What happens when you mix panels on a string?

[khisanthax]

Ideally I would want all the same panels for my project. But after buying 8 panels (1/4 of what I wanted). I realized they're not great and I could do better with used panels. I'm working towards researching three strings but as an option I wanted to know what happens if you mix panels on the same string?

I imagine you add the voc of each of set, calculations from midnite solar, and add the voc at the coldest temp in the area for each set. What about the current? Do you take the set of panels that has the highest current and use that for your calculations?

My inverter has a max current on the PC array of 27A and my current set of panels has 12.5A so I wanted to ask what the pros and cons of this less than ideal approach would be?

Long term I plan to put this set on a separate scc just for the battery but I currently can't purchase all the panels at once. So even if I purchase 3/4 now and then later on can't get the same panels I still might run into this situation. Thanks in advance.

 

[Tomthumb62]

Post specs of current and new panels and we can give better advice.

 

[TomC4306]

Depends on how dissimilar they are and how they are connected.

 

[TomC4306]

I "acquired" some used Sharp 235w panels.
Then I bought used 255w panels thru Facebook mktplace.
Volts and amps are close.
Currently have 3 strings of 3 of each paralleled at a 6 to 1 combiner box.
Separately, 3 255's and 2 255's and a 235 paralleled to a second mppt.

 

[fafrd]

I "acquired" some used Sharp 235w panels.
Then I bought used 255w panels thru Facebook mktplace.
Volts and amps are close.
Currently have 3 strings of 3 of each paralleled at a 6 to 1 combiner box.
Separately, 3 255's and 2 255's and a 235 paralleled to a second mppt.

They should work fine together at reduced efficiency. Assuming the output of the 235W for all panels is a conservative estimate.

If it’s worth the cost to you, you can get optimal output from all panels by adding optimizers on either all smaller panels or all larger panels (believe they boost current which would likely mean using them in the 235W panels but you should read the specs).

Since you no doubt got a deal on used panels, squeezing a bit more out of them by paying for optimizers probably doesn’t make much sense. You can add additional panels for less cost.

Your description is unclear, but as long as only strings of similar panels are combined to a single MPPT, you are already optimal.

If you have a series string of dissimilar panels, optimizers is the only way to get out some more power.

If you have multiple strings, each composed of identical panels but different types on each string, and sharing an MPPT, adding another MPPT is probably a cheaper way to get to optimal output that adding optimizers to one of the strings.

 

[TomC4306]

I have a surplus of solar as it is.
Looking for additional loads to bring on to the system.
Floating by 2pm.

I "acquired" some used Sharp 235w panels.
Then I bought used 255w panels thru Facebook mktplace.
Volts and amps are close.
Currently have 3 strings of 3 of each paralleled at a 6 to 1 combiner box.
Separately, 3 255's and 2 255's and a 235 paralleled to a second mppt.

Edit:
3 panels in series - 3 strings of these with 235 watt panels (right rear array).
3 strings with 255 watt panels.(left rear array)
All 6 strings to combiner box. To mppt#1
1 string of 255 watt panels paralleled to 1 string of 2-255's and 1-235.
Direct to mppt#2

 

[khisanthax]

Sorry, was sick and on vacation, yay, but I'll post the specs. Catching up now.

 

[khisanthax]

They should work fine together at reduced efficiency. Assuming the output of the 235W for all panels is a conservative estimate.

If it’s worth the cost to you, you can get optimal output from all panels by adding optimizers on either all smaller panels or all larger panels (believe they boost current which would likely mean using them in the 235W panels but you should read the specs).

Since you no doubt got a deal on used panels, squeezing a bit more out of them by paying for optimizers probably doesn’t make much sense. You can add additional panels for less cost.

Your description is unclear, but as long as only strings of similar panels are combined to a single MPPT, you are already optimal.

If you have a series string of dissimilar panels, optimizers is the only way to get out some more power.

If you have multiple strings, each composed of identical panels but different types on each string, and sharing an MPPT, adding another MPPT is probably a cheaper way to get to optimal output that adding optimizers to one of the strings.

So, it's better to keep the strings all the same as opposed to strings and parallels an equal mix of two brands?

Can you give ane example if an optimizer? I've read that using an mmpt on a string with that has one panel type but different from other string can help, but would the mppt connect to an all in one or only to the battery?

