Solar Modules Available

[solarsystems]

Greetings to All!

I'm a newbie here to your site, and with my first post will probably violate some appropriate-use rule. But you guys can suggest a better way to start this thread if there is one...

A couple of you are posting about solar powered aircraft and in general appear to have difficulty in acquiring cells. My company specializes in purely custom solar products for non-residential use (modules and electronics) and has access to a _major_ cell provider...

As you're all noting, the cell vendors do not allow cells to be sold on the open market unless encapsulated in modules. I have access to encapsulation technology, cells (the ones you want, not 5% thin film stuff), and a business model that lets me pursue your kinds of volumes and developement needs.

That said, is there an interest from the community to try and work out a 'standard' UAV/RC solar module made from decent cells? How do we go about it using the RCGroups forum without disturbing feathers?

Thanks in Advance,
Thom
Solar Systems, LLC

[zik]

Sounds like a great idea to me. I'd potentially be interested in developing a high efficiency switch-mode power supply and battery management system to go with this if necessary.

[solarsystems]

Sounds like a great idea to me. I'd potentially be interested in developing a high efficiency switch-mode power supply and battery management system to go with this if necessary.

Hi Zik.

Initially, I was thinking just solar modules in a thin encapsulant. If folks need a lightweight peak power tracker and some kind of battery charging solution then maybe we can work it out as a complete package.

Anyone have thoughts on this? In particular a peak power tracker; do you guys use them now?

How much area do you think you have to populate with cells? What kind of working voltage do you want/need? If we provided modules with a prebonded fiberglass/carbonfiber/kevlar rear face sheet, would that work for your aircraft? etc...

Thom

[workshop]

Thom,

I manufacturer voltage producing devices too. :rolleyes:
See www.teslacoil.com

To answer some of your questions:

600-1000 square inches
Radios need 6 or 5.5V at 1A nominal (varies greatly from 75mA to 3A on my plane)

Camera systems need 5.5V or 12V and about 1A steady

Ultra light weight is most inportant so I'd forgo the backings.

Jeff

[CenTexFlyer]

What is the efficiency of the cell you propose?

[zik]

Initially, I was thinking just solar modules in a thin encapsulant.
This sounds very interesting to me Thom. What kind of cells are you thinking of? What weight and power output per area? Are they flexible?

[solarsystems]

There are a couple of cells possible. At one end of the spectrum, we have access to cells that can go up into the upper 20's%. At the other is the thin film stuff at 8-10%.

For the high end cells, I have one that is rated ~28%. It's been cut to size and measures 3" x 1.5" at 0.0065" thick. The mass is 2.3 grams. At 1000 sq inches this could hold 200 cells with aggressive packing for 175 Watts after tabbing, stringing, bussing, diodes, and encapsulation. Price for the cells alone runs 4 digits and doesn't start with a 1, 2, or 3. :eek:

Given the thickness of these cells, I would say that yes they are 'flexible'. There are two ways to look at flexibility, one is will it conform to a given surface, the second is cyclical flexural strength. In general cyclical flexure is a poor idea from a work hardening view. The more the interconnect to cell bond is stressed, the more that it wants to come undone. At first it won't stand out, but over time (especially in this application where the loading is continous) the tabbing ribbon will want to peel off the face of the cell. A true 'flexible' cell is going to be inherently low efficency (<10%) due to the amorphous nature of silicon (or other material).

The above does not rule out the flexibility of the module to drap over simple (ie non-compound) curves such as the upper skin of a wing. The goal is to reduce the radius any one cell must conform to. There are limits to the radius the cell and module can be forced to go over. However, the fact that you need the cells in series to all point in the same general direction limits the amount of curvature the module is exposed to. Given a simple airfoil, the majority of the surface is 'flat' from about the wing spar to the trailing edge. This is the area to cover. Placing cells from the spar to the leading edge puts too much of the cell pointing away from the rest of the module leading to complications electrically.

Encapsulation for these cells is typically done with a material like hot melt glue that doesn't get soft at high temperatures after manufacturing. The front is like teflon and the back is similiar to acetate only white. For the 6.5mil cells, encapsulation will be the majority of the mass and hence we need to establish a weight target for us to develop a process for.

The suggestion for a rear facing of composite material is so the builder can use the modules rear surface as a structural member of the wing layup. This way your assured good bonding to the wing upper skin. The rear sheet could be anything from .2 oz/ft carbon/kelvar tissue through 10 oz/ft fiberglass. It would be up to the community to establish a standard. If you want nothing on the back, that can be arranged as well.

