Adding panels and optimisers to an existing system

[binding_energy]

I have a professionally installed system with one string of 10 panels on the main roof facing west with no shade problems + one panel on the gable facing south (that one has an optimizer fitted due to sometimes being shaded) . A second string consists of 7 panels on a south facing roof (a little distance from the main roof) all of which have optimisers fitted due to a complex shading pattern. System is rated at 7kW and works well, producing around 10kWh per day in the summer at my location.

I'd like to install a further set of panels on the east side of the main roof to catch the morning sun, before the west side kicks in, maybe another 8 or 10 panels.
I'm assuming these will have to be optimised so as not to compete with the west panels or the south panels.

Question

1) Do you think I would be better of connecting the new panels to string 1 (10 west +1 south) or to string 2 (7 south) - and why?

2) Would it be better to connect the new panels in series with the existing ones or in parallel, or even put pairs in parallel and the paid group in series? (data indicates that in the height of summer both strings generated around 300v max and the inverter has a max DC input of 550v)

My inverter only has two inputs so I can only have two strings at that point

I am quite competent to physically do the installation, just curious as to the most efficient way to connect them up.

Thank you for any help you can offer

 

[robbob2112]

not enough information to answer

Panel specs? post a picture or link
MPPT specs - what make/model is it?

 

[binding_energy]

Sorry, I didn't think the details would affect the general reasoning behind east/west or east/south combinations.

However.
Existing 18 panels are all Haitai HTM 410 and the inverter is a Growatt SPH6000 .
I'm not sure what make optimisers are currently fitted but they do not have individual APs. From their visual appearance they are probably Tigo TS4-A-O.
I also have 2 x 5kW storage batteries in parallel, both Growatt AXE 5.0L if that makes any difference to my question

The specs of all these are attached

The 7 south panels on string 2 are on a separate garage that is attached to the house (but prone to shading by both my house and my neighbour's)

Hope this helps

 

[binding_energy]

not enough information to answer

Panel specs? post a picture or link
MPPT specs - what make/model is it?

A bit more info. Today, in late november, string 1 (generally west) generated around 1.5kW max between 10am and 2pm and string 2 (south) generated between 1.2kW and 1.4 Kw between about 8pm and 2pm. String 1 voltage was roughly 275v between 7am and 4pm and string 2 voltage was approximately 190v between 7:30am and 4pm.
Graphs are attached

 

[robbob2112]

Sorry, I didn't think the details would affect the general reasoning behind east/west or east/south combinations.

However.
Existing 18 panels are all Haitai HTM 410 and the inverter is a Growatt SPH6000 .
I'm not sure what make optimisers are currently fitted but they do not have individual APs. From their visual appearance they are probably Tigo TS4-A-O.
I also have 2 x 5kW storage batteries in parallel, both Growatt AXE 5.0L if that makes any difference to my question

The specs of all these are attached

The 7 south panels on string 2 are on a separate garage that is attached to the house (but prone to shading by both my house and my neighbour's)

Hope this helps

Here are the bits needed to figure it

2 MPPT in the inverter - Both are the same - they don't turn on until string voltage hits 120v so your panel strings need to be at least that as early as possible. The place they work best is 360v so that is the target to try and hit.

in series voltage adds and in parrallel current adds - your panel VOC is 37.26v - so a 10 panel string is just about perfect for hitting the correct voltage. Since a string of 18 panels would be 660v and that is outside the range I assume the installed array is 2 strings of 9 panels each using one of the MPPT. You can't have a 9s2p arrangment because the max input current is 12a and in reality your panels Isc is 13.1a - so in theory on a perfect day with a perfect angle your panels as mounted could burn out the built in MPPT -

This seems to be a poor choice of panels matched to the inverter MPPT to me. They should probably have used something like 335w panels or a different 400w panel with an ISC of 9amps or so. But that requires more racking and optimizers and such so the cost would have been higher. We seldom ever see optimal conditions and most panels only produce 80% of rated value at max.

Your existing array of 10 x 410w panels should generate 3.2kW~4kW and the 7 should generate 2.3kW~2.8kW - in summer - Do you know the tilt angle? or the roof pitch if they are mounted flat with the roof? In winter they typically only produce 20% of summer when they are fixed. So your graphs are pretty spot on power wise.

Note on panel angles and an example
In general the ideal angle is your latitude - so I live a bit west of denver and my latitude is 39.2 and that is my ideal year-round tilt angle. In winter it is better at closer to 50 degrees because the sun is lower in the sky. In summer my ideal is 22 degrees. I have a 5/12 pitched roof so they are at 22 degrees mounted flat.... In my case I can't do ground mount because of the wind speeds and for the same reason I can't tilt them up in winter - they would act like a sail and rip the roof off.

