Lockdown 3: Week 12 – A Time to Charge

It doesn’t look like much, but this is the closing up process of the bottom fuselage of my Cap 10. It’s taken a long time to get to this stage.

Weather permitting, we could be flying next week. This week, though, the wind has been trying to lift the roof off the house since the early hours of the morning. 18 mph continuous and 40 mph gusts. While we’re on the subject of flying, now is the time to get out the transmitter, charge it up, cycle the batteries and make sure it still works properly. The same with the LiPos, so I’m putting mine through a charge cycle now. I also notice the BMFA advice about getting some simulator time before flying for real in order to get all the cobwebs out. That’s handy, because my flight simulator is working again!

With all the work deadlines out of the way last week, I’ve got some time to myself again. I’ve fixed the Unity joystick issue with my Taranis. Well, actually, I bodged it so it worked. I mirrored the left and right halves of the axes and reversed them so that I got a value from -1 (full left) to zero (centre) to +1 (full right). Not 0 (full left) to -1 (centre) and +1 (centre) to zero (full right) as you get from Unity. It’s just wrong, there’s nothing else you can say about it. I also had to reduce the ATV to 88%, otherwise I was getting it flicking to the opposite direction at the extremes of travel. In future, I think a better fix is to limit the axis travel and add an offset so that it can’t go negative.

However, I now have a simulator that I can fly! Although “fly” is an interesting term, because it’s not exactly flying like an aeroplane. But, that’s the fun part. You can learn as much about making a virtual plane fly as you can with a real plane. Probably more, actually, because you really have to delve into the physics of flight. I’m going to be going back to my Martin Simons book on Model Aircraft Aerodynamics again to refresh my memory on vortex and profile drag. What I have in the simulator at the moment is much more drag than you see in real life. When I cut the throttle, the aircraft was almost stopping dead in the air. The new wing model did seem to be producing lift correctly, though, which is a big surprise. The hack that was currently in the simulator to get it working is horrendous. So, that’s my first job, to read up on aerofoils again and get my Clark Y section working just like in real life. What is going to set this simulator apart from the rest are the “widgets” that let you see what’s happening so that you can evolve a design before making it for real. I rather had my sights set on a full polars chart on the screen with the current operating point highlighted with a cross. Seeing it stall in theory and practice at the same time would be interesting.

Next week is the Easter Bank holiday, which is a time when I’ve been known to fly my new aircraft for the first time. I don’t think it’s going to happen this year, though, because the Cap 10 is still too far off finished and there are too many things I still don’t know how to complete. I’m probably a week off having a functional, working, aircraft with motor and control surfaces moving, but I’m still very much in the dark over final finishing. The FlightTest guys seem to spray theirs in about half an hour. I can’t see me doing that somehow. The original idea was to use white foam and have a white aircraft, but it looks like I’m still going to have to paint the white foam white in order to hide all the joins where the materials combine. I’m not sure it’s exactly waterproof either, so that’s going to need some testing. I had thought about just flying it as it is, because there’s a good chance it might not work. We’ll just have to see how it goes.

OK, so here’s to flying outside again. It’s probably going to the the flying wing again.

Lockdown 3: Week 11 – Closing Up

The weather wasn’t bad this morning, but there seemed to be a lot more wind than forecast. At least we might get back into the air sometime soon.

I’ve done some work on the Cap 10 this week as I’ve decided to make a point of not starting any new builds until it is finished. Closing up the bottom fuselage is the real sticking point at the moment, but I now seem to be making progress. Most of the front lower fuselage is done, I just need to stick the main part that’s slightly curved and then cut out a hatch. The middle fuselage section is a bit more of a problem due to all the access that’s needed to connect the wings, servos, radio and fuselage tail section bolts. I hadn’t realised in my original design that, in order to actually screw in the bolts, the bottom fuselage piece must not protrude any lower than the green piece of cardboard in the picture. That big piece of foam board above the ruler was supposed to slot into the recesses made by the black foam doublers inside the fuselage. On dry assembly, I discovered that it’s impossible to turn the bolts to attach the rear fuselage as it was designed. This is going to require a small piece of 1/32 ply where the green cardboard is, then a slightly modified rear section of fuselage bottom to accommodate. The marks on the white foam board piece show where I’m going to have to cut out an access hatch to allow for field assembly. It’s an interesting 3D problem, but one which I think I’m now well on the way to solving.

