Micro-Inverters in the Attic

MikeNZ

Member
Hi all,

I am looking to install a PV system on my roof using micro-inverters, but my situation wont allow for these inverters to be mounted on the roof, so i want to install them in the attic below the array and join them together there. Im aware that the DC cabling inside the attic is required to be in conduit or enclosure before it reaches the inverters, but once its converted to AC is Romex ok to join the inverter outputs together?
is there any other rules im missing?
 

GoldDigger

Moderator
Staff member
Hi all,

I am looking to install a PV system on my roof using micro-inverters, but my situation wont allow for these inverters to be mounted on the roof, so i want to install them in the attic below the array and join them together there. Im aware that the DC cabling inside the attic is required to be in conduit or enclosure before it reaches the inverters, but once its converted to AC is Romex ok to join the inverter outputs together?
is there any other rules im missing?
You are potentially overlooking the consequences of Rapid Shutdown requirements in recent Code cycles. One advantage of micro-inverters is that the uncontrolled DC from the panels does not travel outside the array, avoiding some of the rapid shutdown requirements. Bringing the DC inside the attic may trigger those requirements and shutdown at the microinverter itself would not be sufficient.
 

ggunn

PE (Electrical), NABCEP certified
Hi all,

I am looking to install a PV system on my roof using micro-inverters, but my situation wont allow for these inverters to be mounted on the roof, so i want to install them in the attic below the array and join them together there. Im aware that the DC cabling inside the attic is required to be in conduit or enclosure before it reaches the inverters, but once its converted to AC is Romex ok to join the inverter outputs together?
is there any other rules im missing?
Microinverters typically mount up under the module backsheet inside the frame. I don't see how you could mount modules but not modules with micros. Mounting microinverters in the attic would be, IMO, a wiring nightmare. You'd have to either have a roof penetration under every module or home runs between every module and its inverter. It's not something I would ever consider.

Some AHJ's I deal with allow microinverter interconnections with Romex but some also require a #8 ground wire.
 

pv_n00b

Senior Member
Microinverters typically mount up under the module backsheet inside the frame. I don't see how you could mount modules but not modules with micros. Mounting microinverters in the attic would be, IMO, a wiring nightmare. You'd have to either have a roof penetration under every module or home runs between every module and its inverter. It's not something I would ever consider.

Some AHJ's I deal with allow microinverter interconnections with Romex but some also require a #8 ground wire.
I've seen some resi roof mounting systems that won't allow MLPE boxes under the modules. They are zero clearance systems that attach the modules flat to the roof. I don't see them often because with no back of module airflow it really heats up the modules and reduces the output but they are out there. RSS will probably kill them off.

Since you will have a lot of roof penetrations OP make sure your sealing technique is 100%. Better yet, use a string inverter.
 

jaggedben

Senior Member
Just use a racking system that allows them on the roof.

Rapid shutdown doesn't apply if DC is only one module worth.
 

ramsy

Senior Member
Micro Inverters should be listed for 30-40c ambient, not 50-60c in attic space, or garage space.

Same reason smoke detectors won't work, and not required in attics or garages.
 

jaggedben

Senior Member
How so? Plenty of DC modules have Voc > 30V.
I guess if one extended the DC conductors beyond the length of the module leads then 2017 requirements could be triggered. Otherwise I see it as quite unlikely they'll be outside the array boundary. Perhaps my comment was unwarranted though; yes it could be an issue if one installed them just so.

Micro Inverters should be listed for 30-40c ambient, not 50-60c in attic space, or garage space.

Same reason smoke detectors won't work, and not required in attics or garages.
Roof surfaces get really hot too. The Enphase IQ7 has a max temp rating of 65C. There are a bunch of reasons I think it's a bad idea but that's not high on the list.
 

SceneryDriver

Senior Member
Since you will have a lot of roof penetrations OP make sure your sealing technique is 100%.
It's not IF a roof penetration will leak, it's WHEN. Punching that many holes in the roof is just BEGGING for roof leaks that will be impossible to get to, what with being underneath the apparently zero-clearance mounted panels.

You want to put as few holes as possible in a roof, and do it absolutely correctly when you do. Call a roofer for roof penetrations, and let someone else take the liability.

