Thread Tools
Jan 27, 2015, 01:38 AM
JulietKiloMike's Avatar
Thread OP
Build Log

Noob Build: Trirotor Tiltwing


Hi All,
I have embarked on an ambitious school project for my Applied Science Research class--building a VTOL aircraft from scratch--and thought I would start a build thread so that my various trials, failures, and learning experiences are catalogued for all to see (and so that I can learn from the expert builders who frequent this forum). I'm working in a team of two with a classmate, and we are planning to base our design on something vaguely similar to the XC-142, with the exception that we will be using 3 engines instead of 5 (only two engines on the wings).

We have a KK2 board on hand and are planning to use OpenAero firmware when the VTOL phase comes...but right now we are still in the early design phase, and will be working towards a fixed-wing prototype before we do anything fancy. So far, we've picked out an airfoil and done some wind-tunnel testing to find the necessary parameters for our wing, but right now, our aircraft consists of a few boxes of electronics, some spreadsheets, and a CAD file. I'll do my best to post some updates as the project progresses!

EDIT: This project has been completed for a while now, and wow it was quite a ride! I figured that I would move some of the more interesting links up to the front post for easy access.
The final aircraft CAD file, free to download: http://a360.co/1AAyMjs
Video of the aircraft in action:
DTF Trirotor Tiltwing VTOL STOL FFF Testing at Baylands (4 min 51 sec)

Plans for laser-cutting the aircraft parts are attached to this post, and many more videos of the aircraft during its testing process are available online at https://www.youtube.com/user/CoolNamesAllTaken

Massive Development Video Dump:
Foamboard VTOL Maiden Flight and Crash (4 min 7 sec)

Foamboard Trirotor Tiltwing VTOL Aborted Takeoff (1 min 34 sec)

Foamboard Trirotor Tiltwing VTOL Motor Run-up (0 min 16 sec)

Foamboard Trirotor Tiltwing VTOL Wing Tilt Test (0 min 18 sec)

DTF Tritotor Tilwting VTOL Motor Run-Up and Wing Tilt Test (4 min 0 sec)

DTF Trirotor Tiltwing Fixed-Wing Test Flight and Soft Crash (3 min 26 sec)

DTF Trirotor Tiltwing Fast-Forward Flight Test 1, 4-17-15 (2 min 25 sec)

DTF Trirotor Tiltwing Fast-Forward-Flight Test 3, 4-17-15 (1 min 55 sec)

DTF Trirotor Tiltwing Fast-Forward-Flight Test 4, 4-17-15 (5 min 26 sec)

Openaero VTOL KK2 Shutdown (0 min 47 sec)

DTF Trirotor Tiltwing FIRST HOVER TEST :D (0 min 0 sec)

DTF Trirotor Tiltwing Hover Test 2 (0 min 53 sec)

Openaero VTOL tail rotor throttle problem (1 min 3 sec)

DTF Trirotor Tiltwing Table Landing 1 (2 min 3 sec)

DTF Trirotor Tiltwing Table Landing 2 (4 min 29 sec)

DTF Trirotor Tiltwing Flying Into the Science Hole 1 (1 min 18 sec)

DTF Trirotor Tiltwing Flying Into the Science Hole 2 (0 min 27 sec)

DTF Trirotor Tiltwing STOL Sunset Flight (1 min 5 sec)

DTF Trirotor Tiltwing STOL and VTOL Transition Testing (5 min 28 sec)

DTF Trirotor Tiltwing VTOL STOL FFF Testing at Baylands (4 min 51 sec)
Last edited by JulietKiloMike; Feb 11, 2016 at 06:42 PM.
Sign up now
to remove ads between posts
Jan 27, 2015, 01:45 AM
JulietKiloMike's Avatar
Thread OP
Okay, first update!

Here is a link to the current aircraft CAD file: http://a360.co/1D72Rcs

I'm still working on the internal structure, motor mounts, and electronics positioning to get the CG at 1/4 MAC. The aircraft's structure will be composed of 3/16" dollar tree foam board and 1/8" laser-cut plywood. We aren't going for scale details or prettiness...we just want it to fly! Hopefully by the end of this we will have a reasonably practical VTOL RC aircraft that can heave itself off the ground long enough to impress our teacher

The aircraft has a planned wingspan of approximately 42 inches. The fuselage length is still in the works, but should turn out to be somewhat similar, depending on what we need for CG.

