MyGeekShow - Building the Solar UAV Prototype - Blogs

Hugues July 4, 2015 at 12:47pm

Hi Trent, so you are are still alive! Where were you for the last 6months? Lots of people posted this question on your YouTube channel without response. Glad you are back and keep us posted!

Trent at MyGeekShow July 4, 2015 at 12:32pm

Martin - Love it! I don't take those personally at all, and welcome the input.

I was a bit anxious about roll stability, and sure enough, this morning's flight didn't last long : )
You are correct - the box was really just to get this in the air and set a baseline to work from.
Your Re stuff is spot on also, clearly I need to spend some time on airfoil design understanding.
Great links! I've looked at them all and was sucked in to all the data that was there. There are other teams other there that have done similar things too - much to be learned from them also.

GliderUAV July 4, 2015 at 12:31pm

Trent,

Good luck with this new project.

Just a short comment. You say you want to fly as slow as possible, but that may not be best. I don't have proof but I have a feeling that optimal airspeed depends on airframe, motor efficiency and climb performance. It may even be smarter to climb, glide, climb instead of continuously running the motor. And, kill the motor in thermals!

Kees

Martin July 4, 2015 at 12:05pm

I would like to make a few comments regarding the design. Please don't take any of them personally.

No ailerons and no dihedral. Good luck getting the plane to turn. It will probably roll, but the turn radius will be very large.

Although the construction is practical the plane is still pretty heavy compared to molded foam planes.

Aerodynamics - a box fuselage and Hershey bar wings are not the way to go for endurance. Is span is limited then the wing should have elliptic lift distribution for lowest induced drag. That's what matters at low speeds.

Also, the 10 cm chord might be a bit too short. The Reynolds number will be pretty low. If flying at 10 m/s then Re will only be 70 000. Clark Y will not perform well at such low Re. Re should be at least 200 000 at the wing root. 70 k at the tip is ok if it's a multi panel/airfoil wing. The formula for Re is Re=v*l*70000 [v (speed, m/s); l (chord, m)].

I know it's on only a first prototype, but you should keep the above in mind nevertheless.

I would suggest you look at the designs of Mark Drela. They are probably the best designs available in public domain --> http://www.charlesriverrc.org/articles.htm

If you want a plane with no ailerons then Bubble Dancer is a good way to go.

This is one great design. They've done as much as possible to reduce drag -->

http://www.tu.no/migration_catalog/2005/07/24/solong-info/binary/So...

I've read that they only landed the plane after 48 hours because they just got bored.

In conclusion, don't design your own wings (and stabilizers) if you are a novice and want efficiency. There are plenty of very good designs you can use.

Charles River Radio Controllers - Articles & Tips Index

Trent at MyGeekShow July 4, 2015 at 11:20am

Thank you all for your feedback! It is an exciting project with many challenges to overcome. There are quite a few options available to me to achieve a successful end to this mission, I'm confident in time I'll get a full 24 hour flight. It may not be a 24 DAY flight, but I'll confident a 24 hour flight is possible with a reasonable budget and hobbyist level of materials / skills.

johnkowalsky July 4, 2015 at 10:25am

Have you done the math on it ? How much power do you need to sustain a flight (with all the electronics**? Calculate per hour and convert into Wh / 24h. From the point just see how much surface area you need for solar panels

If this is your 50% model you've got roughly 0,6 sq meter on the wings of a 100% scale model.

Assuming 1k W of light X 0,6 sq m X 7,5%* efficiency = 45W. That's for like maybe 6 hours per day max so roughly 270-300 Wh per day. Then you have to take all the losses when charging into consideration.

Frankly it does not look good.

*efficiency of 7,5% - yah I know that there are panels that can achieve 15% but it isn't going to face the sun 100% of times. Personally I think it should be more like 4% in the equation as it would make more sense to use thin film panels.

**electronics - measure the power usage of the APM, rc, and telemetry. I suspect that's going to consititute a large % of power usage.

*** Your video mentions the 180%. That's not realistic at all. You're going to get maybe 6 hours of good sun exposure per 24h which means you need to be able to generate 400% (assuming that charging is 100% efficient). The real figure might be closer to 800%

Gary McCray July 4, 2015 at 9:31am

Hi Trent,

Great and this is a really ambitious undertaking.

I am afraid that getting one to last all night is seriously problematic,

In addition to using 18650s you really need to use high efficiency solar cells which cost an arm and a leg although standard cells have now got up into the 13 percent range, you really need 18 to 25 percent cells.

Sometimes you can get slight rejects for academic purposes from the manufacturers at a considerable savings.

There are some previous blogs on here (a year or 2 ago) by people who have attempted to just make a solar plane work at all and some of them include references to the high efficiency cells.

The biggest thing with getting it to fly over night is to spend all day climbing (and have enough cells to make a reasonable rate of climb possible. And then spend the night descending as slowly as possible.

If you haven't already (you probably have), you should check out the Swiss solar impulse manned flight they not only detail the requirements, but present a working strategy as well.

http://www.solarimpulse.com/

Of course their may be a small problem with the FAA if you endup having to climb to 15,000 feet or so.

I definitely wish you all the best with this, please keep us posted.

Best Regards,

Gary

Rana July 4, 2015 at 7:19am

Wish you all the best !

benbojangles July 4, 2015 at 6:03am

Hello, it's a good idea. From what I can gather from the Solar Impulse, it is essentially a glider, and only uses power for a couple of hours to hoist it back up high & also using high pressure thermals. So, from this thought I think the Panasonic Lions are a good idea, At least triple wingspan using Gary's points about balsa aircraft, and flying in zero wind and full sun. Not an easy concept to prove from where I live, but possible. I guess it's going to have to be the desert for testing. For solar cells, I would learn to make your own cells using using solar wafers and adhering them on the wings, tail & fuselage. Best of luck, happy to assist if I can.

Trent at MyGeekShow July 4, 2015 at 3:52am

Gary! Glad to see you also! I've been considering balsa also, in fact all materials in general (CF, fiberglass, etc). I'm going to see how efficient these materials are from a square cm perspective. I'll definitely check out that link, 30 years in production they say! Thats quite a tenure...

It will be interesting to see about the higher altitudes. Right now I'm going to have to do it on a sunny, low wind 24 hour day. If I really want to get serious, and get higher (above the clouds and low altitude weather conditions) I'll obviously need permission from the FAA or something. It's going to be interesting!

Comments