Heat effect on bendable mono solar panels

[Uncle Woody]

Greetings --

I'm interested in buying some 100W bendable monocrystalline 32 cell solar panels for mounting atop our bimini. I've read several CF threads related to build quality, mounting, wiring, controlling, etc., but have not come across any hard data related to actual voltage drop testing in higher temperature climates (i.e. - southern FL or the Bahamas).

I've also read what Don Casey and Charlie Wing have written on the topic of panel output vs. number of cells and panel temperature. They both state that anything less than 36 cells in a high temp climate runs the risk of voltage output becoming too low to charge batteries. However, their writing is at least 8 years old, and is based on cell efficiency in the 10-12% range with individual cells producing about 0.5V each. Cell efficiency has improved greatly since then, and I'm wondering if their information might be outdated.

Several 32 cell panels I'm considering have optimal operating output in the 18V range (about 0.55 volts per cell), and 19% rated cell efficiency. That output has (supposedly) been achieved at 25C/86F module temperature. Real world panel temps in south FL or the Bahamas during the Dec-May time frame are likely to be significantly higher.

One manufacturer's data sheet suggests a Nominal Operating Cell Temp (NOCT) of 47C/117F. That's a 22 degree C difference. They also list a Temperature Coefficient of the panel's Open Current Voltage (Voc) of -0.061V per degree C. By my calculation, that would result in a 1.34V drop in panel output at 47C. Of course, temps could get much higher, but at 47C the panel would supposedly produce much more than enough voltage to meet the 14.6V charging requirement of my 400Ah Lipo battery bank...leaving me a significant upward temp margin. (Sidebar: I am estimating that proper wiring of the system and use of a good MPPT controller would limit other voltage drop to NMT 0.5V. Realistic?)

My primary question for those of you who have gotten this far in my amateur attempt to describe the theoretical/technical performance in higher temp climates (thanks for hangin' with me so far) is this: Does anyone have practical, real world Voltage output experience with 32 cell panel performance in higher temps? Alternatively, are there any engineer-types who can explain how a 32 cell bendable mono panel would be capable of producing 14.6V or more when they get hot? Is it simply a function of modern cell efficiency?

Much thanks,
Uncle Woody

[capt-couillon]

Assuming you are talking the Renogy 100w panels, I will let you know when I make it to warmer climes. Just purchased one panel, with a 30W controller. (plan for more panels later) Mounting on lazarette hatch cover, and as it overhangs a bit, I plan to attach to a piece of 1/8 alum sheet with a couple of 1/2" standoffs below to allow some air circulation.

Theory is nice, and panel output is effected by temp, but in the real world,ya get what ya get. Most of the current production from established manufacturers have about the same temp/output curve.

Wish I had more room as the current problem is economy of scale... You can buy a 230W panel for about the same price as a 100W if you are not constrained by size.

Anyway, as I head south I will be glad to pass along output info vs ambient temp if you want. I don't really expect temp to be a significant factor. At least not in relationship to cloud cover, shading, heading and azimuth. On a fixed installation, it may be a significant question. On a mobile platform with 360 orientation I believe other engineering factors are far more significant than operating temp.

Beware over engineering and buzz words.

I'm just sayin'

[sailorchic34]

All solar cell have a thermal coefficient related to volt drop per degree C rise over STP (standard pressure and temperature), which is 20 degrees C. A rough rule of thumb is you'll have a 1V drop for each 10 degrees C rise above 20C. This is the primary reason 12V panels output 17-18 volts, to allow for the voltage sag to 15V when hot. That's cell temp, not air temp.

The good news is there is more hours in the day in the summer, though less so near the equator. I found that it sort of balanced out between summer and winter.

Myself I prefer hard panels as they tend to be a little more durable and lower cost per watt.

[Uncle Woody]

Originally Posted by capt-couillon

Assuming you are talking the Renogy 100w panels, I will let you know when I make it to warmer climes. Just purchased one panel, with a 30W controller. (plan for more panels later) Mounting on lazarette hatch cover, and as it overhangs a bit, I plan to attach to a piece of 1/8 alum sheet with a couple of 1/2" standoffs below to allow some air circulation.

