**Update** I will start my next build of a Mini-Tiltrotor. Check back for updates! The components list can be found on my board
A Little Background
Ever since I saw a video (at the bottom of this post) of an RC Tiltrotor on Youtube, I have wanted to build one myself. In addition, I have wanted to build my own autopilot. Why build my own autopilot when there are plenty of cheap ones out there these days? Well... Because it is a lot of fun. Anyway, I've had a video of my tiltrotor on Youtube at night. Since filming it, it burned out an ESC. I finally got around to fixing it, mostly so I could take it apart and scrap it to build a tricopter. Yes, I know, a tricopter is not nearly as cool as a tiltrotor, but it is a much better development platform for autopilots. Don't worry, my tiltrotor days are not over. Once I have a stable autopilot, I will build one that converts to full forward flight. I promise.
Advantages of a Tiltrotor
At the expense of a slightly more complicated autopilot and mechanical complexity, a tiltrotor has half the motors, ESCs, connectors, propellers, and power as a quadrotor. Essentially, a cheap tiltrotor can be significantly cheaper than a comparable quadrotor. And they are cooler. Yup. That's about it.
How It Works
Some of you might be wondering how this thing even flies. I plan on doing a multi part video on how things fly that includes something like this fixed pitch tiltrotor. For now, we will have to use a simplified diagram of the forces to understand how a fixed pitch tilt rotor can work.
Roll is the simplest of the axis to control. By simply varying the thrust of either motor, you create a torque on the aircraft causing it to roll
Yaw is also pretty straight forward. You tilt one motor forwards and the opposing motor backwards. With the motors tilted in opposite directions, part of the thrust of thrust vector creates torque on the aircraft.
I've saved pitch for last because it is the most difficult to understand. The biggest problem to understanding how pitch control works on a fixed pitch tilt rotor is the thought that the aircraft "hangs by the props." Imagine this. How about we take the aircraft up to the ISS and fly it up in zero gravity. In zero gravity, tilt the rotors completely forward. What happens if you spin up the motors to create a force that is not aligned with the CG and offset from the CG? Obviously, it creates a torque about the CG. As you know, that is how the yaw control works. Now, add a force aligned to the CG, at the CG. The torque of the original force that is not aligned still creates a torque about the CG, while the force applied at the CG creates an acceleration on the body. Now, come back to earth. The same physics apply. When hovering flat and level, the tilt rotor will produce thrust exactly opposite of gravity and cancel it out. If you tilt the rotors in one direction or another, a portion of the thrust will create a torque on the airframe.
If you tilt the force a little bit, a portion of the thrust turns into a force that produces a torque on the aircraft, regardless of the motion or orientation of the airframe. Because the the airframe acts as a free body in space, the torque translates into angular acceleration.
If you think about this, this is similar to how yaw works. A portion of the thrust goes to create torque about the CG because the force is offset from the CG.
Note that in order for a fixed pitch tilt rotor to work, the rotors can not rotate about the CG. The rotation point must be offset from the CG in order to create a torque on the airframe. The props could be below the CG and work just fine. The further from the CG, generally the more controllable the aircraft becomes.
The Goal: Transition to forward flight. Notice a problem?
What happens when you want to convert to forward flight? Consider tilting the rotors 90 degrees to face forward. Now, there is a large torque on the airframe. How can we counter the torque of the motor? During forward flight, a fixed pitch tiltrotor must have an elevator or some other control surface. Or, somehow, it could align the thrust in forward flight with the CG to avoid torque.
The airframe was basically thrown together with whatever I had on hand. Originally, I built the body out of DollarTree foam board, but it fell apart from rough landings. So, I busted out my hot glue gun and built a frame out of coroplast from home depot. For the arms, I ended up using some 3/4" square wooden dowels from Home Depot. The wood is cheap, light, and dampens vibration, unlike aluminum square tubes. The wood is also easier to cut and drill into.
The battery needs to be as low as possible, and is mounted directly under the arms. The position can be forward or backward to adjust the CG. Ideally, the CG should be directly under the arms. This allows the aircraft to lift off vertically without a pitching moment. In a hover, if the nose the aircraft is pointed down, the CG is too far forward. If the nose is pointed back, the CG is too far back. If the aircraft flips over at takeoff, odds are that your CG is too far off to even fly.
Everything is held together with hot glue. Hot glue is awesome. It even is how I fixed the pivot mechanism to the arms. In hard crashes, the hot glue joints failed. I was able to reheat the glue or clean it off and apply a new blob. There is a lot of value in knowing where your airframe will fail first.
