Month: November 2015

A Test Flight Missed

It’s blowing a gale today and, with threats of heavy rain, any flying is cancelled. That’s a real problem because I needed to test fly the FPV gear on the quadcopter that I flew for the first time last week. Christened “Spidey”, I’ve spent the whole weekend just trying to fit all the wiring and electronics on to it, so I don’t actually think I could have flown it this morning anyway. As the weather forecast was fairly certain for Sunday morning, I was resigned to not being able to fly, so moved just about every component around on the airframe in order to get the best possible installation.

IMG_20151129_192041

The 250 size FPV quadcopter almost ready. Only one of the blade guards has been added. The FPV camera is just visible at the bottom of the picture near the power lead.

So far today, I’ve installed an FPV camera, modified a power board because it didn’t fit, soldered a brand new power system and moved all of the motor and radio wiring.

I’ve still got a lot of work to do to get it to a flying state, so I’ve got to go back to work now.

 

First Flight: Spider

The "Spider" 250 quad and my regular RS352

The “Spider” 250 quad and my regular RS352

The weather was perfect this week and I managed to successfully test my 250 size “Spider” quadcopter for the first time. I had spun the motors up indoors the night before just to make sure it had the power to hover, but this was the first test flight.

I managed to get in five flights with the quad and one with the RS352 this week, as the quad is running on 1100mAh LiPos while the RS352 uses 1300mAh ones, so I had five packs to use, plus a field charge for the final quad flight. The weather was perfect, bright and sunny with no wind and not as cold as I was expecting. As far as test flights go, this was almost perfect, with the only real problem being that the plastic legs kept falling off. There does seem to be a bit of an issue with the armed setting for the throttle though. I had reduced it a bit following the pre-flight tests, but it still spins the rotors to the extent that you occasionally had to diasarm just to keep it on the ground. It seemed a bit unpredictable though. A bigger problem might be that on the final flight I saw the right rear rotor stop, with the effect that the quad tipped over and landed upside down. Luckily I was hovering just a foot off the ground when it happened, so no damage was done. When I tried to spin the rotors up to check, this motor was refusing to run properly, so I might be looking at a new motor or speed controller, or maybe it was just the BEC cutting in for that motor? Other than that, after getting over the “it’s tiny, completely unstable, too sensitive and which way is it pointing?” phase on the first flight, I was then able to zoom it around and test out some speed runs with left and right corrections in preparation for racing through gates.

In addition to this, there were a number of other people flying this week. The French DH Beaver (white) was there, along with two younger guys who I see occasionally who were flying a small P47D when I arrived. The P47D went really well and was extremely aerobatic, so I must find out who makes it. They also had a FunJet, a profile foam aircraft and a DJI 450 quad which I was interested to see flying. Talking to him, he also ignored the instructions and put the long centre piece front to back instead of left to right as the instructions say. Watching it fly, I hope our one is as stable and controllable as his is. The last one to arrive was the guy with the FlyBaby and a Corsair. The Corsair had a problem with the rudder direction being wrong, so, with it also using a gyro stabilisation system, it couldn’t fly. Finally, after everyone but me had left, some other people turned up with a couple of small children and a white foam Cessna type trainer. They had some problems getting the throttle to work, but it appeared to have fixed itself just as I was having to leave. I saw it in the air, so it obviously got away from the ground OK.

UPDATE: It took me 2 seconds to figure out why the quadcopter motor won’t work – a motor wire has come loose, so I hope it’s just a case of soldering. I was a bit worried because I swapped the case on this motor to turn a CCW into a CW thread. In the back of my mind I was expecting a physical failure in the rear bearings, but it’s fine. Expect a post later about MT1806 motors because I’ve had no end of problems with them.

The DJI F450

The DJI F450

Spidey and the RS352

Spidey and the RS352

IMG_20151122_101300

Abigail and Kate

Abigail is the first major storm to be named by the Met Office, while Kate is the tail end of a hurricane. Needless to say it’s too windy to fly. In short it’s a building day.

 

A smaller spider in the process of construction

A smaller spider in the process of construction

It’s probably a good thing, though, because it looks like we’re flying in the University Agility competition at the UK Drone Show in December, so I need to get a 250 size quad flying fairly soon. Bearing in mind that I’ve been building radio controlled aircraft from scratch for several decades, I can’t believe how much trouble I’ve had getting this thing to work. I just assumed that all the equipment would work, but I tend to buy the very best quality equipment on the basis that I’ll get a lifetime of use out of it. A lot of the quadcopter parts are fairly low quality. The EMAX MT1806 motors look good, but I’ve had no end of trouble getting them to work reliably. Connected directly to an ESC and FrSky D4R-II receiver, I’ve found one to have failed before getting past the testing phase. Of the four currently on the quad, two are marginal on startup, seeming to skip like the timing is wrong. Now, before blaming the motors, this could be the wiring, or the ESC programming. I’ve got the dead motor on the bench at the moment waiting for some free time to investigate what the problem is. When I say “dead”, there are no beeps coming from it at all when the ESC starts, so I’m expecting it to be a wiring problem. Then, after getting all the motors running, I couldn’t get the radio to work with the flight controller. I actually had two channels locked together for some bizarre reason.