Theory is nice, and panel output is effected by temp, but in the real world,ya get what ya get. Most of the current production from established manufacturers have about the same temp/output curve.

Wish I had more room as the current problem is economy of scale... You can buy a 230W panel for about the same price as a 100W if you are not constrained by size.

Anyway, as I head south I will be glad to pass along output info vs ambient temp if you want. I don't really expect temp to be a significant factor. At least not in relationship to cloud cover, shading, heading and azimuth. On a fixed installation, it may be a significant question. On a mobile platform with 360 orientation I believe other engineering factors are far more significant than operating temp.

Beware over engineering and buzz words.

I'm just sayin'

Thank you for your feedback and the offer to pass along output info, Cap. That would be interesting to see. I'm sure you are right about other factors having an impact on output, too. My primary concern is weather or not a 32 cell panel will be able to maximize performance during mid-day in southern latitudes.

I have room for larger panels (40 cell), but they would be much more subject to shading than the 100W. Life is full of compromises. Just tryin' to make the most informed choice.

BTW, the only set of test results I've seen was on a 36 cell 135W bendable panel. It showed a 10% output difference in voltage between a panel with airspace below and one without. The direct-mount panel was 20C hotter on its surface resulting in the output difference. Current drop was minimal.

Hope you have a safe and enjoyable journey south.

Uncle Woody

[Uncle Woody]

Originally Posted by sailorchic34

All solar cell have a thermal coefficient related to volt drop per degree C rise over STP (standard pressure and temperature), which is 20 degrees C. A rough rule of thumb is you'll have a 1V drop for each 10 degrees C rise above 20C. This is the primary reason 12V panels output 17-18 volts, to allow for the voltage sag to 15V when hot. That's cell temp, not air temp.

The good news is there is more hours in the day in the summer, though less so near the equator. I found that it sort of balanced out between summer and winter.

Myself I prefer hard panels as they tend to be a little more durable and lower cost per watt.

Thank you for sharing your thoughts, Sailor Chic.

Uncle Woody

[Paul L]

I have two 100 watt flex panels on my Bimini with 32 cells. I am at 12 degs N, ie the tropics. The panels output the same percentage of spec as my 2 hard panels on the aft. Both go through Blue sky controllers. After 18 months in the tropics the flex panels have some visual fogging/discoloration on the surface. So far it doesn't seem to affect their performance. I bought the panels off Aliexpress direct from China.

[noelex 77]

I have used a 32 cell panel a long time ago on a previous boat.
For the same area a 32 cell panel will produce more current, but less voltage.

Providing the panel Vmp voltage is above the battery voltage all is good, but if the panel voltage drops below the battery voltage then the output gets much lower. If below Voc nothing is produced.

The optimum compromise for a lead acid battery is 36 cells which is why it has become the industry standard. The higher cell temperatures with a flexible panel (because of the lack of an air space underneath them). Will tend to swing the optimum to a higher number of cells again.

Lithium batteries have a lower charging voltages which would help considerably (do you really want to charge these to 14.6v). I also hope they are not Lipo but LiFeP04.perhaps this is typo and you mean lead acid?

In short a 32 cell panel will still work in tropical conditions, but not as well as a 36 cell panel.

BTW in your calculations I think you are using Voc when you should be using Vmp. If you use Vmp with a typical cell temp of 45-50c and allow some voltage drop for wiring and contacts fuses etc and you will get a more accurate theoretical result for the voltage output.

[Uncle Woody]

Originally Posted by Paul L

I have two 100 watt flex panels on my Bimini with 32 cells. I am at 12 degs N, ie the tropics. The panels output the same percentage of spec as my 2 hard panels on the aft. Both go through Blue sky controllers. After 18 months in the tropics the flex panels have some visual fogging/discoloration on the surface. So far it doesn't seem to affect their performance. I bought the panels off Aliexpress direct from China.

Thank you, Paul. This is very helpful feedback. Appreciate the dual controllers, too.

Uncle Woody

[Uncle Woody]

Originally Posted by noelex 77

I have used a 32 cell panel a long time ago on a previous boat.
For the same area a 32 cell panel will produce more current, but less voltage.

