Home
About the author

Amateur Radio Circuits

General Circuits

Antennas

Projects

Diary
Links to other sites

CONSTRUCTION PROJECTS

These projects are mainly electromechanical. Any electronic control circuits associated with these projects can be found on my "Circuits" page. Antenna designs, although mainly mechanical can also be found in "Circuits"

My current construction projects are Wind Turbines. I have made three to date. The first one was an 8 bladed machine which used about a 2:1 step drive to a Lucas Permenant Magnet Alternator. I left it working in the North of Iceland twenty years ago. I'm told it was still working several years later when some Lucas Apprentices visited the site.

The next one I made was a four bladed design, using a 5:1 bevel gearbox. This allows a car alternator to be used fixed to the tower, the turbine part luffing round the gearbox. This design showed a lot of promise producing 50 amps at 13.8v in a 20 mph wind. Unfortunately, I had no way of estimating how much the blades would bend in high winds and hadn't allowed enough clearance between the blades & lattice tower. One of the blades cracked at the root (half housing joint), so I stopped it running. I tried running it with two blades but the vibration was too much for the gearbox. It is currently being re-built with three of the original blades dovetailed into a steel hub, keeping the fourth as a spare. The gearbox is being beefed up and a better bearing fitted to the output shaft (roller instead of plain). The blades are of an interesting construction. They are made of marine ply with constant chord, 8ft (2.4 m) diameter, attached to oak roots with fairing pieces attached. The ply is mounted so that the inner third of the blade forms a DeHavilland Slot with the root to increase starting torque. Blade twist was introduced after planing the airfoil shape by using torque bars and heat. (See photos. Available soon.)

A lot of thought was put into how to excite the car alternator. The field winding consumes about 40W (3 A @13.8 V), so a speed switch needs to be included to prevent the battery being discharged when the turbine is below speed or even stationary. After considering several ways of doing this, I came up with a better solution: A large diode is connected in series with the alternator and battery. This needs to handle the full 50 A capability of the alternator. The output terminal is wired directly to the field coil. The rotating field coil/melectromagnet assembly is modified as follows: some ferrite magnets are glued with epoxy resin to the ends of the rotor, such that they are attracted to the rotor when it is energised. In the Delco Remy alternator I used, magnets taken from door catches which have their magnet poles across the thin dimension can be used (6 required). There is enough room for them between the end of the rotor and the aluminium housing. On some of the older Lucas alternators an annular (toroidal) magnet of the sort used in mag-mount antennas and TV focussing magnets can be fitted, as there is more clearance, although some machining of the aluminium housing might be necessary. This modification makes the alternator into a compound type. By effectively increasing the residual magnetism it makes the alternator self exciting, with positive feedback. This has two benefits: the efficiency is improved and the cut in speed is less ( previously it was about 800 rpm.). A further benefit can be had if a pulse charger is used in conjunction with the alternator. In low wind speeds the output voltage from the alternator is less than the battery terminal voltage, so no current flows. The pulse charger steps up this voltage electronically, putting some charge into the battery and acting as a desulphator. This latter aspect is particularly important during calm periods, when the discharge is greater than the charge. ( See "Circuits" page for Pulse Charger.)

 

Meanwhile a third wind turbine/alternator combination has been built. This uses a directly driven 1.3 m diameter three bladed turbine and a flywheel type motorcycle alternator (Honda 135 W). This was totally rewound to use all the available poles on the stator (18). The magnet assembly is cup-shaped and has 12 poles. This gives a 3-phase alternator delivering 200 W max. at 15 v at around 500 - 700 rpm ( 480 turns per phase 0.9 mm wire). This machine was designed to mount on a 2 inch diameter aluminium scaffold pole, bolted to the side of the house. It is hoped that this will be quieter than the other model as the gearbox was noisy (it stood on top of the flat roof extension to the main bedroom and even with sound deadening pads under the legs of the tower, my wife still complained about the noise.) This desgn relied heavily on notes from Hugh Piggot and others. His brake drum alternator (500 W) is very similar to the motorcycle alternator. Personally I wouldn't choose this route again, although a second, slightly smaller (100-150 W) using a moped alternator with my own design of stator is planned if this one is successful. The blades for these machines were made using 2mm thick glass/polyester & 5 mm thick ply, sanded to airfoil section and curved & twisted by heat in one operation, using wooden profile formers and torque levers.

