Panel wiring

[ButchDeal]

I just found these: SMA Sunny Boy 240-US-99-10 Inverter

Made in USA, not china. Have to use their AC cable, but the DC side can have the MC4 connectors my panels have.

I thought they were concentrating on their new optimizer solution. Haven't seen much new on the SMA micros in a long time.
Also SMA is a German company not US.

[max2k]

Ha. We had a request for that on a homedepot shed of only 8X10 to have 4 foot cantilever overhang on 3 sides. Customer had looked up the specs on ironridge xr1000 rails and was sure it was ok because ironridge said so.
We tried to explain that that makes the shed into a glider and we are not licensed to manufacture air craft...

if it is well anchored may be it could still pass as a shed . Are there any particular restrictions inspectors enforce in such case like relative ratio of cantilever part vs length of the supporting 'structure'?

[max2k]

I thought they were concentrating on their new optimizer solution. Haven't seen much new on the SMA micros in a long time.
Also SMA is a German company not US.

It looks like they abandoned microinverters solution, all info on SMA site about that direction is from 2013. They're pushing now their own optimizers: http://www.sma-america.com/powerplus.html

On one hand they allow mixing those optimizers with raw panels so you'd need only as many as you have potentially shaded panels. OTOH RSS would still be required resulting in string level shutdown, not module level if I understand this correctly. Looks like SMA is having difficulty to catch up with SE if optimizers / module level RSS are required. The idea of optimizers to be installed on only potentially shaded panels is interesting if they could come up with some data communication over the same DC wires for RSS purposes and very dumb/cheap RSS module for the rest of unshaded panels eliminating need for string level RSS. At least this solution works with their string inverters which would probably mean it would work with any other string inverter: SMA states their optimizers simply lower voltage of the panel boosting its current when it gets shaded. While this approach seems to be simpler than SE it imposes the same limits on max string length from Voc side as for raw strings, about 13-14 panels max.

It seems the company Tigo they subcontracted module level electronics to produces different kind of modules: when shopping I was offered modules just for RSS without optimization. On 24 panel array it resulted in $1,300 IIRC vs $750 for their conventional RSS. Since OC is still on 2014 NEC I went with conventional one for simplicity and lower cost. What bothers me with Tigo offering is their modules consist of 2 parts- some common 'base' and another module type specific extension. All of them share the same base but extensions vary from one type of the module to another. This must require extra connectors inside the modules and depending how that is done can affect reliability.

[J.P.M.]

if it is well anchored may be it could still pass as a shed . Are there any particular restrictions inspectors enforce in such case like relative ratio of cantilever part vs length of the supporting 'structure'?

Aside from aesthetics, it's not so much a matter of restrictions, it's more the stuff of structural engineering. If you anchor a flimsy structure with reinforced concrete footings 12 ft. deep it may still fail. Again, the stuff of structural engineering.

Structural loading determination and requirements to deal with such loadings vary by area and jurisdiction. Many use ASCE 7, or various sections of it for such things.

[ButchDeal]

It looks like they abandoned microinverters solution, all info on SMA site about that direction is from 2013. They're pushing now their own optimizers: http://www.sma-america.com/powerplus.html

On one hand they allow mixing those optimizers with raw panels so you'd need only as many as you have potentially shaded panels. OTOH RSS would still be required resulting in string level shutdown, not module level if I understand this correctly. Looks like SMA is having difficulty to catch up with SE if optimizers / module level RSS are required. The idea of optimizers to be installed on only potentially shaded panels is interesting if they could come up with some data communication over the same DC wires for RSS purposes and very dumb/cheap RSS module for the rest of unshaded panels eliminating need for string level RSS. At least this solution works with their string inverters which would probably mean it would work with any other string inverter: SMA states their optimizers simply lower voltage of the panel boosting its current when it gets shaded. While this approach seems to be simpler than SE it imposes the same limits on max string length from Voc side as for raw strings, about 13-14 panels max.

It seems the company Tigo they subcontracted module level electronics to produces different kind of modules: when shopping I was offered modules just for RSS without optimization. On 24 panel array it resulted in $1,300 IIRC vs $750 for their conventional RSS. Since OC is still on 2014 NEC I went with conventional one for simplicity and lower cost. What bothers me with Tigo offering is their modules consist of 2 parts- some common 'base' and another module type specific extension. All of them share the same base but extensions vary from one type of the module to another. This must require extra connectors inside the modules and depending how that is done can affect reliability.

