Black vs Silver panels?

[GoingElectric]

Important discussion topic.

I had a proposal and was shown the Hanwa Q.PEAK DUO ML-G10 (non BLK) is listing about 5w more output compared the BLK version Q.PEAK DUO BLK ML-G10.

It is simple physics that the cosmetics to hide the circuits also decrease the power output.
This like putting a filter on a camera, you get better color range, but need to compensate in exposure time and/or lower focal length.

The funny thing is I am finding the mfg do offer the non-blacks, but they are harder to find in their online catalog.
PDF files more likely list all versions, such as this Q-cell catalog,

https://media.q-cells.com/v/eyEEJh2F/

[bob-n]

J.P.M. Yes, I picked my words carefully because I don't know many of the important details, just enough to be dangerous.

My testing was done many years ago with a laser. I can't recall the laser wavelength, but we wore green goggles to block it, so it wasn't green. I don't even recall if the laser was visible. We never wanted to see the beam.

I do recall that the thickness of the wafer affected absorbtion, and we attributed it to standing waves in the silicon. It is possible that PV cell makers have engineered cell thickness for optimum efficiency. They certainly engineered the layers in the silicon sandwich (nitride, various oxides, metals, etc.) for efficiency as well as other unimportant things like reliability.

[J.P.M.]

If there is so much light shining through, why does not someone put a silver mirror coating
on the back so the light goes back again, raising panel efficiency approaching double?
Bruce Roe

I believe what Bob-n writes, but there's not as much transmission through a PV cell as a casual read of Bob-n's post might lead a reader to infer.
Note the words "silicon wafers" and "at many wavelengths". Wafers may not be silicon PV cells, and if they are, what thickness and type ?
Also, in which wavelength (bands) did the silicon wafers transmit ? PV cells can't use use wavelengths longer than 1.15 microns. Is part of the reason such wavelengths are unusable due to transparency in some wavelengths and that may have something to do with what Bob-n reports ?
Bob, comment ?

If A photon wasn't snagged on the first pass through a cell due to the cell's transmission characteristics at that photon's frequency, why would it get picked up do so on a second pass ? And, how would the reflected photons find some n type silicon on the front side to free up an electron ?
Besides, seems to me the cell would need to be physically altered to be, in effect, two cells facing in opposite directions with a common p type layer.

I've no data on PV cell transmission or its wavelength response characteristics to support my doubts, but I kind of wonder if much of anything substantial would be gained by back reflection surfaces which, I'd rather suspect would also tend to increase the cell's operating temps.

Respectfully,

[SunEagle]

If there is so much light shining through, why does not someone put a silver mirror coating
on the back so the light goes back again, raising panel efficiency approaching double?
Bruce Roe

Actually there are some panels that have a second set of cells below the first. What ever light goes through the first is harvested by the second. Again the cost of these panels are hard to justify when the additional harvesting is not much but does increase the overall efficiency of a panel.

[J.P.M.]

... I'm just repeating what the chief technology officer of Mission Solar told me. Same cells, different STC, only difference was the back sheet...

I understand what you write.

Except for why the published NOCT for the all back panels is slightly lower for the 320 W all black vs. the 315 W white back on the Mission Solar data sheets, which seems backwards to me, I'd put more faith in the differences discernible in the data sheets. I do note small differences in the dimensions and operating voltages and currents one to the other as is common.

[nwdiver]

Overall, I'm having trouble understanding how the color of a backsurface that sees VERY close to zero irradiance can have an effect on cell temp.

....... I'm just repeating what the chief technology officer of Mission Solar told me. Same cells, different STC, only difference was the back sheet...

[J.P.M.]

Many years ago, I measured light transmission through silicon wafers. It was quite substantial at many wavelengths, in many cases, 90%. PV cells are roughly 20% efficient. Some of the 80% lost must be going right through and either reflecting back or being absorbed. If it is being absorbed by the back sheet, then it is extra heat. The extra heat may not show up in STC measures. But if it is reflected back, it may have an additional opportunity to create electrical energy (standing waves, etc.) and then it would show up in STC.

Energy balance: What comes out = what goes in less what stays in. Assuming quasi steady state operation, if 1,000 W of energy hit a panel and 200 of those W turn into electricity, the remaining 800 W will be accounted for either as reflected from the panel (mostly from the glazing) or turned into heat and carried off by one of the 3 modes of heat transfer.

