Calculations to optimize Solar Panels - 6000XP at 350-ft from Solar Panels

[Steelart99]

Introduction: I started purchasing components for a solar system with an admitted ignorance of all the parameters that needed to be evaluated. I currently have an EG4 6000XP Inverter and an EG4 Wallmount indoor lithium battery (280ah) and I am happy with those choices for my current household load.

My initial calculations (flawed) indicated that I should use 400W panels in two strings of 8 panels to provide 6400W of power. Note: My panels will have to be about 350-ft from my inverter. Yikes!

I ended up purchasing Aptos DNA-108-BF10-400W, bifacial panels (not delivered yet) with plans to set these up as noted above.
Solar panel descriptions and spec sheet at:
https://solarguyspro.com/products/a...ck-up-to-500w-with-bifacial-gain-dna-108-bf10

Issue: When I got a bit more educated from this forum, I found out how much panel operating temperature needed to be taken into consideration. I did the first set of calculations below and realized that I perhaps should have obtained more panels to feed the inverter up to it’s designed input parameters … See the second set of calculations.

Questions: I would appreciate a quick review of my calculations and my thoughts associated with increasing my solar panel count.

• First … am I assessing performance correctly through these calculations?
• Am I correct to assume that I could add 4 more of the same panels (Total = 20 panels) to better utilize the 6000XP?
• Is it correct to say that in a 20-panel configuration, there is no advantage to having bifacial panels as the added power gain would not be used by the 6000XP?

Initial Calculation of Max Voltage (8 panels in series) SKIP to Lines 10, 11:

• Note: My temps range from -27F (-33°C) to about 100F (37.8°C) ---- Colorado
• Note: Temperature Coefficient of Aptos Panels: Voc = -0.30%/°C
• Temp Differential from STC = 25°C - (-33°C) = 58°C
• % increase = -0.30%/°C * 58°C = 17.4%
• Voltage Increase = (% Increase/100) * Voc
• Note: Open Circuit Voltage of Aptos Panels: Vvoc = 37.18V
• Therefore, Voltage Increase = 17.4/100 * 37.18V = 6.47V per panel
• Max Voc = Voc + Voltage Increase
• Max Voc = 37.18V + 6.47V = 43.65V
• 8-panels Max Voltage = 8 * 43.65V = 349.2V
• This is well under the 480V voltage limit on the 6000XP

8- Solar Panel String under normal operating conditions – SKIP to Lines 14-16:

• Note: The Vmp at NOCT for these panels is 31.28V
• String of 8 panels would be 8 * 31.28V = 250.2V
• The 6000xp is looking for a nominal 320V so 250V is a bit low but should work
• If the panels were generating a nominal 400W each, then the total power would be 16 * 400 = 6400W which is below the max utilized solar power for the 6000XP
• If the panels were generating 500W each (bifacial), then the total power would be 16 * 500 = 8000W which is equal to the maximum utilized solar power for the 6000XP

HOWEVER – Option of 10-Solar Panel Strings (Total of 20 Panels) SKIP to Lines 3, 7, 8:

• It appears that I might be able to use 2 sets of 10 panels in series i.e., 20 total panels
• Max Voltage (at -33°C) for 10 panels in series = 10 * 43.65V = 436.5V which is still under the 480V voltage limit on the 6000xp
• The Vmp at NOCT for these panels is 31.28V. A string of 10 panels would be 313V which is close to the nominal MPPT voltage of 320V for the 6000XP
• These panels are bifacial with a nominal 400W max power with just the front side and 500W with max backside power gain (likely to rarely happen)
• Max Power of 20 panels at 400W = 8000W; 20 panels at 500W = 10,000W
• Note: Maximum utilized solar power on the 6000XP = 8000W with a maximum solar input of 10,000W
• Based on the above calculations, 20 panels at nominal power (400W) will max out the utilized solar power on the 6000XP with no added advantage from the bifacial gain at 500W
• However, the gain in Vmax up to 465V by using 10 panels per string makes for more efficient transmission along the 350-ft run from inverter to solar panels

 

[sunshine_eggo]

Nominal MPPT voltage is max output power divided by usable PV input current. It's not a meaningful target. You want to your Vmp in the MPPT operating range.

