Alternative to Conext MPPT 80-600V Charger for (2) Distinct PV Arrays?

[Old_Skewler]

My current system is a 48V Conext SW4048 inverter with 15 kWh energy storage. I need to install (2) distinct solar arrays on my roof, one with (6) panels and the other with (5) panels, different panel quantities and different panel orientation.

The Conext Charger Controller MPPT 80 600 seems to meet the specs but not the bank budget at $1,600 shipped to me. I am looking for alternate charge controllers that could work with my setup but a few questions have surfaced while doing this research:

1. If a Solar Charger controller has two array inputs, could each array have a different power output and/or quantity of solar panels?
2. Would it be easier to find a 2-array solar charger or to install (2) separate chargers in my situation? I am looking for easier and cheaper solution.

Any feedback on this would be welcome.

Thanks in advance!

 

[sunshine_eggo]

My current system is a 48V Conext SW4048 inverter with 15 kWh energy storage. I need to install (2) distinct solar arrays on my roof, one with (6) panels and the other with (5) panels, different panel quantities and different panel orientation.

Are the 6 facing one direction and the 5 facing another?

1. If they are truly separate MPPT, yes. If they are just convenience connections to the same MPPT, it's not different than putting strings in parallel. They would need to be the same Vmp.
2. It depends on the product and price points.

 

[Hedges]

If you adjusted panel quantity (i.e. removed one panel from string of 6 to make it 5) so both strings were same/similar Vmp, then they can be paralleled and fed into one MPPT. If amperage is OK. This makes better utilization of MPPT, operating it for more hours.

 

[Old_Skewler]

Are the 6 facing one direction and the 5 facing another?

Exactly.

1. If they are truly separate MPPT, yes. If they are just convenience connections to the same MPPT, it's not different than putting strings in parallel. They would need to be the same Vmp.

Understood, but begs the follow up question: the Conext MPPT 80 600 is listed as two arrays, so that means two separate MPPT. Am I reading this correctly?

2. It depends on the product and price points.

Understood. I am trying to find an alternative to the Conext so that I can determine if buying (2) MPPTs is better than (1) larger MPPT.

Cheers!

 

[Old_Skewler]

If you adjusted panel quantity (i.e. removed one panel from string of 6 to make it 5) so both strings were same/similar Vmp, then they can be paralleled and fed into one MPPT. If amperage is OK. This makes better utilization of MPPT, operating it for more hours.

But if they are facing different exposures, would this still work? I was under the impression that paralleling strings with different exposures would affect power output and underperform.

Please clarify how this would work before I look into this option further.

 

[Hedges]

What voltage are your strings? Price of that SCC is due to 600V limit.

Understood, but begs the follow up question: the Conext MPPT 80 600 is listed as two arrays, so that means two separate MPPT. Am I reading this correctly?


Understood. I am trying to find an alternative to the Conext so that I can determine if buying (2) MPPTs is better than (1) larger MPPT.

Cheers!

I believe it is 2 string connections but a single MPPT.

But if they are facing different exposures, would this still work? I was under the impression that paralleling strings with different exposures would affect power output and underperform.

Please clarify how this would work before I look into this option further.

Yes it will work great (unless one string has larger percentage shade when other does not.)

The Vmp of panels with different orientation aren't that far apart. SMA once had a paper reporting their field tests, only about 2% reduction in output watt-hours compared to separate MPPT. The savings in MPPT cost makes it a good deal.

Up to maybe 15% shading on one string vs. the other is OK. If 50% shading of one (25% shading of total number of panels) expect to lose 50% not just 25%.

 

[sunshine_eggo]

Understood, but begs the follow up question: the Conext MPPT 80 600 is listed as two arrays, so that means two separate MPPT. Am I reading this correctly?

You need to look at the specs. The wording implies two MPPT.

But if they are facing different exposures, would this still work?

Yes. Their optimal voltage will not change, just the amperage of each string.

I was under the impression that paralleling strings with different exposures would affect power output and underperform.

Not on a quality MPPT.

 

[Old_Skewler]

What voltage are your strings? Price of that SCC is due to 600V limit.

We are looking at Canadian Solar 385W Mono-crystalline panels, CS3N-385MS

Vmp= 36.6V
Imp= 10.52A
Voc= 43.9V
Isc= 11.32A

I believe it is 2 string connections but a single MPPT.

I need to give Schneider a call and verify, I was under the impression their higher end MPPT was capable of handling (2) parallel strings.

Yes it will work great (unless one string has larger percentage shade when other does not.)

The arrays would have a different exposure and shading. I just don't see how combining all into one MPPT would render the best results.

