-
-
I think all the Neon 2's use "double sided cells" in a conventional backsheet construction. They have released a series of true bi-facial panels now, though.CS6P-260P/SE3000 - http://tiny.cc/ed5ozxComment
-
Yup - Guilty as charged - and proud of it.
I'm very biased against the deceit of those who peddle stuff with bogus claims, and biased against the moronic ignorance that's exhibited by most of those who sing the praises of bifacial panels but seem to know little to nothing about the PV process, and less about the solar resource that makes the process work, and yet who continue to spout the illogic, without understanding how the lack of solar resource (the available and useful sunlight) makes most of the claims associated with bifacial panels impossible to attain in any but the most contrived applications, the claims the results of which are then cynically extrapolated to residential rooftop solar. The great unwashed masses of the solar ignorant the swallow the garbage and con men have another source of revenue that continues to separate Joe and Jane 6-pack from their money.
I was investigating resource availability for multifaced (backfaced) panels in the '80's, mostly for thermal applications. For most reasonable and common situations, it's a real stretch to get enough irradiance to the backside of a flat solar device to get much of any useful additional energy out of it without extraordinary measures such as booster mirrors that make the whole thing look like it belongs in orbit. I'd like to see that 30% LG claim tested on a panel that's parallel to a roof deck and located 6" above it. I'd like to see any in that common application. I'd bet a bifacial in that application may even perform less efficiently, and furthermore, I'd bet most folks are clueless as to my reasoning. For most any reasonable and common application, the results are meek unless panels are a lot less than optimally oriented, or tracking devices on the reflectors are used. In the end, it's as easy or easier to use known concentration techniques, such as frontfaced CPC, booster mirrors or tracking parabolic concentrators.
Simply put, unless you manage to violate all the laws of Thermodynamics, it's not possible to boost the efficiency of a device more than is added to it as a % of the input by backfacing that device.
As I wrote, for no price adder, aside from perhaps reducing reliability a bit by increasing complexity some, in theory at least, no particular harm is done to the process. However, peddlers will use alleged increases to extract a premium by deceptively telling people bifacial is a better, more efficient product. I've been around solar long enough to know that claim is a real stretch.Last edited by J.P.M.; 09-22-2017, 11:43 PM.Comment
-
Well, my stuff was done as part of some work I did for an employer who was considering entering the solar thermal market as part of a bid to update one of the early solar thermal arms near Daggett, CA. That employer manufactures power generation equipment, boilers and such. I drew the straw because I knew a bit more about solar than the other engineers. At the time, there was not much hard data on irradiance availability due to albedo, mostly of the backscattered type that might be incident on the backside of solar thermal collectors. NREL got the bids and did nothing more than dismantled the project a few years later.
Let's see your data.Last edited by J.P.M.; 09-23-2017, 12:01 AM.Comment
-
You have to do the homework first and make a prediction of what no backside contribution annual production would be based on the specific set of circumstances I provide and actual meteorological data for the time period evaluated. I have monthly production data as reported by SMA inverter (no revenue grade metering), spanning several years.Comment
-
Ha Ha. I have ground mount and snow. How does this experiment go? I thought they might replace the
nearest part of this array, but supports to hold them would cause shadows. Bruce Roe
NSnview.jpgComment
-
That sounds pretty incomplete. Do you have side by side data for the same panel with and without the bifacial feature ? Without that, you have nothing.
Is your Meteorological data collected on site ? If not it's GHI. How do you convert GHI to POA irradiance ? What do you use to measure irradiance ?
See my addendum to my original response, completed a few minutes ago. Our posts are crossing in the ether.Comment
-
In the end kWh's/Watt are what matters to someone looking to produce kWhs (versus performing a technical analysis of all the why's). If STC rating and system geometry/application are enough to make an estimate of energy production over extended periods of time (a year), it shouldn't be to hard to see there has been a boost from backside contribution if it is over 5% or so.Comment
-
I'm not sure anyone is suggesting using bifacial panels in a residential roof mount application. Most of the bifacials I've seen advertised to date by MFG have been 72-cell panels. And when used in commercial applications they need to be mounted in such a way that as much of the rear face of the panel is available to receive reflected light from the ground -- aka fixed or tracking ground mount at least 3 to 6 feet off the ground. Ideally more like 10-20 feet off the ground such as in a parking lot or walkway canopy.
While I agree the contribution of albedo is only a small percentage -- I'd be shocked if it ever reached 20% of total output -- in applications with snow, sand, white or light colored concrete or tarmac there may be an added benefit.
But the real questions is how much more does a bifacial panel cost versus the identical non-bifacial. For the MFG, the added MFG cost of the bifacial is very small -- the cells themselves are already double sided for the most part, and the surface area is even slightly greater on the back as there are no bus bars. Add in the extra material costs for a transparent back sheet, and mark them up how much?
Bifacial is a "premium" feature to increase MFG margin, but it's targeted at the commercial customer rather than the residential customer.
IMHO, there's plenty of data/information. It may not pass JPM's muster, but there's enough to see the impact of bifacial is limited at best ~5-15% (usually less than 5-8%) in most cases.