Work has rather taken over this week, but I should have some time to myself after Wednesday. I would have made some progress on the simulator apart from working 14 hour days on a Unity 3D project for work. However, my April copy of RCM&E arrived this week and I’ve been reading the article on the RealFlight 9.5 simulator with interest. It really makes me want to get back into writing mine again. Seeing an aircraft you own fly in your own simulator is a really satisfying experience. Also, there was the article on building from plans, but using Depron in place of balsa. I was really looking forward to this after they advertised it in last month’s issue. The article is good, but it’s only three pages and I wanted more. How does he paint the models? I’m still puzzled over how to finish my Cap 10. The structures were interesting as he uses a flat Depron panel from nose to tail and glues half formers on the left and right sides. Some of the planking appeared to be hard balsa or lite ply, but I couldn’t tell how he made up the complicated shapes. His Chipmunk looked incredible, then the silver Spitfire looked even better. It appeared to be a very composite approach with lots of different materials.

In a similar vein, the Micro Aces Scrappee build earlier in the magazine showed how a micro Depron plane is put together. This one had lots of detailed pictures and I can download the instruction manual too. Guess what I found when I went online, though? It was a bit of a shock to see that there is a Scrappee Biplane conversion kit that looks just like a WACO! I couldn’t believe my eyes! The only thing is that Scrappee looks so small in the magazine, but, at 43cm span, it’s a touch bigger than my WACO at 41cm. I think my WACO also has a slightly higher wing loading, but it has a more powerful drive train and a larger LiPo. Anyway, I’m banned from actually building it until the Cap 10 is finished, but you never know, I might have a go at building in Depron from all the bits left over after the Cap 10? Or I might be completely sick of foam board by then, who knows?

OK, I need to get back to work now to make Wednesday’s deadline.

Lockdown 3: Week 10 – Crazy Windy

I’ve done nothing but work this week, which was to be expected given the various deadlines. The weather has been crazy too, with 70mph winds during the week that threatened to tear the roof off the house. The weekend isn’t much better, so flying wouldn’t really have been an option anyway.

Nothing has happened on the Cap 10 build, which is well and truly stalled at the moment. It’s so close to finishing that I really just need to get on with installing all the internal bits and pieces so I can finish up. I’ve managed to do a little on the flight simulator, but only up to the point of deciding once and for all that the joystick interface in Unity is completely messed up. The only option to get a Taranis working as a joystick in Unity is to define the aileron and elevator mixers to have offset of 100 and scale so that you only get positive outputs from the transmitter. It doesn’t seem to understand axes that go from -100 to +100, as it swaps the directions around incorrectly.

I think my RS352 model does looks very authentic in the simulator, using the textures that I photographed quite a few years ago when I first bought the aircraft and it was still in the box prior to assembly. It could do with a few tweaks to the texture map around the tail, but otherwise it’s very life-like. If I get some time to fix the joystick input then it should also fly in the simulator.

The only other thing this week is that I think I might have had a brilliant idea about my indoor autogyro, which is another project that has stalled until I fix the rotor arms back on. It occurred to me that, for an assisted takeoff, it might be possible to use a device that funnels the air from the propeller along and up into the main rotors, causing them to rotate and produce lift. This is a bit like what I was doing with my big Atom Autogyro, where I was using the wind to spin up the rotors on the ground so that I could do a short take-off. Maybe the same sort of thing with either the model’s own motor, or an external fan, rather like a wind tunnel, would enable me to get it working?

OK, that’s all I can manage this week as I’ve got to get back to work.

Lockdown 3: Week 9 – Nearly There

The weather is perfect and it looks like we’re going to be able to fly again soon. Somebody remind me to swap the Operator IDs on my aircraft for the new ones first, though. I appear to have three Operator IDs from the CAA and I’m not sure which mystery number is actually valid, so I should probably sort that out first.

I’m building, honestly!

