You’d be hard pressed to find an aircraft that wasn’t designed and tested without extensive use of simulation. Whether it’s the classic approach of using a scale model in a wind tunnel or more modern techniques such as computational fluid dynamics, a lot of testing happens before any actual hardware gets bolted together. But at some point the real deal needs to get a shakedown flight, and historically a favorite testing ground has been the massive dry lake beds in the Western United States. The weather is always clear, the ground is smooth, and there’s nobody for miles around.
Thanks to [James] and [Tyler] at Propwashed, that same classic lake bed approach to real-world testing has now been brought to the world of high performance quadcopter gear. By mounting a computer controlled thrust stand to the back of their pickup truck and driving through the El Mirage dry lake bed in the Mojave Desert, they were able to conduct realistic tests on how different propellers operate during flight. The data collected provides an interesting illustration of the inverse relationship airspeed has with generated thrust, but also shows that not all props are created equal.
The first post in the series goes over their testing set-up and overall procedure. On a tower in the truck’s bed a EFAW 2407 2500kV motor was mounted on a Series 1520 thrust stand by RCBenchmark. This stand connects to the computer and offers a scripted environment which can be used to not only control the motor but monitor variables like power consumption, RPM, and of course thrust. While there was some thought given to powering the rig from the truck’s electrical system, in the end they used Turnigy 6000mAh 4S battery packs to keep things simple.
A script was written for the thrust stand which would ramp the throttle from 0% up to 70% over 30 seconds, and then hold it at that level for 5 seconds. This script was run when the truck was at a standstill, and then repeated with the truck travelling at increasingly faster speeds up to 90 MPH. This procedure was repeated for each of the 15 props tested, and the resulting data graphed to compare how they performed.
The end result was that lower pitch props with fewer blades seemed to be the best overall performers. This isn’t a huge surprise given what the community has found through trial and error, but it’s always good to have hard data to back up anecdotal findings. There were however a few standout props which performed better at high speeds than others, which might be worth looking into if you’re really trying to push the envelope in terms of airspeed.
We don’t have to tell you that drones are all the rage. But while new commercial models are being released all the time, and new parts get released for the makers, the basic technology used in the hardware hasn’t changed in the last few years. Sure, we’ve added more sensors, increased computing power, and improved the efficiency, but the key developments come in the software: you only have to look at the latest models on the market, or the frequency of Git commits to Betaflight, Butterflight, Cleanflight, etc.
With this in mind, for a Hackaday prize entry [int-smart] is working on a quadcopter testbed for developing algorithms, specifically localization and mapping. The aim of the project is to eventually make it as easy as possible to get off the ground and start writing code, as well as to integrate mapping algorithms with Ardupilot through ROS.
The initial idea was to use a Beaglebone Blue and some cheap hobby hardware which is fairly standard for a drone of this size: 1250 kv motors and SimonK ESCs, mounted on an f450 flame wheel style frame. However, it looks like an off-the-shelf solution might be even simpler if it can be made to work with ROS. A Scanse Sweep LIDAR sensor provides point cloud data, which is then munched with some Iterative Closest Point (ICP) processing. If you like math then it’s definitely worth reading the project logs, as some of the algorithms are explained there.
Mini indoor drones have become an incredibly popular gift in the last few years since they’re both cool and inexpensive. For a while they’re great fun to fly around, until the inevitable collision with a wall, piece of furniture, or family member. Often not the most structurally sound of products, a slightly damaged quad can easily be confined to a cupboard for the rest of its life. But [Peter Sripol] has an idea for re-using the electronics from a mangled quad by building his own RC controlled paper aeroplane.
[Peter] uses the two rear motors from a mini quadcopter to provide the thrust for the aeroplane. The key is to remove the motors from the frame and mount them at 90 degrees to their original orientation so that they’re now facing forwards. This allows the drone’s gyro to remain facing upwards in its usual orientation, and keep the plane pointing forwards.
The reason this works is down to how drones yaw: because half of the motors spin the opposite direction to the other half, yaw is induced by increasing the speed of all motors spinning in one direction, mismatching the aerodynamic torques and rotating the drone. In the case of the mini quadcopter, each of the two rear motors spin in different directions. Therefore, when the paper plane begins to yaw off-centre, the flight controller increases power to the appropriate motor.
Mounting the flight controller and motors to the paper plane can either be achieved using a 3D-printed mount [Peter] created, or small piece of foam. Shown here is the foam design that mounts the propellers at wing level but the 3D printed version has then under the fuselage and flies a bit better.
We’ve seen some cheap quadcopter builds over the years, but this one takes the cake. After seeing somebody post a joke about building a quadcopter frame out of zip ties and hot glue, [IronMew] decided to try it for real. The final result is a micro quadcopter that actually flies half-way decently and seems to be fairly resistant to crash damage thanks to the flexible structure.