Bifacial gains are almost never realized as stated. When conditions are favorable for the backside, they are less favorable for the front side. Thus, the bifacial aspect improves your chances of getting front side only rated power in less than optimal conditions.

Given the distance, you need to optimize for high voltage. 10S is the way to go.

 

[Steelart99]

Nominal MPPT voltage is max output power divided by usable PV input current. It's not a meaningful target. You want to your Vmp in the MPPT operating range.

Bifacial gains are almost never realized as stated. When conditions are favorable for the backside, they are less favorable for the front side. Thus, the bifacial aspect improves your chances of getting front side only rated power in less than optimal conditions.

Given the distance, you need to optimize for high voltage. 10S is the way to go.

Thanks SE. I suspected the bifacial gain was rarely, if ever, a true output. That said, I had not thought about what you'd said about front/back favorable conditions. Good point.

My suspicion was that high voltage was going to be the way to go, but it certainly helps to have those more knowledgeable in solar to valid our thought processes (for us neophytes at solar)

 

[Brucey]

The chances of getting 400W out of one is low, more likely closer to 300W, so with 10kW (nominal) of panels you'll have better utilization of the charger in everyday conditions.

 

[Steelart99]

The chances of getting 400W out of one is low, more likely closer to 300W, so with 10kW (nominal) of panels you'll have better utilization of the charger in everyday conditions.

It's a bit embarrassing that something that should be intuitively obvious ... simply escaped me ... Thanks for that insight Brucey

 

[ghostwriter66]

SOLARK website has a great panel size calculator

 

[shadowmaker]

I have close to identical problems with my vertical setup (roof mounted monofacials are closer at ~85m (~280ft)) and similar climate (maybe 5C colder both summer and winter. My Deye12K (~Sol-Ark15K) has 800V max and MPPT range of 200V-650V. String voltages are almost always over 650V (even summer) and go even over 800V every cold spring morning. Going over MPPT operating range doesn't seem to have any effect on their operation, at least I haven't been able to see it. Going over 808V seems to trip my inverters which causes some heat in my panels reducing voltage in two-three minutes and then everything works again. This happens some 20-30 times every year. I understand that can be critical for some inverters but my Deyes don't seem to mind at all.

To get transfer losses down I have direct bury ground cable with 4x25mm2 (>4AWG) aluminum for every four strings. Calculators say that this should keep losses around 1,2-1,4% with high voltages I'm using. Cu-Al transfer connectors are 2e/pc and alu cable is almost twice as cheap as Cu, so going alu was no brainer for me.

For bifacial gain, you'll be pleasantly surprised when there's snow on the ground.

 

[Steelart99]

I have close to identical problems with my vertical setup (roof mounted monofacials are closer at ~85m (~280ft)) and similar climate (maybe 5C colder both summer and winter. My Deye12K (~Sol-Ark15K) has 800V max and MPPT range of 200V-650V. String voltages are almost always over 650V (even summer) and go even over 800V every cold spring morning. Going over MPPT operating range doesn't seem to have any effect on their operation, at least I haven't been able to see it. Going over 808V seems to trip my inverters which causes some heat in my panels reducing voltage in two-three minutes and then everything works again. This happens some 20-30 times every year. I understand that can be critical for some inverters but my Deyes don't seem to mind at all.

To get transfer losses down I have direct bury ground cable with 4x25mm2 (>4AWG) aluminum for every four strings. Calculators say that this should keep losses around 1,2-1,4% with high voltages I'm using. Cu-Al transfer connectors are 2e/pc and alu cable is almost twice as cheap as Cu, so going alu was no brainer for me.

For bifacial gain, you'll be pleasantly surprised when there's snow on the ground.

I keep going back and forth on Cu vs AL for the long direct burial wires. While I've looked at various suppliers, I have not done a thorough comparison of what is available and the associated costs. When I'm sizing wiring, I'm trying to keep my line loss at 2% max. I'll have two sets of 10 panels, one for each MPPT on the 6000XP inverter so I'm looking at four 350-ft conductors plus a ground wire (needed???). Wiring can get a bit pricey.

Question: Is there an established listing of recommended wire suppliers (and wire types) on the forum?