The Vmp of panels with different orientation aren't that far apart. SMA once had a paper reporting their field tests, only about 2% reduction in output watt-hours compared to separate MPPT. The savings in MPPT cost makes it a good deal.

Interesting finding, for sure.

Up to maybe 15% shading on one string vs. the other is OK. If 50% shading of one (25% shading of total number of panels) expect to lose 50% not just 25%.

 

[Old_Skewler]

You need to look at the specs. The wording implies two MPPT.

Under the data sheet this is the only refence about it. After more consideration, i suspect this is one MPPT despite two PV Solar inputs? Gotta give them a call to clarify.

Not on a quality MPPT.

Understood.

 

[sunshine_eggo]

Under the data sheet this is the only refence about it. After more consideration, i suspect this is one MPPT despite two PV Solar inputs? Gotta give them a call to clarify.

I downloaded the installation manual and the specifications strongly indicate a single MPPT.

 

[Old_Skewler]

I downloaded the installation manual and the specifications strongly indicate a single MPPT.

Oh shit.. Does that mean at minimum I'll need the two arrays to have the same panel quantities?

 

[Hedges]

Oh shit.. Does that mean at minimum I'll need the two arrays to have the same panel quantities?

Yes. (or same/similar Vmp if different models. I've used 24x "12V", 12x "24V", 8x "36V" all in parallel.

We are looking at Canadian Solar 385W Mono-crystalline panels, CS3N-385MS

Vmp= 36.6V
Imp= 10.52A
Voc= 43.9V
Isc= 11.32A

5s or 6s would be 220Voc or 264Voc nominal, ~253 or 303V in cold weather
There are some lower voltage controllers that might work for that, e.g. Midnight Classic 250 for 5s.
Or an array 3s2p would fit a lower voltage one.

All mine are high voltage strings, easy and economical as part of an AC coupled system, but you may have to pay up for high voltage SCC.

The arrays would have a different exposure and shading. I just don't see how combining all into one MPPT would render the best results.

Most cost effective, unless differences in shading make performance penalty too high.
How much power do you expect from a string when it is not only partially shaded but also off angle? Could be the times you lose out, you don't lose much. A shaded array won't pull down performance of unshaded array but might not contribute anything itself.

 

[sunshine_eggo]

Oh shit.. Does that mean at minimum I'll need the two arrays to have the same panel quantities?

Yes.

5 panels is going to be problematic no matter what. This forces 5S meaning you need a MPPT with at least 250Voc input.

A Midnite classic 250 would do it, and its hyperVoc feature means you'll never have a cold temperature issue. This will give you 63A of charging. That should more than handle the 385W*5 array @ 40A of charge current.

If you dropped a panel and went 2S2P, you could go with a 150V controller. A victron 150/35 is a little over $300.

The 6 panels might be more economical in a 2S3P or 3P2S configuration. 2S3P would allow for a 150V controller. A Victron 150/45 is under $400, and it would only be slightly overpaneled.

 

[Hedges]

"We are looking at" Canadian Solar 385W Mono-crystalline panels, CS3N-385MS

Meaning you don't have them yet, can consider other to fit the available space. You could have multiple strings of same voltage but different orientation and fit within limits of a single charge controller.

Midnight Classic 150 handles more watts for fewer dollars than the 250. Because it has Hyper Voc, strings could be up to 150V nominal (on a cold day, would not start to operate until panels warm up and Voc drops below 150V, so some voltage headroom would be appropriate. 3s2p with the panel you're considering would fit.

If you do have 3 or more strings paralleled, use fuse per string.

 

[Old_Skewler]

5s or 6s would be 220Voc or 264Voc nominal, ~253 or 303V in cold weather
There are some lower voltage controllers that might work for that, e.g. Midnight Classic 250 for 5s.
Or an array 3s2p would fit a lower voltage one.

Yes! This is precisely what I need to better understand so that I can better spec this controller. A few follow up questions:

1. If I went this route I would need (2) charge controllers, one per solar array, correct?
2. What should be better in my case: higher current (2S3P) or higher voltage (3P2S) for the (6) panel array?
3. For the smaller array with (5) panels it seems I should be able to use the Classic 250 under HyperVOC mode, whatever that is..

 

[Old_Skewler]

5 panels is going to be problematic no matter what. This forces 5S meaning you need a MPPT with at least 250Voc input.

A Midnite classic 250 would do it, and its hyperVoc feature means you'll never have a cold temperature issue. This will give you 63A of charging. That should more than handle the 385W*5 array @ 40A of charge current.