The first attempts at building the frame failed, as the zip ties (unsurprisingly) were too flexible and couldn’t support the weight of the motors. Eventually, [IronMew] realized that trying to replicate the traditional quadcopter frame design just wasn’t going to work. Rather than a body with arms radiating out to hold the motors, the layout he eventually came up with is essentially the reverse of a normal quadcopter frame.
Zip ties reinforced with a healthy coating of hot glue are arranged into a square, with a motor at each corner. Then four zip ties are used to support the central “pod” which holds the battery and electronics. No attempt is made to strengthen this part of the frame, and as such the heavy central pod hangs down a bit in flight. [IronMew] theorizes that this might actually be beneficial in the end, as he believes it could have a stabilizing effect when it comes time to record FPV video.
He mentions that he’s still struggling to get the PID values setup properly in the flight computer, but in the video after the break you can see that it’s flying fairly well for a first attempt. We wouldn’t recommend you tear into a bag of zip ties when it comes time to build your first quadcopter, but it does go to show that there’s plenty of room for experimentation.
Before you smash the “Post Comment” button with the fury of Zeus himself, we’re going to go ahead and say it: if you want to build a decent quadcopter, buy a commercial frame. They are usually one of the cheaper parts of the build, they’re very light for how strong they are, and replacement parts are easily available. While you could argue the cost of PLA/ABS filament is low enough now that printing it would be cheaper than buying, you aren’t going to be able to make a better quadcopter frame on a 3D printer than what’s available on the commercial market.
Of course, [Paweł] is hardly the first person to think about printing a quad frame. But he did give his design some extra consideration to try and overcome some of the shortcomings he noticed in existing 3D printed designs. For one, rather than have four separate arms that mount to a central chassis, his design has arms that go all the way across with a thick support that goes between the motors. The central chassis is also reassuringly thick, adding to the overall stiffness of the frame.
The key here is that [Paweł] printed all the parts with 2 mm thick walls. While that naturally equates to longer print times and greater overall weight, it’s probably more than worth it to make sure the frame doesn’t snap in half the first time it touches the ground.
Beyond the printed parts, all you need to assemble this frame are about a dozen M3 nuts and bolts. Overall, between the hardware and the plastic you’re looking at a total cost of under $5 USD. In the video below [Paweł] puts the frame through its paces doing some acrobatic maneuvers, and it looks like 5 bucks well spent to us.
The ESP8266, really showcasing its all-round prowess, hosts both a web server for a HTML5 based joystick and a Websockets server so that a client, such as a phone, could interact with it over a fast, low latency connection. Once the ESP8266 receives the input, it uses interrupts to generate the corresponding PPM (Pule Position Modulation) code which the RC receiver on the quadcopter can understand. Very cool!
What really makes this realtime(ish) control viable is Websockets, a protocol that basically allows you to flexibly exchange data over an “upgraded” HTTP connection without having to lug around headers each time you communicate. If you haven’t heard of Websockets you really should look really check out this library or even watch this video to see what you can achieve.
If you ever wish you could be on your quadcopter when you fly it, you will really want to see the video showing the Dubai police department testing the Hoverbike. The Russian company Hoversurf that markets the device doesn’t provide a lot of technical details, but it looks fairly simple. It is basically a motorcycle seat along with a big quadcopter. From the videos about the device, you can deduce that the pilot can control it or you can fly it remotely. You can see one of the videos, below.
There are a few things that worry us here. Of course, the huge spinning propellers as the pilot’s knee level should give you sweaty palms. In the demo, they even show the removal of the propeller guards before the test flight but let’s be honest, those don’t look like they would keep a falling pilot out of the rotors at all anyway. When looking beyond the hype we find it curious that the demo doesn’t show many (if any) shots of the pilot making a turn. The benefit of a vehicle like this to police should be maneuverability and from what we saw the Hoversurf is still limited.
So is it real? Hard to say. The short videos mostly show vertical or horizontal flight with no maneuvering. Is it hard to turn? Is the battery life really short? One other oddity: When we first saw a letter from the US Patent Office on their site, we thought they might have some new technology. However, that letter is simply showing they registered a trademark and doesn’t reference a patent. If there is a patent we want to know what is new and novel here.
Of course, we know it’s possible to build such a machine since we saw [Colin Furze] do it with two rotors instead of four. The US Department of Defense is working on something with a company called Malloy and there are other practical examples. There are also some less practical examples. What we’re really on the lookout for is a product that works so well it will actually be used. You know, like those Segways that airport police use, right?
We hope Hoversurf can bring this to market because we definitely want one. There’s no reason to think they can’t, but we do wish there were more details forthcoming.
By using our website and services, you expressly agree to the placement of our performance, functionality and advertising cookies. Learn more