Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Question about 120 percent NEC rule with Tesla Gateway 2

This site may earn commission on affiliate links.
Lol you actually found those LOTO clips? Yeah I think those clips should be a good compromise between putting lots of blades next to the ESS or putting a sub panel full of OCPDs next to the ESS. Well, unless you're doing an install in Contra Costa County.

In case it got lost in this thread... when I had the CoCo inspector on site to challenge their interpretation of 706.7, he called the home office and was told by Garth Robertshaw (heads up all of Contra Costa's inspectors) that he agreed the LOTO clip works for 706.7(B). But, he said that the clip would only apply if the OCPD with the LOTO clip were within 5 feet of the ESS. If the ESS were on the other side of a wall, then he would force 707 (E). Basically Garth thinks 706.7(B) with LOTO clip is sufficient for the safety of the electrical worker who comes on site to do work. But 706.7(E) is necessary for the safety of a firefighter who needs to de-energize the house. So no matter what the site placard says, he wants disconnecting means next to the ESS.

Basically you'd have to put the TEG2 and internal panelboard right next to the ESS, but then I think that configuration would still need a disconnecting means outside for fire safety since the disconnecting means are supposed to be near the MSP outside. I'm super confused about all this this either way haha.

706.7 (B) Remote Actuation. Where controls to activate the disconnecting means of an ESS are not located within sight of the system, the disconnecting means shall be capable of being locked in the open position in accordance with 11.25, and the location of the controls shall be field marked on the disconnecting means.

706.7 (E)(1) A disconnecting means shall be provided at the energy storage system end of the circuit. Fused disconnecting means or circuit breakers shall be permitted to be used.

706.7 (E)(2) A second disconnecting means located at the connected equipment shall be installed where the disconnecting means required by 706.7(E)(1) is not within line of sight of the connected equipment.

TLDR, molded case OCPDs next to the ESS is probably the better bet for both electrical worker and fire worker safety regardless if the ESS is installed inside or outside in Contra Costa.

Also, regarding that issue H2ofun had in Rocklin (I think it's Rocklin), their checklist uses the term "disconnects" a bit too literally without defining that OCPDs are valid "disconnecting means". So while you may know 706.7 (E)(1) expressly allows molded case breakers as a disconnecting means; the OCPD by strict definition is not a "disconnect". That's why my cool avatar picture was searchable with the term "disconnect".
 
Last edited:
Lol you actually found those LOTO clips? Yeah I think those clips should be a good compromise between putting lots of blades next to the ESS or putting a sub panel full of OCPDs next to the ESS. Well, unless you're doing an install in Contra Costa County.

In case it got lost in this thread... when I had the CoCo inspector on site to challenge their interpretation of 706.7, he called the home office and was told by Garth Robertshaw (heads up all of Contra Costa's inspectors) that he agreed the LOTO clip works for 706.7(B). But, he said that the clip would only apply if the OCPD with the LOTO clip were within 5 feet of the ESS. If the ESS were on the other side of a wall, then he would force 707 (E). Basically Garth thinks 706.7(B) with LOTO clip is sufficient for the safety of the electrical worker who comes on site to do work. But 706.7(E) is necessary for the safety of a firefighter who needs to de-energize the house.



TLDR, molded case OCPDs next to the ESS is probably the better bet for both electrical worker and fire worker safety regardless if the ESS is installed inside or outside in Contra Costa.

Also, regarding that issue H2ofun had in Rocklin (I think it's Rocklin), their checklist uses the term "disconnects" a bit too literally without defining that OCPDs are valid "disconnecting means". So while you may know 706.7 (E)(1) expressly allows molded case breakers as a disconnecting means; the OCPD by strict definition is not a "disconnect". That's why my cool avatar picture was searchable with the term "disconnect".
I am in Auburn, by name, but nevada county by code.

I have no issue living with the blade disconnects. Makes great talking points.

I have more concerns about the new code in July saying ESS PW's have to be at least 3 feet apart. This would mean I could not add anymore in the future. I blew it when I wanted to save a few bucks and do it later. Live and learn.
 
I am in Auburn, by name, but nevada county by code.

I have no issue living with the blade disconnects. Makes great talking points.

I have more concerns about the new code in July saying ESS PW's have to be at least 3 feet apart. This would mean I could not add anymore in the future. I blew it when I wanted to save a few bucks and do it later. Live and learn.

Lol I forget who is in Rocklin... not like I can find these places on a map anyway...
 



Lol I'm resurrecting this thread from the grave. Why? because PG&E originally barred me from installing enough solar panels to charge an EV on account of me not having an EV at the time. At the time, no kWh in my Greenbutton data = Holeydonut sucks azz.

