Month: November 2016

Cold, Grey and Windy


The weather wasn’t exactly great for flying this weekend, but I managed to get in four flights with the RS352. I was joined by a guy with his girlfriend who flew a small high wing electric glider once, before deciding to give up and go home. I don’t know what it was, but it had a green wing and white fuselage. At this point something spooked the deer and they all came over to hang around with us. The picture really doesn’t do justice to how close they were. Then a little bit later I saw an electric bike coming across and its owner produced a tiny 5g helicopter out of his pocket. I had assumed he had the Taylorcraft on the back of the bike, but the rear carrier was empty. Obviously it wouldn’t fly in the windy conditions, but he told me he had intended to bring his big electric glider, but it was too much of an effort to get it on the bike in the now rather marginal weather conditions. He gave up and went home to the warm after a bit, so I was left on my own with the deer again, plus a few tourists.

In terms of the flying, there was some strong turbulence and it wasn’t a day for doing much other than just keeping it in the sky. I did turn up the elevator movement for flight 3, going from 60% to 70% on the ATV as I was trying to get it to harrier. I can’t say I noticed much difference as the weather was really going down hill at this point. I was even using clear glasses for some of the flights, it was that bad (I almost always fly wearing sunglasses). My main concern was the jittering that I’ve been getting on the control surfaces sometimes. On switching the aircraft on for the third flight, the right aileron went all the way up and stuck in position. I switched off and on again and everything was fine. I couldn’t find any fault. Then, after the flight, I put the transmitter on the floor to switch the aircraft off and the jittering started again. Same problem with the transmitter on the floor when switching on for the next flight. It looks like having the transmitter on the wet grass (on a plastic sheet obviously) is affecting the signal. This is not something I’ve seen before, so it needs a bit more investigating. If it persists, I might either swap the radio or the four servos for a different brand as I’ve never been completely happy with the combination of the Futaba R617FS and Hitec HS65MG servos. Either component on its own I have no problems with, they’re both rock solid quality pieces of kit, it’s just the combination of the two in this installation that’s been giving me problems.

That’s not the last piece of flying I’m doing this week, as we’re running another Drone Masterclass on Wednesday evening. I’ve been trying to 3D print frames all week and failing.


Eventually, out of sheer frustration, I’ve decided to take two of the partly complete frames and glue them together to make something that works as a 100mm sized quadcopter. This is intended to fly with my EVO F3 Brushed flight controller to make an FPV micro quad. The only problem I had was working out how to add the motors, given that the 3D print kept failing before it got to that point. I was going to use plastic tubing, but, when I was thinking about how to hold the motor mount in the centre of the prop guard, it struck me that I could use the plastic from an old propeller. I’ve got loads of these green plastic propellers that I’ve broken and was looking for something useful to do with them. It’s a bit of an experiment, but I’ll have to give it a try and see how it works out. This might just be a genius piece of recycling?


Hello Angus

Angus is the first big storm of the season, and, although it’s almost blown itself out overnight, it’s still wet and very windy this morning. In other words, I’ve been building stuff.

First, there was the 100mm quadcopter X-Frame which I 3D printed at work on Friday. This took the entire day to print and then I discovered that the model was completely welded to the raft. I shouldn’t have printed it on a raft, although the idea is that it allows the model to be removed from the 3D printer more easily. The problem here was that the raft didn’t separate from the model and I spent most of Saturday morning hacking all the excess plastic off with my Permagrit file. We need this frame for an event we’re running at the end of November, so it needed to be done.

That then got me to thinking about whether I could build my own design micro quadcopter frame and build it in less time than this one took to 3D print. So, on Saturday afternoon, I built my own one from plasticard and white foam.

Basically, it’s just black plastic stuck onto white foam to give it rigidity. The way I made it was to make two straight arms with black plastic top and bottom. Then I cut a hole through the white foam in the middle of one and pushed the other through to make an “X”. You can see where the plastic overlaps in the picture. With the benefit of hindsight, it might have been better to cut a top and bottom “X” out of the plastic and make the top and bottom a single piece? That would have allowed me to add some more material in the centre where the flight controller will sit, but I quite like it the way it is.

The interesting thing is that this took less time to build than the 3D printed model took to print and what’s more, it’s also lighter.

As you can see, my frame is 3.8g lighter. I just can’t figure out how to make the blade guards at the moment. You can also see that it’s slightly bigger than the blue frame, which is to accommodate the larger and more efficient 75mm props. I got the flight controller and radio working last Sunday evening, which was quite an impressive feat. It took me quite a while to figure out how to bind my Futaba Field Force 8 with its Frsky DFT module. You first have to press the button on the back of the module while switching on the transmitter. This puts it into binding mode, which is confirmed by a constant beep, beep, beep. Then you power on the QX908 receiver while holding down its bind button, keeping it pressed until the green light goes solid. Disconnect the power from the receiver, then turn off the transmitter, power everything back up normally and they should be bound to each other. I was then able to see my control inputs in the Cleanflight application on the computer when the flight controller was attached via USB. Once I have a quadcopter that’s flying with the FPV working again I’ll do a full build article. At this point I discovered that the leads from the motors wouldn’t reach the flight controller and decided that adding plugs and sockets as we do for the HubSan assembly kits would be a good idea while I’m experimenting. I don’t have any left, so I’m going to Maplin tomorrow. This only adds less than 1g in weight, so it’s not a huge penalty. I might rethink the motor attachments, though, as I’m not completely happy with just pushing them into foam. That’s where the fun is in making your own quadcopter frame.

I still have the preparation to do for the “Drones Masterclass” that we’re running in a little over a week, so I’m probably not going to get much more building done today.


