Solar-powered model planes

Anything you like!
Martin
Posts: 744
Joined: 16 Feb 2018, 14:11
Location: Warwickshire

Re: Solar-powered model planes

Post by Martin »

I updated the sketch (attached to opening post in this thread) to fix a few minor bugs and make it more compatible with 5V Arduinos.

It's still best to use a 3.3V 8MHz Pro Mini if you can - and if your receiver and servos are also happy to run from 3.3.V then using a Polulu regulator and powering everything via that is best. But if you're running more than 10 cells you can get away with using a 5V Arduino if that's all you have available. Obviously you don't power a 5V Arduino from a 3.3V Polulu, but if you're running 10 or 12 cells you can get away with powering the Arduino direct from the solar cells into the Raw or Vin pin, and running your receiver and servos from the ESC BEC or similar.

I altered the code to use the Arduino's internal voltage reference (1.1V) rather than VCC. That makes the code universal for 5V and 3.3V Arduinos, and also reduces the unwanted feedback effects if the solar cell voltage ever drops down close to (or below) the VCC that the Arduino is supposed to be running from.

THIS MEANS YOU NEED A DIFFERENT PAIR OF RESISTORS! The 'standard' recommended resistors are now a 68k from the solar cell + wire to A3 and a 10k from A3 to GND. This gives a working voltage range of 1.1V x (68 + 10) / 10 = 8.58V which is plenty for up to 12-cell operation. If you run more than 12 cells, increase the 68k and adjust the line in the sketch:

#define R1 68000.0

to suit.
Martin
Posts: 744
Joined: 16 Feb 2018, 14:11
Location: Warwickshire

Re: Solar-powered model planes

Post by Martin »

My 10-cell model made a duration flight of over an hour last week. Duration was only limited when I became too bored to continue flying. A couple of days later the record was broken when I flew it for another hour and then Mike took control of the transmitter for an extra 15 minutes or so.

Later, Mike used his 12-cell 'delta' to just beat (or at least equal) the record - he would have flown for another few minutes to make it more certain but was forced to land due to lack of power caused by increasing cloud cover and low sun angle as evening approached.

With a well-charged transmitter battery and a team of pilots to take turns with the flying, duration of 8 hours or more could be easily achieved on a sunny summer day.
Martin
Posts: 744
Joined: 16 Feb 2018, 14:11
Location: Warwickshire

Re: Solar-powered model planes

Post by Martin »

Another slight mod to the Arduino sketch attached to the opening post. We found that suddenly throttling back causes the motor/propeller flywheel effect to work the motor as a generator and briefly push the voltage up higher than the true solar cell voltage. This then subsequently causes the motor to be ramped down in power, or even stopped, because the controller thinks that the voltage is close to or less than the "85% of maximum voltage seen".

To fix this, the sketch now only samples and remembers the highest voltage seen during the first ten seconds after switching on. This works well as the solar voltage changes very little between heavy overcast and full sun providing the motor isn't running - so during the first ten seconds after switching on, when the motor isn't running, the voltage will have risen as high as it ever will with those particular solar cells.
Ralkbirdy
Posts: 44
Joined: 25 Feb 2018, 13:10

Re: Solar-powered model planes

Post by Ralkbirdy »

Time for a 2019 solar plane, found some more cells at a good price so got 50 for some spare ones.

https://www.ebay.co.uk/itm/SOLAR-CELLS- ... ect=mobile

This time I am doing a normal plane along the lines of a dlg. I have a wound carbon tube for the rear and am working on a smoother wing. The tailplane and fin are 3mm depron with a balsa edge and carbon sheet spar. The one side is covered in glass cloth with acrylic varnish. The trailing edge is sanded nearly flat and then the other side is covered, leaving about 5mm of glass cloth TE from both sides to give a fine edge.
Tail and fin
Tail and fin
The body is a half rolled 0.4mm ply structure with 3mm ply plate to bolt the wing to. The motor is a old kontronic gearbox with a tiny 2300kv motor, it should drive a 11x8 folding graupner prop.
Body
Body
20190429_204253.jpg
Ralkbirdy
Posts: 44
Joined: 25 Feb 2018, 13:10

Re: Solar-powered model planes

Post by Ralkbirdy »

The wing has a 0.8mm ply spar top n bottom over a 1/4 balsa spar. The lower surface is open 1/16 balsa with balsa LE / TE, the top surface is 3mm depron. The ailerons are built up using 3x1mm carbon TE and some more carbon sheet capping. This now seems strong with good torsional stiffness and a sharp TE. Hopefully this will be a better section with improved gliding.

With the geared motor I hope that it will have good power in sunshine and improved performance in slight clouds. All following Martin's good advice!
20190429_201123.jpg
20190429_201132.jpg
20190429_212423.jpg
Ralkbirdy
Posts: 44
Joined: 25 Feb 2018, 13:10

Re: Solar-powered model planes

Post by Ralkbirdy »

I have completed this new solar plane, it is about 500g. I found that the cog was further back than I wanted, so I moved the fin and elevator forward by shortening the boom. Then trimmed the tailplane attachment, finally cut holes in the fin removing some depron and covering it with mylar. This reduced the fin from 12.5g to 10.5g.
20190502_195941.jpg

I had a brief flight in modest sunshine which showed that it goes well but too short to do much else. The cog is at the end of the balance range, a dive test just kept diving. I have 3 possible props, this 13x10 is the largest and heaviest. The adaptor should be a spinner but I only have a 3.2mm collett and need to get a 4mm set. The extra nut is just a spacing washer for this odd but useful 4mm collett from another 4mm prop adaptor.

The red fin is a new balsa carbon lifespan job, bit mean at only losing another 1g down to 9.5g, so now trying a depron one without any balsa frame.
20190502_201436.jpg
20190502_200016.jpg
Post Reply