Now that our house has a 12 month old Model 3. And as a surprise to no one, the home has increased kWh consumption compared to when we didn't have an EV. Through 8 months 2022, the home has used the same kWh total as all of 2021.

So, I'm exploring whether I can get a bit more solar added to my house by talking to some installers. The new array is going to be sub optimal... faces north, is shaded by a big ol tree and probably only generates 2,500 kWh* per year. But moar energy is moar energy amirite? PG&E is charging $0.25 per kWh for off-peak energy under EV2-A, and this is going to go to about $0.40 during the next GRC.

PG&E can't deny the extra solar now can they? CAN THEY?????? PG&E? YEAH THEY CAN. I never actually got an answer to the original question about PG&E's issue of this TEG 2 and the 120 percent rule (at least, not an answer that my stupid brain can comprehend). But I do expect the solar installer who has to contact PG&E about my home to add a few panels is going to get a swift kick in the face when PG&E sees the home address that the panels are being applied.

The issue that caused me to post this thread in the first place is encapsulated with @Vines ' example line diagram in the [reply] portion of this message... but I repasted here:

1661653810773.png


Consider the integrated panelboard is capable of getting 63 amps exported onto it from the 3x Powerwalls. And, the PV Gen (solar) is able to push 27 amps at peak from the solar. This is is 90A (not the sum of the breakers; but the 21.6 kW @ 240v generator size) at peak feeding the generation panel.

And PG&E saw my backup load center has 200A worth of goodies to power. In the past, PG&E kept fighting with me and Sunrun that the TEG2 should be assessed for the 120% rule just like it were the busbar on a main service panel that was serving 200A while also getting backfed from a generation panel on the other end of the busbar.

So the 120 percent rule calc would be... TEG2's 200A rating x 1.20 = 240A. Then minus the 200A breaker rating means the maximum PV+ESS allowed is 40A.

Since the aforementioned generation panel is 90A, they claim the system would potentially overload the TEG 2 when the home were drawing 200A while the generation panel was pushing 90A back upstream.

Somehow, PG&E let me get go ahead and proceed even with the potential "setting my house on fire" risk by adding this text to the submitted line diagram design:

NOTE: PRODUCERS STORAGE DEVICE(S) WILL
NOT CAUSE THE HOST LOAD TO EXCEED ITS
NORMAL PEAK DEMAND. NORMAL PEAK
DEMAND IS DEFINED AS THE HIGHEST
AMOUNT OF POWER REQUIRED FROM THE
DISTRIBUTION SYSTEM BY PRODUCERS
COMPLETE FACILITIES WITHOUT THE
INFLUENCE OR USE OF THE ENERGY
STORAGE DEVICE(S).

THE INSTALLED BATTERY IS NOT CHARGING
FROM THE GRID.

THE BATTERY IS EXPORTING TO THE GRID
UNDER NEM AND TOU.


The problem is, I never understood what that above, fancy language actually did to make the 120 percent rule suddenly be resolved. I'd like to explain to the designer adding new panels to my system why this language is so important, and what it actually means for 705.12 (2017). Like why is it that the TEG2 can handle Vines' line diagram and not have to necessarily adhere to the 120 percent rule even though the Tesla Energy Gateway 2 busbar is only 200A rated?


* edit: remember, my concept of ROI doesn't align with you normal humans. I just want more solar so I pay an installer instead of paying PG&E. My entire system is one gigantic negative ROI. And I work in finance. So I know how stupid this all is from a cash-perspective.
 

Attachments

  • 1661653793664.png
    1661653793664.png
    510.8 KB · Views: 407
Last edited:
The problem is, I never understood what that above, fancy language actually did to make the 120 percent rule suddenly be resolved. I'd like to explain to the designer adding new panels to my system why this language is so important, and what it actually means for 705.12 (2017).

There's no rational answer to that question, because from your description PG&E was confused about what the 120% rule means and their original objection was not a proper application of the rule.

Like why is it that the TEG2 can handle Vines' line diagram and not have to necessarily adhere to the 120 percent rule even though the Tesla Energy Gateway 2 busbar is only 200A rated?

The short answer is that the TEG2 is apparently not considered to be "distribution equipment, including switchgear, switchboards, or panelboards" which is what the restrictions in (2017) NEC 705.12(B) would apply to. I'm not actually convinced that's correct.