So...any comments or advice so far? I've had some experience with fixed-wing RC and quadcopters, but this is my first VTOL (and my second scratchbuild ever). Wise words or friendly chatter are equally appreciated!
Jan 27, 2015, 09:09 AM
Registered User
It seems mostly fine as a concept. You should take a look at Randy's XC-142 thread here: https://www.rcgroups.com/forums/show....php?t=2008164
To actually hover with the fuselage level you will need the wing to tilt more than 40 degrees, though. That said, apart from the greater number of motors the configuration is pretty similar. The electronics will be identical, only driving one ESC per wing instead of two. For simplicity's sake you might want to use a fixed T tail in place of a tilting low tail, like in the BA 609. But if the tail is generally clear of the wake of the wing rotors it shouldn't matter much in any case, and a simple fixed tail will do fine. Or you could make the tail all moving and just tweak the mix in the KK2 board.
Jan 27, 2015, 10:34 AM
Registered User
Ran D. St. Clair's Avatar
Yes, it looks a lot like the XC-142, which is not a model that i would recommend as a first VTOL. If you want to give yourself the greatest possible chance of success then I recommend a Separate Lift Thrust (SLT) type. They have nothing that tilts, and are basically a quad (or tri-copter) strapped on top of an otherwise conventional airplane. You can start with most any descent flying airplane including flying wings, but aircraft with a tail tend to be more stable. The VTOL-FunCub is a classic SLT design and is described in this thread.

https://www.rcgroups.com/forums/show...1958738&page=2

If you continue with your current design concept then ditto what Brandano said.

The design isn't sufficiently complete to really know how you intend it to work, but I would be concerned that you don't have any provisions for pitch control in hover. Tilt wings can be tricky in that the CG required for hovering and the CG required for forward flight are only barely compatible if you do it right. This assumes fixed pitch propellers. Variable pitch props are a different matter.

I suggest you poke around this site and limit your choices to aircraft with successful flight videos.

You will need a flight controller of some sort eventually. The OpenAero-VTOL firmware on a KK2 board is one possible choice.

https://www.rcgroups.com/forums/show....php?t=1972686
Jan 27, 2015, 11:59 AM
Registered User
With most models, the lighter you build it, the easier it is to hover / fly, and the less damage the ground inflicts on it when it crashes....which will always happen at some stage during testing....I like your laser cut ply fuse, but I would cut away more ply and / or use thinner ply....

Why don't toddlers break an arm when they fall over?............because they are light enough to bounce!.......Mass, momentum etc....H
Jan 27, 2015, 12:02 PM
Registered User
Ran D. St. Clair's Avatar
Quote:
Originally Posted by CaribbeanBlue
With most models, the lighter you build it, the easier it is to hover / fly, and the less damage the ground inflicts on it when it crashes....which will always happen at some stage during testing....I like your laser cut ply fuse, but I would cut away more ply and / or use thinner ply....

Why don't toddlers break an arm when they fall over?............because they are light enough to bounce!.......Mass, momentum etc....H
Good advice. When in doubt, make it lighter, not stronger. Foam board is a good prototyping material. You probably don't even need the plywood, but if you use it 1/16" thick is plenty.
Jan 27, 2015, 12:48 PM
JulietKiloMike's Avatar
Thread OP
Quote:
Originally Posted by Ran D. St. Clair
Good advice. When in doubt, make it lighter, not stronger. Foam board is a good prototyping material. You probably don't even need the plywood, but if you use it 1/16" thick is plenty.
We only have 1/8" plywood on hand, but we might switch to balsa to save weight. We're using 1/8" plywood for our wing spar and tail boom, but I'm not sure that we need the extra strength for some chunks of the wing box.
Jan 27, 2015, 12:56 PM
JulietKiloMike's Avatar
Thread OP
Quote:
Originally Posted by Ran D. St. Clair
Yes, it looks a lot like the XC-142, which is not a model that i would recommend as a first VTOL. If you want to give yourself the greatest possible chance of success then I recommend a Separate Lift Thrust (SLT) type. They have nothing that tilts, and are basically a quad (or tri-copter) strapped on top of an otherwise conventional airplane. You can start with most any descent flying airplane including flying wings, but aircraft with a tail tend to be more stable. The VTOL-FunCub is a classic SLT design and is described in this thread.

https://www.rcgroups.com/forums/show...1958738&page=2

If you continue with your current design concept then ditto what Brandano said.

The design isn't sufficiently complete to really know how you intend it to work, but I would be concerned that you don't have any provisions for pitch control in hover. Tilt wings can be tricky in that the CG required for hovering and the CG required for forward flight are only barely compatible if you do it right. This assumes fixed pitch propellers. Variable pitch props are a different matter.

I suggest you poke around this site and limit your choices to aircraft with successful flight videos.

You will need a flight controller of some sort eventually. The OpenAero-VTOL firmware on a KK2 board is one possible choice.

https://www.rcgroups.com/forums/show....php?t=1972686
That's a very interesting idea! I hadn't seen that kind of VTOL before, and that might be a good path to take if we hit some serious snags in our current project path.