Thanks for the feedback folks, I'm glad there is some interest. Please feel free to ping me with suggestions, questions, and 'what-ifs'. Bear with me if I can't get you pricing yet... :)

Thom

[air]

Hi Thom,
Those cells sound like they have an excellent power to weight ratio.
If my back of napkin calculations are correct, the 1000 Sq inches of cells should weigh about 657g for that power output of 175Watts.
Have you considered placing the cells within the body of a wing with transparent covering (between the ribs)?
If the cells were just lightly supported within the structure and connected using flexible interconnects it might be possible to isolate them very well from the stresses the wing itself encounters during flight.
It would reduce the percentage of wing area available for use by the cells though.

[dmgoedde]

Have you considered placing the cells within the body of a wing with transparent covering (between the ribs)?

I would be VERY interested to see a working solution. I tried this 10 years ago, using colorless and transparent MonoKote. Comparing pre-covering to post, the covering absorbed about 66% of the radiation the cells used; the current dropped to about 33% of no-covering test. This didn't suprise me, as the covering is polymerized hydrocabons, and these absorb strongly in the IR.

Maybe a much thinner covering (Saran wrap???) would absorb less of the required wavelengths?

Dean

[solarsystems]

Maybe a much thinner covering (Saran wrap???) would absorb less of the required wavelengths?

The trick here is really to get the covering in contact with the cell surface to reduce the number of optical interfaces. Going from air to encapsulation to cell is what these things are designed for. In some cases efficency can increase for an encapsulated cell over the raw cell.

Running the cells on the upper surface also allows air cooling of the cells. This is a significant advantage over cells running in an insulated volume.

The goal is probably to build the right structure to support the module which is light but _locally_ stiff. This means no one cell sees excessive flexure. The gap between cells can absorb lots of flexure if the cell is kept "flat".

Does someone have a wing profile that we could use as representative? With that, we can start working some of these issues..

Thom

[Ceros]

There are a couple of cells possible. At one end of the spectrum, we have access to cells that can go up into the upper 20's%. At the other is the thin film stuff at 8-10%.

For the high end cells, I have one that is rated ~28%. It's been cut to size and measures 3" x 1.5" at 0.0065" thick. The mass is 2.3 grams. At 1000 sq inches this could hold 200 cells with aggressive packing for 175 Watts after tabbing, stringing, bussing, diodes, and encapsulation. Price for the cells alone runs 4 digits and doesn't start with a 1, 2, or 3. :eek:

Given the thickness of these cells, I would say that yes they are 'flexible'. There are two ways to look at flexibility, one is will it conform to a given surface, the second is cyclical flexural strength. In general cyclical flexure is a poor idea from a work hardening view. The more the interconnect to cell bond is stressed, the more that it wants to come undone. At first it won't stand out, but over time (especially in this application where the loading is continous) the tabbing ribbon will want to peel off the face of the cell. A true 'flexible' cell is going to be inherently low efficency (<10%) due to the amorphous nature of silicon (or other material).

The above does not rule out the flexibility of the module to drap over simple (ie non-compound) curves such as the upper skin of a wing. The goal is to reduce the radius any one cell must conform to. There are limits to the radius the cell and module can be forced to go over. However, the fact that you need the cells in series to all point in the same general direction limits the amount of curvature the module is exposed to. Given a simple airfoil, the majority of the surface is 'flat' from about the wing spar to the trailing edge. This is the area to cover. Placing cells from the spar to the leading edge puts too much of the cell pointing away from the rest of the module leading to complications electrically.

Encapsulation for these cells is typically done with a material like hot melt glue that doesn't get soft at high temperatures after manufacturing. The front is like teflon and the back is similiar to acetate only white. For the 6.5mil cells, encapsulation will be the majority of the mass and hence we need to establish a weight target for us to develop a process for.

The suggestion for a rear facing of composite material is so the builder can use the modules rear surface as a structural member of the wing layup. This way your assured good bonding to the wing upper skin. The rear sheet could be anything from .2 oz/ft carbon/kelvar tissue through 10 oz/ft fiberglass. It would be up to the community to establish a standard. If you want nothing on the back, that can be arranged as well.

Thanks for the feedback folks, I'm glad there is some interest. Please feel free to ping me with suggestions, questions, and 'what-ifs'. Bear with me if I can't get you pricing yet... :)

Thom


..... and, in simple terms, if incorporated to the upper surface of a wing (stuck down with resin ?), would it be possible to spray a clearcoat over these cells, which could then be sanded down and polished up to achieve a sealed encapsulation a uniform wing surface - or would clearcoating them be a no-go?

