flying

Wet Play Day

It rained all yesterday. It’s absolutely tipped it down from about 6am this morning. And the weather forecast is for the rain to be cleared by 50 mph winds. It’s not a day for going out, so I’m stuck inside experimenting with the PID settings and FrSky Taranis programming on my brushed micro quadcopter.

 

I did make one flying video this week – well, actually the edit was last Sunday evening, while the flying shots were from the end of August. This was the 3D printed butterfly with the mirror finish that I made a few weeks ago. With the red and blue lights, I was obviously going to fly it in the dark, so here is the flying shot with the lights on and the shadow on the ground, together with the view you get flying it in the dark. It’s really great up until the point when the battery starts to go, the lights flash on and off and the strobe effect means you can’t see anything. I need to make some more “in the dark” videos as you can get some great effects, but this one needs a bit more editing before I upload it.

Anyway, this week marks the end of my current cycle of work, before I draw a line under everything and start again from fresh in October. I’ve done far too much work in August and September, so I’m going to resolve to work on my own projects more from now on. There’s just one last software release to get out for Thursday and then I’m free. In fact two hours after that I’m having a meeting with a company about drones, so who knows where that might lead? OK, let’s do this software development on a Sunday thing one last time.

Advertisements

Road Closures

The roads are closed this weekend because of a cycling event, so there’s no flying today. The weather didn’t look too bad either, but then it got very windy later on.

Anyway, I’ve been playing around with my brushed microquad indoors this morning. I’ve finally got around to doing some testing with tuning the PID values. When I first flew this and compared it to the HubSan X4s we fly, the X4 was about 10 times easier to fly, despite the more advanced quad using an Evo 3 flight controller and running Cleanflight. What you’re supposed to do is to set I and D to zero, then increase the P until it starts to oscillate. I actually left the I and D on their default settings, and increased P from 80 up to 240 before I got the tell-tale oscillations that the PID controller was overshooting because I had an over-damped flying machine. That was much more than I anticipated, so it shows you how soft the default setup was. I’m still not there yet as my guess is that the P value wants to be around 200, then I need to set about finding the best I and D, which I find a lot harder to do. The P is the correcting force directly proportional to the error, while the I is the integral and the D is the derivative. It’s mainly the P that gives you the locked in direct feel that you want, but you can get the P a lot higher if the I and D are correct. If you want to see what the oscillation looks like, I’ve tried to video it below, but the camera doesn’t capture it particularly well:

The oscillation is most noticeable in the first few seconds of the flight. Note that the video is slowed down to 40% of normal speed.

That’s the non-scientific approach to PID tuning, which started me thinking about how to calculate a good starting point mathematically. One of the problems is that different flight controller software scale the PID values differently, so they’re not directly compatible with each other. The values aren’t dimensionless, but it’s very difficult to tell how they’re scaled as it’s often down to 10 bit normalised numbers coming back from the respective sensor chips. Also, sometimes they think in degrees and other times in radians. Here they’re big numbers (i.e. 80 to 240), while on my flight simulator I used radians, so they were much smaller. Either way, you should be able to calculate a rough ballpark figure to start from, which would help a lot. It’s just quite difficult to achieve in practice and I’m still thinking about it. In fact, autotune would make a lot more sense, where the flight controller detects how it’s oscillating and adjusts the values itself.

Well, that’s my thought for the week and there are still so many things I haven’t done this weekend. Like fix the head on my autogyro ready for when I can fly again.

Watching the Clouds Go By

There was a 20mph wind today, so no flying this week. It was looking quite overcast and ominous too, until the wind had blown all the clouds away and left us in bright sunshine.

