Like many other segments of the RC Hobby, after building a few kits, some people like to try designing and building their own models from scratch. Modelers have been doing this for decades with fixed wing aircraft by scratch building from plans, building their own designs or kit-bashing and modifying existing designs. When it comes to helicopters, because of the complex mechanics that are used in them, unless you own, or have access to, a machine shop, it is pretty tough to design and build your own models. Multirotors on the other hand, are more like aircraft in this respect, and do offer modelers the option of quickly and easily designing and building their own multirotor aircraft. For modelers on a tight budget, scratch building can often be a good way to get into multirotors for a minimum investment. In this installment of Multirotor Flight, we will take a look at some of the different options that are available, as well as some general rules of thumb for designing and building your very own multirotor aircraft.
Virtually all multirotors share the same common group of electrical components. These components include a flight controller board, a battery, motors, speed controllers and props. The multirotor frame basically serves as a way to mount these components together in a specific alignment, with the motors spaced far enough apart so the props can rotate freely. This basic arrangement can take the form of a Tri-copter, Quad-copter or Hex-copter platform, and each type has its specific benefits. The most common and simplest of these configurations is the Quad-copter platform, so we will focus our attention on that design for now.
The cool thing about multirotors is that you can use virtually anything to build the frame, as long as it is light enough to be carried by the thrust of the propellers. People have actually taken 4 motors and mounted them to the corners of a Fed-Ex shipping box and built a multirotor from it! Obviously this is not a very efficient design, because over 1/4 of the prop arc on each motor is blocked by the corners of the box, but it does show how simple a quad can be built.
When considering quad frames, there are basically three different types, the X-style, where all the motors are spaced equally from a central body as shown in Figure 1, a Spider style, which is shown in Figure 2, and an H-style, which is shown in Figure 3. The Spider style frame is most commonly used for FPV or aerial photography, since this design provides a way to mount a camera out ahead of the front props to eliminate them from the view of the camera. The commonality of all three of these designs is that the four motors basically form a square, with the electronics mounted somewhere in the middle.
The frame itself can be constructed from a wide variety of materials including wood, plastic, metal, fiberglass, carbon fiber or any combination of these materials. An extremely simple and inexpensive frame can be constructed by taking a wooden yard stick from a home improvement store, cutting it in half, and epoxying the two pieces together into an X shape. The motors can be attached to the ends of the arms with wood screws, and the remaining electronics can be mounted with hook and loop fastener or double sided servo tape. Other wood designs can be constructed from 3/8” to 1/2” square pine or spruce stock combined with 1/8” plywood for motor mounts and center plates. Figure 4 shows an example of this type of frame with a small plastic food storage box added to the center to house all of the electronics.
Lightweight metal is also a good material to make rugged, easy to assemble frames. Most home improvement stores carry 3 foot, 4 foot and 8 foot lengths of square aluminum tubing in several sizes, and this can be used to assemble multirotor frames. Center plates and motor mounts can be fabricated from either sheet aluminum or fiberglass G-10 or FR4 material, and everything can be held together with sheet metal screws or pop rivets. These types of frames are extremely crash resistant, and are easy to fabricate with common hand tools. Figure 5 shows an example of a typical metal quad-copter frame.
Once you have your frame materials picked out, the actual design is only limited by your imagination, but remember, for best operation you do want the motors to end up in as close to a square pattern as possible. Once you pick a frame shape, you need to decide on a frame size. For multirotors, the frame size is usually listed as the center to center distance in millimeters between two motors on opposite corners of the frame. Most commercially available flight controller boards are designed to work best with mid-size multirotors with a frame size of between 400mm and 800mm. For a first try at scratch building, something in the 500mm to 600mm size is a great place to start.
Once you decide on a frame size, the next thing you want to do is select the proper motors for your quad. In any quad design, you want to make sure that you have enough power to fly properly and efficiently. In order for any multirotor to fly well, you need to have a thrust to weight ratio of at least 2 to 1. If your machine weighs 40 ounces ready to fly, including batteries and all payloads, then you want to make sure that the combined thrust of all the motors at full throttle is at least 80 ounces. At this point you may ask, “How do I know how much thrust my motor will produce?” If you purchase good name-brand motors that have published propeller data charts, this is easy. You simply look up a motor and prop combo, and see how much thrust it produces. If you choose motors with no test data from the manufacturer, then you have to take an educated guess as to the thrust they will produce.
When picking out motors, low Kv models that spin larger props are always more efficient and quieter than higher Kv motors spinning smaller props. For the best thrust efficiency, you should use the largest prop your motors can safely run. Most mid-sized multirotors run well on 3-cell Li-Po batteries, and motors with Kv values in the 800 to 1000 RPM/volt range do very well on 3 cells. If you want to use a 4-cell Li-Po battery, then motors with a Kv value in the 650 to 800 range would work best.