 

[khisanthax]

I have a surplus of solar as it is.
Looking for additional loads to bring on to the system.
Floating by 2pm.


Edit:
3 panels in series - 3 strings of these with 235 watt panels (right rear array).
3 strings with 255 watt panels.(left rear array)
All 6 strings to combiner box. To mppt#1
1 string of 255 watt panels paralleled to 1 string of 2-255's and 1-235.
Direct to mppt#2

I've read that the voltage of each string needs to be the same as other strings when in parallel otherwise they all get dragged down to the string with the lowest voltage, is that accurate? Or are all your paralleled strings similar in voltage?

 

[fafrd]

I've read that the voltage of each string needs to be the same as other strings when in parallel otherwise they all get dragged down to the string with the lowest voltage, is that accurate? Or are all your paralleled strings similar in voltage?

If string Vmp matches perfectly, you will be optimal when strings are combined in parallel and controlled by a single MPPT.

When there is a mismatch in string Vmp, when the strings are combined in parallel and controlled by a single MPPT, the low strings will be pulled up to higher than Vmp (so current will drop under Imp and power will drop under Pmp) and the high strings will be pulled below Vmp (so currebt will increase above Imp towards Isc and power will drop under Pmp).

You can get a swag of how much potential power you are losing i this situation by taking te average of the two string voltages, deciding by string length to determine individual panel voltages, and then looking at the panel power curves to eyeball how far off you will be from Vmp and how much current or power will drop near that voltage.

The other way to mix and match panels is to compose identical swings of both panels if Imp is relatively close for both panel types.

Let’s say you have panels with identical Imp but very different Vmp (for example a mixture of 60-cell and 71-cell panels).

You are actually better off making all strings serving one MPPT contain 2 of panel and 1 of the other panel in this situation. If the Imp is identical for both panel types, this will be optimal (and you can estimate how much power is lost if Imp is not identical by estimating average Imp and eyeballing what that translates to in terms of string voltage drop using the I-V curves).

In your case, 1 MPPT could be served by 3 parallel strings of 1 + 2 in series and the leftovers could then allow you to serve a second MPPT with 3 parallel strings of 2 + 1.

Again, this is only for the case where Imp of the two panel types is relatively close.

Matching current / Imp as closely as possible is most important for panels in series and matching string voltage / Vmp_string is most important for strings / panels in parallel.

 

[fafrd]

So, it's better to keep the strings all the same as opposed to strings and parallels an equal mix of two brands?

I just added a post on this - it’s most important to match Imp as closely as possible for all panels in a string (doesn’t matter if Vmp differ) and most important to match Vmp of strings (total Vmp of all panels in the string) of strings that combine in parallel and are served by a single MPPT (and doesn’t matter if string Imps differ).

Can you give ane example if an optimizer?

Tigo (pretty much the only one that will work with any MPPT).

I've read that using an mmpt on a string with that has one panel type but different from other string can help, but would the mppt connect to an all in one or only to the battery?

I don’t understand your question.

If about optimizers, they match current at the expense of voltage.

So, for example, if you have one mismatched panel with smaller Imp or that is partially-shaded reducing current,the optimizer will increase current to match the Imp current flowing through the other panels and will reduce voltage far below Vmp to compensate, so the other panels in the string are not choked off (current starved).

So Vmp x Ishaded becomes Imp x V shaded where Vshaded is below Vmp and the total string voltage drops to (N - 1) x Vmp + Vshaded.

Using more independent MPPTs is generally a better way to get more out of mismatched arrays (in panel type or string length or orientation) than spending a buck of money on optimizers.

But, for example, if there is only a single sub-sized panel, adding a single optimizer on that one panel can be the best way to add it to a larger already-existing string.

 

[TomC4306]

I've read that the voltage of each string needs to be the same as other strings when in parallel otherwise they all get dragged down to the string with the lowest voltage, is that accurate? Or are all your paralleled strings similar in voltage?

Voc 37.7 and 37.0
Vmp 30.5 and 30.1
Isc 8.92 and 8.5
Imp 8.36 and 7.81

9 panels of the first paralleled to 9 panels of the second.
Works fine.

 

[fafrd]

Voc 37.7 and 37.0
Vmp 30.5 and 30.1
Isc 8.92 and 8.5
Imp 8.36 and 7.81

9 panels of the first paralleled to 9 panels of the second.
Works fine.

