable by the way, although it can be pretty tricky
i would probably put together a guide on all this, assuming it works how it should
That is only true if the gyro and the sticks are controlling the same axis. If you put the gyro on one axis with no input, it will fight everything you do.

RR, I disagree that the little AS3X planes are bad for beginners, they make really good trainers, in spite of being four channel jobs. You need to be ready for a four channel plane, but the gyro assists the beginner a lot, allowing them to focus on flight path instead of keeping the airplane stable. A KK board can do the same thing, on all axes at once, just like the AS3X. It's normally used for multi choppers, but it can also be used on planes, or as a FBL controller.
jasmine2501 is offline Find More Posts by jasmine2501
Reply With Quote
Old Mar 27, 2013, 03:13 PM
Registered User
United States, MA, Shrewsbury
Joined Mar 2013
86 Posts
Quote:
Originally Posted by jasmine2501 View Post
That is only true if the gyro and the sticks are controlling the same axis. If you put the gyro on one axis with no input, it will fight everything you do.
nah, OpenAero is pretty smart, it can tell when you want to turn (generally), but it does need some tinkering to really work right. also the turns will feel a bit sluggish compared to a true 4 channel but all trainers tend to be sluggish anyway.

also, stabilizers don't have enormous control authority, they smooth the flight corridor, not dictate it. (well, they can dictate it if you set your gains and stuff too high)

Quote:
Originally Posted by jasmine2501 View Post
A KK board can do the same thing, on all axes at once, just like the AS3X. It's normally used for multi choppers, but it can also be used on planes, or as a FBL controller.
assuming you flash the right firmware, and the default kk2 fixed wing firmware isn't very good.
joekitch is offline Find More Posts by joekitch
Reply With Quote
Old Mar 27, 2013, 03:26 PM
An itch?. Scratch build.
eflightray's Avatar
South Wales U.K.
Joined Mar 2003
12,895 Posts
Joe, I assume you do fly 4 channel aerobatic low wing type models with no dihedral.

You must have at some stage flown more or less in a straight line and given just rudder inputs.

Generally the plane will travel straight on but yawed slightly, until the yaw starts to instigate some sort of turn, nose down roll.
If you give alternative left and rights, you can still fly a fairly straight line and 'wag the tail'.

The point I'm trying to make is the plane doesn't want to turn like a RET with dihedral.

Perhaps you can get a 'stabilizer' system to eventually do what you want. but as already pointed out by RR, why ?. If you're going to fit some sort of stabilizer to make a model easier to fly, it might just as well be the full house 4 channel.

I would love to see one working if you can achieve it, just to prove it does work as good as a RET. Perhaps your next project ?
eflightray is offline Find More Posts by eflightray
Reply With Quote
Old Mar 27, 2013, 03:38 PM
Registered User
United States, MA, Shrewsbury
Joined Mar 2013
86 Posts
Quote:
Originally Posted by eflightray View Post

I would love to see one working if you can achieve it, just to prove it does work as good as a RET. Perhaps your next project ?
that's what i'm thinking, yes.
i'd really like to take some really finnicky airframe like a small warbird and toy with KK2 and open aero until it's as docile as an HZ supercub

the only downside to a system like this is that every plane is different, so there would need to be an "ideal configuration" for each plane.

and on that note stepping down to something more common and universal like a Twinstar II would probably work a lot better. my hope is that eventually manufacturers will bundle these kinds of systems with RTFs and have the pilot literally "set the difficulty" of flight from beginner to advanced
joekitch is offline Find More Posts by joekitch
Reply With Quote
Reply


Thread Tools

Similar Threads
Category Thread Thread Starter Forum Replies Last Post
Discussion KK2.0 LCD Board as a quad + gimbal stabilizer??? crd Multirotor Electronics 34 Jun 05, 2014 08:14 AM
Discussion KK2 board problem kenbad Multirotor Electronics 5 Jan 12, 2013 11:14 AM
Discussion Fried my KK2 board!!!! Rubiclone Multirotor Electronics 2 Jan 12, 2013 12:11 AM
Discussion KK2.0 Board as standalone gimbal stabilizer DYORD Multirotor Talk 0 Jan 08, 2013 08:27 PM
Found wanted kk2 board boss777 Aircraft - Electric - Multirotor (FS/W) 3 Jan 07, 2013 07:25 PM