Voltage differences in strings

[Britman]

Hey Guys
I want another string and have been advised on another forum that the voltage difference is too much

1 string of 270w x4 (1080w) panels in series
4x43.6voc = 174.4v.
1080/174.4=6.19a
And
1 string of 375w x6 (2250w)in series
6x48.8voc= 292.8v
2250/292.8= 7.6a

What are the guidelines and is it a percentage of the voltage that can be different (ie within 10% ? etc)

What happens is you have a string that's theoretically matched but is say east facing and live time voltage difference could be significant??

 

[sunshine_eggo]

Hey Guys
I want another string and have been advised on another forum that the voltage difference is too much

You have been advised correctly.

1 string of 270w x4 (1080w) panels in series
4x43.6voc = 174.4v.
1080/174.4=6.19a
And
1 string of 375w x6 (2250w)in series
6x48.8voc= 292.8v
2250/292.8= 7.6a

Your calculations are incorrect.

Current is determined by power / Vmp. It is also listed as Isc for short circuit and Imp for max power.

What are the panels' Imp values?

What are the guidelines and is it a percentage of the voltage that can be different (ie within 10% ? etc)

If strings are in parallel on the same MPPT, the strings must have a very similar Vmp. Typically within 5-10%. Your performance penalty can be estimated by the % difference in Vmp.

What happens is you have a string that's theoretically matched but is say east facing and live time voltage difference could be significant??

If the difference is too great, the lower voltage string may serve as a short circuit path for the higher voltage array to discharge into providing no power to the MPPT. Even if it doesn't, you'll never get a drop of power from the lower voltage array. Facing matters very little. Good ambient light is enough to produce voltage. It's the intensity of the light that produces current/amps.

Those panels are already poorly matched from the start. If you need to use both arrays, your optimal course of action would be to add another MPPT to handle the other string. The only other option I see is to install one of the 375W panels in the 270W string to get Vmp close to the other. The 375W panel will take a small hit due to the lower current of the 270W panels.

Can your existing MPPT even handle the added power? MPPT are rated by output, so you have to divide total PV power by battery voltage to get current. You'd need 278A for 12V, 139A for 24V and 69A for 48V.

 

[MisterSandals]

What happens is you have a string that's theoretically matched but is say east facing and live time voltage difference could be significant??

Before attempting to mix and match your panels, it'd be helpful to know your SCCs limitations. What is the max input voltage? Will assume a single MPPT. Is it an AIO or grid tied or ??

Are there restrictions which prevent you from connecting some 270W panels to some 375W panels (like they are in significantly different locations)?

 

[Britman]

Before attempting to mix and match your panels, it'd be helpful to know your SCCs limitations. What is the max input voltage? Will assume a single MPPT. Is it an AIO or grid tied or ??

Are there restrictions which prevent you from connecting some 270W panels to some 375W panels (like they are in significantly different locations)?

Thank you , No restrictions on connecting the panels to the most efficient output.
Max Mppt Op Voltage = 450vdc
Max PV input 6000w
Max Charge 100a
AIO growatt 5000es
270w panels vmpp=36v - ISC= 7.9a
375w Panels vmpp =40v - ISC= 9.9a

 

[MisterSandals]

1 string of 270w x4 (1080w) panels in series
4x43.6voc = 174.4v.
1080/174.4=6.19a
And
1 string of 375w x6 (2250w)in series
6x48.8voc= 292.8v
2250/292.8= 7.6a

Messing with numbers...
270W panel Vmp is 36V and Imp is therefore 7.5A
375W panel Vmp is 40V and Imp is therefore 9.37A

Mixing and matching panels...

Lets try matching volts with equivalent strings:
(2x 38) + (3 x 40) = 196V x 7.5A = 1470W (each string) 2940W total

Adding 1 375W panel to 270W 4S string: Best i see, produces 3239W
(4 x 38) + (1 x 40) = 192V x 7.5A = 1440W
(5 x 40) = 200V clipped when put in parallel to 192V x 9.37A = 1799W

This gives you 2 5S strings that you combine in parallel (5S2P).
I think its actually pretty good considering panel differences.
Hopefully someone checks my math.

 

[sunshine_eggo]

Messing with numbers...
270W panel Vmp is 36V and Imp is therefore 7.5A
375W panel Vmp is 40V and Imp is therefore 9.37A

Mixing and matching panels...

Lets try matching volts with equivalent strings:
(2x 38) + (3 x 40) = 196V x 7.5A = 1470W (each string) 2940W total

Adding 1 375W panel to 270W 4S string: Best i see, produces 3239W
(4 x 38) + (1 x 40) = 192V x 7.5A = 1440W
(5 x 40) = 200V clipped when put in parallel to 192V x 9.37A = 1799W

This gives you 2 5S strings that you combine in parallel (5S2P).
I think its actually pretty good considering panel differences.
Hopefully someone checks my math.