Windmill Version 1

The blades were finished using a thin layer of glass fibre and polyester resin, with a top coat of resin and sanding filler (David's Isopon/Plastic Padding.) Final finish polyurethane paint. Update:(21/01/2002) I have since scrapped this turbine design as it partly de-laminated during an overspeed in a recent gale; I think the root cause was down to the method of construction of the shaft & bearing assembly. The bearing was a press fit on the end of the shaft, nearest the taper for the magnet assembly, with no shoulder to stop the bearing being pushed along the shaft. Vibration set up by the mounting pole resonating at a particular rpm turned the magnet assembly into a hammer which pushed the shaft through the bearing. This did two things. The magnet left contact with the stator and the blades moved closer to the supporting pole. The end result being severe overspeed ( possibly in excess of 2000 rpm) and the blades probably clipping the pole. the shaft will be modified with a welded collar and the blades made much stiffer. I'll probaly go back to using marine ply with a sheath of GRPon the leading edge and possibly over the whole blade.(21/01/2002)

Windmill in use at Burton Dassett

UPDATE: The machine above has gone through several evolutions, including being run as a downwind turbine. This was due to a mistake in that the marine ply blades were shaped and assembled such that it ran anti-clockwise, with the consequence that the rotor hub tended to unscrew itself when run as an upwind turbine. As a downwind turbine everything was fine except that I didn't think it weathercocked satisfactorily, so I added a sub-fin. There wasn't much learance and due to a rocking motion in high winds the blades struck the sub-fin and consequently failed at the roots. I repaired them by cutting off the roots, boring the hub and each blade and inserted 1" steel tubes with cross bolts to hold it all together. It failed at high speed again ( more or less the same failure mode ) because the grain structure in the root area was mostly in the wrong direction. I went back to my original hub which had delaminated. The roots of this were OK so I replaced the outer blade sections with new plywood and sheathed it with glass fibre tape & resin/filler as before. These are the blades you see in the photos. The machine has been running successfully now for nearly two years in this configuration with only a few problems. The shaft supporting the tail fin was replaced about 9 months ago because one of the welded brackets fatigue failed. It has been replaced by a clamped bracket & no problems since. This machine was designed with a dual purpose. Most of the year it would be at home where it would help power the house lights. Occasionally it would be taken out to a suitable hilltop where it would power an Amateur Radio Station. It has done both successfully, although output at the home location was less than anticipated. It has been moved from the side of the house because although the machine itself runs very quietly, the pole rattled in its brackets and all attmpts to deaden the sound failed. It is mnow mounted on a 10 m high guyed 2" mast, near the end of the garden away from the house. The photos show it on top of a 700 ft hill on a 10 ft pole, powering the radio station.

I also have planned a 500+ W constant speed machine using a 930 rpm induction motor as a generator to provide 240 V 50 Hz "mains". This requires building an articulated hub for use with a 3 m three bladed turbine and 2.75:1 bevel gearbox made from a breast drill. I got this idea from a much bigger machine I saw in the Faroes which used a car back axle (about 3.8:1 ratio) and a 10 m diameter turbine ( if only I had the real estate !) and a ground mounted alternator -- several kilowatts.

In the meantime I have helped Bob G4GEE erect a commercial turbine (Rutland 913) on a 2 inch aluminium mast. This is a six-bladed 90 W (300 W in a gale) design of small diameter (600 mm). The alternator in this is suprisingly heavy, weighing about twice as heavy as my 200 W design. It turns easily even in light winds, though it doesn't produce any power worth talking about until the wind speed is about 10 mph. This would benefit from a pulse charger. Both Bob and I decided the layout of the control box left something to be desired from the assembly point of view. Most of the design effort seems to have gone into the turbine/alternator assembly, which was very well made, but Iwould expect that considering the asking price (around £500). I've spent less than £50 on each of mine in materials, but I don't charge myself for labour. (About 40 hrs for each turbine, including building the lattice tower.)

73 de John G8SEQ