Yes, I agree.

That’s an almost perfect match in Vmp.

The difference in Imp is large enough that mixing panel types in a series string would not be advisable (could reduce output from the larger panels by as much as ~6.6%…).

But if you have no choice but to form a last string out of a mixture of both panel types, it’s not the end f the world. The larger panels in that mixed string will provide power closer to that of the smaller panels, but everything will work just fine.

 

[khisanthax]

I currently have the eco-worthy 195w panels:

vmp: 20.2v, Imp 9.65a, voc 24.5v, Isc: 10.14

The ones I'm looking to purchase are Rec 370w:

vmp: 37.4, Imp: 9.9, voc: 44.1, Isc: 10.55

The current AIO has a 27A max for the PV array. So I'm pretty sure I can't use these as three strings. Any other options that wouldn't bring everything down to a crawl?

 

[fafrd]

I currently have the eco-worthy 195w panels:

vmp: 20.2v, Imp 9.65a, voc 24.5v, Isc: 10.14

The ones I'm looking to purchase are Rec 370w:

vmp: 37.4, Imp: 9.9, voc: 44.1, Isc: 10.55

The current AIO has a 27A max for the PV array. So I'm pretty sure I can't use these as three strings. Any other options that wouldn't bring everything down to a crawl?

Imp of these two panels is close enough that it’ll probably be better to combine in series strings if you don’t have separate MPPTs for the two arrays.

What is the max voltage rating you of your AIO?

As you noticed, you can only feed two strings in parallel to that AIO, probably 3 of the 195W panels, no more than 2 of the 370W panels, and possibly 1 370W and 2 195W panels per string (depends on max voltage specification).

How many of each panel type are you aiming at and how many MPPTs? (With current and voltage specs)?

 

[khisanthax]

Imp of these two panels is close enough that it’ll probably be better to combine in series strings if you don’t have separate MPPTs for the two arrays.

What is the max voltage rating you of your AIO?

As you noticed, you can only feed two strings in parallel to that AIO, probably 3 of the 195W panels, no more than 2 of the 370W panels, and possibly 1 370W and 2 195W panels per string (depends on max voltage specification).

How many of each panel type are you aiming at and how many MPPTs? (With current and voltage specs)?

The max voltage is 450v. If I understand right, the reason I can only use two string in parallel is because using three would go over the current rating for the PV array? I already have 8 of the 195w and none of the 370w, but so my goal was 8 of the 195w and at least 16 of the 370w in order to get close to the 6kw max that the AIO can use. Aside from the AIO I didn't plan on using any MPPT's, I think that's what I'm trying to figure out now. What I would need in addition to the AIO to get close to that 6kw rating.

It sounds like to make it work with what I have, I would need to have additional mppt's to get close to that out put of 6kw?

 

[fafrd]

The max voltage is 450v.

OK, so subject to calculating worst-case Voc (at coldest annual temps), that means as many as 18 of the smaller panels or as many of 10 of the new ones.

If I understand right, the reason I can only use two string in parallel is because using three would go over the current rating for the PV array?

Yes.

I already have 8 of the 195w and none of the 370w, but so my goal was 8 of the 195w and at least 16 of the 370w in order to get close to the 6kw max that the AIO can use.

If you split the 8 195W into two strings, that’s ~100Voc out of a max of 450V. Now each 370W you add to each string consumes another ~45V of your headroom.

You had ~350V of headroom, so 7 additional 370W panels per string for 14 additional 370W panels is the most you’ll be able to add without getting another Solar Charge Controller / MPPT (and you need to calculate worst-case Voc to assure you can add 7 370W panels per string).

Aside from the AIO I didn't plan on using any MPPT's, I think that's what I'm trying to figure out now. What I would need in addition to the AIO to get close to that 6kw rating.

I’m not sure what that 6kW rating is - is that the maximum array power your AIO recommends?

Since you are limited to 2P with that string / panel current, 8 x 195W = 1560W plus 14 x 370W =5180W gets you to 6740W is probably the max you can achieve with that AIO.

Even if you are limited to only adding 6 370W panels instead of 7 due to coldest-temperature, you are still achieving 6kW of solar (exactly ).

Adding more would require adding another SCC to charge the battery in parallel with the AIO.

It sounds like to make it work with what I have, I would need to have additional mppt's to get close to that out put of 6kw?