The longer answer is that the TEG2 (ignoring the interior panelboard, which has to be protected like any other panelboard) was designed for the interconnection of multiple power sources. So they hopefully have accounted for and done the testing for the worst case scenario. Which would be something like a 125A non-backed up panel, a 200A backed up panel, a 200A grid connection, and 125A from PV and PWs, with everything actually running at its rating.

BTW, the soon-to-be published 2023 NEC, not likely to be adopted by CA until I guess 2026 (as the 2020 NEC is I believe being adopted Jan 1, 2023) has language in 705.12(B) referencing "distribution equipment with . . . specific listing and instructions for combining multiple sources." So once that's in effect, the TEG2 or its successor might reasonably be required to have such a listing.

Cheers, Wayne
 
Typically when we are designing the internal Gateway subpanel, we do not put non-battery loads in it. Therefore we wouldnt use the 120% rule at all in your design shown above, we use the 100% rule 705.12.B.2.3.c in your TEG integrated subpanel and main panel. which would allow 200A of generation breakers to be installed on the 200A TEG busbar.

The main panel is the same, except with a 225A busbar and 200A subfeed breaker you have additional 25A of non-backup loads there per code.

PGE can be unpredictable and do not often to follow the NEC and aren't as well trained in it. I would not follow their code advice especially when directed at a homeowner. I had a PGE inspector come fail our interconnection inspection at a 52 kW residential install because the PV was producing power when he power cycled the AC disconnect briefly.

Apparently, someone taught him that PV had to wait 5 minutes to reconnect whenever grid power drops out, but he hadn't read the newest version of UL 1741 SA. The standard now allows for short power dropouts, and the PV to reconnect immediately rather than the somewhat arbitrary 5-minute wait period after any dropout.

The 5-minute wait makes sense and still applies when restarting a dark grid, so things can equalize before adding another uncontrolled source.
 
The short answer is that the TEG2 is apparently not considered to be "distribution equipment, including switchgear, switchboards, or panelboards" which is what the restrictions in (2017) NEC 705.12(B) would apply to. I'm not actually convinced that's correct.

The shorter answer is that in the 2017 NEC, section 705.12(B)(2)(3) on busbars explicitly covers only "busbars in panelboards." And a panelboard has to have OCPD in it to be a panelboard. So as long as there's no main breaker in your GW2, under the 2017 NEC, the busbar restrictions (of which the 120% rule is just one option, as Vines pointed out) clearly doesn't apply to the GW2. In CA, that means until the end of 2022.

However, under the 2020 NEC, which CA adopts January 1, 2023 (IIRC), the comparable section is now 705.12(B)(3), and the limitation to panelboards has been removed, so it applies to all busbars of "distribution equipment . . . capable of supplying multiple branch circuits or feeders, or both". Which at first glance would include the GW2. So at that point in time, refer to my previous answer.

For limited additional discussion, see a thread I started at mikeholt.com because of this question: (2020) 705.12(B) and Bussed Gutters, Etc. I agree with the commenter who suggested that the details were not necessarily thought out or considered in the drafting of the 2017 or 2020 NEC.

Cheers, Wayne
 
As always, thanks @wwhitney and @Vines !

Uhhhh so let me try to spin your ultra-fancy-pants explanations in a way that may make more sense to someone outside of TMC. Is the following somewhat accurate?

With solar installs, the 120% rule is commonly used for main service panels which were originally designed only to have energy sources from the utility grid. When the main service panel has breakers for home loads and is also used as a backfeed device for solar or batteries, the 120% rule applies.

However, the Tesla Energy Gateway 2 only has breakers mounted on its generation panel that is rated for 200A. The breakers for loads being serviced at the home are in a separate load center, which is also 200A rated. Since the generation breakers and the load breakers are not on a common bus, the 120% rule does not need to apply.

*********************

BTW, my TEG2 has the 200A Eaton breaker installed because PG&E said it was "service equipment". Vines didn't draw it that way in his example diagram, because there's no reasonable way someone as awesome as him would design a system as FUBAR as what is on my wall right now.

This is how I ended up with the convoluted mess where both my MSP has a 200A breaker and the TEG2 has a 200A breaker... plus they're both neutral bonded to ground. PG&E and Sunrun couldn't make up their minds on which was service equipment because of how my fence goes in between the MSP and TEG2...

PG&E insisted the TEG2 is now my service equipment since it's the first thing with loads that comes after the meter socket. Sunrun's designers were confounded by my 5 foot tall fence (like a fence you can easily see over if you're 5.5 feet tall)... and said whatever was on the PG&E side of my fence had to be replicated on my side of the fence.