Motor, ESC, and Battery Selection
- hexTronik DT750 Brushless Outrunner 750kv
- Hobbyking SS Series 25-30A ESC
- Turnigy 2200mAh 3S 30C Lipo Pack
- Zippy-K Flightmax 4000mah 3S1P 25C Lipoly Battery
- Turnigy Multistar 20 Amp Multi-rotor Brushless ESC 2-4S
** Update **
HobbyKing recently released a cheap multi-rotor ESC with fast update time and low lag, supposedly
I plan on doing a writeup on multi-rotor craft with some actual theory about motor and prop selection, but for now I'll jump to the conclusion. To increase flight times, you want low Kv motors with larger props. The hexTronik DT750 Brushless Outrunner 750kv is a perfect motor for mid to large sized copter. Two of these motors with the right props together could lift about 4lbs. Plenty for a 2lb tiltrotor. The cheap SS ESCs will work fine. The ESCs have a noticeable delay, but they work with the right gains. If you get really ambitious, there is alternative firmware for these ESCs that allows better 'copter control. Can't go wrong with cheap and dirty, though.
As for battery selection, you need a peak of about 40A with the DT750 and an 11" prop. The minimum battery you would need is a 2200mAh 20C. A 4000mAh 3S battery will give you significantly more flight time, and you should still be able to hover at around half throttle. In fact, I have flown this tiltrotor with a 5000mAh 3S battery and it handled fine. You just have to ensure that you can still get your CG in the correct place.
- GWS HD1060 3-BLADE, BLACK, STANDARD PITCH
- GWS HD1060 3-BLADE, BLACK, REVERSE PITCH
- Slow Fly Electric Prop 11X4.7SF (4 pc)
- Slow Fly Electric Prop 11x4.7R SF (4 pc - Green Right Hand Rotation)
- Slow Fly Electric Prop 11x4.7SF RH Rotation (4 pc)
As shown in video
Cheaper alternative (and tested with the DT750 to work quite well)
With the 11x4.7 SF props and two of the 750Kv motors, on a 3S lipo, the tiltrotor hovered around 150W and easily lifted the tiltrotor at about half throttle. I don't have a scale, but I estimate the final tiltrotor weighed in around 1.5lbs. With a 47Wh battery you are looking at getting around 10-15 minute flight times, depending of your flight characteristics. These props, with the DT750 motors, will give you good disk loading and efficiency. And, with HobbyKing stocking them in 4 packs, you will have plenty of spares in case you break one.
- Front Wheel Steering Arm & Mount Set 40mm (5sets)
- Corona DS329MG Digital Metal Gear Servo 3.8kg / 32g / 0.11s
Pivot (Nose Wheel Steering Assembly)
I had planned on having a complex design laser cut out that included ball bearings and all sorts of fancy hardware. I became frustrated not having the stuff to build my designed tiltrotor, so I hacked together a cheaper alternative. The pivot is actually a cheap nose wheel steering assembly hot glued to some pieces of cheap wood from Home Depot. I took a hacksaw/dremel to an old bent 4mm nose wheel assembly for the axles. I bent the ends into hooks and hot glued small 1.5in square 1/8in thick pieces of wood to the hooks for my motor mounts.
In order to get the steering arm to bite well on the axles, I dremeled a flat spot where the set screw sits, and used plenty of loctite to make sure that grub screw wouldn't budge. For a servo, I used a standard size servo zip-tied to the underside of the arm. Some double sided tape in between the servo and the arm can keep the servo from shifting around and provide padding. I used a ball link to connect the servo horn to the steering arms, but a clevis or just a bent pushrod would probably work fine. The less slop, the better. As listed, a good fast digital servo is what you probably need. The standard size may be overkill, and you can probably get away with a smaller one. However, better too big than too small, right?
Although the ESCs provide 3A linear voltage regulators, and you have two of them, with the digital servos you risk burning out one of the regulators. I ended up burning one of them out because I had been too lazy to wire up the UBEC I had laying around. You may think that if you plug 2 BECs in parallel, you'd get 6A! Not quite. The SS ESCs have 2 1.5A linear regulators in parallel. So, if you plug 2 SS ESC BECs in parallel you actually end up with 4x 1.5A regulators in parallel. Unfortunately, if you put a 6A load on the regulators, there is no guarantee that the load will distribute evenly across the 4 regulators. One of the 1.5A regulators could get the full 6A load.
What happens when you burn out the regulator on the ESC? The ESC stops working and you lose a motor. On a tiltrotor, there is no chance of recovering after losing a motor. Better not to risk it and just throw a UBEC in there.
- HobbyKing Multi-Rotor Control Board V2.1 (Atmega168PA)
- USBasp AVR Programming Device for ATMEL proccessors
Alternative to building one yourself
As mentioned at the beginning of the post, I designed and built my own autopilot. My next effort will be a write-up about the autopilot. However, there are cheap commercial autopilots that can handle a dual motor fixed pitch tiltrotor. I first got the idea from watching the following video, which uses an older KK controller board.
I haven't personally tried the KK board on a tiltrotor yet, but you should be able to go here for the tools and firmware for the KKMulticopter boards to make it work with a tiltrotor. I have a HobbyKing board, v2.1, sitting in my garage that I mean to compare to my own autopilot one of these days. Also, don't forget the programmer, which, at its price is a steal.