IMG_20180909_173331.jpg

 

I’ve been flying the micro quadcopter indoors with my Taranis QX7S this week. First problem: “why is nothing working?”. I forgot it was bound to the transmitter at work for the robotics week event. I spent half an hour crawling around on the floor clicking switches and going through the programming until I realised. However, I have finally managed to figure out how to adjust the stick length on the transmitter. They were just (VERY) tight, so I ended up gouging stick shaped bits out of my fingers trying to get the throttle stick to budge at all. The right stick wasn’t so bad, which is how I worked out that the top just unscrews and then the bottom can be screwed up and down to tighten against the top in the desired position. Exactly how my Futaba sticks work.

I’ve also made another butterfly this week. I left the lights on this one, though, so I made a video of it flying in the dark. I’ll put it online once I have a chance to edit it a bit.

IMG_20180907_134857_crop.jpg

 

That’s all I’ve managed to get up to this week as work is taking over again. I was supposed to be having next week off, but it’s been cancelled. I was going to write the flight controller for the micro:bit drone, but that will have to wait until another time.

 

It’s Rained All Day

When I first got up this morning, the weather didn’t look too bad, but the forecast was for the wind to pick up and bring in persistent heavy rain. Sure enough, by 11am the heavy stuff hit and it’s been on and off like that all day. I wasn’t ever planning to go flying this week because the weather forecast looked so bad. I’ve got a lot of work to complete for a presentation next week and a butterfly quadcopter to finish building.

 

I was hoping for some Sun this morning, as that would have shown up the mirror finish on the quadcopter frame much better than the artificial light you can see in the photos above. The camera is picking up some strange effects because of the light and the mirror surface on the PLA. I’ve been wanting to print another quadcopter with a mirror finish for a while, and this one was done on an Ultimaker 3 printer by cleaning the bed thoroughly and printing directly onto the glass bed, which is heated to 60 Celsius. I’ve also printed with a brim on the model to help with the adhesion. I’m very happy with the results as you can see the mirror finish on the top surface with all the filament runs visible. It’s almost like a white carbon fibre the way it looks.

OK, I’ve got to get back to work now as I’ve still got a lot of code to write today, but here’s hoping that next week brings better weather.

 

Aborted Flying

I got to the flying field this morning, put the aeroplane on the grass and the wind immediately flipped it upside down. We used to have a saying that when the leaves were blowing across the field then it was too windy to fly. I thought it was a bit windy when I left the house, but by the time I had got to the field it was almost gale force. The forecast was for 11-13 mph, but this felt worse out in the open. It was trying to rip the aircraft from my hand before I even put it down and all my bits and pieces were blowing about, so I decided common sense was the order of the day and went home.

Weirdly enough, there was a guy in the middle of the field when I arrived, but I didn’t recognise him and couldn’t tell what he was (trying to?) fly. I think it was probably a drone as you can usually see a fixed wing aircraft from that distance, so at least the computer could fly it for him. Flying my 450g RS352 in that sort of wind isn’t a particularly good idea, but it will fly. That’s the thing about drones, they can fly in higher wind speeds because of the auto stabilisation, but then the fixed wing response should be to just go bigger and heavier. “The right aircraft for the right conditions” is the correct approach, so, with the right aircraft fleet, you can fly every day (except rain). I think I need to expand my fleet. I’ve been looking at big electric gliders for a while, but don’t have the space for one.

That’s all that’s happening this week. I’ve got a HubSan to convert into a butterfly, an autogyro to fix and a micro:bit quadcopter to write a flight controller for, but work has been taking over again. Once August is out of the way, I should have some spare time on my hands.

Can You Believe It’s Raining?