We are using the XC-142 design in order to specifically avoid the use of variable-pitch props (I'm not a helicopter guy). What isn't shown in the snapshots of the incomplete CAD drawing is a tail rotor. We are planning to use a mini-quad motor on our tail for pitch control...as long as we have the CG in front of Center-of-lift of the wing and slightly behind the center-of-lift generated by the motors on the wings, we should, in theory, have controllable pitch.

Here's some more links to our parts list and components list, if you're interested: https://docs.google.com/document/d/1...it?usp=sharing

https://docs.google.com/spreadsheets...it?usp=sharing

Thank you for the great advice!
Jan 27, 2015, 01:42 PM
Registered User
Ran D. St. Clair's Avatar
Quote:
Originally Posted by JulietKiloMike
That's a very interesting idea! I hadn't seen that kind of VTOL before, and that might be a good path to take if we hit some serious snags in our current project path.

We are using the XC-142 design in order to specifically avoid the use of variable-pitch props (I'm not a helicopter guy). What isn't shown in the snapshots of the incomplete CAD drawing is a tail rotor. We are planning to use a mini-quad motor on our tail for pitch control...as long as we have the CG in front of Center-of-lift of the wing and slightly behind the center-of-lift generated by the motors on the wings, we should, in theory, have controllable pitch.

Here's some more links to our parts list and components list, if you're interested: https://docs.google.com/document/d/1...it?usp=sharing

https://docs.google.com/spreadsheets...it?usp=sharing

Thank you for the great advice!
The XC-142 has the CG 1 unit of measure behind the lift motors and 16 units of measure in front of the tail rotor. This 16:1 ratio is extreme and problematic for adequate pitch stability in hover. An 8:1 ratio or less is recommended. This ratio also applies to the total disk area of the wing mounted rotors, as compared to the tail rotor. For example, if you used 2 each 9" diameter props up front, and one 4.5" diameter prop in the rear, you would have the desired 8:1 ratio.

The CG also needs to be correct for forward flight though. That implies abuut 33% of the mean aerodynamic chord, (25% is a bit nose heavy). Typically the CG is right on the wing spar,which is also the wing pivot point. As you rotate the wing to vertical the heavy motors will move up and back which will move the CG back slightly, like 1/4". If the motors hang 1" below the wing then their center of lift will be 1" ahead of the wing when the wing is vertical for a total of 1 1/4" ahead of the CG. At 8:1 that puts your tail rotor 10" behind the CG which makes for a very short tail. I will let you do the math from here, but hopefully you will see that you are headed for a CG problem. You can hang the motors farther below the wing, like 3", but then the props are probably hitting the ground when the wing is flat. Tilt wings are not the easiest way to go... I did it to make a scale aircraft. You have no such limitation.

Also, plywood is not usually the best material for a wing spar. I recommend you get yourself a carbon fiber tube. Lots of materials can work well, but plywood is heavy and half the plies run the wrong direction for strength in tension or compression. An I beam or tube/Box is also recommended over a solid cross section for the best strength and stiffness per weight. A box or tube is best for torsional stiffness which is typically needed for a wing carying motors.
Jan 27, 2015, 01:54 PM
Registered User
Ran D. St. Clair's Avatar
2 motors on the wing or 4?

Your batteries are on the small side. You will only get about 5 minutes flight time, which means about 3 minutes with a safety factor. That will stress them and end their life early. Don't believe the C ratings that they advertise. Larger batteries will reduce the stress and give you more safety margin.
Jan 27, 2015, 03:14 PM
JulietKiloMike's Avatar
Thread OP
Quote:
Originally Posted by Ran D. St. Clair
The XC-142 has the CG 1 unit of measure behind the lift motors and 16 units of measure in front of the tail rotor. This 16:1 ratio is extreme and problematic for adequate pitch stability in hover. An 8:1 ratio or less is recommended. This ratio also applies to the total disk area of the wing mounted rotors, as compared to the tail rotor. For example, if you used 2 each 9" diameter props up front, and one 4.5" diameter prop in the rear, you would have the desired 8:1 ratio.

The CG also needs to be correct for forward flight though. That implies abuut 33% of the mean aerodynamic chord, (25% is a bit nose heavy). Typically the CG is right on the wing spar,which is also the wing pivot point. As you rotate the wing to vertical the heavy motors will move up and back which will move the CG back slightly, like 1/4". If the motors hang 1" below the wing then their center of lift will be 1" ahead of the wing when the wing is vertical for a total of 1 1/4" ahead of the CG. At 8:1 that puts your tail rotor 10" behind the CG which makes for a very short tail. I will let you do the math from here, but hopefully you will see that you are headed for a CG problem. You can hang the motors farther below the wing, like 3", but then the props are probably hitting the ground when the wing is flat. Tilt wings are not the easiest way to go... I did it to make a scale aircraft. You have no such limitation.