 

[shadowmaker]

I keep going back and forth on Cu vs AL for the long direct burial wires. While I've looked at various suppliers, I have not done a thorough comparison of what is available and the associated costs. When I'm sizing wiring, I'm trying to keep my line loss at 2% max. I'll have two sets of 10 panels, one for each MPPT on the 6000XP inverter so I'm looking at four 350-ft conductors plus a ground wire (needed???). Wiring can get a bit pricey.

Question: Is there an established listing of recommended wire suppliers (and wire types) on the forum?

All four strings are pushing 11kWp through individual 4x25mm2 alu cables. With snow on the ground my inverters clip at 12kW for two hours on a good day. In one of those cables 16mm2 Cu wire is included for array grounding. I also have ~75m (~250ft) from my inverters to my main panel. We have 3-phase system here so single 4x35mm2 (2AWG)+16mm2 Cu direct bury cable is enough. Of course I need two such cables to be able use backup power when grid is down. I think I paid something like 2500e for my cables. Still almost half price compared to Cu. Just remember to use some kind no-ox paste with every possible connection, alu or Cu.

Here in Finland every alu cable sold needs to meet strict specifications so you buy where you can get it cheapest.

 

[Steelart99]

I just found some interesting pricing for AL and CU wire for carrying my PV loading. The 8AWG Cu is actually cheaper than the 6AWG AL. I don't know if this is "good" pricing or not ... still researching.

FWIW, my calculations were based on 350-ft, 436V, 15A, 2% Voltage Drop (https://solarwiresizecalculator.com/)

8 AWG CU (With 5% discount: $0.75 /ft)

8 AWG 19 Strands Copper Building Solar Photovoltaic PV Wire 2KV UL 4703

Specifications : Size AWG : 8, Weight : 0.084 lbs per ft, Number of Strands : 19, Outside Diameter : 0.25 inches, Insulation : XLPE, Insulation Thickness : 0.075 inches, Voltage : 1kV/ 2kV. Standards : Can be used as Type USE-2 per UL 854, Can be used as type PV per UL 4703, RoHS Compliant, UL...

nassaunationalcable.com

6 AWG AL (With 5% discount: $0.81 /ft)

6 AWG Aluminum Solar Photovoltaic 2KV PV Wire

Specifications : Size (AWG or KCM) : 6, No. of Strands : 7, Insulation Thickness (mils) : 85, Nominal Diameter (mils) : 339, Appox. Weight (lbs./kft.) : 55, DC Resistance at 20°C : 0.6609. Standards : UL Listed PV wire under UL 44 and UL 4703. Conductors : The PV cable conductor is an 8000...

nassaunationalcable.com

 

[HRTKD]

Similar copper PV wire at TEMCo Industrial is $.716/ft when buying 250' and $.718/ft for 500'. So your find is a competitive price, but the longer length of cable will make a difference in final cost.

For my system, I would not use Aluminum cable.

EDIT: clarified that I wouldn't use aluminum cable.

 

[Steelart99]

Similar copper PV wire at TEMCo Industrial is $.716/ft when buying 250' and $.718/ft for 500'. So your find is a competitive price, but the longer length of cable will make a difference in final cost.

For my system, I wouldn't not use Aluminum cable.

TEMCo came back and told me they could do 250' or a 500' roll, but would not cut/sell at 350'

HRTKD, I'm not sure I understand ... you would not use AL or you "wouldn't not" use AL ... which means you "Would" use AL?

 

[HRTKD]

Err, I would not use Aluminum wire.

 

[mZooma]

Get your wire yet? I took my time and found a couple sources, of stranded copper 10 gauge to run this distance. Put it through three-quarter PVC then sunk it 1 foot .I’ve got two strings of 12 laying flat. At Peak they provide 4000 W over 300 foot away..

 

[Steelart99]

Get your wire yet? I took my time and found a couple sources, of stranded copper 10 gauge to run this distance. Put it through three-quarter PVC then sunk it 1 foot .I’ve got two strings of 12 laying flat. At Peak they provide 4000 W over 300 foot away..

Sounds like you had a similar long distance to deal with as I did. Its nice to have that done, isn't it.

I did get my wire and had quite the "experience" putting it through 350' of 1-1/2" conduit in a 24" deep trench. 4 strands of 8 AWG and one strand of 10 AWG.