Find a different racking system that allows the inverters to be mounted behind the panels, and make the roof penetration somewhere that's NOT directly under the array, so it can be serviced later.


SceneryDriver
 

ggunn

PE (Electrical), NABCEP certified
Find a different racking system that allows the inverters to be mounted behind the panels, and make the roof penetration somewhere that's NOT directly under the array, so it can be serviced later.
We frequently put Soladeck boxes under the array for aesthetic reasons, but it's just one penetration per array that's not into a rafter and getting to it is pretty easy.
 

ramsy

Senior Member
..The Enphase IQ7 has a max temp rating of 65C..
Found it. Enphase IQ7/Plus.
Ambient temperature range -40ºC to +65ºC

Amazing. Never imagined roofs getting that hot behind a panel, in the shade.
At 65c wire must be 150%+ oversized, per NEC Table 310.15(B)(2)(a)
 

jaggedben

Senior Member
Found it. Enphase IQ7/Plus.
Ambient temperature range -40ºC to +65ºC

Amazing. Never imagined roofs getting that hot behind a panel, in the shade.
At 65c wire must be 150%+ oversized, per NEC Table 310.15(B)(2)(a)
Usually with micros one is just plugging a listed module into a listed inverter so there's no wiring subject to NEC rules. We solar guys are also used to using 10awg wire for circuit currents less than 15A.
 

ggunn

PE (Electrical), NABCEP certified
Usually with micros one is just plugging a listed module into a listed inverter so there's no wiring subject to NEC rules. We solar guys are also used to using 10awg wire for circuit currents less than 15A.
True, but microinverter trunk cables typically have #12 conductors.
 

MikeNZ

Member
You are potentially overlooking the consequences of Rapid Shutdown requirements in recent Code cycles. One advantage of micro-inverters is that the uncontrolled DC from the panels does not travel outside the array, avoiding some of the rapid shutdown requirements. Bringing the DC inside the attic may trigger those requirements and shutdown at the microinverter itself would not be sufficient.
the micro inverters will be located within 3' of the point of entry, this is still within the array boundary limits according to 690.12(B) (1), so this means the micros can act as a rapid shutdown source. which is one of the main reasons we have decided to locate them there.
 

MikeNZ

Member
Microinverters typically mount up under the module backsheet inside the frame. I don't see how you could mount modules but not modules with micros. Mounting microinverters in the attic would be, IMO, a wiring nightmare. You'd have to either have a roof penetration under every module or home runs between every module and its inverter. It's not something I would ever consider.

Some AHJ's I deal with allow microinverter interconnections with Romex but some also require a #8 ground wire.
the reason we are installing these in the attic is our particular system has no space at all on the roof for the inverters, its a thin film BIPV setup. A home run per string would mean around 10 pairs of armoured and labelled cables heading down to multiple string inverters which is just too uneconomical to consider. Surely loops between micro inverters in the attic is less of a nightmare than a home run from each penetration point?
 

MikeNZ

Member
I've seen some resi roof mounting systems that won't allow MLPE boxes under the modules. They are zero clearance systems that attach the modules flat to the roof. I don't see them often because with no back of module airflow it really heats up the modules and reduces the output but they are out there. RSS will probably kill them off.

Since you will have a lot of roof penetrations OP make sure your sealing technique is 100%. Better yet, use a string inverter.
we have a very good design for sealing each point of entry but thank you for the concern, and as in an above reply, a string inverter is too uneconomical to consider for our particular system.
 

MikeNZ

Member
I guess if one extended the DC conductors beyond the length of the module leads then 2017 requirements could be triggered. Otherwise I see it as quite unlikely they'll be outside the array boundary. Perhaps my comment was unwarranted though; yes it could be an issue if one installed them just so.



Roof surfaces get really hot too. The Enphase IQ7 has a max temp rating of 65C. There are a bunch of reasons I think it's a bad idea but that's not high on the list.
you are correct, we are going to be using the IQ7+, temperature has been a heated (pun intended) discussion point around here, but we are pretty sure we have it sorted. After reading my other replies to concerns, are there any more reasons this may be a bad idea you can think of?
 
Top