The MeteoGroup forecast promised heavy rain from early on this morning, but it didn’t materialise until around 12 o’clock. Before that, though, we had 30 seconds of rain, then it stopped, then another 30 seconds of rain, in a repeating pattern throughout the morning until it finally became heavy. It was windy too, so it wasn’t really a day for flying outside. All I’ve managed so far is circuits of the living room with the modified HubSan X4 to cycle some of the LiPos from our drone workshop the other week. It’s all very well having lots of 1C 500mAh LiPos, but somebody has to take care of them between workshops. What I used to do was to charge them, fly them, then charge them back up with the USB charger, periodically taking them off charge to check the voltage until they reached a stable 3.81 volts storage level. All this takes time (looking at the stack of 10 batteries in front of me), so I’m thinking about using the Overlander RC6-VSP charger on the storage mode. I’ll have to make up a special lead and, obviously, there’s no balance cable as they’re 1C, so I’m not sure whether the charger will let me do this. Anyway, the original method means I get lots of flying practice with the HubSan which I was using this morning to improve my ability to maintain a stable hover whatever orientation the quadcopter was in. I’m getting quite good at manoeuvring around with it pointing towards me, but I’m a fixed wing pilot at heart and still don’t feel entirely comfortable when it’s not moving forwards. Even with the glowing blue eyes I forget which way round it is when I’m hovering, but I’m getting a whole lot better at it now.

IMG_20180812_123856.jpg

The grey quad in the foreground was the one I was flying this morning. It’s modified from the HubSan X4 in the box in the background.

Talking about HubSan X4s, I’ve got to do another conversion to turn the X4 into one of our butterflies for a special project. I’m thinking about keeping the corner LEDs on for this one so it glows in the dark. Now that Maplin has gone, I’m going to have to find another source for connectors. I would really like to make the LEDs removable, but it’s going to need some really micro connectors so I don’t add a lot of weight.

Finally, I think it’s about time we did some instructions to show other people how we do these modifications and also publish the frame designs. Ultimately, what I would like to do is to put together kits so that other people can run Drone Masterclasses like we’ve been doing for the last couple of years.

It’s Been a While

It’s been a while, but I’m back in the air again. The heatwave continues though.

IMG_20180805_101927.jpg

 

That’s how hot the sun feels, but it’s apparently 31 degrees in the shade. There were a few people around today and the drones didn’t make an appearance until later on. There was the Nigerian guy with his son there when I arrived, but he claims he hasn’t seen me for the whole of this year, which I can’t believe. They had a mini super 60 type model, a Bixler which his son flew around really well, an F35 foam model, an F16 and a drone. I just had the RS352, but managed to get in 4 flights once I had figured out which battery was which and what state of charge they all had.

IMG_20180805_101941.jpg

This is the mini super 60 style model and, yes, it really is that hot.

Here’s the F16 going overhead:

Anyway, we then had a new guy with a drone come and hang out with us, plus a husband and wife who walked right through the middle past us and went to fly 200 metres away under a tree, then the other guy with his wife who I often see flying a Mavic (mainly her actually). I waved to them, but they still don’t want to talk to us, but they always fly from the same spot about 10 metres away, which is really handy if that’s the direction the wind is in. I don’t think he knows why we launch the way we do.

So, the guy with the drone hanging out with us in the middle turned out to be absolutely lethal and a complete moron. He’s a really nice guy and I like him a lot, but we ended up talking about insurance. Yes, it’s a condition that you must have insurance to fly here. No, claiming that you didn’t know is not a defence. The responsibility is always with the pilot to know what the regulations are. Also, it’s £35 a year, so you would be stupid not to. That part was OK as he was a really great guy, so could see the sense in what we were saying. The problem then came with the flying part. I got a bit worried when he said that he was actually frightened of flying his drone. Surely you would fly something a lot less powerful if you were a beginner and unsure as to what you were doing? Learn on something simple and move up as your expertise increases. Anyway, when he tried to fly it he nearly took our heads off. His first mistake was trying to fly with the goggles. He seemed to go to full power immediately, the drone spun completely out of control and tumbled in our direction, hitting the grass and coming to a stop right in front of us. He had no idea what it just did because he had the goggles on. So, here’s the moronic bit. When he took the 3 bladed props out of the packet, there were no markings on them so he assumed they were all the same. He couldn’t understand why his other drone had A and B props and this one didn’t, but never questioned it any further. Of course, when we looked at the props we could see left and right handed ones immediately, which we pointed out to him. The other slightly strange thing is that all his motors are counter-clockwise ones. The only difference between clockwise and counter clockwise motors is actually the thread on the prop adapter, so that’s fine as he was using nyloc bolts to hold the props on. Anyway, it all got put back together correctly, but I left him to it at this point and went home, so I’ve no idea whether he managed to fly it or not. He was still frightened of it, though, which is not a good way to fly a 100mph drone. In the good old days, Darwinian natural selection was the dominant force. Beginners would get a Spitfire, crash it and not come back. Once you remove the skill element people can just press a button and fly without ever knowing what they’re doing, so they do really dangerous things and the drones survive as they’re carbon fibre.

