diy solar

diy solar

New, more efficient solar panels coming? Hopefully.

Sennen

Solar Enthusiast
Joined
Dec 7, 2023
Messages
260
Location
USA
They seem to think it will be much cheaper to produce than current panels, too.


 
Lower price doesn't matter much anymore because we're buying PV panels for $0.20/W in the US (think they're $0.10/W in China).
Cost is beginning to be dominated by mounting hardware. And it is eclipsed by labor.

More efficient so less area, less mounting hardware, less labor would be a savings.
But only if not more cost per watt. And just as durable, so we don't have to pay for installation twice as often.

Labor, permits, etc. are the majority of cost today, like 2/3 of total for residential at least.
We pay $3/W for rooftop GT PV, with parts being about $1/W or less.

Compare to Australia, $1/W turnkey installed.

There are some initiatives to streamline permitting.

 
Certainly not against higher performance panels, and lower cost is always welcome.

Last batch from Signature were $0.30 / watt, so I've got no complaints.
 
Heck I can buy panels on Ebay that must be more efficient than 36% based on size now. Not sure how they do it but 400w from 11in. by 11in panel. For the low price of $49! That would be just a bit more than a penny a watt not including the free controller and accessories.

No need to wait.
 

Attachments

  • 1711886956138.png
    1711886956138.png
    371.5 KB · Views: 23
  • 1711887036379.png
    1711887036379.png
    601.5 KB · Views: 23
Heck I can buy panels on Ebay that must be more efficient than 36% based on size now. Not sure how they do it but 400w from 11in. by 11in panel. For the low price of $49! That would be just a bit more than a penny a watt not including the free controller and accessories.

No need to wait.
I don’t believe the ad is accurate. I’ve seen ads like this that claim a high wattage that doesn’t ad up with the voltage and current specs. If you could get a 400 watt panel in an 11 inch square panel for $49 that would be amazing and also huge news 😎
 
Lower price doesn't matter much anymore because we're buying PV panels for $0.20/W in the US (think they're $0.10/W in China).
Cost is beginning to be dominated by mounting hardware. And it is eclipsed by labor.

More efficient so less area, less mounting hardware, less labor would be a savings.
But only if not more cost per watt. And just as durable, so we don't have to pay for installation twice as often.

Labor, permits, etc. are the majority of cost today, like 2/3 of total for residential at least.
We pay $3/W for rooftop GT PV, with parts being about $1/W or less.

Compare to Australia, $1/W turnkey installed.

There are some initiatives to streamline permitting.

I don’t believe the ad is accurate. I’ve seen ads like this that claim a high wattage that doesn’t ad up with the voltage and current specs. If you could get a 400 watt panel in an 11 inch square panel for $49 that would be amazing and also huge news 😎
You could of course construct a stacked solar 'cell' comprised of say 2 layers of perovskite/sillicon PV mounted onto a 'polished nickel substrate reflector'. Assuming the layers have 20% conversion efficiency each then total conversion efficiency would be a little under 60%. (20 + 16 + 12.8 + 10.2). so a 1.8m x 1.2m panel would give about 1200W (compared to current 400W for same size) figures are approximate. Cost would be about 20% more for 200% more power. An added benefit is that the heating due to thermal effect would be reduced by 40 to 50% so the panels would run cooler resulting in lower efficiency loss over time and increased lifetime.... That's Win, Win, Win, Win imho.....

More than 2 layers give diminishing returns so more than 3 or 4 layers is of no economic value.

Oh, i almost forget adding a fresnel lens to the top layer, helps capture more of the incident light and reduces the emitted (reflected) light increasing the efficiency by a further 8%....
 
Last edited:
You could of course construct a stacked solar 'cell' comprised of say 2 layers of perovskite/sillicon PV mounted onto a 'polished nickel substrate reflector'. Assuming the layers have 20% conversion efficiency each then total conversion efficiency would be a little under 60%. (20 + 16 + 12.8 + 10.2). so a 1.8m x 1.2m panel would give about 1200W (compared to current 400W for same size) figures are approximate. Cost would be about 20% more for 200% more power.

Are you sure about 20% more cost?

One P/N diode junction in single-crystal cells is sliced silicon ingots with implant baked in. Assembled panels are about $0.10/W in China, $40 for 400W?
Two layers means thin-film processing on top, vacuum deposition. At what cost? About the price of a flat-screen TV glass?
 
