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

Santa Clara County retroactively Changing ESS Rules

This site may earn commission on affiliate links.
24" is the recent edit. I suspect some jurisdictions might polish it a bit and make it their own.

From what I know this design will be compliant now or January 1, 2023. It seems like a reasonable ask. You will need to p
PowerWall 2 Best Practice | Tesla Motors Club
rovide an interconnected heat detector still unless it already exists.
So San Jose would permit a stack of 3 powerwalls these days with no bollard?

All the best,

BG
 
So San Jose would permit a stack of 3 powerwalls these days with no bollard?

All the best,

BG
Let me put it this way, that is the system I would design in that city. I won't pretend to know the AHJ 100%, but as drawn, this is a code-compliant installation in my opinion. I do not have the authority to make a more definite statement.

The wild card is jurisdictions are changing their minds constantly. Next week it might be different. An inspector might have a bad day, or not be well informed.

Last month this stacking approval changed in County of San Mateo. I literally had to do the inspection myself and answer every question the fire marshall had. They tried to cancel the permit issued a year ago because "we are supposed to know the code and that 3' was required between the ESS as of July 1 2021." After an hour or so, they performed the final signoff because they agreed we followed the code to the letter. They stamped and approved the UL 9540A testing in that case and our plans which clearly showed the intended field conditions.

The month before, it was Menlo Fire suddenly disallowing the Tesla UL 9540A testing and therefore stacking, so this problem is getting worse. There are fewer bay area AHJ who accept the Tesla UL 9540A testing these days than those that deny it. I have tried to educate people as best I can but a specific few in the fire community won't approve Tesla 9540A testing on a technicality and also are convincing other officials the same.

There are 4 levels of testing in the UL 9540 Standard Cell - Module - Unit - Installation

If your Li-ion battery could pass the cell or module level testing, then no unit level testing is needed since no flammable gasses would have escaped. This ESS could be marked "For use in residential dwelling units" (AKA the Unicorn Mark) and would not be location restricted. You could put it in your kids bedroom right over your headboard or in your kitchen as far as the fire code is concerned.

The Santa Clara County fire service pushback is that "Cell and Module level data was not provided by Tesla" but unless you are aiming for a unicorn mark (or possibly indoor installations where the buildup and composition of flammable gasses is a concern) that data has little relevance. Nobody is installing raw Tesla powerwall cell or modules in their garage since the code requires UL 9540 listed systems.

The data that matters is the unit-level data, and that data shows that a powerwall that is pushed into thermal runaway in the test lab will not propagate to adjacent surfaces, nor be hot enough to ignite dry wood from heat flux. If the PW unit had some installation restrictions, such as not being allowed on an untreated wood siding wall or under a wooden overhang or eave, this would come up in the installation level testing and be noted in the report and installation manual as a restriction.

Other ESS manufacturers are primarily using iron phosphate, and are releasing media all about how safe their iron phosphate chemistry is, compared to the dangerous NMC Tesla chemistry. This plays right into the fears of the fire service, and so far Tesla hasn't been very successful in convincing anyone otherwise.
 
If your Li-ion battery could pass the cell or module level testing, then no unit level testing is needed since no flammable gasses would have escaped. This ESS could be marked "For use in residential dwelling units" (AKA the Unicorn Mark) and would not be location restricted.
Per UL9540A, that mark requires passing the cell level testing. Module level performance means that thermal runaway in one cell does not propagate to thermal runaway in neighboring cells, I believe.

Do you happen to know whether the LFP storage options include any that pass at the module level rather than at the unit level?

Do you know why Tesla hasn't registered their UL 9540A test results in the UL database? Login Enphase has their results registered, although the free ul.com account doesn't seem to let me download them.

Cheers, Wayne
 
Per UL9540A, that mark requires passing the cell level testing. Module level performance means that thermal runaway in one cell does not propagate to thermal runaway in neighboring cells, I believe.

Do you happen to know whether the LFP storage options include any that pass at the module level rather than at the unit level?

Do you know why Tesla hasn't registered their UL 9540A test results in the UL database? Login Enphase has their results registered, although the free ul.com account doesn't seem to let me download them.

Cheers, Wayne
Thanks for the clarification re Cell and modules, you are correct.

Unfortunately I can't speak to manufacturers specific testing, I have seen Tesla's but not others. I also cannot speak to why Tesla hasn't uploaded results.
 
24" is the recent edit. I suspect some jurisdictions might polish it a bit and make it their own.

From what I know this design will be compliant now or January 1, 2023. It seems like a reasonable ask. You will need to provide an interconnected heat detector still unless it already exists.
Ok, so just making sure I've got the latest rev correct...so danger zone in the new code would be up to 36" of the drive path (defined by the garage door line) on either side...extending 24" off the back wall. I should be able to make this happen with a 3xPW stack.

Semi-related question, if going solar...with 3x PW...does that mean I would have one PW+ and two PW2's? Can you stack a PW+ with a PW2?