Regardless of POV, the reality is my MSP is just a glorified and oversized meter socket now. And sitting between the MSP and TEG 2 is a 200A blade disconnect for good measure because firefighters wear thick gloves. And there are OCPD’s and disconnects everywhere. But I can’t really tell if the “service equipment” or “panel board” language actually applies to my TEG2 or not.
 

Attachments

  • 1661747697445.png
    1661747697445.png
    738.3 KB · Views: 36
Last edited:
In re "what is the service OCPD?" is there a single OCPD such that if you shut it off, all the loads on the premises lose power? If so, the upstream most (closest to the utility transformer) such OCPD is your service OCPD. That is the only place neutrals and grounds may be connected; anywhere downstream of that service OCPD, it is prohibited to reground the neutral. Also, your grounding electrodes (ground rods, Ufer, etc) need to connect to the grounded (neutral) service conductor at the service OCPD or upstream (I think PG&E doesn't allow upstream).

As to the first question, that would require a bit more thought. It's at least a little accurate, not so much.

Cheers, Wayne
 
Uhhhh so let me try to spin your ultra-fancy-pants explanations in a way that may make more sense to someone outside of TMC. Is the following somewhat accurate?

With solar installs, the 120% rule is commonly used for main service panels which were originally designed only to have energy sources from the utility grid. When the main service panel has breakers for home loads and is also used as a backfeed device for solar or batteries, the 120% rule applies.

However, the Tesla Energy Gateway 2 only has breakers mounted on its generation panel that is rated for 200A. The breakers for loads being serviced at the home are in a separate load center, which is also 200A rated. Since the generation breakers and the load breakers are not on a common bus, the 120% rule does not need to apply.

Let me try to edit that:

With solar/ESS installs, the 120% rule is commonly used for panels that have more than one source of supply, but weren't explicitly designed for that. When the main service panel has breakers for home loads and also a breaker for interconnecting solar or ESS, the 120% rule is the typical way to ensure the panel busbar is protected.

The Tesla Energy Gateway 2 may contain an optional panel within it, but that panel gets treated just like any panel outside the Gateway's enclosure; it's just a case of two different pieces of equipment sharing a case for reasons of convenience. The TEG2 ex internal panel typically has no breakers, so there is nothing to apply the 120% rule to. And it is a piece of equipment that has been designed for connection to multiple sources of supply.

The TEG2 internal panelboard when used may be supplied by multiple sources of power, in which case the 120% rule is one option for protecting its busbar. When it is used as a generation panel, containing only breakers for solar/ESS, then typically a different rule is used instead.

Sorry I didn't get it that concise. Maybe you can improve on it.

Cheers, Wayne
 
  • Like
Reactions: kayak1
In re "what is the service OCPD?" is there a single OCPD such that if you shut it off, all the loads on the premises lose power? If so, the upstream most (closest to the utility transformer) such OCPD is your service OCPD. That is the only place neutrals and grounds may be connected; anywhere downstream of that service OCPD, it is prohibited to reground the neutral. Also, your grounding electrodes (ground rods, Ufer, etc) need to connect to the grounded (neutral) service conductor at the service OCPD or upstream (I think PG&E doesn't allow upstream).

As to the first question, that would require a bit more thought. It's at least a little accurate, not so much.

Cheers, Wayne


Lololol, I have so many disconnects on my wall... my avatar is now a disconnect.

If you ignore the Tesla Powerwalls for a second, I have 3 different OCPDs that if opened, would shut off power from reaching my home loads. They are wired in series too; basically they're daisy chained 200A to one another; just each one is in a different panelboard (or whatever you call the Gateway).


*************************************************

So on the topic of the 120 percent rule and the TEG2... I'll just pare it down a ton and see what the installer says. They'll probably tell me I'm weird hah.


With solar/ESS installs, the 120% rule is commonly used for panels that have more than one source of supply, but weren't explicitly designed for that. When the main service panel has breakers for home loads and also a breaker for interconnecting solar or ESS, the 120% rule is the typical way to ensure the panel busbar is protected.

In its basic configuration, the Tesla Energy Gateway 2 (TEG2) has no internal breakers on a shared busbar, so there is nothing to apply the 120% rule to. However, the TEG2 may be configured with the addition of an optional panelboard mounted within its casing as a generation panel. This generation panel is rated for 200 amps, and only contains breakers for solar/ESS (no home loads and no energy sources from the grid). So the 120% rule still does not apply.


***************************************************

Thanks Wayne!
 
If you ignore the Tesla Powerwalls for a second, I have 3 different OCPDs that if opened, would shut off power from reaching my home loads. They are wired in series too; basically they're daisy chained 200A to one another; just each one is in a different panelboard (or whatever you call the Gateway).

