Fathers’ Day 2018

There’s no flying this week because the weather has gone all weird again. It’s been raining for the last couple of hours, then gone very windy. Never mind, though, as I’m snowed under with work again and haven’t done anything else this week.

Here’s a sneak preview of something I’ve been working on for a while:



I’m not sure whether the BBC MicroBit’s magnetometer and accelerometer can be made to stabilise the quadcopter, but it’s certainly got enough power to lift. As something you can use to teach kids aerial robotics and programming on, it might be just the thing. It’s still very much a work in progress though.


Autogyro Flies!!!!!!



The ATOM Autogyro flew again!

OK, so I had two attempts where it ran along the ground, jumped into the air, wanted to go vertical again and turned around to the left, heading back towards us before I managed to effect a landing (sort of). There was one other aborted take-off when I recognised that the head speed was too low as it was rolling left. Then after that a four minute flight. I didn’t realise until afterwards, but that is the first time it’s ever taken off from the ground. All the other flights have been hand launches.

My strategy this time was simple. Put it on the ground and wait for the wind to spin the rotors up to flying speed. Then apply throttle and run along the runway trying to make it hop into the air, then back down again. The reality was that the wind refused to blow when I wanted it and I stood there waiting for so long this morning that I clocked up 1h5m on my radio’s timer. I had several goes on and off while the wind did its thing, but I think the thrust line might be wrong based on the two near misses. The first time it ran along the runway, jumped into the air and immediately wanted to go vertical, apart from me adding a lot of down and backing the motor power off very quickly. Taking off from the ground rather than doing a hand launch meant that my fingers were already on the sticks as it lifted. This resulted in it turning around to the left and heading back towards us, but I just about managed to hit the ground softly enough for a second go. The next time the wind was favourable, the second attempt was largely the same as the first, but now I had the rotor head adjusted on the elevator trim to be much further forward. Also, it was noticeable that I could control it when the power was off, so the trick would seem to be not to use full throttle. When I had a third go, it sailed away and climbed into the air with me in control but under less power. What followed after that was a four minute flight where I nearly stuffed it into the ground on at least three occasions. It ended up with full left aileron trim and full down elevator. The height loss when you turn on the ailerons is huge, so I was adding in rudder in order to bring the nose around. I have to say that there were a lot of control inputs going in just to keep the thing in the air, but it is a very different type of aircraft to fly. What I need now is to be able to get it into the air consistently so that I can trim it to fly in a more stable manner. All in all, though, it was a good day and I’m very happy.

As for the rest of the morning, I also flew a UMX Beast for somebody else and absolutely loved it. The ailerons aren’t powerful enough as they had been turned right down, but it was very easy to fly in that trim. Loops were very nice. I only got one flight with my own RS352, though, because of all the autogyro testing. Other than that, we had a guy with a Parrot Flying wing (autonomous flight) and foam Cessna High Wing aircraft. Then another guy with a custom built F450 drone, another with a Mavic and a third with a DJI Phantom. The best aircraft belonged to the guy with the foam jet style aircraft, Eddie the eagle and a foam Vampire ducted fan. I love Vampires. You can just see it in the background of the image above, along with Eddie, who also looked great in the air.

Well, I’m happy. I flew the autogyro and it’s back in one piece, so I can make a few trimming changes before next time. Just for the record, the only difference between this and the last flight was the addition of the extra shim under the blades to give a more negative angle (and less lift). I’ll look at the thrust angle next and do a re-check on the balance point (hang angle) just to make sure. I’ve also got some video footage of the flight, so that might get posted once I’ve done some editing.

The Autogyro Wind Blew Too Late



I took the autogyro out this morning, but there was absolutely no wind first thing, so I ended up not being able to fly it. If the breath of wind there was at the end of the session had been there at the beginning, then it would have been perfect. There wasn’t enough wind to spin the rotors until late on and rushing a quick test flight when I was pushed for time didn’t seem like a sensible idea, so the autogyro went home in one piece.