I didn't post the numbers since I didn't yet have the complete panel data, but I came to the same conclusion:

The only other option I see is to install one of the 375W panels in the 270W string to get Vmp close to the other. The 375W panel will take a small hit due to the lower current of the 270W panels.

You used 38V instead of 36V, so 184V instead of 192V vs. 200V.

?

Given the higher voltages, I would seriously consider an MC4 blocking diode in each string.

 

[MisterSandals]

You used 38V instead of 36V, so 184V instead of 192V vs. 200V.

Yeah, thanks that a good catch. I was running scenarios with my guesses for Vmp and missed fixing that after OP stated the actual Vmp.

Adding 1 375W panel to 270W 4S string: Best i see, produces 3239W
(4 x 38) + (1 x 40) = 192V x 7.5A = 1440W
(5 x 40) = 200V clipped when put in parallel to 192V x 9.37A = 1799W

Becomes
(4 x 36) + (1 x 40) = 184V x 7.5A = 1380W
(5 x 40) = 200V clipped when put in parallel to 184V x 9.37A = 1724W

Total on paper: 3104W

 

[GooseF16]

if anyone ever doubted the genius of this forum.. remove that doubt.. this short thread is awesome! nice work ya'll..

 

[Britman]

That's Great!
So to summarize so i understand - Having 2 equal strings would net 2880w on paper
By adding one of the 375 to the smaller panels pushes that to 3104w
The volt difference is 16v ( is that an acceptable difference?) between the 2 strings

Sunshine can you explain the purpose of the diode and a link maybe ?

Thank you all ...

 

[sunshine_eggo]

That's Great!
So to summarize so i understand - Having 2 equal strings would net 2880w on paper
By adding one of the 375 to the smaller panels pushes that to 3104w
The volt difference is 16v ( is that an acceptable difference?) between the 2 strings

Sunshine can you explain the purpose of the diode and a link maybe ?

Thank you all ...

Amazon.com : mc4 diode

You would simply place ONE of the above MC4 blocking diodes in each string.

If extreme shading occurred on the lower voltage string, that string could serve as a short circuit for the other string causing the higher voltage string to dump all of its current into the lower voltage string rather than being drawn by the MPPT. The blocking diode allows current to flow in only one direction preventing any backfeeding between strings. It's likely not needed, but in this case with ~200V, I would be concerned they might be needed at some point.

 

[MisterSandals]

You would simply place ONE of the above MC4 blocking diodes in each string.

Would you recommend putting this on the + wire, closest to where the 2 strings are joined in parallel?

 

[sunshine_eggo]

Would you recommend putting this on the + wire, closest to where the 2 strings are joined in parallel?

(+) yes. Anywhere in the string that's convenient.

 

[Britman]

Gents you have been great Thank you .. I will post pics when up and running

 

[Bud Martin]

BTW, when you have them up and running, check the temperature of that MC4 blocking diode when you have full current flow, they can get hot, and if it's a hot day it can get so hot that it deforms the body of the MC4.
Some of them I took apart has poor crimping job because you are not supposed to crimp on solid wire, MC4 terminal is made for stranded wire, if they do then they should crimp and solder but that will cause too much work.

 

[sunshine_eggo]

BTW, when you have them up and running, check the temperature of that MC4 block diode when you have full current flow, they can get hot, and if it's a hot day it can get so hot that it deforms the body of the MC4.
Some of them I took apart has poor crimping job because you are not supposed to crimp on solid wire, MC4 terminal is made for stranded wire, if they do then they should crimp and solder but that will cause too much work.

Interesting.

I just bought this one to check it:

https://www.amazon.com/dp/B07SWTJ4FH

Indicates the diode can be easily replaced. Might have some sort of connection with tension/pressure on it.

 

[Bud Martin]

It will be interesting to see how well it is constructed when you take it apart.

 

[MisterSandals]

I just bought this one to check it:

https://www.amazon.com/dp/B07SWTJ4FH

What are the considerations as to which to choose? 10A, 15A or 20A
They are all the same price. Do want to match it as you would a fuse or is bigger always better? Or?

 

[RCinFLA]

Hey Guys
I want another string and have been advised on another forum that the voltage difference is too much

1 string of 270w x4 (1080w) panels in series
4x43.6voc = 174.4v.
1080/174.4=6.19a
And
1 string of 375w x6 (2250w)in series
6x48.8voc= 292.8v
2250/292.8= 7.6a

You will get killed mostly by the lower voltage array.

