ruijsink.nl history

mm-hist.htm; updated: 01 January 2007

MicroMag

History of the MicroMag System

Since the beginning of my modeling career I was interested in smaller models, the biggest internal combustion motor I flew was a 1.5 cc Webra Record diesel, the others were the 0.8 cc Webra Piccolo diesel and the Cox PeeWee 020. I started competition flying with some A-1 gliders but mainly in Coupe d' Hiver rubber. In this class I became twice the national champion. Both classes are the smaller version of the World Championship classes A-2 and Wakefield (F1A and F1B).
When I started RC flight money was a bit of a problem, so I tried to build some gear of my own. Not too successful at that moment but I made some flights with the illustrious galloping ghost system. This is basically a single channel system where the transmitted tone is pulsed. Right-left control is achieved by changing the on-off ratio, where e.g. right can be on and left is off. Changing the pulse frequency could be used to control up-down. The Rand actuator would make big strokes when the frequency is low and short strokes when high. The middle kite like plate translates this oscillation into a something that could move the elevator. Then there was a third arm that would move when the pulse was (dis)activated for more than half a second then the oscillations of the motor changed into spinning a few revolutions of the output arm activating the third arm that could operate the throttle.

The Rand single channel, three function Galloping Gohst actuator.

The magic of the Galloping Ghost system was that it worked! And it was lighter than any other 'multi' radio system at the time. But, the control was far from subtle. The single channel variant was used by several suppliers in the 60's like Adams and Elmic. These used an actuator with a magnet that could move in the field of an electromagnetic coil. Along this technology, mr. Hilmar Bentert from Berlin had made a 3.5 gram receiver and a 15 magnetic actuator published in 1963.

A photo of Bentert's receiver and actuator as published in Aeromodeller March 1964

This inspired me to make one myself my actuator weighting 9 grams. The control was marginal when the rest of the system was kind to me. At that time prices and weights of commercial radio came down. This made it less obvious to make it myself. In these years my notion started to grow that building your own electronics is nice and not a lot more expensive than the ready bought stuff.
In 1972 Dieter Engels published the do-it-yourself PicoProp system in the German Magazine 'Modell'. This was a single channel analogue proportional radio with a special transmitter and a feedback servo. He integrated the good but simple radio and the servo electronics on one PCboard, where also the clever servo mechanics made from a Swiss TO5 motor with 1:141 reduction box and the remains of a carefully disassembled potentiometer. This complete unit weighed about 30 grams, very light for that time. He already realized that in order to achieve low weight the number of battery cells had to be reduced, his system was powered by 2 button cells of 225 mAh. at that time a lightweight at 27 grams.
I made this system and flew it in my first R/C electric flights in 1974.

The PicoProp single channel board with gearhead for the servo.

It is a pity that I don't have the system itself anymore or even a photo of mine. The model, with the fuselage rebuild is still flying, now with a Speed-300 and a 3:1 reduction driving a 6 x 5.5 Graupner speed prop. The model weighs now 350 grams with 6 x 1000 mAh NiMH doing more than 20 minutes. Radio is a Webra Rx and Cannon UltraMicro Servo's (yes the first under 10 grams!)

The Sub Mini as it still flies now.

Around 1985 I started again with the LM1875, an IC from National Semiconductor meant for toy R/C. It had an AM superhet radio with decoder for 2 channels and circuitry to adopt two extra switched channels. Super for lightweight R/C at the time, I thought. I designed very nice receivers for this IC that worked but not very well. Some small commercial receivers have been made with it but none were a real success as far as I know. The Rx was 4.5 gram with conventional components on the top side and SMD components on the other side of the PCB.
In 1994 Herbert Hoger published a small FM receiver of 3.5 grams. It was by far the lightest 9channel Rx. I had seen. I build one and it worked. Herbert and I got into contact and exchanged some ideas especially on the circuits to power magnetic servo's. I designed PCB along the lines of his receiver and my servo electronics.

The LM1875, the Herbert Hoger receiver and my version with servo electronics

I build it and was not entirely satisfied. The Rx had no IF filtering at all and a ceramic discriminator, a combination that is not good in blocking interference. Also the IC for the Rx was rather big as it was intended for double conversion and only a small of the IC part was used. In that time I came into contact with Wijnand de Joode a very good FreeFlight indoor scale builder and flyer. His dream was to build and fly Peanut scale models (33 cm span) on radio control. We became good friends, where his ability to build really superb scale models and my experience in radio control proved to go together well.

An action photo of the FIKE-E in flight, Me at the controls and Wijnand de Joode on my right.
The Fike is a very fine flyer, although it doesn't have any dihedral the model turns with grace.