Are you sure about 20% more cost?

One P/N diode junction in single-crystal cells is sliced silicon ingots with implant baked in. Assembled panels are about $0.10/W in China, $40 for 400W?
Two layers means thin-film processing on top, vacuum deposition. At what cost? About the price of a flat-screen TV glass?
What % of the Assembled panel cost is down to the cost of the silicon PV segment when you factor in the Assembly , Test, Shipping, Taxes, Profit margin, etc...

So what you do is take a single cell and stack another on top, (chip on chip has already been done), In this case there is no need for the back metal on the top slice (just grind and polish), but the bottom slice should have a top solderable metal (already in use in some FABs). You bond the two slices together either by ultrasonic or thermal welding (basically melt the top metal of the bottom slice), again this is an existing die attach process (nothing new here).

Furthermore, you can use a thinner slice from your ingot (equals more slices)

Now build your panels using the stacked cells....(the process is the same)...

The geometries are very large so no critical alignment issues.

Also the top of a solar panel is a sheet of protective glass (just like your TV panel.
 
Last edited:
What % of the Assembled panel cost is down to the cost of the silicon PV segment when you factor in the Assembly , Test, Shipping, Taxes, Profit margin, etc...

So what you do is take a single cell and stack another on top, (chip on chip has already been done), In this case there is no need for the back metal on the top slice (just grind and polish), but the bottom slice should have a top solderable metal (already in use in some FABs). You bond the two slices together either by ultrasonic or thermal welding (basically melt the top metal of the bottom slice), again this is an existing die attach process (nothing new here).

Furthermore, you can use a thinner slice from your ingot (equals more slices)

Now build your panels using the stacked cells....(the process is the same)...

The geometries are very large so no critical alignment issues.

Also the top of a solar panel is a sheet of protective glass (just like your TV panel.
Sounds interesting, though outside of my expertise. My question though is if this idea is feasible, why aren’t they manufacturing panels like this?
 
Sounds interesting, though outside of my expertise. My question though is if this idea is feasible, why aren’t they manufacturing panels like this?
You mean perovskites?

Cost; they are far more expensive than single crystal silicon.
Efficiency; they are not that much more efficient. Good pervoskites can hit 25% these days, but single crystal silicon can hit 23.5%.
Manufacturability. No one has manufactures perovskites in large quantities yet,
 
Heck I can buy panels on Ebay that must be more efficient than 36% based on size now. Not sure how they do it but 400w from 11in. by 11in panel. For the low price of $49! That would be just a bit more than a penny a watt not including the free controller and accessories.

No need to wait.
Well shucks , based on these panels size and power , I could get approx 120,000 watts of PV power on my RV roof….I missed that boat…story of my life…
 
Agreed Perovskites are more costly, than Silicon PV, but my calculation was based on a layer efficiency of 20%

Top Layer 20% (of the incident 100%) = 20%
Bottom Layer 20% (of remaining 80%) = 16%
Reflect back
Bottom Layer 20% (of remaining 64%) = 12.8%
Top Layer 20% (of remaining 51.2%) = 10.24%
Total conversion (Assuming no other losses) = 59.04%

I do expect there to be some losses in the reflection and transmission steps, but this is offset by the present efficiencies (>23%).

Also there are some simple techniques (nothing new) that could be used easily and cheaply to increase the efficiency further.

A couple more things that are in development...

Reel to Reel 'printing' of PV on polymer 'like newsprint', although not currently as efficient as existing PV, it is only a fraction of the cost.

Transparent PV on Polymer windows, (again not as efficient, but potentially useful on High Rise office blocks.
 
Sounds interesting, though outside of my expertise. My question though is if this idea is feasible, why aren’t they manufacturing panels like this?
Why have they only just started making bifacial panels in volume ?,
Why have then combined PhotoElectric and ThermoElectric effects in the PV Panels.... (Solar & Thermal) ?

My Guess is its commercial thing, "Profit Margin 'v' Competitive Advantage 'v' Market Forces, 'v' Demand".
 
Why have they only just started making bifacial panels in volume ?,
Why have then combined PhotoElectric and ThermoElectric effects in the PV Panels.... (Solar & Thermal) ?