After that I’ve got a picture of a glider flying to the moon to calm me down a bit.

IMG_20180805_102553.jpg

 

Rain and Cycle Races

Flying was never an option this weekend as all the roads are closed for a cycling event and I’m right in the middle of the course. It probably wouldn’t have made any difference, though, as the current heatwave has given way to rain and thunder storms.

IMG_20180729_133600.jpg

The FPV quad (orange), one of the modified HubSan quads (silver), two new bug frames and a whole load of kit that I have to test for next time.

Last week I ran another Drones4Good workshop, with a new bug frame which came off the 3D printer about an hour before the start. That means I never had the opportunity to do the first flight of the prototype and, as none of the students wanted to fly it, the frame is still unflown. Maybe I’ll try that out later, as you can see from the different prop guards on the two above that the models all evolve a bit as they’re tested and flown. Then there are all the batteries to charge, cycle and store.

Anyway, this week I’m running an event at the IET, but we’re not allowed to fly any drones. I tried for a drone cage, but they were most insistent. That means we’re taking the quadcopter simulator (with Pokemon to rescue) and a 3D printer. As micro:bit are going to be on the stand next to us, I’m wondering about trying to get my micro:bit quadcopter to work again. The project got shelved as the last drone workshop got close and I got swamped with work, but I have a choice this afternoon of whether to write a flight controller for a micro:bit drone, or make some improvements to the quadcopter simulator.

Oh, Boy, It’s Hot Again

IMG_20180722_124510.jpg

 

It was 30 degrees at 8am again this morning. I would have had to take the bike this week, but with my current shoulder problem, I can barely get a backpack on, let alone lift the bike. I thought about going flying, but elected to stay indoors as I’ve still got a lot of work to do for the drone workshop we’re running tomorrow afternoon.

You can see part of the kit in the photo above. I was flying the unmodified HubSan around the living room last night to cycle the LiPos so they’ve had a couple of charges after lying dormant since March. One of the flight controllers failed, so we have 5 sets of kit for the custom 3D printed frames, one unmodified HubSan for flying practice and one FPV drone (orange) to show them what it looks like using the Goggles. The Flysight screen gives everyone a view of the FPV and the MicroBit on the orange frame in the bottom left is a bonus. I haven’t had time to finish the software to make it fly, so it stays home this time. It’s a shame really, but that’s really for more of a coding session.

In addition to the HubSan, I’ve been flying the FPV drone with my new Taranis QX7S transmitter (blue). I’m not really sure about this transmitter and I only really bought it because it allows me to test at home without having to transport the transmitter back and forth. When we run the workshop tomorrow, this gets swapped out for the Taranis X9D currently in the office. As for my QX7S, when I opened the soft carrying case and got it out yesterday, the black hand grips on the left and right sides were peeling off. They’re only made from sticky plastic, which is cheap and nasty compared to the integrated rubber or plastic versions you find on Spektrum or Futaba radios. The OpenTx system isn’t at all easy to use – long press/short press, did they ever read a guide on user interface design? Then there’s the sticks with those big lumpy bits on them which I hate. I also can’t for the life of me figure out how to lengthen the short stubby sticks. They’re supposed to unscrew, but I’m frightened of exerting too much force and snapping them off. I’m sure they’ve accidentally glued them.