From lots of measurements and heat transfer correlations in the body of heeat transfer knowledge, I'm pretty sure not too much of what hits my array in the way of sunlight passes through the panels.

In my "you could just" younger days, I conjured up the idea that since only solar energy with wavelengths below 1.15 * EE-6 m (about 75 % or so of the total energy emitted by the sun is below that wavelength) is useful for silicon to convert sunlight to electricity, a surface coating that acted as a bandpass filter or as a selective reflector not unlike solar thermal selective absorber surface coatings, but sort of in reverse, might be able to reflect energy not useful to the solar conversion process before it got turned into heat and in so doing enable lower cell temps for the same POA irradiance.

[bcroe]

Many years ago, I measured light transmission through silicon wafers. It was quite substantial at many wavelengths, in many cases, 90%. PV cells are roughly 20% efficient. Some of the 80% lost must be going right through and either reflecting back or being absorbed. If it is being absorbed by the back sheet, then it is extra heat. The extra heat may not show up in STC measures. But if it is reflected back, it may have an additional opportunity to create electrical energy (standing waves, etc.) and then it would show up in STC.

If there is so much light shining through, why does not someone put a silver mirror coating
on the back so the light goes back again, raising panel efficiency approaching double?
Bruce Roe

[SunEagle]

Many years ago, I measured light transmission through silicon wafers. It was quite substantial at many wavelengths, in many cases, 90%. PV cells are roughly 20% efficient. Some of the 80% lost must be going right through and either reflecting back or being absorbed. If it is being absorbed by the back sheet, then it is extra heat. The extra heat may not show up in STC measures. But if it is reflected back, it may have an additional opportunity to create electrical energy (standing waves, etc.) and then it would show up in STC.

That is why bimodal solar panels can generate more power from the reflected light from below. Unfortunately the cost of those panels doesn't seem to off set the efficiency increase.

[bob-n]

Many years ago, I measured light transmission through silicon wafers. It was quite substantial at many wavelengths, in many cases, 90%. PV cells are roughly 20% efficient. Some of the 80% lost must be going right through and either reflecting back or being absorbed. If it is being absorbed by the back sheet, then it is extra heat. The extra heat may not show up in STC measures. But if it is reflected back, it may have an additional opportunity to create electrical energy (standing waves, etc.) and then it would show up in STC.

[J.P.M.]

The white vs black back sheet definitely has an effect but it's small. I got a tour of the Mission Solar factory with their CTO. He mentioned that the 315 and 320w panels had identical cells. The only difference was the 315s used a black back sheet while the 320s were white. The cooler temperature from the increased albedo (and maybe a little more light scatter IIIRC) gave it an extra 5w. So < 2% difference. Pretty negligible and IMHO the improved aesthetics is worth such a small price.... probably why almost all residential panels are black/black now.

The albedo an object or a surface receives is not a function of the color of that surface as your statement seems to infer.
In the same location and physical environment, a white surface sees just as much irradiation due to albedo from the surroundings as does a black surface. The albedo is f(surroundings), not the properties of the receiving object.

For reasons explained below, the backside of a nearly flush or small roof to panel gap roof mounted array will see little, if any, albedo. In any case, albedo (reflected irradiance) coming off an object will not reduce its temperature. Depending mostly on the properties of the surface of the reflecting object, that albedo may have the effect of making the object's temp. lower than if it was a blackbody (or as may seem counterintuitive higher, depending on the surface characteristics such as any wavelength dependent emittance that makes % absorbance > % emittance), but that's not the same as a temp. reduction, and that's not semantics or separating fly crap from pepper. So, to say "The cooler temperature from increased albedo ..." makes no sense. Emittance and albedo are not the same.

Also, for solar arrays sitting parallel to a roof with very small (roof to panel clearance/array length (or) width) dimensions, the relatively small contribution of albedo on the backside of an array to total solar production even in ideal situations is so small for common roof top applications that it's likely to be unmeasurable and can be ignored. Same thing goes for array temps.

Consult sources on thermal radiation heat transfer, particularly the "Handbook of Heat Transfer" by Rohsenow and Hartnett and you'll see what I'm talking about. Other authors: Siegal & Howell, Sparrow & Cess, Wiebelt, and others. See in particular the areas dealing with something called "view factors".

Bottom line: For most rooftop applications, or those applications with relatively small clearances between the back of panel and a parallel surface such as a roof, there will be little, if any measurable solar radiation in the form of albedo that gets to the backside of a panel or an array. To the degree that is a statement that describes an application, it's a true statement.