I spent all yesterday going through the autogyro’s systems to make sure everything was spot on. One difference between this setup and last time is the addition of the extra shims under the blade roots, so the blades are at a more negative angle to the head plate rotation plane. Also, while reading the original RCM&E article, I noticed that the designer put a 1 degree left side thrust angle on the motor using washers. I didn’t have this two weeks ago, but I can’t remember if the successful flights last year had it in or whether I decided to remove it? I added the left side thrust this time just as a precaution, as it won’t make a huge difference. The balance, measured using the “hang angle” technique, will make a big difference, as will the head plate angle (elevator), so this is what I spent most of the time checking. I actually had full down elevator trim on the radio, so I adjusted the mechanical linkages and moved this back to zero on the radio to give myself some margin for adjustment.

My plan for flying was simple. I was going to sit the aircraft on the runway and pull the elevator fully back to tilt the head back and let the wind spin the rotors up. Then I was going to release the elevator and try moving forwards slowly with the rotors spinning to try and “hop” like they suggest when trimming an autogyro. The idea is to fine tune the head angle so you can add a little speed and slight back tilt for it to gracefully take to the air. It was a disappointment not to be able to fly it.



I managed three flights with the RS352, which sustained some damage on the journey to the field. The flight box moved and took a bite out of the right aileron, but nothing that a smear of PVA won’t fix. It flew cuban eights well in the completely still, hot, air. We don’t get these conditions very often and it was very busy at the field this morning. I saw the guy whose Hurricane I flew ages ago, with a foam drone and a Multiplex Easy Star, which I seemed to keep finding in the same bit of air as me. We also had three or four hand launch gliders, a pure bungee launch one, the high wing Cessna type trainer from the other week and a few drones, but they like to keep to the limits of the field and refuse to talk to us gyronuts.

There’s not much else to say, except that it was so hot I think I’m starting to melt. This afternoon I need to make a brushed motor controller for a micro drone, which means I’m soldering in the heat. If it works, then I’ve got an interesting aerial robotics test bed to play with.

Oh, well, I’m disappointed not to fly the autogyro, but that was the sensible option and it gets to fly another day. Let’s hope next week’s weather is better.

Autogyro Repair, the Sequel



Well, true to my word, I repaired the autogyro back to flight ready status in a week. Unfortunately, the weather conspired against me. We had violent electrical storms the night before and forecasts of storms the next morning. Although it did rain quite hard, I’m not sure that the storms every really developed, but one place I don’t want to be is standing in the middle of a field holding a radio transmitter when there’s electrical activity overhead.

As for the repair, I stuck the spruce longerons back together again.



Then the fuselage sides got glued back together where the LiPo had moved forwards in the crash and separated them. You can just see in the picture below that I’ve also had to add a plywood plate where the spruce longerons meet the red fuselage. It’s just visible on the right of the picture and, if you compare to the area with the sellotape in the picture above, you should be able to make out the new 2mm plywood strengthening plates.



I’m a bit worried about making this area too strong, as it will eventually just break at the point where the spruce meets the fuselage. After that, I’ll be trying to dig the longerons out of the fuselage to replace them as they extend right up to the rear former.

Apart from sticking Pete the pilot’s head back on, the only thing left was the motor.



I probably could have bought a replacement shaft for the motor, but I chose to make my own out of a piece of 3.2mm rod.



You can see my new motor shaft at the bottom of the picture. It’s not quite perfect like the machined original above, but I made it and I like it better. Actually, it made me wonder what the commercial motor shafts are made from, as the original was obviously designed to fail rather than bend the motor bell in an impact. That didn’t exactly work and a shaft costs £3 while a brand new motor costs £10. My solid steel shaft is a lot stronger than the original, so one more hit on the ground like before and it’s going to be a new motor. I can live with that.



So there we are, one fixed autogryo and I’ve now got a whole week to properly test that everything works before committing it to the air again. Next time out I’m going to increase the shim under the rear of the blades to 0.8mm, which should reduce the lift and help with the launch.

That’s it for this week, but I did video the autogyro’s reconstruction last night, so I might post that later depending on how it came out. Let’s hope the conditions next week are more conducive to autogyroing.