I started to design my own Rx along the Motorola Mc3361. I found some very small ceramic IF filters and LC discriminator filter. This Rx with the servo electronics was a direct success. All parts were carefully picked for low voltage performance as the weight of the batteries was (and still is) the remaining barrier for smaller and lighter planes.

Two sides of the first generation of the MicroMag receiver together with the current version.

The first plane was a PonyMax build by Wijnand de Joode. It is a Czech homebuilt with simple lines and good proportions, 44 cm span and 40 grams. The Kenway KR-1 motor was powered by 3 cells Sanyo 50 mAh.

The Pony Max, build by Wijnand de Joode, flown by Rick Ruijsink

This power combination is still used in the majority of my models. Its first flight was at the 1996 fair in Sinsheim, where the European Slow flight really took off. Walter Scholl flew his superb 100 cm span 110 gram Bleriot, and Rainer Mugrauer astonished everyone with the flight performance of his Mucke. He flew under 9 tables with a model that could also fly a circle with a radius of not much more than its wingspan.

Rainer Mugrauer flying between the legs of nine tables

There was so much interest in the radio that I was forced to plan production. That proved to be more time consuming than I expected. The design work on servo's and radio seemed relatively easy, now all the difficult parts were working properly. One of the problems that I encountered with the first hand made pre-production models was one of compatibility. All the electronics except the speedcontroller were integrated on one PC board, which meant that the channel allocation was fixed. As Graupner/JR always has throttle on channel #1 and e.g. Futaba always on # 3, it would have been required to make two versions. With the fixed crystals this would explode the number of variants that I and/or a dealer would need to satisfy his customers. I also had one client for whom I made the Futaba version and when he moved to Graupner I had to change his receiver back. The other snag was that the neutral timing of the channels had to be adjusted by hand through the placement of individually picked SMD resistors, as the non-computer transmitters have a trim range that cannot cope with the tolerance in the normal timing resistor-capacitor. A few of the pre-production radio's went to the R/C CO₂ guru in the USA Henry Pasquet.

Henry Pasquet with a handful of CO₂ models.

His positive and critical comments were very welcome, he brought me in contact with Bob Selman, an enthusiastic indoor flyer who is a talented, professional microprocessor programmer.

Bob Selman with his Daisy

We discussed the problems (I don't like the euphemism 'challenges') and Bob promised me we could solve them and he was right. It took some time but it was worth the time. We could, after a lot of bench and later flight tests (one at midnight under the light of a street lantern) realize all the goals.

The original Daisy.....

...... as a testbed for the microprocessor development.

Furthermore one 8 pin IC did replace: one 14 pin, one 16 pin IC, four transistors, two capacitors and ten resistors. The old receiver was 15 x 24 mm and 2.3 grams the new is 16 x 18 mm and 1.8 grams and is fully self configuring.

The new MicroMag reciever and speedcontroller from both sides on my fingers.

The servo's were improved by using custom made neodym magnets and specially orthocyclic would coils. This winding technique ensures the optimum hexagonal wire packaging. This yields the highest magnetic field for a given mass of copper. The amount of copper in relation to the mass of the magnet was optimized as well. Two sizes of magnetic actuators are now in production. The small one is 0.7 gram, the big one is 1.7 gram. The actuators are fully proportional, the current in the coil (which follows the stick position) generates a moment on the magnet that is proportional to the current.

Two small and two 'big' actuators mounted on a small balsa tray

An integration of the speedcontroller and the Rx has been considered and tried, but sensitivity to interference increases, and the 'heavy' motor wires will be longer, in fact increasing the weight. Also the flexibility for the producer, dealer and buyer/user is degraded, so this idea was abandoned.
A separate small speedcontroller that is able to handle 1.5 amp was designed. Part of its intelligence is built into the receiver, so these two go together. You cannot use this controller with a normal receiver, you cannot use the MicroMag receiver with a normal speedcontroller. You can use the MicroMag receiver without speedcontroller in a CO₂ model or a glider.

Update 01 January 2007

Some years after the introduction of the MicroMag system other suppliers came with their own systems based on magnetic actuators. To name a few http://www.jmp-solutions.com, http://www.microinvent.com, and http://www.microplanesolution.com/index.htm for 27 to 72 MHz systems for use with conventional transmitters. Or on http://www.plantraco.com/hobbies/product_butterfly.html you will find a system on 900 MHz. The ultimate light system based on 27 th 72 MHZ AM transmitters cn be found at http://microflierradio.com.

MicroMag is not available anymore. We are currently considering a new generation of micro radio control systems.