Designing with bifacial modules

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Consulting Electrical Engineer - Photovoltaic Systems
I have designed a few PV systems with bifacial modules, but they have all been low tilt ballasted systems so I ignored the bifacial aspect. Now I am designing a ground mounted system and I am not sure what to do about it. It's getting hard to find modules that are NOT bifacial modules these days and I am not sure how to treat them. Thoughts?
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Consulting Electrical Engineer - Photovoltaic Systems
I answered my own question. Maximum voltage is the primary variable to get right, and Voc does not change with illumination of the back side of a bifacial module (duh). Never mind.
 

Elect117

Senior Member
Location
California
Occupation
Engineer E.E. P.E.
Would you use the bifacial gain current Isc or the rated Isc? I see most of the time they use the normal Isc value but I have never been able to confirm with someone who specializes in solar design.
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Consulting Electrical Engineer - Photovoltaic Systems
Would you use the bifacial gain current Isc or the rated Isc? I see most of the time they use the normal Isc value but I have never been able to confirm with someone who specializes in solar design.
I don't know, exactly, but I nearly always have enough conductor ampacity headroom that it wouldn't make any difference to the design.
 

Elect117

Senior Member
Location
California
Occupation
Engineer E.E. P.E.
I don't know, exactly, but I nearly always have enough conductor ampacity headroom that it wouldn't make any difference to the design.

It wouldn't make much of a difference on conductor selection but for inverter oversizing ratio and DC fuses it could? Maybe I am overthinking it. I don't do solar design but have had to review other people's and I didn't know what to do with the bifacial gain information.

Like is it just a small bonus or is it something that can play an impact on design parameters.
 

ramsy

Roger Ruhle dba NoFixNoPay
Location
LA basin, CA
Occupation
Service Electrician 2020 NEC
Like is it just a small bonus or is it something that can play an impact on design parameters.
My read is that marketing is boosted with higher efficiency ratings, and vertical orientations that avoid snow load. Nameplate values still determine conductor size, fuses, & inverters.
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Consulting Electrical Engineer - Photovoltaic Systems
My read is that marketing is boosted with higher efficiency ratings, and vertical orientations that avoid snow load. Nameplate values still determine conductor size, fuses, & inverters.
Yes, but the data sheets for many bifacial modules show different nameplate values for front side only and front side + back side illumination. Voc does not change but Isc does.
 

ramsy

Roger Ruhle dba NoFixNoPay
Location
LA basin, CA
Occupation
Service Electrician 2020 NEC
Yes, but the data sheets for many bifacial modules show different nameplate values for front side only and front side + back side illumination. Voc does not change but Isc does.
Seems like NEC 690.8(A)(1)(a) gives us 2 options:
(1) sum all Isc values together x 1.25, or
(2) engineering supervision w/ highest 3-hr Amp average, if within 70% of (1).

To sum Isc of both sides appears to be overkill, except for vertical orientation, since some studies measure the highest contribution of the backside below 10%.

Doesn’t help that bifacial is not defined, or considered in the 2020 NEC.
Got a Nameplate to show us?
 

wwhitney

Senior Member
Location
Berkeley, CA
Occupation
Retired
They also call it minimum at test conditions which also throws me off.
That's just the binning that they do after production. If you buy a 475W panel, you might get one that tests out anywhere from 475W-479.9W (as per the data sheet spec of +5W, -0W). So if you want to be conservative, you should use the data sheet Isc values for a 480W panel, as those will also be maximum Isc values for the 475W panel.

Cheers, Wayne
 
Last edited:

ramsy

Roger Ruhle dba NoFixNoPay
Location
LA basin, CA
Occupation
Service Electrician 2020 NEC
BSTC is what they call the bifacial values.
The Isc shown for BSTC column appears near 10% of adjacent column, perhaps not overkill If it includes both sides.
They also call it minimum at test conditions which also throws me off.
Me too. Also reference IEC standards —perhaps not harmonized with NFPA-70 maximums needed for Article 690.
 

Elect117

Senior Member
Location
California
Occupation
Engineer E.E. P.E.
That's just the binning that they do after production. If you buy a 475W panel, you might get one that tests out anywhere from 475W-479.9W. So if you want to be conservative, you should use the data sheet Isc values for a 480W panel, as those will also be maximum Isc values for the 475W panel.

