My first PV System

[Logic28]

I have a question for you guys and gals:

I’m building a ground mount 15 modules PV System with a dedicated detached solar shed for all my Victron goodies.

I have prepared both a combiner box with the three lines each with its own twin fuse, twin surge protector/breakers/reverse diodes plus a separate box with the three isolating switches initially intended to be installed on one of the PV posts.

However, looking for the appropriate cable, I came out with the best solution being an armoured 7 ways 10mm each using 6 of them, 2 per string.
Now, is it worth installing the combiner box out in the open next to the isolators or, since I will be adopting individual lines anyway, should I perhaps move the combiner box comfortably inside the solar shed 15 metres away?
In which case, since it’s all travelling inside a 75mm double wall trunking, why not using standard black and red solar cables?

Any comment will be very much appreciated
Many Thanks

 

[Bluedog225]

I’m sorry to tell you that you’ve been scammed. That’s a sheet of osb, not a solar panel. We’ve all been there. ?

 

[Logic28]

I’m sorry to tell you that you’ve been scammed. That’s a sheet of osb, not a solar panel. We’ve all been there. ?

Nice one mate! I was waiting for the first one to say that ?
Twelve here and three more on their way now

 

[Hedges]

I have prepared both a combiner box with the three lines each with its own twin fuse, twin surge protector/breakers/reverse diodes plus a separate box with the three isolating switches initially intended to be installed on one of the PV posts.

However, looking for the appropriate cable, I came out with the best solution being an armoured 7 ways 10mm each using 6 of them, 2 per string.
Now, is it worth installing the combiner box out in the open next to the isolators or, since I will be adopting individual lines anyway, should I perhaps move the combiner box comfortably inside the solar shed 15 metres away?

You can combine at either end. Multiple wires in a conduit are derated to lower current, but individual wires have much greater ampacity than what one PV panel/string puts out.

I brought MC cable into a waterproof box and spliced to normal wires which ran in conduit, a pair of wires for each 3000W PV string, one per 2500W 600V max inverter. Later when I changed inverters, I was able to reconnect two, three, or however many in parallel.

I don't have a disconnect at the array; maybe we are supposed to with more recent codes. I disconnect at the combiner box or inverter, verify Voc not Vmp so I know zero current, then unplug MC cables at array if I want to work on wires connected to PV side of disconnect.

I don't believe reverse diodes serve any purpose. Virtually no current flows backwards through a PV string at Vmp.

Disconnects should be rated for DC at Voc and Isc. If they are polarized, I don't believe they should be used for more than two PV strings in parallel (PV normally pushes current one direction, but one shorted string in parallel with others means current flows other direction.) In this case, a diode could make polarized breakers OK.

 

[Logic28]

You can combine at either end. Multiple wires in a conduit are derated to lower current, but individual wires have much greater ampacity than what one PV panel/string puts out.

Thanks for your very descriptive response. My obvious first question is why would that be? Just trying to understand the physics of it

 

[Logic28]

I brought MC cable into a waterproof box and spliced to normal wires which ran in conduit, a pair of wires for each 3000W PV string, one per 2500W 600V max inverter. Later when I changed inverters, I was able to reconnect two, three, or however many in parallel.

I have three identical strings of 2000W each (11/13A) eventually combined, so you would suggest using individual wires inside the twin wall ducting underground even though they would be unprotected by the MC? I was just thinking about any intrusion penetrating the ducting like potential rodents like moles which we seem to have in abundance in this area.

 

[Logic28]

Disconnects should be rated for DC at Voc and Isc. If they are polarized, I don't believe they should be used for more than two PV strings in parallel (PV normally pushes current one direction, but one shorted string in parallel with others means current flows other direction.) In this case, a diode could make polarized breakers OK.

I have implemented three DC dedicated disconnects also in order to work more easily on the combiner unit regardless of whether it is next to them or further away inside the solar shed.
I guess my only concern about moving the combiner 15 mt away was about protection for the three lines (fuses and RCD) being better upstream rather than downstream.

 

[Logic28]

I don't believe reverse diodes serve any purpose. Virtually no current flows backwards through a PV string at Vmp.

And finally, as far as the reverse diodes, I don't see how I could prevent any of the strings/moduled partially or fully shaded from being seen as resistance by the other strings which are eventually connected in parallel.

 

[Hedges]

Thanks for your very descriptive response. My obvious first question is why would that be? Just trying to understand the physics of it

The NEC ampacity table assumes 30 degrees C ambient, and quotes the current in three cable conductors that would raise temperature 60 degrees to its 90C insulation rating (or lower temperatures, for lower rated insulation, or because switch/breaker terminals have 75C rating.

If you had more wires carrying current, more power dissipated with less surface area so greater temperature rise. Therefore less current allowed.
Obviously "raceway" and "cable" are different from metal conduit or PVC conduit. I haven't seen how to address their greatly different thermal resistance.

https://www.usawire-cable.com/pdfs/nec%20ampacities.pdf

I have three identical strings of 2000W each (11/13A) eventually combined, so you would suggest using individual wires inside the twin wall ducting underground even though they would be unprotected by the MC? I was just thinking about any intrusion penetrating the ducting like potential rodents like moles which we seem to have in abundance in this area.

"Thin wall ducting", I'm not sure what material or configuration. I'm familiar with rigid steel, PVC, thin steel EMC (not for burial), direct burial cables. What kind do rodents get inside? Follow local wisdom on how to protect wires. MC gets chewed too. Underground, make sure wires are rated for wet. Sometimes, lower temperature rating wet.

I've found separate wires for each PV string convenient when I later rearranged. I suppose you could do rearranging to some extent at the outside combiner. I have multiple physical arrays, each with multiple strings. I pull more wire, but conduit is getting stuffed. I recommend oversize conduit, and leaving a cord or spare wire to pull more.