6kW = 2 x [(4 x 195W) + (6 x 370W)] and your 450V AIO should be able to handle that 2P x (4S + 6S) string…

 

[khisanthax]

OK, so subject to calculating worst-case Voc (at coldest annual temps), that means as many as 18 of the smaller panels or as many of 10 of the new ones.

Yes.

If you split the 8 195W into two strings, that’s ~100Voc out of a max of 450V. Now each 370W you add to each string consumes another ~45V of your headroom.

You had ~350V of headroom, so 7 additional 370W panels per string for 14 additional 370W panels is the most you’ll be able to add without getting another Solar Charge Controller / MPPT (and you need to calculate worst-case Voc to assure you can add 7 370W panels per string).

I’m not sure what that 6kW rating is - is that the maximum array power your AIO recommends?

Since you are limited to 2P with that string / panel current, 8 x 195W = 1560W plus 14 x 370W =5180W gets you to 6740W is probably the max you can achieve with that AIO.

Even if you are limited to only adding 6 370W panels instead of 7 due to coldest-temperature, you are still achieving 6kW of solar (exactly ).

Adding more would require adding another SCC to charge the battery in parallel with the AIO.

6kW = 2 x [(4 x 195W) + (6 x 370W)] and your 450V AIO should be able to handle that 2P x (4S + 6S) string…

I'm going to make fun of every single person who ever said you don't need math in school!

With 8 370w in the series it comes to 404.8v VOC @ -20c, so I'd probably be only able to use 7.

4 195w panels at -20c the VOC comes to 104.8. Sizing

7 370w panels at -20c VOC is .54.2, so that goes over. Sizing

6 370w panels seems to be the magic number at 303.6v VOC. Sizing

You did great at calculating that. So, let's say for those winter months where I get less light I wanted a few extra panels to bump things up, and yes I mean the maximum array power, sorry I didn't phrase that right, how do I use an additional SCC?

I think I'm conceptualizing something wrong. Is the underlying assumption that the panels charge the battery and the battery provides power to the load? I was conceptualizing that the panels would provide power to the load directly and charge the battery, but the battery would only be used at night when the panels aren't generating power.

Is this how the additional SCC helps? It adds more charging for the battery, which in turn has more power for the load during the day?

 

[fafrd]

I'm going to make fun of every single person who ever said you don't need math in school!

With 8 370w in the series it comes to 404.8v VOC @ -20c, so I'd probably be only able to use 7.

4 195w panels at -20c the VOC comes to 104.8. Sizing

7 370w panels at -20c VOC is .54.2, so that goes over. Sizing

6 370w panels seems to be the magic number at 303.6v VOC. Sizing

Yeah, sounds like 6 new 370W panels added to each string of 4 of your existing 195W panels is about the limit at -20C (which is darned cold - where are you based???).

You did great at calculating that. So, let's say for those winter months where I get less light I wanted a few extra panels to bump things up, and yes I mean the maximum array power, sorry I didn't phrase that right, how do I use an additional SCC?

During winter months Voc is at its highest, so you cannot get more panels by adding to the existing strings / MPPT. It’s the shorter days that primarily reduce daily solar production so you need another charge controller to get additional charge into your battery despite the shorter days.

I think I'm conceptualizing something wrong. Is the underlying assumption that the panels charge the battery and the battery provides power to the load? I was conceptualizing that the panels would provide power to the load directly and charge the battery, but the battery would only be used at night when the panels aren't generating power.

During daylight hours, solar power is powering loads as first priority and charging the battery when there is additional solar power after charging loads as second priority.

When loads come on that exceed available solar power including at night when there is no sun, the battery will provide the energy to power loads beyond available solar power.

A new SCC will only be sending additional charge to the battery during the day, so in practical terms, when the AIO is receiving enough solar power to power loads total power charging the battery will be increased with an additional SCC and when solar energy into the AIO is insufficient to offset all load, rather than filling the gap with battery energy, the AIO will first fill the gap with energy / charge from the new SCC and will only discharge the battery if the SCC is not supplying enough DC energy to fully fill the gap.

So you’ll end every day (in winter at least) with more stored energy than you are getting without the added SCC.

Is this how the additional SCC helps? It adds more charging for the battery, which in turn has more power for the load during the day?