I flew the autogyro again today for the first time since its crash in October. It didn’t go so well.





These are the before shots. I actually liked the previous all blue colour scheme better, but this one looks more like the RCM&E review model in its red, white and blue.

I was a bit worried that there wasn’t enough wind to turn the rotors for a hand launch, but there was plenty to get up to flying speed. The problem stemmed from the launch, as I’ve damaged my right shoulder and had to hand launch left handed for the first time. Looking at the launch video, it’s not so much a left handed problem as a consciously not throwing it forward too fast problem. The YouTube clips I’ve seen of autogyro hand launches show a gentle push with the arm at 45 degrees tilting the rotor disk to get it to spin up, followed by a slow transition from 45 degrees to level with the horizon and a push forwards. I was worried that my previous launches were more javelin style aircraft launches, in other words, get as much forward speed as possible. This isn’t necessary with an autogyro, as the speed is in the rotor disk. You do have to throw it flat and with some forward speed, which I didn’t do well enough.


This is a fraction of a second after launch. This isn’t going to end well.

So, a fraction of a second after launch, I had an autogyro prop hanging in a vertical manoeuvre. All I could really do was try and stabilise it laterally, allow it to climb as much as possible and try to flip it around in a stall turn. Another five feet and I would have done it, but it ended with a fairly soft impact directly on the nose.

Spy satellite pictures show more of the descent back to Earth:

vlcsnap-2018-05-20-14h24m49s325.png vlcsnap-2018-05-20-14h24m58s307.pngvlcsnap-2018-05-20-14h25m09s555.png vlcsnap-2018-05-20-14h25m18s156.png vlcsnap-2018-05-20-14h25m28s024.png vlcsnap-2018-05-20-14h25m43s694.png vlcsnap-2018-05-20-14h25m52s060.png vlcsnap-2018-05-20-14h26m03s337.png vlcsnap-2018-05-20-14h26m16s930.png


OK, so I had a RunCam filming the hand launch and the images above are from a GoPro Hero 4 Session dug into the grass pointing skywards to the left. I wish I had captured more of the short flight, as you can see what I mean by “stabilisation problems”. I’m not just prop hanging, the head is pointing ground-wards at an alarming angle.


The damage isn’t too bad considering. Like I said earlier, I had it stabilised by the point of impact, I just needed more height to fly. It’s snapped the motor shaft, split the fuselage where the LiPo moved forwards and snapped the two spruce tail sticks. I am determined to fix this quickly and get it back in the air next week.

This is the most difficult damage to fix, as it’s right up at the point where the spruce longeron meets the fuselage. I’ll have to have a think about this.



The split fuselage is easy to fix:


Oh, and I nearly forgot, Pete the Pilot isn’t feeling too well:



It must have been that negative g manoeuvre that I pulled.

So, that was the autogyro. I also had three flights with the RS352. There was one other guy with a high wing foam aircraft practising his touch and goes, the lady with the drones and a couple of guys with some really cool HLG gliders.

OK, where did I put that book on how to fix autogyros for the stupid and impatient?

Stuck Indoors



The roads are closed due to an event today, so I can’t get out to fly. I thought the weather forecast said that it was going to rain anyway, but it never materialised and it’s now hot and sunny again.

Never mind, I’ve spent the time filming a micro quadcopter build and setup with my new Taranis QX7S. The blue thing has just flown its first aircraft, although that was only the micro quad you can see in the picture and then only about 2 inches above the worktop to prove that it was working. The plan is to add the FPV and give it a proper go with the goggles later this evening.

The specifications are as follows:

4×8.5mm brushed motors

500mAh 25C LiPo

FrSky XM 1 gramme SBUS receiver

F3 EVO Brushed Flight Controller

CleanFlight Flight Controller Software

The FrSky bound to the XM receiver without any problems, but it was a real pain to get the auxiliary channels working for the arm and flight mode functions. You have to add a mix for the two switches and assign them to channels 5 and 6. The other thing to remember is to set the “Brushed” mode for the flight controller. This used to be something you had to set on the command line with “set motor_pwm_rate=32000”, but it’s now on a drop down option on the menu. With the receiver type set to “SBUS”, the flight controller should talk to the radio without any problems. NOTE: I’m using the FrSky LBT (Listen before talk) European firmware, which was pre-flashed in both the transmitter and receiver module at the factory.