OK, so the one closest to the utility transformer is your service disconnect, and it's the only place where (a) N and G are bonded and (b) GECs can be landed. If either (a) or (b) is happening in the other redundant disconnects, then that's a code violation, and whoever did that needs to come back out and fix it for you.

In practice, it's not really a big deal if they are all next to each other. You'll get some neutral current on some of the enclosures of the 2nd and 3rd disconnect, which isn't allowed. But neutral current on the enclosure/conduit between the first disconnect and the meter (if separate) is often unavoidable and accepted.

On the second paragraph of your explanation, some edits:

In its basic configuration, the Tesla Energy Gateway 2 (TEG2) has no internal breakers on a shared busbar, so there is nothing to apply the 120% rule to. However, the TEG2 may be configured with the addition of an optional panelboard mounted within its casing. When this panelboard is used as a generation panel and only contains breakers for solar/ESS (no home loads), it would not comply with the 120% rule, but it does comply with one of its alternatives (the "sum of all breakers" rule).

Cheers, Wayne
 
Do you have the Tesla design documents for your current setup? That could help us all understand how it's currently wired.


I went through Sunrun, and they don't make it a habit to send customers their design docs because customers aren't supposed to be involved in the nitty gritty stuff. Har har.

Anyway, here's the crappy diagram had to come up with on my own. I actually need moar disconnects if I have to conform with that latest 060559 doc that Vines just shared. Crazy!

1661799536642.png
 
As already mentioned, the neutral/ground tie should only occur in once place. In your setup, it should be in the Main Panel and not in the TEG2.

My setup is similar with multiple disconnects between the meter and the TEG2. I have a 200A main breaker which feeds a 125A breaker which feeds a 200A blade disconnect which feeds the 125A TEG2 main breaker.
 
As already mentioned, the neutral/ground tie should only occur in once place. In your setup, it should be in the Main Panel and not in the TEG2.

My setup is similar with multiple disconnects between the meter and the TEG2. I have a 200A main breaker which feeds a 125A breaker which feeds a 200A blade disconnect which feeds the 125A TEG2 main breaker.

Yeah, I know the TEG2 shouldn't be neutral-bonded-to-ground but nobody's going to make that change hah. Even the county inspector that I had was a super stickler type. He told me my air conditioner condensers were too close to my fence and was about to red tag me unless I added a gate. And he said my Powerwall disconnects weren't line of sight... and we went back and forth about whether a locking device on an OCPD was sufficient (I lost by the way). And he said I didn't have a weep hole on one of my LBs. He even said one of the grounding rods wasn't pushed in far enough.

But when it came to the redundant neutral to ground stuff, he just said 'whatever' and left.
 
  • Funny
Reactions: jgleigh
Yeah, I know the TEG2 shouldn't be neutral-bonded-to-ground but nobody's going to make that change hah. Even the county inspector that I had was a super stickler type.
Who are you going to believe, the county inspector that was actually on site, or some random guy on the internet who can give you all the code citations?

: - )

Cheers, Wayne
 
Who are you going to believe, the county inspector that was actually on site, or some random guy on the internet who can give you all the code citations?

: - )

Cheers, Wayne


Yeah the same guy said the LOTO clips weren’t code compliant for the line of sight thing.

Btw @Vines you actually getting inspectors to give the ok on those LOTO’s?

 
Yeah the same guy said the LOTO clips weren’t code compliant for the line of sight thing.

Btw @Vines you actually getting inspectors to give the ok on those LOTO’s?

LOTO clips are not in my opinion a code-compliant means of locking out a breaker, since they are not a permanent means that requires a tool to remove. Permanent means typically are part of the breaker or clip to the breaker behind the deadfront.

Per 2017 NEC 705.22, all interconnected power supplies (both PV, ESS and others) need the equipment disconnect to be lockable. Practically, I almost never see AHJ requiring lockability for PV equipment disconnects. There is some disagreement among the code community whether 2017 NEC 706.3 removes that requirement for ESS since 705.22 is not on the list.

Looking forward to NEC 2020, it seems clear that per 706.15 equipment disconnects for ESS need to be lockable.
 
LOTO clips are not in my opinion a code-compliant means of locking out a breaker, since they are not a permanent means that requires a tool to remove. Permanent means typically are part of the breaker or clip to the breaker behind the deadfront.
Holeydonut called them LOTO clips but the link was to a post of yours about a clip from the breaker manufacturer.

Cheers, Wayne
 
  • Like
Reactions: Vines