The EG4 18Kpv CAN handle 18/21kW of PV on its inputs

[SignatureSolarJames]

All, I just want this thread out there if people have concerns about PV capacity annd how the MPPTs are supposed to work on the 18Kpv

We have:
-1x 25A input
-2x15A inputs

Both of these are max voltage open circuit (VOC ) of 600v, always derate for colder temperatures as the name plate VOC on a panel is at 77 degrees F, in texas I typically leave 40v of space, some climates need more, there is a temperature coefficient on each panel spec sheet that shows what % of voltage is increased per degree Celsius ( just use the 9/5 or 1.8 rule to rationn the F to C if you need)

Panels are facing a huge shift right now, the base cell sizes are frankly quite similar across manufacturers. 4 years ago the market was run by the M1 cell, then the split M6 cell arrived and this year the M10 cell will debut and is already in a lot of newer modules you may have noticed are in the 44 inch width now vs the classic 39/40 inch wide.

All this is to say that panels amperage per module is going up and will continue to increase
The M1 cells made panels around 9 amps, the m6 broke the 10 amp mark and the m10 is at the 11.5amp range

Here are 3 types of panels in each cell category and how I would string them:

-Canadian Solar 540W M10
-Canadian solar 400w all black M6
-Aionrise 360W all black M1

Canadian solar 540W m10
This is a 13Amp panel, so 1x string on the small inputs and 2x on the large (26A vs 25 is safely under the 1.2x clipping recommendations)
That leaves us with 4 strings of panels at 49VOC, the max string length is 11 in texas and may be 10 in colder places, let’s go with 10
4x10x540 = 21.6kW of PV

I running voltage is 400v then total amps of 13+13+25=51A x 400v = 20400 dc watts utilized, above the 18k needed for max utilization

Note the 410w class solarever and others coming soon are in this range too

Canadian solar 400w m6
11.44 amps per panel VOC is 44 so 12 string is 530VOC, safe for most everybody
This is closer as I would say that clippage is high if you try to double 2x strings on the 15A inputs, we always want to keep under 1.2 ratio here as you can run the numbers and find it is 1% total loss or less, going to 1.3 in a sunny climate takes another 1% off yearly production
So 4x 12 x400 w = 19.2kW of PV, if you use this style of panel

Aionrise 360w M1
OK, this gets more interesting, because the amps are 9.9 per panel (the highest M1 I have seen) you can actually put 3x on the 25 input and be under 1.2 amperage ratio (30/25A)
So 5 strings of 360, 11 long (540VOC) =55x360= 19.8kW of solar

If you are in a low light area or if you are adding more solar just as a hedge on cloudy days you can go above 1.2 ratio on the amperage and the total amount of solar will go higher. As an off-gridder I have been guilty of this a lot and it is a good idea many times if you are just trying to put more solar in vs adding batteries.

 

[FilterGuy]

Both of these are max voltage open circuit (VOC ) of 600v, always derate for colder temperatures as the name plate VOC on a panel is at 77 degrees F, in texas I typically leave 40v of space, some climates need more, there is a temperature coefficient on each panel spec sheet that shows what % of voltage is increased per degree Celsius ( just use the 9/5 or 1.8 rule to rationn the F to C if you need)

Do you have any recommended tools for the users to calculate the voltage rise in cold temps? I generally point them to this resource:

Adjusting Solar Panel Voc for temperature

When designing you solar panel system, it is important to adjust you solar panel Voc for temperature in order to ensure you do not over-voltage the PV inputs of your solar charge controller. This paper shows how to calculate the Temperature...

diysolarforum.com

Canadian solar 540W m10
This is a 13Amp panel,

I assume "13Amp Panel" is referring to the Imp rating of the panel. This is different (Lower) than the Isc used for calculating wire size. If I am wrong, please let us all know.