I first rented a 24" trencher which did so-so on parts of my property. Engine keep cycling so the power to the chain keep cycling too. Didn't work well at all. Got about half my money back due to the issues. Oh yeah, it's an hour drive in each direction for me to pick it up and return it (4 hours of drive time total).

So I went back and rented a 36" trencher. I started that up and got about 5' before it threw its chain off. The chain alone probably weighed 100-lbs. After about 2-hours of exhausting wrestling, I was finally able to get the chain back on ... then at about 20' of trenching, it threw the chain again and jammed the whole bar deep down in the trench. Another hour to get it dug out and sorta loosened up, then another hour to get the chain on again. I did more "maintenance" on the damn thing than the rental company ever did ... the chain bar had never been lubricated as far as I could tell and I beat the hell out of the bar to get it loosened up enough to put the chain back on. Eventually, after about 9 hours, I got the 350' trench "done". But of course, I needed another 2 days to hand dig out all the dirt that fell back into the trench. Our soil around here is really hard to get through.

In the end, I was credited 3/4 of the rental cost on the 36" trencher due to all my issues.

My wife and I set up and pulled the wires in about 4 hours.

This is fun ... Right?

 

[JBoffgrid2022]

I hope it's not too late but you'll definitely want to make sure you haven't packed too many wires inside of your conduit. Check your setup with this calculator:
https://www.cerrowire.com/products/resources/tables-calculators/ampacity-calculator/

I have nearly the same setup but x3. I have 48 panels distributed between three inverters going through (3) 2" underground schedule 40 conduit. The run is only 150' and my electrician and I decided to go 8AWG wire to reduce any current drop. It's overkill but we wanted to make sure we had room for expansion if we ever need it down the road. Direct message me if you'd like any photos or information on our install.

 

[Steelart99]

I'd done the conduit fill evaluation before buying the conduit. I only have four 8AWG wires and one 10AWG wire in a 1-1/2 PVC pipe. Per the calculator, I could have 21 8AWG wires. I'm good. I also went with 8 AWG to reduce the current drop.

Wow ... you have a lot going on with your system

 

[JBoffgrid2022]

Not the best photo I have of recent work but it shows the rack has been back filled and graded! Damn that was a lot of gravel.... lol

 

[Steelart99]

You have quite the large array there! Amazing how much work these systems end up being. I know I've spend 10-times as long on mine as I thought would be needed.

I've been working on my ground mounts for my 20 panels. I have a lot of tall trees and I spent a lot of time mapping our tree shadow patterns to determine a good placement. I ended up having a N-S, narrow pattern for my layout to avoid shadowing and the issues with low winter sun angle. I have five rows of 4 panels with about 10' between the rows. Row spacing was needed to avoid Southern panels shading the panels behind.

Today I was aligning and torqueing down all the fasteners on my EG4 BrightMount ground mounts. I'll have two circuits of 10 panels each. The ground mounts are mounted on concrete pillars that are 30-36" underground and up to about 30" above ground. Everything on a 5-deg slope (in two directions) which made the layout a bit "fun".

Couple of pics, including one showing me trying to reattach the trencher chain. Last picture was today's efforts to get the ground mounts bolted down to the concrete pillars.

 

[JBoffgrid2022]

I feel your pain brother!
I went with a Sinclair rack because the state electrical inspector here in Maine required it to be both UL listed and signed off by an engineer. The cost wasn't terrible either for the footings and racking system. I was able to assemble the rack within two short days once the footings were poured. Overall, very happy with results! The rack is just under 100' long and stands around 20' tall when angled at 55 degrees. In the summer I will lower the rack to 15 degrees. I may wait to see how my production goes before I change the angle due to them being bifacial. I even thought of laying white crushed rock under the array to see if that would help with solar production lol.

 

[Steelart99]

Ha ... I just lay snow underneath to help with my bifacial solar panels

 

[JBoffgrid2022]

I'm having to do sooooo much extra work to pass inspections because this is a new build that will have to pass both local and state electrical code....
At least I found an electrician to help me! I called several companies that said they had "no interest" in working on any solar projects....
If I didn't already have a thriving painting company would I go into solar with focus on off grid application. So many people are turning to this up here due to the high electrical/utility costs. My electrical bill here in Maine is anywhere from $500-$800/month....