My Guess is its commercial thing, "Profit Margin 'v' Competitive Advantage 'v' Market Forces, 'v' Demand".
Yup, slow roll out to make everyone buy everything twice. :)
 
Agreed Perovskites are more costly, than Silicon PV, but my calculation was based on a layer efficiency of 20%

Top Layer 20% (of the incident 100%) = 20%
Bottom Layer 20% (of remaining 80%) = 16%
Reflect back
Bottom Layer 20% (of remaining 64%) = 12.8%
Top Layer 20% (of remaining 51.2%) = 10.24%
Total conversion (Assuming no other losses) = 59.04%
They don't add like that. A 20% efficient cell doesn't transmit the other 80% as usable light energy. In tandem cells (like perovskite + silicon) you lose most of the energy when it goes through the transparent cell, so your gain gets smaller and smaller. Keep in mind that the most efficient three-junction cells on the planet only get to 39% - and they are tens of thousands of dollars per watt, usable only for spacecraft and the like.
 
Why have they only just started making bifacial panels in volume ?,
Why have then combined PhotoElectric and ThermoElectric effects in the PV Panels.... (Solar & Thermal) ?
They've been making bifacial panels in volume for at least five years now.

Combined thermo and photovoltaic panels don't work well. To get a thermovolatic cell to work you need high temperatures on one side and low temperatures on the other. So not only do you need active cooling of the back, you have to run the front hotter, and silicon PV cells don't work well at high temperatures.
 
They don't add like that. A 20% efficient cell doesn't transmit the other 80% as usable light energy. In tandem cells (like perovskite + silicon) you lose most of the energy when it goes through the transparent cell, so your gain gets smaller and smaller. Keep in mind that the most efficient three-junction cells on the planet only get to 39% - and they are tens of thousands of dollars per watt, usable only for spacecraft and the like.
I agree there are some losses but the laws of thermodynamics and conservation of energy still apply so, if 20% of the photons are converted into electrons in the first layer, the other 80%(ish) are transmitted to the next layer down (because they cant just disappear). The fundamental problem with solar panels is they get very hot because the majority of the photons are absorbed in the black bottom layer which is neither photoelectric or thermoelectric. Extracting more of the energy as electrons will inherently reduce the operating temperature and there may be other technologies that haven't been considered as a means of doing this. As i said previously reflecting the transmitted light so it passes through the layers twice should double (almost) the conversion rate, indeed a bifacial panel is basically this but with a non transparent interstitial layer.
 
They've been making bifacial panels in volume for at least five years now.

Combined thermo and photovoltaic panels don't work well. To get a thermovolatic cell to work you need high temperatures on one side and low temperatures on the other. So not only do you need active cooling of the back, you have to run the front hotter, and silicon PV cells don't work well at high temperatures.
 
They've been making bifacial panels in volume for at least five years now.

Combined thermo and photovoltaic panels don't work well. To get a thermovolatic cell to work you need high temperatures on one side and low temperatures on the other. So not only do you need active cooling of the back, you have to run the front hotter, and silicon PV cells don't work well at high temperatures.
Rather than thermovoltaic + PV, I fancy water-cooled PV. Just to give a bit of a preheat addition to the house heating in winter; keeping
the PV running cooler is a bonus. Sadly:
  • anything involving a plumber is uneconomic around here
  • the PV cooling becomes less effective in midsummer, just when it's needed
 
Rather than thermovoltaic + PV, I fancy water-cooled PV. Just to give a bit of a preheat addition to the house heating in winter; keeping
the PV running cooler is a bonus. Sadly:
  • anything involving a plumber is uneconomic around here
  • the PV cooling becomes less effective in midsummer, just when it's needed
I have been thinking about using water cooling for my PV panels, and using the heated water for my jacuzzi/spa, bonus is to use some of the extra solar to pump the water around. I envisaged attaching microbore copper tube to the back of the panels, but in hindsight it may freeze up in the winter and probably cause leaks when it thaws out in spring...
 
I agree there are some losses but the laws of thermodynamics and conservation of energy still apply so, if 20% of the photons are converted into electrons in the first layer, the other 80%(ish) are transmitted to the next layer down (because they cant just disappear).
Right. They are converted to heat.

Consider sunlight hitting a piece of black metal. Do the photons go through it? Nope. They convert their energy into heat.
Extracting more of the energy as electrons will inherently reduce the operating temperature and there may be other technologies that haven't been considered as a means of doing this.

Right. That's a good effect of higher efficiency.
 
Back
Top