PV string sizing: Cutting it too close?

HX_Guy

Another question here, hope you guys don't mind and can give me some constructive criticism.
Ideally I would like to setup my system in 5 strings of 10 panels, however using this PV String Sizing Tool it looks like the system would be at the Inverter's lower limit of operation and if the temperature goes above 45.6º (114º F) which is often does, would the system shut down?



The installer suggested a 4 string x 12 panel setup, but I said I would prefer the 5 x 10, which he said we could do as well.
With the 4 x 12 setup, the numbers are much more within range.



I'm pretty sure I know what the right answer is but figured I'd confirm here. Thanks for any input.

niel

Re: PV string sizing: Cutting it too close?

from my rough calculations you should be fine at 46 degrees c by about 30v unless i missed something. of course if it goes higher in temp than that the voltage continues to drop so you need to know the max temp recorded for your area. the drop is expressed as a percent (.34%) of the panel voltage or vmp. your 10 pvs in a series string will have 309.3v vmp. if the max temp is say 50 c (122 degrees f) then .0034 x 309.3v x 50 c = 52.581v lost or 256.719v to your inverter and would still be 26v or so above the minimum.

HX_Guy

Re: PV string sizing: Cutting it too close?

Thank for the reply Niel. According to the specs, the drop is actually .46% for Vmp, the .34% is the Voc drop. At 50º C, it would be a 71.139v lost or 238.16v so only 8.16v above minimum. Is that save enough? And the panel temperature, baking in the sun, would be quite a bit higher than the ambient temperature, does that matter?

I'm not quite sure how the program I'm using is getting different results. At 50º C, it shows a voltage of 223.9.

HX_Guy

Re: PV string sizing: Cutting it too close?

I read this in another Fronius PDF and it's a great point.

"When designing your systems, keep this question in mind: 15 years from now, on a hot day, will you still achieve at least 230Vmpp from the aging modules to produce the needed voltage to stay within the Maximum Power Point range?"

I'm going to talk to my installer about upping the number of panels to 13 in 4 strings which even at 118º, each string would be at over 300V, well above the minimum of 230V. There will be more shade but the system will also be bigger (13.0kW) so hopefully that will make more power when not shaded so we don't need the banked kWh from the winter months.

Cariboocoot

Re: PV string sizing: Cutting it too close?

Looks to me like you could get about 10% V drop from 46C temps. To drop below the 230V min MPPT value you'd need 25% Voltage drop from temperature. That would be a 55C increase, at which point your roof would be on fire (80C = 176F).

niel

Re: PV string sizing: Cutting it too close?

good point on the vmp %. it certainly won't hurt upping the number of pvs in the strings, but do still be conscious of the max voc as i see it at about 533v for -1c with 13 pvs in series. getting a bit too close that way. 4x12 may be a good overall option, but would reduce power by 2 pvs over your base 50 pv design. up to you.

HX_Guy

Re: PV string sizing: Cutting it too close?

Niel, where are you seeing the max VOC stuff? I can't see to locate it (it's probably staring me int he face).

I think my ideal setup given what I have to work with would this this...4 strings of 12 in this layout. Couple problems though...first and I may not be able to overcome it is that I'm not really following the fire setbacks on all hips. Everything is pretty ok except for the blue string, there should be an 18" setback on any hip where there are PVs on the adjacent roof plane and I obviously have zero. The installer said the inspector doesn't normally go up on the roof and measure, they just do a walk around and as long as you'r not breaking the rule everywhere it's usually fine.



The other potential layout would be this, but then there would be serious shading issues with the yellow and green strings for about 4-5 months of the year.
The first setup really minimizes it and during the important production times between 10am - 4pm, there really wouldn't be any shade. This second setup would actually require 4 x 13 strings and it would still produce less than the first because of the shading according to my calculation.

niel

Re: PV string sizing: Cutting it too close?

the voc is affected by temperature too as you don't want the pvs after a record cold to be hit with a strong morning sun. that is the voc x .34% x the degrees c below 25 + total voc of all pvs in the string.

1 pv voc = 37.68v
x 13 pvs in string = 489.84v voc
x .34% = 1.665456v
x 26 degrees c diff = 43.301856v
+ 489.84voc = 533.141856v

this is about 16v away from blowing out the inverter.

HX_Guy

Re: PV string sizing: Cutting it too close?

I see, thanks...I hadn't even considered that end of it.

I have to refigure out the layout anyway because if they aren't lenient at all with the fire code setback, I really can't do the designs that I was wanting. Sucks.