I have implemented three DC dedicated disconnects also in order to work more easily on the combiner unit regardless of whether it is next to them or further away inside the solar shed.
I guess my only concern about moving the combiner 15 mt away was about protection for the three lines (fuses and RCD) being better upstream rather than downstream.

Fuses should be located where they protect downstream wires (or panels). If fuses are in combiner in panels, probably no need to protect home run wire. Any source that can put excessive current through it? Generally fuses per string are only needed for more than two parallel strings into one MPPT.

And finally, as far as the reverse diodes, I don't see how I could prevent any of the strings/moduled partially or fully shaded from being seen as resistance by the other strings which are eventually connected in parallel.

I measured a PV panel in shade, and its Voc was similar to Vmp of other panels in the sun and powering inverter. So I don't think it sinks any current, except when other strings are unloaded or partially loaded (trying to produce Voc because system off or between Vmp and Voc because batteries are getting full and SCC isn't drawing all.)

I think power dissipated by the diode is noticeable (whatever percentage of string Vmp), and power savings by blocking leakage is about zero.
The anti-backfeed diode obviously needs to be able to handle Imp continuously without overheating. Too bad bypass diodes usually aren't sized like that (can burn up and damage panel if shaded when other panels get direct sun.)

There are "ideal diode" circuits which use a MOSFET, avoiding voltage drop of a diode.

 

[Logic28]

There are "ideal diode" circuits which use a MOSFET, avoiding voltage drop of a diode.

Yes, that is what I am intending to use as precautionary measures when I can find the adequate model for the voltage, thanks again for all your valuable input on this subject.

 

[HRTKD]

I brought MC cable

A term I was unfamiliar with. I had to look it up. Metal Clad Cable.

Armorlite® Type MC | Southwire

Choose Armorlite® Type MC Cable by Southwire. UL Listed 600 Volt cable with aluminum interlocked armor for commercial and industrial buildings.

www.southwire.com

 

[fblevins1]

Why do I get the feeling your mount project is going to be just a tad more functional than mine? I needed to get the panels up and going for the lab so about 50 dollars worth of cement blocks and heavy duty tie wraps. Have a good laugh. It was pushing 4PM in Monterey so the sun was pretty low and off to the right (picture left)

 

[Hedges]

A term I was unfamiliar with. I had to look it up. Metal Clad Cable.

Armorlite® Type MC | Southwire

Choose Armorlite® Type MC Cable by Southwire. UL Listed 600 Volt cable with aluminum interlocked armor for commercial and industrial buildings.

www.southwire.com

Apparently I lied.

I brought PV wire with pre-installed MC connectors to the junction box.

That "MC cable" looks suspiciously like the "BX" I'm familiar with.

Why do I get the feeling your mount project is going to be just a tad more functional than mine? I needed to get the panels up and going for the lab so about 50 dollars worth of cement blocks and heavy duty tie wraps. Have a good laugh. It was pushing 4PM in Monterey so the sun was pretty low and off to the right (picture left)

I intend to have PV as a utility coming to junction box in my new office/lab/workshop. For use tinkering with inverters, instead of unrolling wire across the living room floor as I have been doing. I already use dryer or stove type outlets to plug in AC cable of inverters, and twist-loc for 3-phase of a couple voltages.

 

[WindWizard]

Why do I get the feeling your mount project is going to be just a tad more functional than mine? I needed to get the panels up and going for the lab so about 50 dollars worth of cement blocks and heavy duty tie wraps. Have a good laugh. It was pushing 4PM in Monterey so the sun was pretty low and off to the right (picture left)

It looks awesome for a temporary solution. I would worry that the wind may cause havoc with your solar panels.

 

[fblevins1]

It looks awesome for a temporary solution. I would worry that the wind may cause havoc with your solar panels.

It was/is a concern, I am hoping the weight of the cement blocks will keep them down, I did put two more blocks on the headwind side to help keep a wind sail from forming under the panels. Oh crud, my son just noted that the headwind blocks are casting a shadow onto the panels. Whoopsie

 

[Hedges]

A 1/2" expanding anchor bolt in concrete is good for 5000 lbs. Which is like a ton of cinder blocks. 2.5 tons, actually.

But I prefer to put in an expanding insert with female thread, so whatever is attached can be removed without leaving a stud sticking up.

(When all you've got is a rotohammer ...)

 

[timselectric]

That "MC cable" looks suspiciously like the "BX" I'm familiar with.

But 100 x easier to work with.

 

[WindWizard]

I have prepared both a combiner box with the three lines each with its own twin fuse, twin surge protector/breakers/reverse diodes plus a separate box with the three isolating switches initially intended to be installed on one of the PV posts.

@Hedges has given you some good advice.

I have found that diodes are not necessary on my system. My system is arranged as 3S X 5P X 2. I put fuses on each set of 3 panels and did all the combining at the panels and brought in 2 seperate sets of MC4 8 AWG. All of my wires are enclosed in a 1 1/2" PVC pipe in the ground. The PVC pipe is fully enclosed so no rodents can get into them. The 3 panels can produce a VOC of around 150 Volts. I have two Victron 250/100 controllers that handle the load very well. It can produce around 10kw of power for up to 80kw per day.

 

[Hedges]

"But 100 x easier to work with."

How so?
I've used wire that came with spiral metal jacket around it, and I've bought the flexible metal conduit and stuffed wire through it.
Some is steel, some is aluminum.
I don't know the difference between MC and BX.

 

[WindWizard]

A 1/2" expanding anchor bolt in concrete is good for 5000 lbs. Which is like a ton of cinder blocks. 2.5 tons, actually.

I like to just take a hammer drill and put some holes in the cement and then buy just regular concrete bolts and drill them in.