I stand by my original, grossly simplified assertion, the cheapest ($/Watt) panel that isn't crap will be the most cost effective in terms of kWh/$. Does anyone know how much more MFG will be charging for bifacial? If it's more than ~5-10% you'll never recoup this added cost.Comment
-
In the end, I'd suggest the cost/kWh calculated by some sensible cost analysis methods, which, at the end of the analysis seems to be what your calling kWh's/Watt, as well as the proper array sizing as determined by the application, and other things such as ease/practicality of service, all overridden by safety considerations, all need consideration and, except for safety, need to be balanced against the application requirements and one another. For example, (slightly ??) less cost effectiveness, however that's determined, may be acceptable if it means maintenance will be easier or less frequent as the duty requires.
A side point: As for the kWh/Watt number, just how is a bifacial panel specified ? Do you know? Seems a bit confusing to me. If a front surface panel is rated at 300 W, What rating would a bifacial panel carry ? And, how would that STC rating be done ? How would efficiency be calculated ? and on what surface area. Any ideas ?
To your comment about STC rating and system geometry: Those are two of the necessary, but not sufficient parameters necessary, and far from a complete set of parameters and information that are necessary and required before an estimate of annual system production can be done. And to repeat, unless side/side comparisons are done with equipment that's identical with the exception of one being bifacial, results won't mean much.
On the other hand, if backside irradiance can be compared to front side irradiance, either by measurement and/or calculation, with proper and agreed upon treatment of the irradiance for directional and some wavelength dependent properties made, and a simplifying assumption made that output, at least to the small increases we're talking about here, is linear with input, then performance
enhancement becomes a linear function of irradiance enhancement from adding back side irradiance to front side irradiance.
For all the bifacial applications I've seen, and for what usually becomes the "underside" of a solar device or array, is where most marketing and simplified analyses resort to generalizations, and that's where the smoke/mirror, dog/pony shows and deception/half truths usually start.
I say that because I believe I've got as good or a better handle on solar irradiance and thermal radiation energy transfer as anyone, and probably more than most, at least in the solar energy arena. The whole subject is not easily amenable to closed form analysis. Long, complicated story - much more complicated than can be dealt with here.
Briefly:
Understand that all backside incident irradiance on a panel or array will be from 3 sources, same as front surface irradiance: Direct, diffuse and reflected. However, all 3 of those segments will be much less for the back surface and, for a common residential array orientation, close to or == zero. If so, and I' believe it is, both academically and experientially, the energy harvest contribution will be the same == zero, or close to it. That will also be the amount of enhancement added by the bifacial portion of a bifacial panel for most residential applications. Most all of the beam irradiance will be early and late in the day and small, made that way by a low solar elevation angle, and made much smaller yet by a high angle of incidence for more southerly facing orientations. The sky diffuse portion will, for most common residential array orientations be close to zero, with tilted array backside diffuse being slightly more but still slight, and albedo contributions smaller yet. The total annual irradiance from all 3 components on the back side of a 30 deg. tilted array at 30 deg. latitude is ~ 0.035 of the front side, with ~ half of that backside beam irradiance being incident for only half the year and then only for solar azimuth angles <90,>270 deg., and at such high actual or effective incidence angles that reflection losses from any backside receiving surface will make net energy production --- >>> 0.
As for albedo enhancement by and from the surroundings in general, and for any receiving surface, look at SAM for some very approx. guidance as to what happens to front surface only (non bifacial) energy production when/as surface albedo is increased.
According to SAM, at 30 deg. tilt for an array at 30 deg. latitude, south facing, every 1 % increase in surface albedo increases system output by ~~ 0.00046 or less. What that may indicate, as a very rough 1st approx., is that a 70 % surface albedo, such as snow might produce, will result about a 3% or so increase in annual output and again. and for a back surface contribution, to a fist approx. something like a 0.03 * 0.035 = 0.001 incremental increase in output of a bifacial panel over a standard single faced panel as a result of a very reflective surface. For albedos of, say 0.20 or so, which BTW, to my measurement and experience is a good deal higher than what I've approx. measured many times, added basic backside side enhancement will be closer to zero yet. BTW, my measurements back in Buffalo over more winters than I care to remember indicated about, and very roughly, a 6% - 8% irradiance enhancement on a vertical surface after fresh and copious snowfalls. To me, given the difference in surface tilts, and the thermal output differences I seem to remember, those numbers make some approximate sense.
Most PV and other models don't have the ability to estimate the incident solar energy that may strike the backside of a panel or an array. I had to write my own stuff going back to first principles using what I learned and then confirmed by measurement. As a measure of how long ago that was, the first time I did it was using Fortran IV . A couple of iterations and a lot of years later, I used BASIC and now EXCEL to tweat/redo/enhance the algorithms. I believe my stuff gives reasonable numbers but I've not verified the back surface numbers by pyranometer measurement- yet - not quite pissed off enough.
Since I've not done a side/side comparison of similar bi and non-bifacial panels, I can't/won't comment on your speculation about how easy it may be to see a 5 % increase in system output from the use of bifacial panels. However, and to repeat, the total annual incident insolation on the backside of what's a common residential array orientation will be close to or = zero., and in probably most cases, less than the uncertainly in the instruments ( Good and professional pyranometyers, for example are probably +/- 5%, and only then if kept in good and frequent recalibration.Therefore, for arrays with those (more common) orientations (>>95% ?? of residential arrays ??), any extra $$ paid for a bifacial panel will be a waste of money, and any bogus claims of output enhancement by peddlers will be no more than more ways to separate the solar ignorant from their assets and if for no other reason than instrument uncertainties, impossible to verify in a practical sense.
Bottom line: Caveat Emptor.
Take what you want of the above. Scrap the rest.Comment
Tweet
Comment