Well, that’s it for now. I’m going to give the quadcopter a go indoors later, so we’ll have to see how that turns out. Hopefully normal flying will be resumed again next week.

Bank Holiday Flying

The weather is perfect this bank holiday weekend. It’s 26 degrees in the Sun with no wind. In fact, I was expecting so little wind that I didn’t bother with the autogyro as I didn’t think there was enough to spin the rotors and allow me to do the hand launch trick. It turned out that wind is the one thing you can rely on in this country, as it was a touch more turbulent than the flat calm predicted.

It was just me and my friend with a big thermal glider first thing. We watched a red kite soaring around effortlessly above us and listened for little skylarks calling to each other and doing their characteristic death dive down in to the long grass. After that a lot of drones turned up and it was chaos. Before that, though, there was a new guy with a small foam (<1 metre) thermal soarer type of aircraft. It was one of the cheaper ones as the TX had no rates and we couldn’t turn down the elevator as much as it needed. Being this small, it was more agile than he wanted as a beginner, so he went through a sequence of launching, climbing away, turning the motor off, flying, dropping a wing, panicking not knowing what to do and then hitting the ground quite softly. The front was glued and taped back together when required and off he went again. Possibly I should have flown it for him, but the nose was a bit of a wreck and it looked a handful. His persistence and slow improvement was something to watch as he really wanted to be able to do it for himself. His friend had a drone, which seemed to know how to fly all on its own. This was an F450 with a big black box of electronics that kept it airborne. After that we had two kids with 250 sized FPV drones, an Inspire, the usual guy with his wife and a Mavic and then at the end the two guys with a collection of wings, high wing soarers and helicopter. In addition to this my friend on the bike came over with an aeroplane. I haven’t seen him for ages and he brought a very old high wing white foam cabin model on 27MHz. I think it was a Hobby Zone Mini Super CUB, as the wings were held on with bands and it had been in the garden for 3 years. We didn’t have enough bands, so taped the wings on with white tape and flew it anyway. I absolutely loved it. The only thing was the throttle, which sprung back to the centre position (zero), so all the throttle control was from the centre position to full forwards. Then, every time you let go, the throttle would spring back to zero. Despite this, it felt like I was flying an old fashioned balsa and tissue plane, even if it was foam. I ambled around the sky for a good 10 minutes before making an absolutely perfect landing, running along the wheels to a stop right in front of me.

In addition to the Cub, I got three (maybe four?) flights with the RS352. I can do a really good right spin now, but left spins don’t work nearly as well. In the course of flying, I nearly took out a model helicopter and one of the drones a couple of times. We really need to do something about this as the drones are invisible and have no awareness at all of where they are and who is around them. Also, we nearly got hit by a high wing model as it was being launched. The guy threw it and the model just dropped the right wing a bit and very slowly did a 180 degree turn back towards us, skimmed our heads as we ducked and crashed into the ground. We also had a drone nearly hit someone flying from a chair as the pilot flying FPV didn’t know where he was. Apart from that I got some really good flying in and really loved the opportunity to fly the old fashioned cub.

Finally, another friend turned up just as I was leaving with a brand new foam Corsair. He hadn’t flown it yet and was obviously nervous, but this was the same type as I flew for another guy about a year ago. As I said to him, with the AS3X technology, anybody could fly a warbird and this one was an absolute dream to fly. I saw him with it in the air as I was on my way home, so it went up OK.

That’s all for this week, sorry there are no pictures, but I forgot to take any with all the stuff that was going on. I’m on an enforced rest next week, but might have some interesting news to tell.

