Do TIGOs actually work?

[pv_n00b]

Yeah, well, try telling that to an inspector who insists otherwise, that turning off a single inverter for maintenance in a multi inverter system must activate all the DC connected to that inverter even though there is nothing in the NEC to support his position.

The NEC is pretty clear about the purpose of RSD. If the AHJ wants to ignore the NEC and just make up stuff then that gets dealt with on a case by case basis. AHJs are always looking for fun and entertaining ways to AHJ. It has to be a boring job.

 

[pv_n00b]

Wow really but then how is it safe for fire fighters whom might be about to chainsaw thru the same wires?

Firefighters have a pretty good set of PPE on when they are working. I have yet to see a PV tech wearing full firefighter kit while they are troubleshooting a PV system.

 

[ggunn]

There may still be plenty of DC voltage on the conductors as discussed here.

Not if the standards set forth in the NEC are adhered to. It is not enough to simply disconnect the DC modules from each other and the home runs back to the inverter; the residual voltage left after the disconnection must be bled off in order for the voltage to be less than 30VDC in 30 seconds outside the array and 80V inside the array. I don't know if the SunSpec protocol stipulates whether the inverters or the RSD boxes on the modules perform this function, but earlier SolarEdge inverters used to have to have resistors added to them across the + and - terminals to meet this requirement.

 

[jaggedben]

Not if the standards set forth in the NEC are adhered to. It is not enough to simply disconnect the DC modules from each other and the home runs back to the inverter; the residual voltage left after the disconnection must be bled off in order for the voltage to be less than 30VDC in 30 seconds outside the array and 80V inside the array. I don't know if the SunSpec protocol stipulates whether the inverters or the RSD boxes on the modules perform this function, but earlier SolarEdge inverters used to have to have resistors added to them across the + and - terminals to meet this requirement.

See the actual discussion I liked to. It's not even that simple. Also 80V isn't necessarily a safe criterion by itself for maintenance work.

 

[tortuga]

See the actual discussion I liked to. It's not even that simple. Also 80V isn't necessarily a safe criterion by itself for maintenance work.

The systems Phil is referring to (if I understand him right) as well as the ones I tested were an array on a roof then a pipe run to a single inverter, the brand of rapid shutdown was Tigo. So would that inverter not fall under 690.12(B)(1) outside the array boundary? The thread you linked mentions and discusses the 1-3 options for 690.12(B)(2) 'inside the array boundary'.

 

[ggunn]

See the actual discussion I liked to. It's not even that simple. Also 80V isn't necessarily a safe criterion by itself for maintenance work.

I did. The 80V applies to conductors between the module and the RSD attached to it. I don't see anything we can do about that, but anyway, rapid shutdown doesn't have anything to do with maintenance.

 

[pv_n00b]

I did. The 80V applies to conductors between the module and the RSD attached to it. I don't see anything we can do about that, but anyway, rapid shutdown doesn't have anything to do with maintenance.

The 80V value was selected not for any specific safety reason but it was decided by the CMP that 80V would allow the Voc of any single module that might be used on a roof in the future. There are higher Voc thin film modules but they are rarely used on roof installations.

 

[pv_n00b]

Not if the standards set forth in the NEC are adhered to. It is not enough to simply disconnect the DC modules from each other and the home runs back to the inverter; the residual voltage left after the disconnection must be bled off in order for the voltage to be less than 30VDC in 30 seconds outside the array and 80V inside the array. I don't know if the SunSpec protocol stipulates whether the inverters or the RSD boxes on the modules perform this function, but earlier SolarEdge inverters used to have to have resistors added to them across the + and - terminals to meet this requirement.

SunSpec is only for control of the RSD MLPE. It does not specifically address RSD and how it should work.
Prior to RSD there was no requirement for how long the DC stage caps in the inverter would keep the DC terminals energized after being disconnected from the PV array. Most instruction manuals recommended that people wait for a specified period after disconnecting the PV array from the inverter before doing any work on the inverters. Some of the bleed-down systems took 5 or 10 minutes to bleed the caps down. After RSD came out it took a while for inverter manufacturers to catch up and modify the inverters to bleed-down faster to comply with RSD. During the intermediate period, there were a lot of interesting things that were tried to make RSD work without the inverters playing along. But now UL-listed string inverters all play nice with RSD and bleed down the DC caps within the allotted time.
The future looks to be murky though with 3741 listed systems being fielded. The assumption that every system with RSD reduces the array voltage to 80V or less will quickly not be as common as 3741 systems that leave the array at full voltage. The results could be shocking.

 

[ggunn]

The 80V value was selected not for any specific safety reason but it was decided by the CMP that 80V would allow the Voc of any single module that might be used on a roof in the future. There are higher Voc thin film modules but they are rarely used on roof installations.

