Solar PV Array Switching

[C4NYS]

Hello, Due to my geography and an attempt to keep my hardware costs down I have come up with a 3 Array system, Morning - Midday- Afternoon. Except for the number of panels each array is hardware identical. One expensive device I would like to conserve with is the Charge Controller. Has anybody ever seen an autonomous switching device that could be used so I would only need one Charge Controller during a solar cycle?
Bruce

 

[sunshine_eggo]

Hello, Due to my geography and an attempt to keep my hardware costs down I have come up with a 3 Array system, Morning - Midday- Afternoon. Except for the number of panels each array is hardware identical. One expensive device I would like to conserve with is the Charge Controller. Has anybody ever seen an autonomous switching device that could be used so I would only need one Charge Controller during a solar cycle?
Bruce

using split arrays to optimize power delivery throughout the day, this is the first time I've seen a switch proposed.

Why shut off power from potential sources? Your morning array will still produce midday and evening. why waste that with added cost and complexity?

Simply size your MPPT to handle the total output of your 3 arrays.

You can simulate each array individually at your exact location in PVWatts (in my sig, line #1). Download hourly output to .csv file and then combine the three files to yield your aggregate array. Size a single MPPT for that, never switch, and enjoy converting more photons to electrons!

 

[Zwy]

using split arrays to optimize power delivery throughout the day, this is the first time I've seen a switch proposed.

Why shut off power from potential sources? Your morning array will still produce midday and evening. why waste that with added cost and complexity?

Simply size your MPPT to handle the total output of your 3 arrays.

You can simulate each array individually at your exact location in PVWatts (in my sig, line #1). Download hourly output to .csv file and then combine the three files to yield your aggregate array. Size a single MPPT for that, never switch, and enjoy converting more photons to electrons!

I've often thought of a possible DC transfer switch in order to switch an array to a different system. I will have 2 separate systems here when I finish, one at the shop, the other at the house. Current array is behind the shop and was the best spot on the whole acreage. I'm currently debating where the pallet of panels on the way will be installed. Having a DC transfer switch to change where the power goes might be handy. Shop needs more PV in summer and less in winter. The house could use more in winter due to heating needs with heat pumps, less in summer.

Eaton double pole switches, well most of them, are rated to 250V DC. I have to wonder how the DC arc gets broken as I have one in my basement for a transfer switch and there is no arc chute but there is a spring loaded switching mechanism for fast breaking of current. One could use an IMO ahead of a double throw to break any arc if one desired to switch during the day.

 

[Rocketman]

If you wanted to switch PV - look at four Tyco Contactors Ev200AAANA. Someone was selling used ones (from EV’s) pretty cheap. One for each leg going to a mppt. Then use a dpdt switch to turn the two to the garage off while simultaneously turning the two to the house on.

I will update if I remember the vendor.

Edit - battery hookup .com

 

[Zwy]

If you wanted to switch PV - look at four Tyco Contactors Ev200AAANA. Someone was selling used ones (from EV’s) pretty cheap. One for each leg going to a mppt. Then use a dpdt switch to turn the two to the garage off while simultaneously turning the two to the house on.

I will update if I remember the vendor.

Edit - battery hookup .com

TE Connectivity EV200AAANA 500a 0-900vdc Tyco 12/24v

These were recovered from EV battery modules. These are VERY EXPENSIVE online. If you have any project that needs a high current relay this is your chance to save some money. Make a cheap 30a BMS into a 500a BMS. This will work for batteries in the 9-36v range so perfect for Lifepo4 and Lithium...

batteryhookup.com

Thanks, I'll keep that in mind. I would need a 24V transformer added for the coil but it appears these are rated on switch contacts to 900V with 2000A at 320V.

 

[sunshine_eggo]

Those EV200 relays have a 9-36V coil input. Works great on my 12V system.

While those will work, it's still generally not recommended to disconnect an array while active, so one would need to turn off the MPPT, switch, turn on the MPPT especially since it's being done daily.

I'm going to loop around to the idea that it makes sense to:

1) add complexity
2) add cost
3) lose available power

Rather than size the MPPT for the aggregate array.

I don't see how this makes sense at all. The respective arrays must have the same series Vmp, but that can be accomplished simply by varying the number of strings.

I eventually plan to have (all panels acquired):

5940W South array (3S6P 330W)
2565W East array (3S3P 285W)
2665W West array (3S3P 285W)

All three groups have nearly identical Vmp around 115V.

I wouldn't think of switching them.

 

[Hedges]

Just connect all PV strings in parallel (through fuses or non-polarized breakers.)
You know all those inverter and SCC manufacturers who say all panels, all strings, have to have same orientation?
They're wrong.

All strings in parallel have to be same/similar Voc and Vmp.
All panels in series have to be same/similar Isc and Imp, and same/similar orientation.

 

[chilly2]

Hello, Due to my geography and an attempt to keep my hardware costs down I have come up with a 3 Array system, Morning - Midday- Afternoon. Except for the number of panels each array is hardware identical. One expensive device I would like to conserve with is the Charge Controller. Has anybody ever seen an autonomous switching device that could be used so I would only need one Charge Controller during a solar cycle?
Bruce

I have a two array setup. One captures morning, and one late morning to sun down. Each array feeds one of a pair of 60A charge controllers that communicate with each other to charge a single 5kWh battery bank.

As they're completely separate and feeding the same battery there's no point having an on off switch on any of them. Each generates or it doesn't.

I'm exploring putting the morning array on a tracker so it can turn and capture evening also, but it's more for fun as I don't see a fast ROI on that.

