wnd's weblog

My second helicopter

At the end of April I finally fulfilled one of my long-time dreams of buying a "real" radio-controlled helicopter. Sure, I already had one of those IR-controlled mini helis, but those things are really nothing but toys. This one would have to be fully controlled; altitude, yaw, pitch, and roll. In other words, this would have to be at least four channel helicopter. Since it would be my first true RC helicopter, I decided that a co-axial helicopter would be a good way to get started.

While I thought I had already decided to get Walkera Dragonfly 5#10 (which looks just great in my opinnion), I decided to get something bigger. Something bigger would fly better in outdoors, where the air would seldom be still. After serious consideration I decided that Walkera Dragonfly 53#1 would fit my needs -- and wallet -- the best. It would be good compromise between price, performance, and looks. It's not a beauty, but it's metallic. After spending a week looking around I decided to buy the helicopter through eBay. eBay was the place where I could get the helicopter from within EU to avoid customs and VAT.

It took just four days to get the helicopter. I placed my order on Sunday night, cleared out the hassle about extra batteries in Monday morning, and received a package from Germany on Friday. How nice is that. I would then spend a week just spinning up and spinning down the helicopter. I would try to keep the tail on spot and get used to the controls. I would make the blades produce just enough lift to make the helicopter effectively weightless, and try to hold the helicopter still. When I felt confident enough, I increased power and ascended from the floor.

I was now flying, but the helicopter would keep yawing counter clockwise. In addition to that, the thing wouldn't stop move sideways to right. Trimming didn't help, and in fact, trimming only meant that part of the stick's range was dead zone -- the servos didn't move one bit there. I tried to explain the problem with the fact that the top rotor was always spinning up later, so obviously it would produce less torque. How this would lead to banking to right was a mystery to me. I spend days trying to adjust center of gravity by moving the battery and adding weights, but nothing helped. Adding weights made the helicopter flyable, so after a couple of weeks I could hover for five minutes without losing control and having to land and spin down for safety.

After searching the Internet and posting to an RC forum for help, I came across some instructional videos on Google Video about adjusting RC helicopters. Inspired by the videos I fastened the blades of my helicopter and checked blade tracking, which proved to need some attention. I never got that far, but in progress of not getting that far I ended up adjusting the length of aileron servo ball link rod.

Adjusting the aileron servo ball link turned out to be the best thing I've done with my helicopter so far. It did not only eliminate the dead zone for the sticks, but also allowed me to keep the trimmers in near centre. In other words, the helicopter would now hover easily in one place without yawing or banking significantly without constant battle to keep it that way. Naturally the helicopter still wouldn't fly on its own, but hovering was now much easier -- and I'm sure weeks of training wasn't the only thing to thank here.

So... Now I have an RC helicopter that hovers easily, flies for fifteen minutes per charge, and that could basically be flown outdoors. I'm still not confident to go outdoors, but day after day I dare to fly higher, and feel more in control. I can't wait.

Cherry trees in Helsinki, 2008

Spring/summer is a bit early this year in Finland. Not everyone likes it but I'm not complaining. Last Sunday was just splendid day to get out to sit in a park. In addition to beautiful weather and good company, cherry trees were in full flower. Actually these things are not real cherry trees but they're close enough for me. I just wish one day I could get to Japan in April...

I also uploaded some other photos to random 2008 set. I'll have to put a link there on "photos" page at some point.

Big graphics card is big

About a day ago I finally received my new graphics card. GeForce 9600GT would be more powerful than my old and faithfull 7800GT, but more importantly, 9600GT wouldn't have any moving parts -- it would be completely silent. When I got home, I opened the box and was surprised how big the thing was. I soon shut down my computer and opened the case. I removed the old GFX-card only to find out my the graphics card wouldn't fit in, thanks to larger-than-average chipset heatsink. Now what?

So the thing won't fit in as-is. Forgetting the whole thing is out of the question so the only option is to make it fit. After carefully considering all options, I decided to find out what would happen when a chipset heatsink and side cutters would meet. To my surprise my cheap Chinese side cutters won. I disassembled my computer to remove the motherboard from the case and popped the lately replaced chipset heatsink off the motherboard.

I then made some measurements of the chipset fan and my new GFX-card. I realised that I could either chip almost half of the fins, or move the heatsink a bit aside, and cut only two rows of fins out of the way. Since the chipset core was only about one eighth of the casing area, I decided in favour of cutting two rows on fins and relocating the heatsink.

I had to hack Debian's nvidia-kernel-source package to get the latest nVidia drivers (171.06) installed neatly. Version 169.12 of the drivers would fail completely with 9600GT. My system is now running its second day with the new graphics card and mutilated chipset heatsink. So far everything looks good -- and sounds just brilliant. I can warmly recommend Geforce 9600GT with passive heatsink for discriminating Linux user.