Cheers, Wayne

Would you use the bifacial Isc or the nominal one?
 

ramsy

Roger Ruhle dba NoFixNoPay
Location
LA basin, CA
Occupation
Service Electrician 2020 NEC

wwhitney

Senior Member
Location
Berkeley, CA
Occupation
Retired
Would you use the bifacial Isc or the nominal one?
Good question, although the point of my earlier post was that this "minimum" issue is a separate one.

I guess it would depend on the details of the installation. If there was zero chance of backside illumination, I'd use the STC figures. Otherwise, I'd investigate what "at BSTC: 1000 W/m² + φ × 135 W/m², φ = 70 % ± 5 %, 25 ± 2 °C, AM 1.5 according to IEC 60904-3" means and whether it adequately covers the possible backside illumination for the installed configuration. If it does, then the BSTC Isc would be the ticket.

Cheers, Wayne
 

Elect117

Senior Member
Location
California
Occupation
Engineer E.E. P.E.
The Isc shown for BSTC column appears near 10% of adjacent column, perhaps not overkill If it includes both sides.

Me too. Also reference IEC standards —perhaps not harmonized with NFPA-70 maximums needed for Article 690.


In this instance they installed 1,628 of them and said the size was 789.58kW DC. It wouldn't have had an impact on most of the design in their case but I didn't know if it is something to be aware of. Like possibly under sizing protection as things get combined or the inverters.

I guess some of the 125% multipliers could offset the issue.
 

Elect117

Senior Member
Location
California
Occupation
Engineer E.E. P.E.
Good question, although the point of my earlier post was that this "minimum" issue is a separate one.

I guess it would depend on the details of the installation. If there was zero chance of backside illumination, I'd use the STC figures. Otherwise, I'd investigate what "at BSTC: 1000 W/m² + φ × 135 W/m², φ = 70 % ± 5 %, 25 ± 2 °C, AM 1.5 according to IEC 60904-3" means and whether it adequately covers the possible backside illumination for the installed configuration. If it does, then the BSTC Isc would be the ticket.

Cheers, Wayne

In my instance the cells were angled north /south on a flat roof so the likely hood of having sunlight under the cells (other than the outskirts) was very unlikely.

Short of using the engineering route for NEC, I still don't know what I would expect someone to use.
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Consulting Electrical Engineer - Photovoltaic Systems
Of course, the most pressing issue would be the voltage, which when I (prematurely) posted my question I did not snap to the fact that it would not change (duh again) if the back side of the module were illuminated. Most modules tell you what the maximum string fuse size is, so individual Isc is probably not an issue unless you are combining strings and approaching the current limit of an MPPT input. But I usually err to the side of caution when I am designing a system, so I look at the most current a module can produce and design to that unless I know for sure that the back side will always be shaded.
 

wwhitney

Senior Member
Location
Berkeley, CA
Occupation
Retired
Of course, the most pressing issue would be the voltage, which when I (prematurely) posted my question I did not snap to the fact that it would not change (duh again) if the back side of the module were illuminated.
Per the datasheet in post #9, it does change, but only by 0.19V per panel, or about 0.4%. So unlikely to be an issue, unless the string Voc based on STC without backside illumination was extremely close to your limit.

Cheers, Wayne
 

ggunn

PE (Electrical), NABCEP certified
Location
Austin, TX, USA
Occupation
Consulting Electrical Engineer - Photovoltaic Systems
Per the datasheet in post #9, it does change, but only by 0.19V per panel, or about 0.4%. So unlikely to be an issue, unless the string Voc based on STC without backside illumination was extremely close to your limit.

Cheers, Wayne
I did not look at that one, but the ones I did look at did not show a change. It's not an issue for me because I would not design to within 0.4% of the maximum allowable string voltage, anyway.
 

ramsy

Roger Ruhle dba NoFixNoPay
Location
LA basin, CA
Occupation
Service Electrician 2020 NEC
In this instance they installed 1,628 of them and said the size was 789.58kW DC.
That appears to match power class 485. The corresponding BSTC column is 530.5

So they are not following the NEC to sum both sides, by not using BSTC, much less Max values without American Equipment Nameplate Standards.

BSTC w/ Wayn’s Max estimate +5W in post #10 is near ~535 per module.

1,628 modules = 870.98kW

The 1.25 multiplier should be for continuous use, not for Data Sheet errors.
 
Top