Scientists invent double-sided solar panel that generates way more electricity | Hacker News Reader

Scientists invent double-sided solar panel that generates way more electricity

independent.co.uk

52 pointsthunderbong28 comments

Bifacial modules are nothing new, it's been commercially available to domestic markets in my country for a while. It's not popular for domestic use because they are often mounted flat on a pitched roof. I think the value of bifacials come in effect in solar farm contexts, where the back of one row of modules is illuminated by the sun reflecting off the row behind it.

The specific novel-ness in this article relates to the new perovskite chemsitry approaching commercial production. Making a new solar cell chemistry bifacial does not really come over as very novel imo?

Also, the article says "generates VASTLY more electricity" (emphasis mine), but then in the body it says it's 20% more compared to single face panels?

> Also, the article says "generates VASTLY more electricity" (emphasis mine), but then in the body it says it's 20% more compared to single face panels?

A 20% increases in solar electricity production fits the bill for vastly IMO. If I was 20% taller I would be VASTLY taller.

I just installed ground mount bifacial panels at my house. The marketing said that they generate “up to 30% more” electricity being bifacial. I concur that there are significantly more hours of production (albeit at lower amounts during these shoulder hours) than if they were monofacial.

All of that with the cost of only doubling the amount of solar cells.

Unless solar cells are almost free, that's not a vast improvement. That's only cost-positive if the cells are less than 1/5 of the total cost of your solar farm. I don't think any place qualifies.

The support structure is also a large cost in solar power farms. So this would not double cost. Also, the backside will not need as much cleaning

If it gave a 35% improvement, it would be an almost universal small gain. If it gave you a 25% improvement, it would be a good candidate for one place or another that face unusually high land or labor costs.

But no, at 20%, I don't believe anybody will want this. Maybe in a decade that changes, but not right now.

1) Solar panel hardware costs are typically in the ballpark of 20-40% of solar farm costs.

2) You've assumed the panel costs double, which is likely false.

3) solar generation costs have been falling exponentially and upgrades are incorporated into various installations and new developments all the time.

Probably will provide significant benefits at higher latitudes. Especially where there is snow and panels are mounted 45degrees plus.

But here near the equator where view of the sky is the only thing that matters, they are a non-starter. Perhaps in solar farms where panel mounting is 4m or higher?

> All of that with the cost of only doubling the amount of solar cells

Why should that change if vastly is the correct adjective or not? If I overpaid by a factor of 10 to be 20% taller I would still be vastly taller, regardless of how much I paid.

Adult male height mean is 70 inches. Standard deviation is 3 inches. A 20% increase is 4.67 standard deviations above the mean, which is sensibly describable as a vast improvement in terms of human height. But your statement about "vast" for height is only true because of the particular distribution. It doesn't generalize.

> Adult male height mean is 70 inches. Standard deviation is 3 inches.

If I was 20% taller I would be well over 7ft tall.

> because of the particular distribution. It doesn't generalize.

and IMO it fits solar electricity production increases.

> If I was 20% taller I would be well over 7ft tall.

ok?

> and IMO it fits solar electricity production increases.

I think it's distracting more than anything. We can, at best, debate whether the distribution of power generation is similar to the distribution of height to get any usefulness of out of this.

> debate whether the distribution of power generation is similar to the distribution of height to get any usefulness of out of this.

and IMO they are, thus my original comment.

Bifacial panels don't double the number of cells. There is just a single layer of cells that is exposed on both sides.

> single layer

Bifacial panels have two layers of transistors on the silicon subtrate. One on the front and one on the back. Monofacial panels have just one layer. See for example https://en.wikipedia.org/wiki/Bifacial_solar_cells#/media/Fi...

They have twice as many transistors, but they don't have twice as many cells.

I don't think it doubles the quantity of solar cells. I think it is a single cell that is reactive on both sides. I don't think they extend the shoulder hours I think it extends the overall surface area to gather more indirect light.

Mounted over snow, water, white roof would be the optimal set-up.

Perovskite panels have been available for several years https://columbusenergy.com/

EDIT: wrong company - this is correct https://sauletech.com/technology/

Have they dealt with the lead escaping the encapsulation problem yet?

They seem to be using two different percentages which is confusing. More details can be found in "Highly efficient bifacial single-junction perovskite solar cells"[0]

Of the new panels; the front is 23% efficient and the back side is 20% efficient (see also: "an efficiency of about 91-93% of the front") so overall you collect 43%, vastly more than 26% ("current solar cell technologies") or even 30% (silicon-perovskite solar cells in lab conditions) of the available power.

[0]: https://www.sciencedirect.com/science/article/abs/pii/S25424...

>the front is 23% efficient and the back side is 20% efficient (see also: "an efficiency of about 91-93% of the front") so overall you collect 43%

That doesn't seem quite realistic, since the back side isn't getting full sunlight like the front unless the surface behind the panel is a perfect mirror. Depending on what's reflecting onto it, it could still end up worthwhile.

The efficiency is most likely a measure of solar radiation striking the surface of the bottomside not the topside.

Correct, which is why you can't add them and claim 43% efficiency total. If each side was 60% efficient, you wouldn't add them to 120%.

Neat. I'm assuming these would be fairly limited to niche installations, since the cost would almost certainly not be competitive to current panels in an installation where you have the ability to just add more area (not to mention the difficulty of perovskites in general).

I'm trying to think of places these might be worth it -- I guess space constrained areas where ground albedo is high?

Maybe I'm misunderstanding - are the solar panels that 'face down' relying on light that passes through the top panels, or just on ambient light, or is some sort of reflector required underneath? If so, is this going to be a problem for agrivolatric installations?

Frustrating that the study this article is reporting on (linked at the end) https://www.cell.com/joule/fulltext/S2542-4351(23)00224-6 is paywalled. To save others' time, the "download full issue" obtains a .zip of PDFs, but not including the relevant article, only other articles in the issue that presumably someone paid to make open access.

> up to 20 per cent more power overall

Compare with the bifacials used in this (6 mo old video) real-world comparison.

https://www.youtube.com/watch?v=mzqjCsZKUYQ

Why not solar spheres?

Reply on news.ycombinator.com