This is a 13Amp panel, so 1x string on the small inputs and 2x on the large (26A vs 25 is safely under the 1.2x clipping recommendations)

I assume the ratio you are talking about is the Array-Imp/Controler-Usable current. (Similar to the DC-Watts/AC -watts ratio used with string converters).

we always want to keep under 1.2 ratio here as you can run the numbers and find it is 1% total loss or less, going to 1.3 in a sunny climate takes another 1% off yearly production

When I first read this my reaction was "huh???". It sounds like it is saying the annual production goes down but that is probably not what you are trying to say. The annual production goes up as more panels are added and the ratio goes up, but the amount of unnrealized production during clipping also goes up. Over-paneling like this is often used to compensate for cloudy or winter conditions. It can also be used to compensate for panels at less than ideal angles or just the fact that panels never seem to operate at their rated capacity.

Overview of Over-paneling an MPPT controller

To get to the paper, click on the orange button at the top of the screen. There is a misconception that putting more panels on an MPPT controller than the controller is able to use will damage the controller. However, in almost all cases it is...

diysolarforum.com

 

[SignatureSolarJames]

Do you have any recommended tools for the users to calculate the voltage rise in cold temps? I generally point them to this resource:

Adjusting Solar Panel Voc for temperature

When designing you solar panel system, it is important to adjust you solar panel Voc for temperature in order to ensure you do not over-voltage the PV inputs of your solar charge controller. This paper shows how to calculate the Temperature...

diysolarforum.com

I assume "13Amp Panel" is referring to the Imp rating of the panel. This is different (Lower) than the Isc used for calculating wire size. If I am wrong, please let us all know.

Yes, Imp

I assume the ratio you are talking about is the Array-Imp/Controler-Usable current. (Similar to the DC-Watts/AC -watts ratio used with string converters).

correct. panel nameplate vs controller capacity is safely 1.2 ratio

When I first read this my reaction was "huh???". It sounds like it is saying the annual production goes down but that is probably not what you are trying to say. The annual production goes up as more panels are added and the ratio goes up, but the amount of unnrealized production during clipping also goes up. Over-paneling like this is often used to compensate for cloudy or winter conditions. It can also be used to compensate for panels at less than ideal angles or just the fact that panels never seem to operate at their rated capacity.

Overview of Over-paneling an MPPT controller

To get to the paper, click on the orange button at the top of the screen. There is a misconception that putting more panels on an MPPT controller than the controller is able to use will damage the controller. However, in almost all cases it is...

diysolarforum.com

correct, a 10kW array at 1.0 ratio in Texas will make 15548kWH per year whereas a 10kW array with a 1.2 ratio will make 15547kWh per year

 

[Cmy]

How does a Solarever 455 fit with the EG4-18? Higher Voc/Vmp, lower Isc/Imp than the 410 panel. Does the inverter care if three strings are used and the fourth empty for future growth? Using Alex Beale's calculations https://footprinthero.com/solar-panel-voltage-calculator to get:

455	Total W	Watts/S	Voc 49.35	MAX V 8F	Vmp 41.51
7	4st 12.7	3185	345.45	390.5	297.57
8	4st 14.52	3640	394.8	446.3	340.08
9	4 st 16.38	4095	444.15	493.5	382.59
10	3st 13.65	4550	493.5	548.4	425.1
11	3st 15.01	5005	542.85	603.2	467.61

I am 85 miles SW of Sulphur Springs and used 8F from the coldest afternoon of the 2021 Texas 'freeze'. Little solar at the predawn coldest temps.

Starting with pallet of 31 455s and adding where needed.

11 panels - 3 strings Too much? 15kW is nice, was looking for 14kW
10 panels - 3 strings Fits but one panel left over
9 panels - 4 strings Better four than three strings?, pvWatts shows 16.38kW yields 22.16kW annually. Monthly net metering, excess gifted.
Fewer than 9 panels wastes too much Voc capacity but was planning on the Sol-Ark 15 3250W string limit. The EG4 18 makes my install easier at a lower cost, I hope Will's initial test results are valid for future production.

 

[FilterGuy]

Does the inverter care if three strings are used and the fourth empty for future growth?

I am pretty sure the answer is no.

11 panels - 3 strings Too much? 15kW is nice, was looking for 14kW

According to the table provided, the array Voc would get above the 600V limit so yes, it would be too much.

