Help me understand solar panel ratings and when/how I need to acomodate for "over voltage" situations.....

[cdherman]

I am trying to get my head around an Hybrid/AIO system. Looking at the EG4 18k for instance, it uses up to 500v, but can handle up to 600v without damage. OK, seems straight forward enough. I will have a long run for the DC lines back to the AIO, so I want max voltage.

Then I look at panel spec sheets. These Canada solar 400w units have the following ratings: Wattage 400, VOC 36.8, ISC 13.85A
Canada Solar 400w

So the ignorant guy in me says I can take 500v / 36.8 = 13.58, round down to 13 panels and that string should be safe.

But someone, somewhere in my readings, notes that one must accommodate for "over production events" like light reflection off of snow or cold temps. I have tried PVWatts to get an idea if/when this could occur at the location where the panels will be, but to no avail.

Experience tells me that in the location the system is going, snow cover is seldom a problem, but can occur. This system would be oriented to the west, 270 deg. I would think that would also reduce events of "double insolation" but I could be wrong, thinking afternoon in January on a cold snow covered day as the sun is setting. But in that event, the panel orientation, while correct in terms of axis, would be VERY flat, like 30-40 tops (reference 90 as optimal for production), rendering the reflected light off the snow nearly parallel to the panels. The panels are going on a roof around 20 ft high, which further "flattens" the angle, if my aging brain is seeing my long lost trigonometry correctly....

Other variable I see mentioned that creates over production: Very cold temps. Now, these panels will be on western high plains. It does get very cold there, rarely a day with no clouds and it won't break 0F. Rare. But it happens. Polar vortex out of Canada and it gets darn cold. But happens again when the sun will be hitting the panels at really low angles.

In short, how does one arrive at an optimal string size to avoid over voltage situations, but also mindful that higher voltages allow better DC transmission with smaller (and cheaper) cables over a distance (150 ft)?

 

[GXMnow]

In the specs for the solar panels, you should look for the voltage temperature coefficient. The VOC voltage is at 25C. As the panels get colder the voltage will rise. My panel increase by 0.3% per degree C colder. At 0C, that is a 7.5% increase. 36.8 volts x 1.075 = 39.56 volts. Putting 13 of them in series pushes the voltage to 514 volts. And that is still 32F. Depending on where you are, it might get a lot colder than that. The typical rule of thumb is to allow 20% from the panel VOC to the max MPPT input voltage.

The big question is what does the 18K do between 500 volts and 600 volts? Does it shut down to protect itself, or does it just fail to do proper MPPT and not collect maximum power? Most solar charge controllers have a voltage range where the MPPT can track well, and then some headroom where it will still function without damage but loses some efficiency.

When the batteries become full, it is possible the MPPT will stop pulling current and the panel voltage can go to the full open circuit level. On my smaller DC system, I have seen a few times where the voltage and even the power can go above the STC specs when there are nice white clouds in the sky. The panels get full direct sunlight, AND the sun reflects off the clouds and adds additional light onto the panels. It does not happen often, but all it takes is once and you could pop the MPPT input with too much voltage, current, or power if you don't leave any headroom. With a 500 volt MPPT range, I would probably design for 400 volts and be on the safe side, especially if you get sub freezing weather.

 

[PreppenWolf]

I'm in NH where it can get pretty cold. We've had a night of -34F in the last 10 years. If it was me I'd trim it to 12 panel series and parallel 4-5 strings.

 

[cdherman]

OK, next question then. This whole business of volt rising with cold temps. This occurs, even if the amps are down, due to poor angles, clouds?

 

[zanydroid]

OK, next question then. This whole business of volt rising with cold temps. This occurs, even if the amps are down, due to poor angles, clouds?

A function of temperature. It's modeled in the spec sheet as a voltage vs temp coefficient. Plug and chug it in your design.

The other effects predominantly impact current, though they will also shift Vmp so you want some buffer on that vs starting and operating range.

 

[zanydroid]

how does one arrive at an optimal string size to avoid over voltage situations, but also mindful that higher voltages allow better DC transmission with smaller (and cheaper) cables over a distance (150 ft)?

Use the AHRI or similar weather data to find the extreme, and then add a buffer because there have been new extremes lately.

Plug your current and voltage into a voltage drop calculator to convince yourself that your selected string voltage for safety is perfectly fine. Keep in mind your voltage drop at 400V working voltage is over 2x better than at 240V

As well for residential you can't go over 600Voc anyway in a design so there's minimal engineering wiggle room.