SMA purchased Tigo

[max2k]

SMA purchased Tigo

so there's hope their next gen offering would look like better integrated solution

[ButchDeal]

so there's hope their next gen offering would look like better integrated solution

possibly but they have a lot to learn on the integrated optimization solution and asic side of things to catch up to SolarEdge.

[max2k]

possibly but they have a lot to learn on the integrated optimization solution and asic side of things to catch up to SolarEdge.

nah, all they need to do is to run their own small residential install company so their engineers would get up the roof once in a while and see how things work / don't work in the field. That would get things moving quickly. ASIC is just one of the possible ways to skin proverbial cat.

[ButchDeal]

nah, all they need to do is to run their own small residential install company so their engineers would get up the roof once in a while and see how things work / don't work in the field. That would get things moving quickly. ASIC is just one of the possible ways to skin proverbial cat.

I am talking about R&D on the technical side not the installation and documentation stuff which is a lot easier to improve.

They have a lot to do to catch up .
SolarEdge has the most efficient inverters by far as well as very reliable optimizers that fully integrate.

SMA so far has done little technically with the tigo system. The next step would have them communicate and integrate with the inverter, then have dedicated (cheaper / more reliable) inverters without the built in MPPT. This second phase will likely require new optimizers as the tigo does not maintain constant voltage like the SolarEdge optimizers...
Then increase the efficiency based on the constant voltage,
thats just to get where solarEdge is now....

[max2k]

I am talking about R&D on the technical side not the installation and documentation stuff which is a lot easier to improve.

They have a lot to do to catch up .
SolarEdge has the most efficient inverters by far as well as very reliable optimizers that fully integrate.

SMA so far has done little technically with the tigo system. The next step would have them communicate and integrate with the inverter, then have dedicated (cheaper / more reliable) inverters without the built in MPPT. This second phase will likely require new optimizers as the tigo does not maintain constant voltage like the SolarEdge optimizers...
Then increase the efficiency based on the constant voltage,
thats just to get where solarEdge is now....

I think all major inverters are efficient enough nowadays and I wouldn't even pay attention if it exceeds 96-97%. Constant voltage is not really required except to enable longer strings. I think SMA/Tigo approach to optimization when they just maintain constant current flowing through optimizer leads to simpler on the roof modules and opening possibility for mixing optimizers with raw panels to reduce the cost. It sounds though Tigo overcomplicated its implementation negating that advantage.

IMO any manufacturer needs to maintain close connection with the field, the rest can be solved one way or the other. SMA has such channel but it doesn't sound it gets high enough priority within the company. The moment they change that and allow field to drive internal product development we might see more relevant offerings.

[ButchDeal]

I think all major inverters are efficient enough nowadays and I wouldn't even pay attention if it exceeds 96-97%.

HDWAVE from Solaredge is CEC 99% efficient. SMA has a CEC efficiency of 96.5%
OH but thats just a little bit, but thats like loosing an entire modules production on a 10kw array.

Constant voltage is not really required except to enable longer strings. I think SMA/Tigo approach to optimization when they just maintain constant current flowing through optimizer leads to simpler on the roof modules and opening possibility for mixing optimizers with raw panels to reduce the cost. It sounds though Tigo overcomplicated its implementation negating that advantage.

ok clearly we are talking two different things.

I am not talking about simplisity of installation with tigo, solaredge or string. I am talking about simplicity of the elctronics and reliability of the equipment. SMA has multiple complicated MPPT inputs on their inverters, SolarEdge doesn't have any, simpler and less parts, more reliable.
SolarEdge is very simple to INSTALL you can't get much damn simpler. but that is completely different.
As for the install part, solaredge has things like integrated auto rapid shut down, integrated module level monitoring
SMA could add these two things and integrate the monitoring, and simplify the optimizer communication, but that likely would add complexity to their already complex systems.
I would hope that they are starting over with a simpler system that has the features built in, instead of trying to bubble gum and duct tape more to the existing systems they have.

IMO any manufacturer needs to maintain close connection with the field, the rest can be solved one way or the other. SMA has such channel but it doesn't sound it gets high enough priority within the company. The moment they change that and allow field to drive internal product development we might see more relevant offerings.

possibly but a better design instead of just tacking two companies and their products together would get them a much better design. That takes R&D.

[max2k]

ok clearly we are talking two different things.

nope, the same- when optimizer needs to only maintain current passing through by lowering its voltage it simplifies design a lot: fully irradiated panel would look like no optimizer is there while fully shaded would look like short. Too bad it seems they lost beauty of this idea in the implementation. I agree, duct taping 2 company offerings was not a smart move unless they saved on the idea above and currently busy developing normally integrated product.