[solarsystems]

..... and, in simple terms, if incorporated to the upper surface of a wing (stuck down with resin ?), would it be possible to spray a clearcoat over these cells, which could then be sanded down and polished up to achieve a sealed encapsulation a uniform wing surface - or would clearcoating them be a no-go?

Ceros,
Unfortunately, we can not provide raw cells (ie not already encapsulated). However, that's not to say that you're too far off the mark. The process we have in place would leave a glass-like finish over the cells. This surface could be laid face down in a wing mold, and the rest of the wing built up over it (think of the module as a thick decal/paint job).
If you have an existing wing, the module can be bonded to the surface with light pressure and the edges feather-filled in.

Does someone have a set of wing molds we can use? I could do a layup over the modules and set the video on youtube for everyone to see what we're talking about...

Thom

[Ceros]

Fine, I've got that - no prob's.

Okay, so you lay out the cells, put in all the diodes, wire it all up and lead your wires out of the wing where-ever you want (by the root I presume). Then you lay up over the back of the cells the carbon fibre/glass fibre, cure it all.

Done - now you take the wing out of the mould and what you have is a surface similar to what one would find on a curved shower wall that had being surfaced with small tiles.

Would it not be possible to now "skim" the whole lot with a thin clear-coat, that would feather in the gaps between the tiles, and secondly, re-introduce the curved profile over the skin - or is it not possible for whatever reason to cover the cells with anything - however thin and clear it was?

If not - where does the problem lie: is it something to do with loss of cell efficiency because of the clearcoat (however thin and optically clear it could be made), or does it have something to do with coefficients of expansion, cracking of the clearcoat and or similar problems.

If it was posssible what drop in cell efficiency/performance could one expect to experiance (%), and how would that be related to opacity of the clear coat (I would think that even a 100 - 250micron clear coat is going to introduce some drop in cell performance - however opticaly clear the clearcoat could be formulated).

The downside of course that one would have to accept is that once all sealed up, getting access to any cell would result in damage to the clearcoat and any wiring sealed in it - which means: crash and its not a case of repair but a case of remake and replace.

I would certainly be interested in contributing a mould (and material) if the above was viable

[solarsystems]

I would certainly be interested in contributing a mould (and material) if the above was viable

Ceros,
I think we're getting close. The first steps you described I agree with. The difference is that instead of opening the mold and finding tiles with gaps like grout lines in ceramic tiles, you would find a smooth surface.

I propose using an optical-grade epoxy as the first layer in the mold, followed by .75oz glass and then the cells with the wiring pre-wired. I think this could be accomplished by using a resin-transfer-method approach. The cells are flexible enough to handle a reasonable curve, so they would follow the airfoil contour. At the end of the first bag cycle, the cells would be encapsulated in the mold almost as if they were paint.

You're right in that the cells are pretty well fixed in the wing. The leading edge of the cells should be reenforced against impact with a layer of extra glass. Something between the leading edge of the wing and the leading edge of the cells would provide some protection. The cells are sensitive to breakage, but a broken cell isn't always a dead or even significantly degraded cell. As long as there is a path from the individual pieces of the cells to the rest of the circuit, then the performance hit is minimal. It's when silicon is isolated from the galvanic circuit that things get ugly.

We could try and think of ways of isolating the cells from the structure so they can be removed. One possibility is to place a 1 mil teflon sheet over the back of the cells to keep the next layer from bonding directly to the back of the cell. This way only the edges of the module get epoxied to the wing. If the wing cracks up, rip the wing off the modules around the edges and salvage the cells. This might also limit the amount of loading the module sees...

About efficency; a thin layer of encapsulant can actually improve cell performance as the cell is designed to optically interface to something besides air. It really doesn't matter what it interfaces with, so long as it's something with an index of refraction between air and silicon. If the layer gets too thick, then the opacity of the layer because an issue. UV degregation is another important factor. At first the thin layer is OK, but as it sees more UV from the Sun, the layer may yellow or fog badly. Most epoxies and fillers fall into this catagory.

Thom

[Ceros]

Well, all I'm concerned about is maintaining a clean curve to wing surface - and if you can cover the cells with an opticaly clear epoxy, then thats all I need to know - thats good enough for me.

Exactly how you go about it is something best left for those in the know to deal with i.e. your good self.