Based on the graphs you posted while I am typing I would add the single panel to the south roof with an optimizer. Ideally you wouldn't have to deal with shade. You would get more power out if it on the west roof without the shade but from the above calculation I wouldn't stress the MPPT more than it is.


For any more panels in a separate array you would need an independant MPPT connected to the batteries as the ones you have are used. Stick with the Growatt brand and it should integrate into your system and monitoring

 

[binding_energy]

Here are the bits needed to figure it

2 MPPT in the inverter - Both are the same - they don't turn on until string voltage hits 120v so your panel strings need to be at least that as early as possible. The place they work best is 360v so that is the target to try and hit.

in series voltage adds and in parrallel current adds - your panel VOC is 37.26v - so a 10 panel string is just about perfect for hitting the correct voltage. Since a string of 18 panels would be 660v and that is outside the range I assume the installed array is 2 strings of 9 panels each using one of the MPPT. You can't have a 9s2p arrangment because the max input current is 12a and in reality your panels Isc is 13.1a - so in theory on a perfect day with a perfect angle your panels as mounted could burn out the built in MPPT -

This seems to be a poor choice of panels matched to the inverter MPPT to me. They should probably have used something like 335w panels or a different 400w panel with an ISC of 9amps or so. But that requires more racking and optimizers and such so the cost would have been higher. We seldom ever see optimal conditions and most panels only produce 80% of rated value at max.

Your existing array of 10 x 410w panels should generate 3.2kW~4kW and the 7 should generate 2.3kW~2.8kW - in summer - Do you know the tilt angle? or the roof pitch if they are mounted flat with the roof? In winter they typically only produce 20% of summer when they are fixed. So your graphs are pretty spot on power wise.

Note on panel angles and an example
In general the ideal angle is your latitude - so I live a bit west of denver and my latitude is 39.2 and that is my ideal year-round tilt angle. In winter it is better at closer to 50 degrees because the sun is lower in the sky. In summer my ideal is 22 degrees. I have a 5/12 pitched roof so they are at 22 degrees mounted flat.... In my case I can't do ground mount because of the wind speeds and for the same reason I can't tilt them up in winter - they would act like a sail and rip the roof off.

Based on the graphs you posted while I am typing I would add the single panel to the south roof with an optimizer. Ideally you wouldn't have to deal with shade. You would get more power out if it on the west roof without the shade but from the above calculation I wouldn't stress the MPPT more than it is.


For any more panels in a separate array you would need an independant MPPT connected to the batteries as the ones you have are used. Stick with the Growatt brand and it should integrate into your system and monitoring




View attachment 258503
View attachment 258500
View attachment 258501

Thank you very much for a detailed reply. There are a couple of points upon which I would like clarification though please.

You say my existing array or 10 panels. Its' actually currently 11 panels, 10 facing west and one facing south (with an optimiser) on one string The other string is 7 x south. Are you suggesting that I remove the single south facing panel from string 1 and put it on string 2 so I have 10 x west on one string and 8 x south on the other? I could do that, just need a bit of extra cable and some plugs. It does seem sensible to put all south ones on the same string even if they are physically separated from each other

And for the additional panels that I would like to put on the east roof to catch the early sun in the winter (I could probably get another 10 or 11 on there although I don't know if I want to pay that much). Are you suggesting I could put them all on a third string with its own stand alone MPPT connected directly to the battery? (I assume the growatt inverter already has two MPPTs internally as I don't have any external ones and the two pairs of cables from my existing two strings plug straight into the inverter)

If so that sound like a sensible idea and quite cheap and straight forward to do. Something like the 3kW Growatt MIC 3300TL-X is only £384, coupled with a ShineWifi dongle so that I can monitor it

I'm guessing there would be no conflict between the Growatt inverter charging and the separate external MPPT also charging as the inverter is primarily used for DC to AC conversion. Also I hope that a separate dongle would just add to the data sent via the single dongle on my inverter so I would be able to monitor 3 strings at once .

Thanks again for your help

​

 

[binding_energy]

not enough information to answer

Panel specs? post a picture or link
MPPT specs - what make/model is it?

My latitude here in a small village in the East of England is 52.41 degrees.
Tilt is pretty much 45 degrees as they are fixed flat against the roof (seems to be how it's done in England). 45 degrees is my roof angle, and that's probably to shed all the rain we get over here compared to Denver!

 

[robbob2112]

All panels in a string need to be in the same plain - facing them a different direction while in series will drag down the current produced.... if the installer did that on purpose either they didn't understand things so well, or they deliberately did it to drag the current under the max limit on the inverter. I don't think I would change that without some careful monitoring since the input current already maxed out. You would still be over so I don't know what the intent of the installer is. If they did do that they did a bad thing.