Yes, thanks for the confirming it. It took me a long time but I understand now what you mean, I went over the Midnite literature just now.

If you dropped a panel and went 2S2P, you could go with a 150V controller. A victron 150/35 is a little over $300.

That is very interesting.

The 6 panels might be more economical in a 2S3P or 3P2S configuration. 2S3P would allow for a 150V controller. A Victron 150/45 is under $400, and it would only be slightly overpaneled.

So far I went from a $2K Schneider 600V unit to (2) Midnite 250V @ $800 each to possibly (2) Victron at $800 both!

I can see a light at the end of this tunnel.

Thank you very much for taking the time to explain this.

 

[Old_Skewler]

Meaning you don't have them yet, can consider other to fit the available space. You could have multiple strings of same voltage but different orientation and fit within limits of a single charge controller.

I actually have (4) Q-CELLS-Q.PEAK-DUO-L-G7-405W, donated from a PV Farming client. These are sitting on my basement, unused. I decided to sell these (4) and buy new matching panels and start fresh with the solar install.

What other panel would you recommend? What am I looking for, that would help me set them up in a way that I could use a single controller? Ideally I would like to find a single controller for less than $800 that could handle about 10 to 11 panels mounted on two roof areas.

Should I be looking for lower voltage panels? lower current? Or just lower wattage?

 

[Hedges]

Should I be looking for lower voltage panels? lower current? Or just lower wattage?

Consider your available area/space, and see what panel sizes tile it best, while making strings of desired voltage.
5s was undesirable. 3s2p or some other configuration might use all area and give same voltage as other orientation.
Newer panels are 20% or more efficient, meaning 200W/m^2. Older panels around 130W/m^2.

If you can fit multiple strings of two orientations within the parameters of Midnight Classic 150, that may be attractive. Or maybe some solution from Victron.

Having more than one SCC gives redundancy in case one fails.

"Hyper Voc" means Voc is acceptable up to voltage spec of SCC (e.g. 150V) plus battery voltage (e.g. 48V). I think they simply recognize over-voltage and don't start switching the buck converter, so transistor sees Voc - Vbat. (If it switches transistor and inductor, voltage flies down to negative voltage of battery, and transistor sees full Voc.)

2. What should be better in my case: higher current (2S3P) or higher voltage (3P2S) for the (6) panel array?
3. For the smaller array with (5) panels it seems I should be able to use the Classic 250 under HyperVOC mode, whatever that is..

Don't say "3P2S", say "2S3P", difference being groups are connected in series first, then parallel. (batteries some times get done either way, but PV panels always series first. Except connections within some PV panels, especially "half cut" ones.)

I think you meant or 3S2P. I think 3S2P or 3S3P or something like that, on a Classic 150, would be better.
The Classic 250, you pay more dollars for fewer watts, but can go higher voltage. Higher voltage can be a benefit for long wire runs.
Since yours is on the roof, I think you can have more strings in parallel, higher current, a bit heavier wire.

I actually have (4) Q-CELLS-Q.PEAK-DUO-L-G7-405W, donated from a PV Farming client. These are sitting on my basement, unused. I decided to sell these (4) and buy new matching panels and start fresh with the solar install.

You don't necessarily have to have matching panels, although it is normally done within an electrical array. Consider what Vmp they would be in 4s or 2s2p, if that would work with the other panels. And how well they fit.

 

[Hedges]

I think you meant or 3S2P. I think 3S2P or 3S3P or something like that, on a Classic 150, would be better.

Well, it is possible 2s3p is better than 3s2p. SCC is more efficient when input voltage is closer to battery voltage.
But need voltage to remain high enough in hot weather. And with partial shading (a section of a panel bypassed through its diode.)
You can study the curves published for Classic to decide. Few vendors give that much information.

 

[Old_Skewler]

Meaning you don't have them yet, can consider other to fit the available space. You could have multiple strings of same voltage but different orientation and fit within limits of a single charge controller.

Midnight Classic 150 handles more watts for fewer dollars than the 250. Because it has Hyper Voc, strings could be up to 150V nominal (on a cold day, would not start to operate until panels warm up and Voc drops below 150V, so some voltage headroom would be appropriate. 3s2p with the panel you're considering would fit.

If you do have 3 or more strings paralleled, use fuse per string.

Okay, after some back and forth I believe we settled this part of the design:

(12) Canadian Solar 400W panels wired 6S2P coupled to (1) Victron 250/75 MPPT. And I will list my (4) QCells for sale or leave them for a future project. Here is the Victron Calculator results:



It would be amazing if the above was done correctly and there is no other surprises regarding the panel selection, configuration and MPPT.

Thanks again!