Yes, but there is a way around that. SolarEdge P1101 optimizers are designed to handle two modules in series with a maximum temperature corrected series Voc of up to 125V, permitted under 690.12(B)(2)(1) as a listed hazard control system.

 

[ggunn]

SunSpec is only for control of the RSD MLPE. It does not specifically address RSD and how it should work.
Prior to RSD there was no requirement for how long the DC stage caps in the inverter would keep the DC terminals energized after being disconnected from the PV array. Most instruction manuals recommended that people wait for a specified period after disconnecting the PV array from the inverter before doing any work on the inverters. Some of the bleed-down systems took 5 or 10 minutes to bleed the caps down. After RSD came out it took a while for inverter manufacturers to catch up and modify the inverters to bleed-down faster to comply with RSD. During the intermediate period, there were a lot of interesting things that were tried to make RSD work without the inverters playing along. But now UL-listed string inverters all play nice with RSD and bleed down the DC caps within the allotted time.
The future looks to be murky though with 3741 listed systems being fielded. The assumption that every system with RSD reduces the array voltage to 80V or less will quickly not be as common as 3741 systems that leave the array at full voltage. The results could be shocking.

The future is murky for parts of the NEC? Horrors! Say it ain't so!

 

[pv_n00b]

Yes, but there is a way around that. SolarEdge P1101 optimizers are designed to handle two modules in series with a maximum temperature corrected series Voc of up to 125V, permitted under 690.12(B)(2)(1) as a listed hazard control system.

Because they listed the P1101 under 3741. Systems listed under 3741 have no specific PV array voltage requirement, so the PV array will be anything from 0V to the module or MLPE string Vmax after shutdown initiation. There will be no way to tell just by looking at the system what someone might be exposed to. The only safe assumption now is to treat every array like it has full voltage after an RSD initiation.

 

[ggunn]

Because they listed the P1101 under 3741. Systems listed under 3741 have no specific PV array voltage requirement, so the PV array will be anything from 0V to the module or MLPE string Vmax after shutdown initiation. There will be no way to tell just by looking at the system what someone might be exposed to.

I know, and I am glad that the P1101 is so listed. I use them a lot; I think that the added risk is minuscule since there is not that much difference between 80VDC and 125VDC and the conductors are just running between two physically adjacent modules.

 

[tortuga]

The 80V value was selected not for any specific safety reason

Yeah i'll add the arbitrary voltage of 100 volts from energy storage systems to that list of voltages selected for no safety reason at all. 706.20(B) I think is the 2020 code ref, 706.30(A) in the 2017.
People forget that the NYC 110/220V DC Edison system operated just fine for a very long time. The last DC poco portions were shut off in when 2007?
I'd say change those numbers to 250VDC L-L and 125V L-G as that is a long time rating for DC and batteries going back 100+ years pre dating our AC systems.

 

[pv_n00b]

I remember reading an article maybe 10 years ago about PG&E ending DC service in San Francisco to the few remaining legacy customers. Old electrical systems are kept alive by old elevators. It's so expensive to bring a 100-year-old elevator system up to current code that owners will do just about anything to keep them running. I know at least one building that has an old elevator system that still requires an attendant to operate it, they only have it running during special events. I can only imagine how unsafe it is by modern standards.

 

[pv_n00b]

I know, and I am glad that the P1101 is so listed. I use them a lot; I think that the added risk is minuscule since there is not that much difference between 80VDC and 125VDC and the conductors are just running between two physically adjacent modules.

Yup, and I see other 3741 listed systems with the full 1,000V Voc available in the array after shutdown. So 125V is nothing.

 

[ggunn]

Yup, and I see other 3741 listed systems with the full 1,000V Voc available in the array after shutdown. So 125V is nothing.

What systems are those?

 

[pv_n00b]

What systems are those?

SMA CORE1 + Sollega. IronRidge should have the BX system with the CORE1 out soon if not already out.

 

[gene6]

I know little about solar, and i get you all are discussing 690.12. my stupid layman's question is why not just treat these conductors as service conductors? under art 230? Seems like the same thing?

 

[jaggedben]

I know little about solar, and i get you all are discussing 690.12. my stupid layman's question is why not just treat these conductors as service conductors? under art 230? Seems like the same thing?

Their totally not the same thing at all in multiple ways. How do you think they are the same? Also how would 'treating them like service conductors' address saftey issues for firefighters?

 

[gene6]

Their totally not the same thing at all in multiple ways. How do you think they are the same? Also how would 'treating them like service conductors' address saftey issues for firefighters?

Did I not mention I am an idiot and don't know what I am talking about? LOL
No really thought the only similarity is they you have unfused conductors entering a building right?
So why not just treat them as service conductors?