9 panels - 4 strings Better four than three strings?

"Better" depends on the circumstance. It requires an extra pair of wires, but by spreading the panels out means all the strings will have a lower current. This will at least reduce the voltage loss and may allow for smaller wires on the home run.

In addition, if there are any shading issues, 4 strings may allow a layout that reduces the impact.

 

[Cmy]

Thanks for the response. Initially given the guideline that Voc total <90% of MBBT limit, then started reading about going slightly over limits. Your May 2 comment mentioned paneling to 500Vmpp and using the 100V remaining for cold temp reserve. James used 1.2 as a safe panel to inverter capacity ratio - almost the difference between Voc and Vmp. Just trying to learn, I may be taking things out of context.

Panels are going on a metal out building with a 200A breaker box. West facing (280 degrees) roof is 50'x35' (5 degree slope) allowing more panels than I need. Surprised me when pvWatts showed all 32 455W panels on the east side produced 19.59kW where all on the west side was 19.70kW. That allows two strings on each side, same end as the breaker panel, with shorter wire runs making up for any production loss. Even with 12AWG wire resistance is 1% or less and closer to 0.5% with 10AWG.

Thanks again for the help.

 

[TK1979]

This is good to know, one question though, if running 2ea 18KPVs, is it best to split the strings evenly between both? Or load one up? Any benifits/ cons either way?

 

[FilterGuy]

This is good to know, one question though, if running 2ea 18KPVs, is it best to split the strings evenly between both? Or load one up? Any benifits/ cons either way?

The answer to this is very dependent on the circumstances.

As a general rule of thumb, more MPPT controllers is a good thing.

However,
If you have a long run from the arrays to the controllers, 6 runs of cable (with 6 PV disconnects) might be a reason to not use all the mppts.)
It might be better to run fewer but higher voltage strings.

If you use both inverters and you need to take one of them off line, the other will still have some solar.

If you have some panels that get shade, it might be best to isolate those panels to their own MPPT.

 

[Zerocool00]

How would you configure the EG4-18K with these panels? I have 43 total 270W Trina panels and I am perplexed at what is best way to arrange the strings per MPPT. The other factor is my ground mount will be 475ft (wire run) away from the inverter. I have looked at the Sol-Ark panel sizing calculator, but not sure if what I am configuring is optimal for the 18K. Max/Min ambient are for my area in MS.


Appreciate the input!!

Michael

 

[FilterGuy]

How would you configure the EG4-18K with these panels? I have 43 total 270W Trina panels and I am perplexed at what is best way to arrange the strings per MPPT. The other factor is my ground mount will be 475ft (wire run) away from the inverter. I have looked at the Sol-Ark panel sizing calculator, but not sure if what I am configuring is optimal for the 18K. Max/Min ambient are for my area in MS.

With that minimum ambient temp and Voc temp coefficient, the highest STD Voc would be 600V/1.132 = 530V. (See this for how I got that)

That means the longest string is |530/44.3| = 11 panels

The easiest setup would be:

PV 1: Two strings of 11 panels - 22 Total​

PV 2: One string of 10 panels - 10 Total​

PV 3: One string of 10 Panels - 10 Total​

-----------​

42 Grand Total​

This configuration also meets the inverter specs for Isc and Imp

 

[Gavin Stone]

I agree with the above. But considering the length of the run I would probably combine the first string at the array.

 

[TK1979]

I agree with the above. But considering the length of the run I would probably combine the first string at the array.

Does this mean 4 wires?

 

[robby]

Does anyone even have a picture of their system achieving 18Kw of PV?

 

[EastTexCowboy]

Does this mean 4 wires?

He means combining the 4 wires down to 2 wires at the array, i.e. parallel. It would save you one of the wire runs. If I were doing it, I think I absolutely would do what Gavin recommends but I'd still pull the fourth set of wires, just to have a spare pair in case you ever need them. There's a cost of course, but pulling more wires later on that long of a run would not be fun. Just make sure you size your wire to handle the parallel circuit if you do this.

 

[Gavin Stone]

What he said. ??