And one last question before i committ myself - just how much weight do you anticipate solar cells and incorporating them will add to the wing as a perentage of the overall wing mass/weight?

[solarsystems]

Ceros,
We're still working out the how part. I think it comes down to having a set of molds that we can do the vacuum bag layup in. We have done a previous encapsulation job for a sailplane-type craft. It was done with something that I think you guys might think is too heavy. I'm trying to work out if we can get a mold and do the encapsulation in the builders mold and send it back ready for the next stage of wing layup. This would ensure a 'perfect' clean surface (depending on the quality of the builders mold!).

As for the mass fraction, it depends on the cell selected, the encapsulant selected, and the wings mass. The usual encapsulant is similar to the stuff in house top modules, but _really thin_ (<=1/32"), no glass, and about 20g/cell (4"x4") (includes cell, encapsulant, diode, and wiring). I'm hoping to use clear epoxy and drive this down to something on the order of 10-15 grams.

We're all still very much in experiment planning mode here! :D

Thom

Well, all I'm concerned about is maintaining a clean curve to wing surface - and if you can cover the cells with an opticaly clear epoxy, then thats all I need to know - thats good enough for me.

Exactly how you go about it is something best left for those in the know to deal with i.e. your good self.

And one last question before i committ myself - just how much weight do you anticipate solar cells and incorporating them will add to the wing as a perentage of the overall wing mass/weight?

[skate5290]

I am very much interested in this solar solution. Are there any new developments?

[solarsystems]

I am very much interested in this solar solution. Are there any new developments?

Skate:
It looks like we're settling on putting together single cells with thin epoxy encapsulation on the upper surface and a thin plastic film on the back. Under the film, the wiring will be soldered in and brought out.
This will allow designers to select the quanity and layout while letting us use the thinest possible encapsulation without retooling.

I'm going to be putting together the epoxy system here in the next couple weeks. Bear with me.. :)

Thom

[aeroanalysis]

There are a couple of cells possible. At one end of the spectrum, we have access to cells that can go up into the upper 20's%. At the other is the thin film stuff at 8-10%.

For the high end cells, I have one that is rated ~28%. It's been cut to size and measures 3" x 1.5" at 0.0065" thick. The mass is 2.3 grams. At 1000 sq inches this could hold 200 cells with aggressive packing for 175 Watts after tabbing, stringing, bussing, diodes, and encapsulation. Price for the cells alone runs 4 digits and doesn't start with a 1, 2, or 3. :eek:


Thom

Thom,

I sent you an e-mail a while ago and haven't received a reply so I will ask again here. Could you please explain why someone interested in PV cells would be buying cells from you, and not purchase them direct from Sunpower, for example? Do you offer a discount from the manufacturers price? Is it your integration services you are offering?

-Daniel

[solarsystems]

Thom,
I sent you an e-mail a while ago...
-Daniel


from Thom Kreider <solarsystemsllc@gmail.com>
to "aeroanalysis @ RC Groups" <dnhatfield@ucdavis.xxx>
date Wed, Apr 16, 2008 at 10:53 AM
subject RE: PV for solar UAV's
mailed-by gmail.com

"Daniel,

No offense taken, you're absolutely right. Part of the reason to go through us is that cells are hard to source from the manufacturers these days, and Sunpower in particular. We have a history of acquiring cells from Sunpower and think we can continue barring any issues on their end.

The other reason is that at the vehicle level, the solar array should be just another system to integrate vs. develop from sratch. The vehicle builder is an integrater of subsystems that work, and we provide working solar subsystems.

In a collegiate environment; some of the point of building the system is integration and some is subsystem development. We respect that and encourge that effort.

Unfortunately, our agreement with xxxxxxxx doesn't allow us to sell raw cells; only xxxxxxxx is able to and they will not by their choice. As you noted, demand is high and none of the mainstream, medium efficiency manufacturers are selling cells regularly.

Is UC building a vehicle now? The last I talked to UC, it was about solar car racing...

Thom

Via mobile phone"

[aeroanalysis]

Thom,

Thanks for the prompt reply. Could you give me a ROM on a 1000 Watt installation of 25%+ efficient cells? How do the costs compare for triple junction, versus improved and ultra improved triple junction?

You can send your response in an e-mail to aeroanalysis@yahoo.com if you like.

-Daniel

[solarsystems]

You can send your response in an e-mail to aeroanalysis@yahoo.com if you like.

-Daniel

Email sent!

Thanks,
Thom