 

[Zwy]

Those EV200 relays have a 9-36V coil input. Works great on my 12V system.

While those will work, it's still generally not recommended to disconnect an array while active, so one would need to turn off the MPPT, switch, turn on the MPPT especially since it's being done daily.

With something like the EG4 MPPT's, it would turn on automatically as soon as the MPPT sees power.

If batteries were full at other location, there wouldn't be any load present anyway.

I'm going to loop around to the idea that it makes sense to:

1) add complexity
2) add cost
3) lose available power

Rather than size the MPPT for the aggregate array.

In my case, I'm just looking to move power to a dump load most likely. I'm planning on adding thermal storage to my house heating system and can utilize any unused power to heat the thermal tanks.

Hello, Due to my geography and an attempt to keep my hardware costs down I have come up with a 3 Array system, Morning - Midday- Afternoon. Except for the number of panels each array is hardware identical. One expensive device I would like to conserve with is the Charge Controller. Has anybody ever seen an autonomous switching device that could be used so I would only need one Charge Controller during a solar cycle?
Bruce

Morningstar made a MPPT that would switch power according to needs for grid tie with battery/inverter backup power.

In your case however, I would just add more MPPT's.

 

[Warpspeed]

I have two identical sets of panels, one facing East, the other west.
Both connect through series diodes onto the same solar controller and it works fine.
Power is pretty constant right throughout the whole day.

 

[Zwy]

I have two identical sets of panels, one facing East, the other west.
Both connect through series diodes onto the same solar controller and it works fine.
Power is pretty constant right throughout the whole day.

Details?

I get it, strings in parallel.

 

[Warpspeed]

I don't know what other details you need ?
Panels face East/West bolted to equilateral triangular frame, panels tilted sixty degrees in both directions.
Outputs combined through diodes to give pretty constant power output throughout the whole day.

 

[gotbeans]

Just connect all PV strings in parallel (through fuses or non-polarized breakers.)
You know all those inverter and SCC manufacturers who say all panels, all strings, have to have same orientation?
They're wrong.

All strings in parallel have to be same/similar Voc and Vmp.
All panels in series have to be same/similar Isc and Imp, and same/similar orientation.

I see it argued about all over on youtube / blogs / websites
When you're connecting these arrays in parallel, are you using a diode at the end so if one of the arrays is "off" it isn't draining power from the others?
Inspector had me put them in many years ago at my old place so I did here too.

I have two identical sets of panels, one facing East, the other west.
Both connect through series diodes onto the same solar controller and it works fine.
Power is pretty constant right throughout the whole day.

oh this is what I am asking about above?

 

[Warpspeed]

Hedges is exactly right.

If the panels are of the same type, and have the same number of panels in series in each string, the VMP voltages of strings will be the same (or very nearly the same).

The panels will contribute current in proportion to how much solar they receive, they need not all be contributing equally.

At different times of day with different panel orientations, the situation might be continuously changing as the sun moves across the sky.
All you need do is connect a series diode to each string, then combine all strings into just one solar controller.

If you have a mixture of panel types, with very different VMP voltages, this is not going to work. You need separate controllers.

The original poster (Bruce) has a matched sets of panels, so the diode trick will definitely work for him.
I have matched panels in my own East/West array, and the diode trick works perfectly.
In the morning the East panel puts out max power.
Around mid day East and West both contribute about 50% equally.
In the afternoon the West panel puts out full power.
Over a full day the power hardly changes.

 

[AntronX]

If both arrays use same amount of panels of same type I don't think you would even need diodes. Panels facing away from the sun will still produce about 1 - 5% of their rated power from ambient light.

 

[gotbeans]

Yea that's what I have at both places.. same exact panels & count and I set them in large square orientations so one half gets shade before/after the other half.
With 1 diode at the end of them and 1 fuse

If both arrays use same amount of panels of same type I don't think you would even need diodes. Panels facing away from the sun will still produce about 1 - 5% of their rated power from ambient light.

yea but from what I've seen the panels draw power and radiate infrared light if they have no light coming in.. so they'll take from the other array
(and thus the arguments I see all over begins)

 

[AntronX]

yea but from what I've seen the panels draw power and radiate infrared light if they have no light coming in..

That happens at night if directly connected to batteries. During the day they will produce their rated open circuit voltage into the evening. You can try directly paralleling your arrays and measure current direction. You should see your shaded array contributing 0.5 - 1A.

 

[gotbeans]

Yea I've already read identical replies as yours hundreds of times as I said earlier.. was just asking about the diodes. I'll keep them. (especially since others recommend it and I see their knowledge on other posts)
I have 2 in entirely different parts of the yard, one gets entire shade a few hours before other
the panels producing 400v is totally capable of sending power through the ones producing 130v, both during the day unrelated to battery

 

[Warpspeed]

There will be a slight discharge at night without diodes, but usually the solar controller prevents back feeding power at night.

While in theory, series diodes should not be required if all the panels are matched and identical, I have tried it both ways, and the diodes do seem to improve power production. I am not sure why that is, or maybe its my imagination, but I always seem to get better results with diodes.

 

[gotbeans]

There will be a slight discharge at night without diodes, but usually the solar controller prevents back feeding power at night.

While in theory, series diodes should not be required if all the panels are matched and identical, I have tried it both ways, and the diodes do seem to improve power production. I am not sure why that is, or maybe its my imagination, but I always seem to get better results with diodes.

Yea night I don't even worry about. It's having 2 sets of arrays that are connected in parallel back at the arrays themselves.. then 1 wire runs to the inverter itself.
My new system here has 4 of these. 2x2 parallels into 2 inverters