Usually people use #10 or #8 AWG. #12 often gets in trouble with ISC rating of some panel formats.

 

[cdherman]

OK, I've been reading. Voc is open, so larger than Vmp, where the mp stands for maximal production. With the above panels, plugged into a Midnight Solar calculator, I get a VOC at -30C/-22F of 42.1V (547V for 13 panels, within capacity of EG4) and a Vmp of 38.4.

But how is Vmp measured? Does cold change it? Since its lower, I guess its moot. Seems the MN Solar calculator adjusts it using the same temp coeffcient of -0.26%

But I am trying to wrap my head around this correction once and for all, because I want to push the V of the lines back to their limit, but of course not burn things up.

As for knowing the low temps, I think I have that. My great grandparents bought the farm in 1901. Daytime lowest high are storied events. Like it "never got over -15F that day". But its never been under -20F during daylight, IIRC and that is likely not to change, sadly.

 

[cdherman]

Ugh..

Found this very well written bit out there with regards to Voc versus Vmp

Photovoltaic panels are usually characterised by their short circuit current (Isc) and their open circuit voltage (Voc). It is important to make sure that the combined open circuit voltage (Voc) of a PV panel or array does not exceed the voltage rating of the connected controller or inverter. Yes you are correct that the controller/inverter will load the panel/array but its the electrical load connected to the panel/array which determines its operating point. For example, if batteries are connected, the battery voltage sets the operating voltage. In a grid-connected PV system, the inverter loads the PV array at its maximum power point.

When charging batteries or supplying the utility grid, electric current flows and the controller/inverter can regulate the maximum power point of the panels/array I–V curve keeping the voltage at the Vmp level as Ohm’s Law states that Pmp = Vmp*Imp (watts). However, as the batteries become fully charged, or the utility grid requires less power, less current is supplied and the panel/array voltage starts to rise from Vmp to Voc. As I reduces, V*I reduces until P = 0 turning “off” the controller/inverter. But even if there is no power, there is still voltage from the panels/array while the controller/inverter is waiting for enough power to turn back “on” again. Thus Vp is at its highest output voltage = Voc.

Also, on cold early mornings or late nights there may be not be much ambient light generating current, but the PV panel/array can still output full Voc until there is enough power to activate the controller/inverter and regulate the panels. As a rule of thumb, a voltage correction factor of 1.25 should be multiplied to the rated open circuit voltage to account for extreme low ambient temperatures. For example, a panel with a manufacturer rated Voc of 40V, should be seen as having a maximum Voc of 40 x 1.25 = 50 volts.

 

[teal95]

From my understanding panels pretty much have full voltage if there is any light at all, but there current is very dependent on how bright it is.

 

[Madcodger]

Others have given you great advice and info, so no need to repeat most of it. I'll just underscore that it's over-voltage that kills your expensive equipment, and it's voltage that rises in cold temps, even with only a modest mount of early morning light. You may be producing very little current (amps) because of the angle of your panels, etc., but there's that pesky high voltage, and poof! Bye-bye charge controller...

So, personally, I'd look at the lowest OVERNIGHT temperature in the past 100 years because that usually happens right around dawn, add in a modest safety factor of a few degrees, then use the highest panel voltage number (VOC), and buy the SCC based on the thought of never having even a millisecond of voltage that could hurt it. Electricity is remarkably unforgiving.

 

[Hedges]

14 panels in series.

https://signaturesolar.com/content/documents/CS-Datasheet-HiKu6CS6R-MS-HL_v1.1.pdf

Data sheet 36.8 Voc, -0.26%/degree C, 30.8 Vmp.

If your record cold is -20F, that's -29C.
-29C temperature - 25C nominal = -54C relative to nominal
-54C x -0.26%/degree C = 14.0% higher voltage
36.8V x 1.140 = 41.97Voc cold
600V max / 41.97V per panel = 14.3 panels in series

30.8Vmp x 14 = 431.2V per string, well under 500V max MPPT operating.

 

[highestbid]

I'm in NH where it can get pretty cold. We've had a night of -34F in the last 10 years. If it was me I'd trim it to 12 panel series and parallel 4-5 strings.

With an ISC of 13.85A paralleling even 2 strings would take A well over what the 18KPv is rated for. Or am I missing something? I am still trying to get my head wrapped around the math.

 

[Hedges]

True, even the MPPT rated 25A would be a bit over. Two strings of different orientation could be paralleled for that one so might fit 4 strings.