Here are a couple examples of how much thrust you can get with different size motors. Running on 3 Li-Po cells, a motor with a stator size of 2212 or 2213 and a Kv value of around 950 will produce 24 ounces of thrust from a 9×4.7 prop, and would work on a quad that weighs up to 48 ounces. A 2215 to 2217 size motor with a Kv of around 950 will produce around 35 ounces of thrust from a 10×4.7 prop, and would work on a quad that weighs up to 70 ounces. It is OK to have greater than a 2 to 1 thrust to weight ratio on your model, it will just result in longer flight times due to more efficient operation, but do not take this to extremes. If you get much more than a 3 to 1 thrust to weight ratio, the extra available power can result in difficulty in tuning the PID settings or gain settings in your flight controller board due to the high amount of thrust that is available.
To calculate how much power you need for flying a multirotor, a good rule of thumb is 60 watts per pound to maintain a stable hover. This lets you know approximately how much total power you will need, and gives a starting point for picking out motors and calculating flight times. If you have a machine that weighs 3 pounds, ready to fly with batteries, then you will need approximately 180 watts of power in a hover. If you have 4 motors on your multirotor, then each one will be producing 180/4 or 45 watts of power. If you are running a 3-cell battery pack, which produces around 11.1 volts under load, then each motor will pull 45/11.1 or 4.05 amps of current.
The 60 watts per pound rule can also be used to calculate flight times. In this example, 180 watts divided by 11.1 volts gives a total current draw of 16.2 amps and this value can be used to get a rough flight time calculation based on the size of your battery pack. For the sake of this example, let’s assume that you have a 3-cell 3300mah battery pack for your model. Since a 3300mah battery can also be called a 3.3 Ah battery, a 16.2 amp current draw is equal to a discharge rate of 16.2/3.3 or 4.9C. From earlier columns you may remember that flight time can be calculated by taking 60 minutes, and dividing that by the C-rate of discharge. Using this formula, 60/4.9 is equal to 12.25 minutes. This formula assumes full discharge of the pack, which you really never want to do. You should always leave 20% of the energy in a battery pack at the end of each flight, so if you take 80% of the calculated flight time of 12.25 minutes you get 9.8 minutes of flight time from the 3300mah 3-cell Li-Po battery.
Once the motors and props have been selected, the next item to pick is the speed controller. All of the speed controllers in a multirotor should be identical to one another. They should be the same brand, the same size and have the same firmware. Using different brand speed controllers mixed together can make it very hard for the flight controller to do its job properly. When sizing speed controllers for multirotors, it is a good idea to have the speed controllers be able to handle double the full throttle current of your motors and props that you are using. If you have motors that draw 9 amps at full throttle, use 20 amp speed controllers. If you have motors that pull 14 amps at full throttle, then use 30 amp speed controllers.
There are several reasons for over-sizing the speed controllers. First and foremost is reliability. If a speed controller never sees more than half of its rated current, the chance of having an in-flight failure is practically zero. In multirotors, especially quads, the loss of power to one motor will always result in a crash, so you want to make sure that the motors stay running! Another big reason to over-size your speed controllers is cooling. In many cases, the speed controllers are mounted in the center of the frame, often under a cover or inside a radio equipment box where they get little or no air flow for cooling. If you run the speed controllers at less than 50% power at all times, they will run cool enough to be able to operate with little or no airflow across them.
Finally, the last piece of equipment you will need for your project is the flight controller board. This is where the do-it-yourself idea can be taken to extremes if desired. If a modeler is adept in soldering and building electronic circuits, then a flight controller board can also be fabricated from scratch by starting with an Arduino board, or other microprocessor development board, and adding all the sensors themselves. Going this route would only be for the truly hard-core electronics buff, sine it does require a completely different set of skills and electronic knowledge to pull off successfully. For most people, buying an off the shelf, fully assembled Flight Controller is really the only way to go.
When picking a Flight Controller, there are a lot of options available over a wide range of prices. As it is with most things, you get what you pay for, and this is especially true when it comes to flight controllers. A quick look at eBay or overseas online stores will uncover dozens of different Multi-Wii based controller board clones, some for under $20.00. Figure 6 shows a typical Multi-Wii type flight controller board that is available for about $35.00 on line. This is an area where spending a little extra money will pay for itself in the long run. Many of the cheaper boards do not come with software pre-loaded, and require some knowledge of computer coding in order to get them to operate correctly. Many have very basic computer interfaces, or rely on interfaces from open source projects to operate and program the controller boards. Some of the higher priced flight controllers, such as the DJI Naza series, come with all the software pre-loaded and feature a polished, easy to use computer interface that makes setting up the board a very simple procedure. For first time scratch builders, you should probably stick with a board that you are already experienced with that already works reliably, since that is one less thing that you will have to worry about when setting up and flying your model.
Whether you are on a tight budget, and want to build a multirotor as cheaply as possible, or have some radical new design that you want to build and see if it will actually fly, scratch building can be a very fun process, and offer a new challenge for multirotor pilots. By taking the time to figure out a few basic design rules, a successful project can be easily realized. One of the coolest things about scratch building is the sense of pride that you have at the flying field when someone asks you, “What multirotor kit is that?”, to which you can proudly reply, “It is not a kit, it is my own design!”