There will be however, a small temp. increase in panel temp. from the darker diamonds on the frontside that's the result of a black backsheet as little harbor notes.

Further, and for information if you're interested, any albedo that does sneak behind the array (through the small panel to parallel (roof) surface clearance), and then somehow manages to hit the backside of a panel in such a configuration will do so at an angle of incidence pretty close to 90 degrees making the average projection of that diffuse radiation on the backside of a panel closer to zero yet.
Further to that, just like frontside glazing reflectance increases as incidence angle increases, the reflectance of wavelengths we're talking about here by backsheet material increases as the angle of incidence increases. I don't know the reflectance of albedo as f(cos (incidence angle), wavelength) for the EVA material of most backsheets, but I'd guess that regardless of what it is, it'll be a lot closer to 1.0 than 0 for the high incidence angles we're talking about here.

All that comes down to very little of what's already a small amount of reflected irradiance that's had the additional and compounded reductions of small view factor and high incidence angle to sap what's left of the albedo on the backside of an array of any color will be so small as to be immeasuable.
.
Anecdotally, when I got my array, I also got a Davis weather station with a pyranometer that I can detach. When I designed the array, for several reasons, I made the roof to panel clearance high enough that I can get under the array - somewhere between ~ 30 and 35 or so cm. clearance/tile height/profile, depending on where the measurement is taken. I put the pyranometer in an orientation such that the sensor faced "down" on he underside of my 4 X 4 array at the NW corner of the NW "interior" panel. Under bright sun at several times of the day on a day near the spring equinox, I got readings of zero or 1 W/m^2. that seemed to confirm, or at last not contradict what I'd learned about radiation heat transfer.

On what you were told by the Mission Solar folks, that seems to be different than what they put on their datasheets, which in/of themselves seem a bit confusing, with a lot of the small differences in STC ratings due to small changes in rated currents and rated voltages as is tpical of most PV mfgs. I found, for example data sheets for 310 W and 320 W with both monolithic black and checkerboarded frontsides. Interestingly, to me anyway, both the 315 W and 320 W monolithic black panels had a published NOCT that was slightly lower than the 315 W and 320 W non all black panels.

On the backsheet color, for starters, look at how the STC output of a panel is measured (flash testing). The color of its back sheet is or ought to be, in any logical sense, irrelevant, and so have no bearing on a panels' STC rating. If that's a true statement, saying the increase in STC output of a panel is due to the color of its backsheet doesn't make sense.

Overall, I'm having trouble understanding how the color of a backsurface that sees VERY close to zero irradiance can have an effect on cell temp. when that surface sees very little to effectively no irradiance, or in most any case, calc'ed or measured, << 1% of the frontface irradiance, especially when 1.) all of that backside irradiance is diffuse, 2.) all of that small amount of incident backside irradiance has an incident angle that is, due to the geometry of most rooftop arrays, pretty close to 90 degrees.

[littleharbor]

On monocrystalline panels the white backsheet gives you the typical white diamonds at the corner junction of all cells. while I don't mind the look I do think the black backsheet, with its uniform look on the face of the panels looks better.

[nwdiver]

why and how do you think a white back sheet matters that much with respect to either aesthetics or cell temp./panel efficiency ?

The white vs black back sheet definitely has an effect but it's small. I got a tour of the Mission Solar factory with their CTO. He mentioned that the 315 and 320w panels had identical cells. The only difference was the 315s used a black back sheet while the 320s were white. The cooler temperature from the increased albedo (and maybe a little more light scatter IIIRC) gave it an extra 5w. So < 2% difference. Pretty negligible and IMHO the improved aesthetics is worth such a small price.... probably why almost all residential panels are black/black now.

[J.P.M.]

I think some are confusing black frames with black back sheets. I don.t think the black verses silver frame matters much, the white back sheet does.

Andy

I didn't get that impression. I thought folks were referring to monolithic looking "all black" panels with black frames and no lines visible lines on the front face behind the glazing.

As for the color of the back sheet, since probably most applications folks here may be talking about are residential rooftop applications with backsheets mostly hidden from view as well as having close to zero irradiance on them, why and how do you think a white back sheet matters that much with respect to either aesthetics or cell temp./panel efficiency ?

[PVAndy]

I tried to find 320w 60 cell panels with silver frames. Looks like Residential panels only come in black now.

I think some are confusing black frames with black back sheets. I don.t think the black verses silver frame matters much, the white back sheet does.

Andy