Remembrance Sunday 2016

It’s Remembrance Sunday this week, so what we usually do is to put an aircraft up and wait until we hear the cannon. Not having a Spitfire or Hurricane (or Tempest), I had to resort to flying a Taylorcraft.


The weather was perfect, with lots of sun, no wind and not at all cold. I had one flight with my RS352, then the Taylorcraft, followed by a Hobby King Bixler (EasyStar clone). The Bixler had an issue with its flaps not being connected to anything, causing them to flap around on their own. After the owner had flown it, he asked me to have a fly to see if I thought it was behaving oddly. Not surprisingly, the free flaps were causing it to fly in a rather erratic fashion. It was flyable, but you had to constantly correct it. Also, he didn’t tell me until afterwards, but he had the rudder trim and aileron trim on opposite sticks. I had been feeding in right aileron during the flight, which was actually right rudder. I can honestly say that I’ve never seen that type of setup before and can’t think why on earth you would do it. He said that the idea was to be able to trim it without taking your hands off the sticks, but it’s just confusing. I also did a dive test to determine the centre of gravity, which turned out to be round about neutral. It didn’t quite tuck under in the dive, but it sure wasn’t stable, so we suggested adding nose ballast until he’s comfortable with the flying characteristics. Thinking back, I did fly one of these a few months ago which was very tail heavy and was a real handful to fly. It must just be down to the kits. As for the flaps, I suggested he put the hinges in and glue them so they can’t move. There are cutouts and horn positions if you want to add flaps, but apparently no instructions or obvious way of fixing them if you don’t.

After that I had three more flights with the RS352. In between all of this, there was a guy with a drone filming a keep fit group. He was interesting to talk to as he’s just got his permission for aerial work. Talking to him, he said it was really easy to get and just required a lot of paperwork and a flight test. Near the end of the morning a family turned up with a white Piper Cub (see picture). The older kid was actually quite good at flying it, but it’s the little one launching that got me. The picture shows him having just thrown it into the sky.

We also had the EFlite advance and, just as I was leaving, another guy turned up with a flying wing. I only saw it flying from a distance, but, given the small holdall it came out of and the FPV kit, it may well have been a Bormatec Ninox?

I’m back to building micro drones now, when I really want to be finishing the AutoGyro. You can see the parts I’ve just had delivered above. I’ve got two sets of F3 EVO Brushed flight controllers and the generic FrSky receiver that the Eachine QX90 drone uses. I’ve also got four sets of our modified HubSan quads with 3D printed frames repaired and ready. This is for the “Drones Masterclass” that we’re running at the end of the month. The parts above are intended for an FPV micro drone that we’re going to try running a 3D reconstruction from.

Here’s an example I captured of the ATOM autogyro:

Here’s a link in case the 123d Catch embedding doesn’t work: [ATOM Autogyro]

Hopefully there’s going to be more on this project later.

Not Flying Again

It’s a bright, sunny, crisp Autumn morning today. There is also a very strong wind blowing, but I probably would have gone flying if I wasn’t busy with other things today. The only flying this week is going to be the test with the micro FPV quadcopter in the garden and around the house yesterday.

I’ve switched to a different frame from the one I crashed and broke last week. It wasn’t badly damaged, so I’ve stuck it back together with cyano, but I think I prefer the “X” frame with the HubSan prop guard attached to the motors.



As you can see, the flying weight is 55.8 grammes, but 12g of that is the white 3D printed PLA frame, which can be lightened a bit more. The battery, which is stuck underneath, weighs about the same as the frame, while the black plastic prop guard which came out of the HubSan box weighs 3.5g. Finally, the FPV camera is 4.5g, so the rest is made up by the HubSan flight controller and four motors and props. The camera is only held on by being pushed into the sticky foam tape that’s on top of the flight controller, but it’s a system that has worked really well. The additional wiring, which we added for the Royal Institution project so that the students could construct their own quadcopters, only adds about 1g extra. We’re going to be running another “Drone Masterclass” at the end of the month, so I’ve been busy checking out all the equipment. As the next group is 16-17 years old, we’ve got to add a bit more content, so I’ve been looking into the feasibility of taking pictures with a micro quadcopter and then using that to do a 3D reconstruction. My initial thought was to arrange a load of empty cardboard boxes on the floor to simulate buildings and try to reconstruct it on the computer. This might not be feasible, so I’ve got a bit more experimentation to do.

In the meantime, I’ve been practising with FPV goggles in the garden. I had a go inside to start with, but figured that the extra space and soft grass of the garden might be a better idea. I’m going to have to do something about the goggles, though, as they just don’t work for me. I’m using the Fat Shark Predator V2, but most of the time I can only see the left eye image. My first thought is to make an extension lead for the battery to take the weight off of the goggles. They don’t stay on my head at all and the angle that the screen makes to your eye is critical to seeing a sharp image rather than a blurred one. I might also try out some of the corrective lenses as there is a big difference between the vision in my left eye and right eye.

Despite all these problems, though, I was starting to get the hang of flying FPV. The impression of speed is huge because of the camera view and there is a lot more space around the aircraft than you think. Just don’t fly backwards. When I did try it inside later on that evening, I just about managed to fly forwards, turn left through a doorway, left again, through the next room, left, across the landing and left again to return to where I started. Looking at the gap I managed to get through to do this, it’s amazing that I had that much control with a HubSan flight controller. I’ve got a couple of F3 EVO controllers which run Cleanflight on order which should arrive later this week. My aim is to film my own video where the quadcopter flies through the house at amazing speeds, just missing everything. I’m going to try some filming later today as I’ve figured out how to record from the FPV camera onto the computer. Using the first person view, along with third person RunCam and camera phone videos, I should be able to put together something good. Well, that’s the plan anyway.