Canada Downtime


Grid Drones

I’ve only just got back from Canada after attending the CHI2018 conference, so I really needed to do nothing for a bit this week. The weather was quite bad anyway, so it wasn’t much of an opportunity missed. It’s been a very grey morning, cold, with more wind than they forecast and light drizzle while we wait for the heavier rain to arrive in the afternoon.

Anyway, while I was at CHI I got to see a drone demo which was interesting. I had a chat with the guys while they were setting up and their drones are about 60g with a square frame comprised of carbon fibre cross pieces. They use very thin carbon strips of about 0.5mm thickness and about 5mm wide, but end on. The six sides of the cube have a carbon cross in them so that you are looking at the 0.5mm face and the 5mm depth makes it strong if you push on the side. And that’s exactly what they got a little kid to do. They put him inside the net and let him move the drones up and down and around, which is what they’re designed to do. They are based on the 100mm brushed motor type drone with 8.5mm motors and 55mm props, so they’re not dangerous. Added to that, there is the carbon cube frame, so you can’t put your fingers into the props very easily. The individual drones are position tracked, using the infra-red emitters you can see at the back of the picture (red circles on tripods). This means that they stay in place and adjust up and down to make a surface as you move them up and down. It was all very impressive, so now I’m left wondering whether I can make something like a drone turtle for national robotics week in July? I would like to make something that the kids can program and see move around a path that they can set. I’m off to have a play with some microcontrollers now.

A Grey Day with Clouds Brewing


My Atom autogyro modelled in Blender (work in progress)

The weather didn’t look good this week, so I had decided not to take the autogyro out. It was just that it was very changeable throughout the whole morning, with strong winds, then calm, but always with lots of threatening dark clouds overhead and occasional sunshine peaking through.

When I arrived, I bumped into a guy with a drone on his way home. After that another guy turned up with a Phantom, then the lady with the custom drones, a couple with a DJI Spark (selfie copter) and another Phantom. I had the only fixed wing of the day in my RS352. However, another guy came over on a bike and, after chatting for a while, he said that he was definitely going to bring his aircraft over one Sunday morning to learn. That’s a good result.

Anyway, I had 4 flights with my RS352. My flying was really erratic this morning. I think I must be getting out of practice. I managed some good landings in the blustery conditions, though. Although I didn’t take the autogyro, I have been slowly building a virtual 3D model which I can put into my simulator. That’s the image at the top of the page this week, but it’s still very much a work in progress as I’m not very good at 3D modelling. The only outdoor photos I took this week were a set of selfies of me and the RS352 with the clouds in the background, but I’m not going to inflict them on anybody. They didn’t come out very good.

That’s it for this week, but I’m on an enforced break next week due to being out of the country, so the next flying is in two weeks time.

My Autogyro is Flight Ready



And there he is in all his autorotating readiness. I’ve finally finished repairing my Atom Autogyro after only 6 months and it’s now ready to fly again. Except for the fact that the weather isn’t playing nicely. We’ve had heavy rain earlier in the morning and then that sort of light misty almost not there rain that soaks into everything.

Oh, well, that gives me more time to check out all the autogyro’s systems to make sure everything is working perfectly.


OK, that’s quite a lot of autogyro pictures, but I’m rather pleased with the results. I can’t wait to fly it now.

Talking about flying, now that I’ve got it all back together again, I’ve been wondering how the propeller managed to shear the blade off in fight the last time.


The pictures above show the full forward, centre and full back positions, as near to side on with the propeller vertical and the blade in the directly forward position as I could manage it. By my measurements, even in the full forward position, the blade has to bend down by 13cm in order to hit the prop. The dimensions of the blade and position of the strike just about work, if the first contact came with the prop about 45 degrees on the right side of vertical. There’s what looks like an initial hit on the leading edge of the rotor blade, then the whole thing has sheared off about 2.5cm inwards, so it’s entirely possible that there were two contact points on successive rotations. I’ll never know for sure, but I’m now going to look further into how much the rotor needs to tilt forwards. I think autogyros generally need to keep the rotors in the back position and never forwards of the horizontal, but that’s something I can research during the week.

Let’s hope next week brings some good autogyro weather.