At last! A radio-controlled plane under 1 gram. Here's Martin's Midge - wing span 3.2", weight 0.98 grams, two channels, 27 MHz radio.

Just about everything had to be re-engineered to get below a gram - the only "standard" part is the incredible Leichty Micro 3 channel receiver at a nominal 100mg, see http://microflierradio.com/. Everything else, from the prop to the actuator had to be rebuilt several times to achieve the needed weights. I'm particularly pleased with my 2mg on-off switch that is easier to use than a plug and socket and much much lighter.

The prop is basically the same as on the Mosquito - direct drive 1" diameter by 0.6" pitch carbon, left handed, and thinned down. The motor is based on a 4mm x 8mm, 28 ohm pager with no case, no flux ring, and the magnet ground down. Also the bushing was replaced with a plastic one, with the original brass bearings, and the shaft was replaced with a carbon fiber rod. The battery is a 10maH lithium poly, reduced by careful trimming down to 530mg. The actuator coil is 1/16" ID, 135 ohms with a 1mm diameter by 0.5mm N50 magnet. The airframe is 0.010" carbon rod for the fuselage stick, and 0.005" carbon for the flying surfaces. Covering is 2um mylar, and the rudder is 1/100" balsa.

Getting below 1 gram was much harder than I anticipated. I owe a lot to David de Wit whose 0.94 gram Elfin, http://members.dodo.net.au/~daviddewit/, showed that it can be done, and gave me ideas of what component weights I had to aim for. It should be noted that the Elfin has three channels of control, whereas the Midge has only two. Also the Elfin is controlled by infra-red while the Midge is radio controlled.

I called it Martin's Midge because it sounds like a midge in flight - that little direct-drive prop really screams, but quietly, at 20,000rpm. It takes about 85mA and generates about 0.5g thrust.

The definitive maiden flight was on Tuesday, October 25. After some initial porpoising it settled down to really stable flight at about 4mph. It did several circuits and a figure of eight. Flying time was a couple of minutes, and I landed before it ran out of power, so it could have gone longer. After recharging we had a second flight and captured some of it on video, though this time the flight was cut short by a bad glitch. No damage though.

I've been working towards this for several months and it's a relief to have got there. The next challenge is to reduce the weight just a little further so I can add a third channel, though it's not really necessary. I'm sure I can take weight out of the airframe. The horizontal stabilizer is designed to accomodate an elevator and the receiver already provides 3 channels. The added weight will be only about 20mg so it should be possible, as David de Wit has shown.