Providing the panel Vmp voltage is above the battery voltage all is good, but if the panel voltage drops below the battery voltage then the output gets much lower. If below Voc nothing is produced.

The optimum compromise for a lead acid battery is 36 cells which is why it has become the industry standard. The higher cell temperatures with a flexible panel (because of the lack of an air space underneath them). Will tend to swing the optimum to a higher number of cells again.

Lithium batteries have a lower charging voltages which would help considerably (do you really want to charge these to 14.6v). I also hope they are not Lipo but LiFeP04.perhaps this is typo and you mean lead acid?

In short a 32 cell panel will still work in tropical conditions, but not as well as a 36 cell panel.

BTW in your calculations I think you are using Voc when you should be using Vmp. If you use Vmp with a typical cell temp of 45-50c and allow some voltage drop for wiring and contacts fuses etc and you will get a more accurate theoretical result for the voltage output.

Thank you for your feedback, Noelex77. If I go with the flex panels, I plan on mounting them on 1/4" Seaboard HDPE (comparable to Starboard) with 1"+ of airspace below the Seaboard. That should provide some heat relief.

I have a bank of CALB 200Ah cells (wired 4S2P) with max charge voltage of 3.65V per cell (14.6V for the bank). I realize that's max. Your point is well taken about not needing 14.6V in order to charge. (Lipo is my shorthand for LiFePO4. Sorry for any confusion.)

The mfg data sheet for the 32 cell 100W panel lists Vmp (Optimum Operating Voltage) as 17.7V. However, the data sheet lists the temp coefficient of Voc (open circuit) as -.061V per degree C above STC. I figured it safer to calculate loss against Vmp since that's the lower number ...would rather err on the safe side. Does that make sense?

Thanks again,
Uncle Woody

[noelex 77]

Originally Posted by Uncle Woody

The mfg data sheet for the 32 cell 100W panel lists Vmp (Optimum Operating Voltage) as 17.7V. However, the data sheet lists the temp coefficient of Voc (open circuit) as -.061V per degree C above STC.

I think you should be fine with those numbers. The Vmp is very high for 32 cell panel. Most 36 cell panels are not much higher with a typical Vmp of around 18v (the popular Kyocera panels are 17.9v).
The temperature coefficient is also quite low. So at normal operating temperatures the Vmp will not be much different to the 36 cell Kyocera panels (assuming the specs are accurate). Make sure there is not much voltage drop in the wiring.

Have a good read of the lithium thread if you have not done so already. Mainsail always gives very sensible advice and is advocating a much lower maximium charge voltage. Lithium is the opposite of lead acid and is not very happy at 100% charge. Aiming to squeeze the last few % of charge into the batteries like we do with LA is not sensible. The cautious approach is to adopt lower maximium charge voltages at least untill the dust clears and the ideal charging parameters become clear.

[Matt Johnson]

If using MPPT, why not wire in series and avoid the voltage issue completely?

Matt

[Uncle Woody]

Originally Posted by funjohnson

If using MPPT, why not wire in series and avoid the voltage issue completely?

Matt

Good question, Matt. I had considered wiring 4x50W panels in series to help deal with shading, but had a concern about lower current output. Would a good quality MPPT controller produce nearly as much current from series wired panels as 2x100W would in parallel?

I've researched the Genasun LiFePO4 controllers, and have been thinking in terms of one for each panel in a parallel config. Now you have me wondering if just one of their GV-Boost 105-350 units would do the job for series wired 50W panels. Since the 50W panels are rated at about 2.8A each, wiring them in series would avoid exceeding the 8A input limit of the Genasun LiFePO4 controllers. Would you care to comment further?

Thank you,
Uncle Woody

[OceanPlanet]

Originally Posted by Uncle Woody

Good question, Matt. I had considered wiring 4x50W panels in series to help deal with shading, but had a concern about lower current output. Would a good quality MPPT controller produce nearly as much current from series wired panels as 2x100W would in parallel?