Also, plywood is not usually the best material for a wing spar. I recommend you get yourself a carbon fiber tube. Lots of materials can work well, but plywood is heavy and half the plies run the wrong direction for strength in tension or compression. An I beam or tube/Box is also recommended over a solid cross section for the best strength and stiffness per weight. A box or tube is best for torsional stiffness which is typically needed for a wing carying motors.

Thank you, sir! This is extremely helpful.

I definitely have to do some more math for CG calculations. Right now my tail rotor is around 18" behind CG, which is far from optimal. I'll try to bring it in a lot closer by moving the tail plane forward.

Currently I'm using 10" diameter props on the wings and a 5" prop for the tail. I might have to add some sort of landing gear to get enough prop clearance when hanging the engines beneath the wings. I'll make some tweaks and let you know how it goes.

The carbon fiber tube is a good idea, but I've heard bad things about cutting carbon fiber. Do you have any suggestions for working with it?
Jan 27, 2015, 03:18 PM
JulietKiloMike's Avatar
Thread OP
Quote:
Originally Posted by Ran D. St. Clair
2 motors on the wing or 4?

Your batteries are on the small side. You will only get about 5 minutes flight time, which means about 3 minutes with a safety factor. That will stress them and end their life early. Don't believe the C ratings that they advertise. Larger batteries will reduce the stress and give you more safety margin.
2 motors on the wing: 900Kv SunnySky's swinging 10" props on 4S. According to SunnySky's data sheet, they should be drawing slightly over 20A apiece on full power. Fortunately, we should only be running them at 6A apiece for a stable hover (we're shooting for somewhere between 500g and 1kg AUW right now).
Jan 27, 2015, 03:44 PM
Registered User
Ran D. St. Clair's Avatar
Quote:
Originally Posted by JulietKiloMike
Thank you, sir! This is extremely helpful.

I definitely have to do some more math for CG calculations. Right now my tail rotor is around 18" behind CG, which is far from optimal. I'll try to bring it in a lot closer by moving the tail plane forward.

Currently I'm using 10" diameter props on the wings and a 5" prop for the tail. I might have to add some sort of landing gear to get enough prop clearance when hanging the engines beneath the wings. I'll make some tweaks and let you know how it goes.

The carbon fiber tube is a good idea, but I've heard bad things about cutting carbon fiber. Do you have any suggestions for working with it?
Your prop sizes and motors are reasonable. That makes an 8:1 ratio in terms of the distance from the CG. If you locate the tail rotor as necessary to maintain the correct ratio, and put the horizontal stabalizer in front of the tail rotor, you will end up with a very short coupled design and likely poor pitch stability in forward flight. You would be better off to go with twin booms and put the horizontal stabalizer behind the tail rotor. If your fuselage is wide, like 5 1/2 inches, then you can extend twin vertical stabalizers from the sides of the fuselage in the manner of the WUV.

https://www.rcgroups.com/forums/show....php?t=2307837

I am not really recommending it, as there are better solutions, but it would solve your CG problem with minimal change to the overall design.

Cutting carbon fiber is no big deal. Wrap the tube, rod, or whatever with scotch tape and cut it with a fine tooth saw. A hack saw works in a pinch. Suck up the dust immediately with a vaccume cleaner or do your cutting and sanding under running water in the kitchen sink. You can wear rubber gloves to avoid itchy skin, but I never bother. CF cuts easily and is not really dangerous like some other materials.
Last edited by Ran D. St. Clair; Jan 27, 2015 at 03:49 PM.
Jan 27, 2015, 04:00 PM
Registered User
Quote:
Originally Posted by Ran D. St. Clair
Cutting carbon fiber is no big deal. Wrap the tube, rod, or whatever with scotch tape and cut it with a fine tooth saw. A hack saw works in a pinch. Suck up the dust immediately with a vaccume cleaner or do your cutting and sanding under running water in the kitchen sink. You can wear rubber gloves to avoid itchy skin, but I never bother. CF cuts easily and is not really dangerous like some other materials.
I happened to notice an rcgroups post the other day where the author suggested using a small pipe cutter.
Jan 27, 2015, 05:32 PM
Registered User
if the motor pods are slung under the wing the rotor disk will move forward during the conversion, and give you a little more leeway in placing the CG. This comes at the cost of greater torque on the tilt servo, that is usually already under a very stressful load. A better option to move the GC back would be to increase the horizontal stabilizer size. How do you plan to control yaw? Ailerons in the rotor slipstream, tilting tail rotor or both?


Quick Reply
Message:

Thread Tools