I am really desperate to get my Cap 10 finished, but I’ve been snowed under with work this week and certainly next week and the week after that too. Things have stalled a bit, but here’s my complete mess build picture, above. It’s not usually like that, but, as you can probably see, I’ve been making the velcro wrap-around to hold the LiPo down in that front section. That’s it there on the green cutting mat. It’s upside down and the battery would sit on the velcro loops that are currently facing down, on the short edge, after which the two long edges wrap down and around the LiPo, making absolutely sure it never moves a millimetre. I’d got to the point where I figured that I needed all the gear installed to check sizes, balance and cabling etc. Then I can put the bottom fuselage piece on, cut a hole and make it removable and find a way of attaching it (velcro). I should buy some small magnets, actually, they would have been very useful. The only ones I have at the moment are so powerful that they would rip the foam to shreds. That’s the plan, but it’s very difficult to get back into a building project when progress is so stop start. I’m frightened to seal up the fuselage just in case I’ve forgotten a vital bit. In this case, it was a hole for the receiver leads through the second bulkhead. Not really a showstopper, but it’s easier to do it now. So, I guess, the next thing to do is to get all that radio gear, ESC, LiPo and motor inside with all the cabling to check that it all works out logically and that I haven’t missed anything.

This is Acrostik. Span 80cm, power 3S 2200mAh Lipo. Manic!

I know I was talking about making a small WACO SRE biplane last weekend, but I had another good idea. A mini version of my existing Acrostik would be interesting. This was made from the prototype plan, given to me by the designer to test before it went into RCM&E. Sarik hobbies now have the plan here: [link]. If you’re wondering why the tail on mine looks a bit smaller than the one in their picture, then it’s because there was a bit of a problem getting all of the drawing onto the magazine pull-out. I forget which version ended up getting published, but they both seem to fly the same. The story behind this plane is that the original designer broke his main plane on the Saturday and didn’t have anything to fly on the Sunday, so Acrostik was born out of carbon tube. There’s one along the leading edge and one making the fuselage. The wing is symmetrical with balsa ribs, while the ailerons, tail, elevator and rudder are built up flat pieces that are profiled. I can’t believe he made this in a day, though. This one has been converted from the original Speed 400 and gearbox to a £10 brushless motor. The 2200mAh LiPo is rather over the top, but, even with an extension to push the motor further forwards, you need the weight of the 2200 LiPo in order for it to balance. However, this does give you 15 minutes of flying time. And after 15 minutes of flying this, you might need to sit down for a while to calm your nerves. It rolls at more than 360 degrees per second. In fact, the roll is so fast that I’m worried when I null out the rotation with opposite aileron that the wings are going to stop and the fuselage is going to keep rolling. It’s a lot of fun.

And so, I thought, “how hard could it be to make a mini one?”. My first thought was to build it for a 1300mAh LiPo with one of my 30g drone motors. All up weight would be a bit under 200g, but the numbers for a 40cm wing and 20g/dm2 wing loading don’t work out. The idea is to make a removable wing and tail, which I’ve already sketched some really neat ideas for. The other option is to go for a really mini one at UMX size with a 500mAh 1S LiPo and 4x4g servos. That could work, and I might stand more of a chance of building the whole thing in a day. It would certainly be easier than finishing off my foam board Cap 10. After watching lots of Flitetest videos about how to finish foam aircraft I’m still confused as to how to do the final sealing. It seems to be an evolving art. Never mind, next month’s RCM&E contains an article on building in foam, so that should be good. In the meantime, even if I do get all the bits attached to the Cap 10, you can probably tell that I’ve still got no idea how to finish and seal it.

Things do feel like they are coming back together after a very long break, though. I’ve been doing some simulator stuff in the evenings, but Unity not talking to the FrSky Taranis QX7S joystick is starting to get to me. I had thought that the new joystick input system in Unity would actually work, but, no, it doesn’t. It’s difficult to pin down, but I think the HID data packets that the Taranis sends to the computer are being parsed incorrectly by the Unity input system. It surely can’t be that hard to just read a raw input value for an axis can it? Aileron and elevator come out reversed while throttle and rudder are fine because they’re read as ‘axis’ values while the others are a composite 2D ‘stick’. When I say “backwards”, what I actually mean is that the high value is in the middle and the zero is full left or full right. In other words, if you wiggle the stick in the middle by less than 1mm, it will flip between plus or minus one. I think the only fix is to offset the value as you used to have to do with the Taranis before, so no change there with the new Unity joystick system then. Also, Windows works perfectly if you run the USB joystick calibration, so it’s only programs using Unity as a framework that are at fault. It’s very close to flying once I fix the joystick problem, though, so I should be able to get some virtual practice in soon. Probably a good thing too, as I don’t think I’ve flown anything since October.

It’s taken a long time, but things are now coming back together, which is good.