I've researched the Genasun LiFePO4 controllers, and have been thinking in terms of one for each panel in a parallel config. Now you have me wondering if just one of their GV-Boost 105-350 units would do the job for series wired 50W panels. Since the 50W panels are rated at about 2.8A each, wiring them in series would avoid exceeding the 8A input limit of the Genasun LiFePO4 controllers. Would you care to comment further?

Thank you,
Uncle Woody

It depends on the voltage of those 50W panels. If you put them all in series then the voltage is likely much higher than the batteries, so you wouldn't need the GVB boost controller at all since the reason for their existence is to raise the voltage of the panel up to the battery voltage (if panel V is less than the battery). What is the panel voltage?

However, seriously consider this: if you wire the panels in series, and you shade one of them, it will cut the current output of the whole array. It's much better to wire in parallel to small individual controllers (use GV4 or GV5, if panel voltage >battery), or in parallel to one larger controller (BSE 2000i, WRM15, etc.). The GV4 are not very costly so four of them are not all that much and would provide the least losses in partial shading.

But it all depends on the panel voltage and battery voltage...if those are 8V Solbian panels (SP50?) then the GV4 won't work, you would indeed need the GVB. Or wire in two series pairs and use 2 x GV10 (essentially creating 2 x 16V x 100W arrays).

IF the panels are small-cell higher-voltage panels, then a GV4 or GV5 will do the trick for each panel.

[Uncle Woody]

Originally Posted by OceanPlanet

It depends on the voltage of those 50W panels. If you put them all in series then the voltage is likely much higher than the batteries, so you wouldn't need the GVB boost controller at all since the reason for their existence is to raise the voltage of the panel up to the battery voltage (if panel V is less than the battery). What is the panel voltage?

However, seriously consider this: if you wire the panels in series, and you shade one of them, it will cut the current output of the whole array. It's much better to wire in parallel to small individual controllers (use GV4 or GV5, if panel voltage >battery), or in parallel to one larger controller (BSE 2000i, WRM15, etc.). The GV4 are not very costly so four of them are not all that much and would provide the least losses in partial shading.

But it all depends on the panel voltage and battery voltage...if those are 8V Solbian panels (SP50?) then the GV4 won't work, you would indeed need the GVB. Or wire in two series pairs and use 2 x GV10 (essentially creating 2 x 16V x 100W arrays).

IF the panels are small-cell higher-voltage panels, then a GV4 or GV5 will do the trick for each panel.

Thanks for your feedback, Bruce. I talked with Alex at Genasun this evening, and he agrees that parallel wiring will work best with his MPPT controllers.

The bendable panels I have in mind are rated at 17.7 Vmp (32 cells). That's for both the 50W and 100W panels. I've just come across some 40 cell bendable panels with 120 watts at 22 Vmp. Now I'm trying to decide if I have room enough to mount them. They are 10" longer than the 100W panels, and the extra length might make them more subject to occasional shading. However, the extra voltage should result in a nice increase in current from the MPPT controllers when the panels are not shaded.

Lots of trade-offs in designing a system, aren't there? I'm inclined to go with the most watts I can find room for on the bimini, hook MPPT controllers to each panel (wired in parallel), then relax while letting nature and physics do their thing.

Thanks again for sharing your insights.

Uncle Woody

[OceanPlanet]

Originally Posted by Uncle Woody

Thanks for your feedback, Bruce. I talked with Alex at Genasun this evening, and he agrees that parallel wiring will work best with his MPPT controllers.

The bendable panels I have in mind are rated at 17.7 Vmp (32 cells). That's for both the 50W and 100W panels. I've just come across some 40 cell bendable panels with 120 watts at 22 Vmp. Now I'm trying to decide if I have room enough to mount them. They are 10" longer than the 100W panels, and the extra length might make them more subject to occasional shading. However, the extra voltage should result in a nice increase in current from the MPPT controllers when the panels are not shaded.

Lots of trade-offs in designing a system, aren't there? I'm inclined to go with the most watts I can find room for on the bimini, hook MPPT controllers to each panel (wired in parallel), then relax while letting nature and physics do their thing.

Thanks again for sharing your insights.

Uncle Woody

Yep, so GV4 or GV5 for the 50's, and GV10's for the 100's. We have them in stock...we sell hundreds of them for Alex...