I do think that’s a good idea about adding the extra string for expansion. But yes, paralleling the first strings at the array. So that would be 11+11.

 

[FilterGuy]

With that minimum ambient temp and Voc temp coefficient, the highest STD Voc would be 600V/1.132 = 530V. (See this for how I got that)

That means the longest string is |530/44.3| = 11 panels

The easiest setup would be:

PV 1: Two strings of 11 panels - 22 Total​

PV 2: One string of 10 panels - 10 Total​

PV 3: One string of 10 Panels - 10 Total​

-----------​

42 Grand Total​

This configuration also meets the inverter specs for Isc and Imp

View attachment 156116

If you really wanted to you could actually get it down to two runs from the array:

PV 1: Two strings of 11 panels - 22 Total (Paralleled at the array)​

PV 2: Two strings of 10 panels - 20 Total (Paralleled at the array)​

PV 3: Nothing​

​

PV 2 would have an Isc of 16.5 amps and an IMP of 15.2 amps so it would be ever so slightly over-paneled. This would not hurt the system but there is a small chance that the panels could be slightly underutilized at peak sun. If it did happen, it would probably not happen for more than a few minutes per day during June and July. (Around Summer Solstice).

BTW: To be in code, PV wire sizing has some special rules. See this resource:

Fusing & Wire Sizing guidelines for solar panels

To get the paper, click on the orange'' button at the top of the screen. This resource is intended to help the user understand when they need to add Over Current Protection Devices (OCPDs - Fuses or Breakers) to the solar panel array and what...

diysolarforum.com

 

[FilterGuy]

Does anyone even have a picture of their system achieving 18Kw of PV?

I haven't seen one but it could be done with these panels.

Since these panels could be doubled up on MPP2 and MPP3, there could be 22 panels on each MPPT for a total of 66 panels. That would be just shy of 18K.

It turns out that with these panels, you could put 3 strings of 11 on MPPT 1 so the total comes out to be 88 panels. That would be 23.76K, and that is higher than the overpanel spec of the inverter so you would have to cut back some.

 

[TK1979]

He means combining the 4 wires down to 2 wires at the array, i.e. parallel. It would save you one of the wire runs. If I were doing it, I think I absolutely would do what Gavin recommends but I'd still pull the fourth set of wires, just to have a spare pair in case you ever need them. There's a cost of course, but pulling more wires later on that long of a run would not be fun. Just make sure you size your wire to handle the parallel circuit if you do this.

Good to know, thank you!

 

[TK1979]

If you really wanted to you could actually get it down to two runs from the array:

PV 1: Two strings of 11 panels - 22 Total (Paralleled at the array)​

PV 2: Two strings of 10 panels - 20 Total (Paralleled at the array)​

PV 3: Nothing​

​

PV 2 would have an Isc of 16.5 amps and an IMP of 15.2 amps so it would be ever so slightly over-paneled. This would not hurt the system but there is a small chance that the panels could be slightly underutilized at peak sun. If it did happen, it would probably not happen for more than a few minutes per day during June and July. (Around Summer Solstice).

BTW: To be in code, PV wire sizing has some special rules. See this resource:

Fusing & Wire Sizing guidelines for solar panels

To get the paper, click on the orange'' button at the top of the screen. This resource is intended to help the user understand when they need to add Over Current Protection Devices (OCPDs - Fuses or Breakers) to the solar panel array and what...

diysolarforum.com

This makes a lot of sense, thank you!

 

[Zerocool00]

Thanks for all the information!

I am sourcing all my small(er) parts now and the wire is the next largest expense. I will use the SolArk tool to look at voltage drop with different configurations of parallel strings of 10AWG (3) vs 6AWG. I've found a deal on 10AWG THHN wire that I am trying to make work. Plan would be use run in 2" conduit underground from Ground mount back to solar/well pump shed. I also have to deal with meter base being 75ft away from that, but so be it. I plan to run a parallel run of 2" conduit when I trench for future expansion. I got these Trina panels dirt cheap a few years back so this 11.3kW will be array #1 into a single 18k unit.

Anyone have a good source for low cost wire?