Voc is the panel voltage (likely from 25 degs C spec) where all the illumination generated current is shunted back into the panel so you get zero output current yield, so dividing panel wattage (likely again at 25 degs C) by Voc is totally wrong.

General fairly close approximation is Vmp will be about 0.81-0.85 x Voc and Imp will be about 0.93-0.95 x Isc at full sun illumination.

So for first panel, Vmp is about 0.84 x 174v Voc = 146 vdc. Imp will be 1080w / 146 Vmp = Imp = 7.4 amps
Voc and Vmp will likely be lower when panels gets hot from sun, so don't expect 1080 watts yield unless panels are held at 25 degs C.

Second panel Vmp = 0.84 x 292v Voc = 245vdc. 2250 watts / 245v Vmp -> Imp = 9.17A

MPPT controller will try to maximize the power of the parallel combination.

The lower voltage series array will drag down the voltage on the higher voltage array. This will lower the high array to a little lower than the Voc of the lower array, likely in about the 165v operating range with a full illumination current of about 9.6 amps. Higher power array will yield about 1584 watts.

Lower voltage array will be dragged up towards its Voc, with same parallel compromise operating voltage as the previously estimated 165v. This will drastically drop the output current from the lower wattage array. Estimate output current will be about 3.2 amps yielding a power output from lower voltage array of about 165v x 3.2 amps = 530 watts.

So summing the two arrays, 1584watts + 530 watts will yield about 2100 watts from the optimum combo in parallel.

Bottomline, results will not be much better than the higher voltage array alone, maybe even less.

 

[Britman]

You will get killed mostly by the lower voltage array.

Hi Guys i had promised to give some feedback so here it is ....





Mistersandals estimate " Total on paper: 3104W" was pretty fair considering the 1st string points south and the second points
west with 1 panel south facing ... they are also set at 51' (from memory) which was the optimum at the winter solstice.



Now that I've finished setting it up and playing with different loads i'm ready for Gen II
I have 25Kwh of additional battery capacity and another 5k inverter.
The Plan is to move the panels in front of the fence to the right in the picture and combine with some
more panels i have..

Please throw in your opinions on the configuration (yes i know they are a mixed bag):
6 x ET 375w / Vmp - 40v so Imp= 9.37a
4 x ET 270w / Vmp - 36v so Imp= 7.5a
3 x Sharp 234w / Vmp - 29.2v so Imp= 6.5v
3 x Samsung 235w / Vmp - 30v so Imp=7.8a
1 x Solon 235w / Vmp - 29.5v so Imp =7.9a
3 x Vsun 445w / Vmp - 41v so Imp =10.8a

4 Strings total 2 each going into 2 x 5000es GrowWatt AIO

Inverter =
Max PV array OCV = 450Vdc
Range 120-430 Vdc
Max PV input 6000w

Thank you Thank you

 

[sunshine_eggo]

Hi Guys i had promised to give some feedback so here it is ....

View attachment 159998

View attachment 159984

Mistersandals estimate " Total on paper: 3104W" was pretty fair considering the 1st string points south and the second points
west with 1 panel south facing

Your W array will likely perform better without the S facing panel by a good measure. When the sun is in the west, the current produced by the south facing panel will restrict the W panels.

... they are also set at 51' (from memory) which was the optimum at the winter solstice.

Now that I've finished setting it up and playing with different loads i'm ready for Gen II
I have 25Kwh of additional battery capacity and another 5k inverter.
The Plan is to move the panels in front of the fence to the right in the picture and combine with some
more panels i have..

Please throw in your opinions on the configuration (yes i know they are a mixed bag):
(123V) 3 x Vsun 445w / Vmp - 41v so Imp =10.8a
(240V) 6 x ET 375w / Vmp - 40v so Imp= 9.37a

All of the above in series: 363V @ 9.37A = 3404W max

(144V) 4 x ET 270w / Vmp - 36v so Imp= 7.5a
(90V) 3 x Samsung 235w / Vmp - 30v so Imp=7.8a
(30V) 1 x Solon 235w / Vmp - 29.5v so Imp =7.9a
(90V) 3 x Sharp 234w / Vmp - 29.2v so Imp= 6.5A

All of the above: 354V @ 6.5A = 2301W

4 Strings total 2 each going into 2 x 5000es GrowWatt AIO

Unfortunately, I don't see a way to break any of that reasonably into parallel strings except for those two would work decent in parallel with each other.

Inverter =
Max PV array OCV = 450Vdc
Range 120-430 Vdc
Max PV input 6000w