On the subject of swapping motors around in quadcopters, I had 4 of the plus thread motors and 2 of the CCW ones. Obviously, the failed one was a CCW thread so I didn’t have a replacement to hand. Now, the only difference between the plus and CCW motors is the left hand or right hand thread on the case, so I wondered whether this could be changed over. It turns out that it’s really easy to do, requiring just the removal of a single circlip on the back of the motor to allow the case to be pulled off. The new motor with the old case attached is working fine, so I know it’s not a problem with the magnets. I’ll do another blog post on this showing the steps required once I get some spare time, as it’s a useful trick to know.

Finally, I’ve been buying lots of tools and stuff which I need for the quad. In the picture above you can see my new Turnigy Hex Drivers. I bought these on Amazon (the link is to the HobbyKing site), but I can’t help wondering whether they are the real deal or fake? They cost £6.99, but were delivered in a cheap looking vac-form plastic pack with a cardboard sleeve behind. The plastic and cardboard pack has been but off at the top with scissors and contains no branding except for “BIN B.14.2.1.2 [11458] S638” and a CE mark and bar code on a sticker on the back. It looks fairly good quality, but not exactly the same design as the one on the Hobby King site. Also, the plastic holder for the bits inside has a lot of the plastic left behind and has been badly cut in one place. The spring clip that holds the bits in is also a little loose.

Normally I wouldn’t buy anything from Amazon Market Place for that reason, but we needed something quickly to do the filming. On the plus side, though, at the same time I bought myself a crimping tool for DuPont (Futaba/JR) connectors. This is something I’ve needed for a long time as I have to admit that crimping with pliers really doesn’t work. These are made by Jago and are definitely good quality, being professionally packaged, along with a guarantee. The jaws are switchable for different types of crimping, so I’m looking forward to having a play with these to make some of the connectors I need for the quad.

So, the target for this evening is to have the quad hovering. Unfortunately, this means that the Autogyro project is on hold for a bit.

Congratulations, It’s a Spider

I’ve been working on another project today, so no flying this week.

The DJI F450 quadcopter after a day's building

The DJI F450 quadcopter after a day’s building

As you can probably see, it’s a DJI F450 quadcopter that we’ve been building today. The idea was to film the build as a time lapse, rather like we did with the frame last Saturday (see previous post). We built the frame in about 40 minutes flat last week as we were running out of time, but this week we had a professional camera and somebody who actually knew how to film things properly. I’m hopeless with cameras, so I’ll just stick to what I know, building and flying things. Anyway, we disassembled the frame and did it all again, except that this time we mounted the flight controller, GPS, radio, U/C legs, telemetry and Fat Shark FPV kit. Hopefully the shots we got will look very good, as last week’s test with a GoPro Hero 4 Session camera wasn’t too bad.

I’m a bit disappointed that we didn’t get to switch anything on and make it work, but I really need to sit down with the Mini-Hawk flight controller manual and make sure that I’ve stuck all the right bits in all the right holes before I switch it on and blow everything up. Their quick start guide really doesn’t make it obvious where everything goes and you need to read it in conjunction with 3D Robotics’ PixHawk manual as the hardware is the same, except for being in a smaller physical package. All the connectors are the same, but the Mini Hawk just has rows of pins front and back which the connectors plug into. Once we’ve got it all figured out and working, there is going to be a follow-up post with the details and some filmed segments showing how to do it.

OK, so I said there was no flying this week, but I should probably confess to flying my HubSan Q4 around the office just to show the F450 how it’s done.

That’s it, I’m now off to try and make the other quadcopter work, which is a 250 sized one. I was hoping to have it working for the filming today, but discovered late on Saturday that one of the motors wasn’t working and didn’t have a replacement. Sadly, until this is flying, I’m going to have to put the Autogyro on hold for a week or two. I really desperately want to fly the autogyro, especially as a new plan has just come out in the latest issue of RCM&E – The Cruiser.

Running, Mist and Serious Building

No flying this week as all the roads are closed due to a running event. I’m not sure I would have been able to go anyway, because it has been very misty all day.

Anyway, yesterday we were filming the build of an F450 quadcopter:

The F450 frame, ESCs and motors

The F450 frame, ESCs and motors, plus a lot of other bits

It’s amazing just how much stuff you need to build a quadcopter. Normally, I can just dip into my cupboards full of tools and test gear, but this time I had to take it all with me in a rucksack. Just look how much stuff I packed into it:

The contents of a quadcopter building rucksack

The contents of my rucksack

So, after the F450 came the 250 size quad that I’m also building. I spent all of the evening soldering power wires onto the wiring loom in preparation for soldering the motors to the ESCs. This needs the power to be connected and the radio working, but I’m not quite at that stage yet. I’ve also been trying to create time-lapse movies of the build sequence, but the app on my Nexus 4 phone keeps crashing, so you only get to see me as far as the first ESC connection.

Never underestimate the amount of wiring in a quadcopter

Never underestimate the amount of wiring in a quadcopter

As this is an ‘X’ frame rather than an ‘H’, the wiring is really dense. I’m also hoping to glue the tail onto my autogyro later, but at the moment I desperately need to get the quadcopter working, so that has to come first.

It’s already evening and I’ve still got so much to do…