So, I’m almost there now with 5 sets of kit which I take directly off of the working quadcopters and bag up for the kids to rebuild tomorrow. I still need to test the FPV and tune the FPV drone a bit if I get time. The HubSan is actually much more stable in the air, despite the superior flight controller in the other one (EVO 3 brushed). Mind you, if last time is anything to go by, the kids don’t want to see it fly, they just want to hold the camera drone, point it at themselves and take selfies on the Flysight screen. Lucky there’s an arm switch on it, but that’s why we use 8.5mm brushed motors, because they’re safe.

OK, I’m in final packing up and planning mode now. Tomorrow should be a blast.

A Year of Autogyroing

IMG_20180715_104326.jpg

 

It was too hot again, but I wanted to fly the autogyro because this is the weekend a year ago when it flew for the first time. You can see just how hot it’s been the last few weeks from the picture above. The grass is dead and the ground is completely dried up and cracking. Anyway, as it happened there wasn’t enough wind to spin the autogyro’s rotors, so I didn’t risk a flight, although it was right on the cusp of being possible. I elected to take the safe option and go home with it in one piece rather than risk a flight. This time out I’ve removed the two washers behind the motor mount giving it left thrust and changed to give me one washer of down thrust instead. That’s washers top and bottom on the right side moved to left and right on the top. This is in response to the fact that I ended up with full left aileron trim and full down elevator trim last time it flew. Unfortunately, I never got to test the change, so it’ll have to wait until next time.

Back to today’s flying and there was already a guy flying a foam Cessna type trainer when I got there, plus the guy with the DJI Phantom from the other week and somebody else who brought a Hobbyzone UMX Corsair up on his electric bike. My first flight of the day was with his Corsair, but, launching left handed, my first throw went straight into the ground. I’m obviously getting too relaxed about flying these UMX aircraft, as they normally fly perfectly straight from my hand. I got the owner to throw it for me on the next launch, with almost exactly the same result. This aircraft has AS3X, so it’s supposed to be able to fly itself. The third flight worked out better and I got it into the air, but it was a wild little beast. There was much too much aileron movement – I knew that from before the first launch, but it was adjusted down mechanically as far as it would go and there was no way of changing it on the very basic transmitter. I found that I could fly it around, just coaxing it with the ailerons, but too much aileron input caused the nose to drop and it headed downwards fast. In getting used to flying it around, I tried to add a click of down elevator, then another, before the wing dropped again and I stuffed it into the long grass. There was no damage because of the cushioning effect of the long grass I was flying over, but I was a bit worried about finding it again, so didn’t take my eyes off of the point in the grass. It was easy to find in the end, nose down in the long grass with the dark blue tail sticking up in the air, not even having made it all the way down to the ground. It was just stuck in the top of the grass. We didn’t fly it again after that, but I would like to get it sorted out properly because the F4U Corsair looks absolutely fantastic in the air.

After that I had four flights with my RS352, plus a bit of messing around with the Autogyro when the wind looked like it was getting up. It was just teasing me, though, because as soon as I had the autogyro set up and ready to fly, the wind dropped to nothing. When I flew the RS352, the wind would get back up. I decided it wasn’t an autogyro friendly day, but then the RS352 wasn’t going all that well either. I don’t know whether the heat was causing a lot of turbulence, but they were odd flying conditions for what you would expect to be flat, calm.

We had one success story of the day, where the Italian guy who saw the previous flight of the Autogyro a few weeks ago managed to find his lost drone out in the long grass. I hate losing aircraft like that, so it was really good to see him get it back. All in all it was a very quiet day, with just the couple who fly the Mavic turning up near the end. They were the only ones left when I had to call it a day.

IMG_20180715_104339.jpg