Whether the instantaneous energy to offset all load is coming all from the AIO MPPT, from the AIO and the new SCC, or from the AIO, the new SCC and the battery is immaterial - you’ll end up with more energy in the battery come sunset if you have the additional SCC.

You said your goal was to add 16 370W panels and only 12 could be added to the 2 strings supplying the AIO.

So let’s say you’d like to add another 4 370W panels with a new SCC.

A 2S2P array of those panels will need Vmax of at least 110VDC and 4x370W = 1480W, so you probably want an SCC that can handle 1500W of charge power without clipping to get the most from your 4 additional panels.

You never stated what your battery voltage is but I’m guessing it’s 48V.

As an example, this Epever 60A SCC handle up to 150Voc_max and can charge a 24V battery with up to 1500W of solar power (60A @ 24V): https://www.amazon.com/dp/B081RRJSS...=1&sp_csd=d2lkZ2V0TmFtZT1zcF9waG9uZV9kZXRhaWw

I pointed to that one because that is what I am using to charge my 24V LiFePO4 battery with 3 x 380W panels and I’ve been very happy with it’s caoability.

With a 48V battery, you could get by and save a bit with a 30A or 40A SCC rated for 150Voc_max and 48V charging such as this one: https://www.amazon.com/EPEVER-Contr...9032074&hvtargid=pla-1622291890451&psc=1&th=1

 

[khisanthax]

Yeah, sounds like 6 new 370W panels added to each string of 4 of your existing 195W panels is about the limit at -20C (which is darned cold - where are you based???).

During winter months Voc is at its highest, so you cannot get more panels by adding to the existing strings / MPPT. It’s the shorter days that primarily reduce daily solar production so you need another charge controller to get additional charge into your battery despite the shorter days.


During daylight hours, solar power is powering loads as first priority and charging the battery when there is additional solar power after charging loads as second priority.

When loads come on that exceed available solar power including at night when there is no sun, the battery will provide the energy to power loads beyond available solar power.

A new SCC will only be sending additional charge to the battery during the day, so in practical terms, when the AIO is receiving enough solar power to power loads total power charging the battery will be increased with an additional SCC and when solar energy into the AIO is insufficient to offset all load, rather than filling the gap with battery energy, the AIO will first fill the gap with energy / charge from the new SCC and will only discharge the battery if the SCC is not supplying enough DC energy to fully fill the gap.

So you’ll end every day (in winter at least) with more stored energy than you are getting without the added SCC.

Whether the instantaneous energy to offset all load is coming all from the AIO MPPT, from the AIO and the new SCC, or from the AIO, the new SCC and the battery is immaterial - you’ll end up with more energy in the battery come sunset if you have the additional SCC.

You said your goal was to add 16 370W panels and only 12 could be added to the 2 strings supplying the AIO.

So let’s say you’d like to add another 4 370W panels with a new SCC.

A 2S2P array of those panels will need Vmax of at least 110VDC and 4x370W = 1480W, so you probably want an SCC that can handle 1500W of charge power without clipping to get the most from your 4 additional panels.

You never stated what your battery voltage is but I’m guessing it’s 48V.

As an example, this Epever 60A SCC handle up to 150Voc_max and can charge a 24V battery with up to 1500W of solar power (60A @ 24V): https://www.amazon.com/dp/B081RRJSS...=1&sp_csd=d2lkZ2V0TmFtZT1zcF9waG9uZV9kZXRhaWw

I pointed to that one because that is what I am using to charge my 24V LiFePO4 battery with 3 x 380W panels and I’ve been very happy with it’s caoability.

With a 48V battery, you could get by and save a bit with a 30A or 40A SCC rated for 150Voc_max and 48V charging such as this one: https://www.amazon.com/EPEVER-Contr...9032074&hvtargid=pla-1622291890451&psc=1&th=1

I had to double check my temperature conversion unit, but yes -4f is what we could get in Ronkonkoma long island in NYC.

Damn, that blew my mind. So, even if a second scc is connected to the battery the power from the scc or battery is pretty much the same since it'll flow into the battery and then to AIO and the load. So, as long as you don't exceed the charge rating of the battery and connected cables, you could add another scc and panels, theoretically.

I do have a 48v battery. Would the linked scc work for my battery even though it's only 24v and not 48v?

All in all, it sounds like additional panels with a separate scc is a good way to increase how you can use the current panels and charge the batteries.