Solar panels are basically like a diode in general electronic - ok they actually are a diode - but in this case the sun's energy forward biases the diode so current will flow and they have a voltage generated per cell - Your panels have 108 cells - in 6 rows of 18. There are 3 sets on the top end and 3 sets on the bottom end connected in series with bypass diodes between them. Each cell adds 0.345v in series of 18 - so 6.21v - then those are in series. So you have 6 series sets * 6.21v = 37.26 -- your Voc with rounding ...

The bypass diodes are used when there is shading or more important when a string of cells isn't producing the voltage it should... When you have the sun in the west the south facing panel ends up in bypass on all the strings ... bypass diodes are intended for temporary use when a plane flies over, or clouds or the like, not continuous operation. The optimizers takes over this function and gets some current out of the panel when under normal circumstances it would generate no power and in fact use power - 13.1a * the voltage drop of 0.4v per diode x 6 - so 31w of power would be wasted and the diodes would burn out in the end.

Does your monitoring show the current received from the arrays?

If you intend doing more panels you need a clamp meter that does DC current. With that you can use it on the wires to see what the panels produce in maximum sun. Read in my signature link for the tools I have section - it has several cheaper clamp meters that work as well as better quality ones.

So before I changed anything I would need to understand the actual conditions at max power - current and voltage which equals watts, but would need spot readings verse trend lines on a graph.

I think in an ideal world the single south panel would be connected to rest of the south panels so you would have 8 and 10 verse 11 and 7... and you would collect another 350w to 400w from the strings.


Yeah, your built in strings are using the MPPTs- so a separate MPPT and a separate string of panels. Growatt offers a number of stand alone MPPTs you could use.

I looked it up and the SPH6000 can't be stacked so no expansion by adding another one to get more MPPT that way - some inverters can be added in parallel to get more wattage and more MPPT, but not your model.

 

[binding_energy]

Brilliant, Thank you for your reply. It all seems much clearer now. I used to be an electrician so already have a couple of clamp meters that will measure DC current. Unfortunately the monitoring software doesn't show current but using the 7 panel, all south facing, string at 12:30pm during a day in August this year (max power output) I get a ballpark PD of 200V and a corresponding power of about 3080W. That gives that give an approx current of 11.6A from that one string, close to the max of 12A for the inverter's MPPT on that string. Maybe that's why the installer put the 8th south panel into the west string, to bring down that string's current (12.1A at the same time as the numbers above).

I'll do some more calcs around the cost of what I'm proposing and also take some real measurements so I have a more accurate idea of what is happening at the moment. If I may I might come back in a few months once I have a firmer idea of the existing system and what difference my proposed addition would make.

I'll also study your 'answers' thread, looks fascinating, especially rule 11 !

Thank you

 

[robbob2112]

Brilliant, Thank you for your reply. It all seems much clearer now. I used to be an electrician so already have a couple of clamp meters that will measure DC current. Unfortunately the monitoring software doesn't show current but using the 7 panel, all south facing, string at 12:30pm during a day in August this year (max power output) I get a ballpark PD of 200V and a corresponding power of about 3080W. That gives that give an approx current of 11.6A from that one string, close to the max of 12A for the inverter's MPPT on that string. Maybe that's why the installer put the 8th south panel into the west string, to bring down that string's current (12.1A at the same time as the numbers above).

I'll do some more calcs around the cost of what I'm proposing and also take some real measurements so I have a more accurate idea of what is happening at the moment. If I may I might come back in a few months once I have a firmer idea of the existing system and what difference my proposed addition would make.

I'll also study your 'answers' thread, looks fascinating, especially rule 11 !

Thank you

I need to revise and earlier statement - when you exceed current on a MPPT you just end up clipping - so you loose power - This is called overpaneling and it is fine - the MPPT will only draw what it needs

It is Voltage that you can't exceed - not by a volt or you can do damage

So,
Do move the single south to the other string and even them out - and both will collect more power. If that panel gets shaded it needs an optimizer to match the others. If it is never shaded you can skip the optimizer. The installers probably did what they did so they didn't need to run the wires.

Do wear insulated gloves and also turn off the strings when you do the switch. Be prepared to replace the MC4 connectors - they can sometimes not come apart or can need to be clipped off for other reasons -

The panels didn't specify a brand and they are supposed to match - but if you can't find it - then go with Stabuli brand - it is the original and top quality. You may or may not need a removal tool to get them apart.

While you are up there inspect for heat damage and loose connectors - and make certain the ground wires are attached to the panels - they are allowed to be daisy chained.

 

[binding_energy]

Even more thanks - looks like I've got a job on now !
Regards