Re: the heat interconnected detector...have you seen a commercial model that satisfies this need for both interconnected + unconditioned?
 
Let me put it this way, that is the system I would design in that city. I won't pretend to know the AHJ 100%, but as drawn, this is a code-compliant installation in my opinion. I do not have the authority to make a more definite statement.

The wild card is jurisdictions are changing their minds constantly. Next week it might be different. An inspector might have a bad day, or not be well informed.

Last month this stacking approval changed in County of San Mateo. I literally had to do the inspection myself and answer every question the fire marshall had. They tried to cancel the permit issued a year ago because "we are supposed to know the code and that 3' was required between the ESS as of July 1 2021." After an hour or so, they performed the final signoff because they agreed we followed the code to the letter. They stamped and approved the UL 9540A testing in that case and our plans which clearly showed the intended field conditions.

The month before, it was Menlo Fire suddenly disallowing the Tesla UL 9540A testing and therefore stacking, so this problem is getting worse. There are fewer bay area AHJ who accept the Tesla UL 9540A testing these days than those that deny it. I have tried to educate people as best I can but a specific few in the fire community won't approve Tesla 9540A testing on a technicality and also are convincing other officials the same.

There are 4 levels of testing in the UL 9540 Standard Cell - Module - Unit - Installation

If your Li-ion battery could pass the cell or module level testing, then no unit level testing is needed since no flammable gasses would have escaped. This ESS could be marked "For use in residential dwelling units" (AKA the Unicorn Mark) and would not be location restricted. You could put it in your kids bedroom right over your headboard or in your kitchen as far as the fire code is concerned.

The Santa Clara County fire service pushback is that "Cell and Module level data was not provided by Tesla" but unless you are aiming for a unicorn mark (or possibly indoor installations where the buildup and composition of flammable gasses is a concern) that data has little relevance. Nobody is installing raw Tesla powerwall cell or modules in their garage since the code requires UL 9540 listed systems.

The data that matters is the unit-level data, and that data shows that a powerwall that is pushed into thermal runaway in the test lab will not propagate to adjacent surfaces, nor be hot enough to ignite dry wood from heat flux. If the PW unit had some installation restrictions, such as not being allowed on an untreated wood siding wall or under a wooden overhang or eave, this would come up in the installation level testing and be noted in the report and installation manual as a restriction.

Other ESS manufacturers are primarily using iron phosphate, and are releasing media all about how safe their iron phosphate chemistry is, compared to the dangerous NMC Tesla chemistry. This plays right into the fears of the fire service, and so far Tesla hasn't been very successful in convincing anyone otherwise.

Been having some good dialogue with SJ City (find them very nice to work with). By no means is this signed off on, but I believe below is updated perspective on my situation that will comply with the current SJ code, until 12/31/2022. So as long as the permit is pulled by Tesla before then, it should be OK (other issue is Tesla design team keeps trying to put the PW outside).

Nice part about this design, is that even if we slip into the new code per @Vines diagram, I believe I am still good for 3xPW because the entire stack is (barely) outside of the drivepath as defined by garage opening and it's >24" off the back wall (if the new code was 36" off the back wall I could even manage too).

Anyways, hoping that we will go with the current code and move this along faster, to ensure we get on NEM2.

Yride

1659826895872.png
 
  • Like
Reactions: Patrick66 and Vines
Ok, so just making sure I've got the latest rev correct...so danger zone in the new code would be up to 36" of the drive path (defined by the garage door line) on either side...extending 24" off the back wall. I should be able to make this happen with a 3xPW stack.

Semi-related question, if going solar...with 3x PW...does that mean I would have one PW+ and two PW2's? Can you stack a PW+ with a PW2?

Re: the heat interconnected detector...have you seen a commercial model that satisfies this need for both interconnected + unconditioned?
Answering my own question...kinda...didnt see anything written, but here is a pic on Tesla.com. So it appears you would stack 1 PW+ and 2PW...

1659926842392.png
 
Answering my own question...kinda...didnt see anything written, but here is a pic on Tesla.com. So it appears you would stack 1 PW+ and 2PW...

View attachment 837953
Yea, the PW portion of the PW+ is just a standard PW2. The “+” is the smaller box above which combines the functionality of the gateway and an inverter into a single unit. That would be paired with a meter ring disconnect to isolate the house from the grid, or another gateway where the meter disconnect isn’t allowed.
 
Answering my own question...kinda...didnt see anything written, but here is a pic on Tesla.com. So it appears you would stack 1 PW+ and 2PW...

View attachment 837953
Regardless of what that marketing material shows, I think this configuration would be disallowed due to a violation of NEC 110.26.A.3 required working space. This picture also does not align with what I have heard of the Powerwall+restrictions. In fact it doesn't even seem to adhere to the clearances required between stacked units to dissipate the heat.
 
Yea, the PW portion of the PW+ is just a standard PW2. The “+” is the smaller box above which combines the functionality of the gateway and an inverter into a single unit. That would be paired with a meter ring disconnect to isolate the house from the grid, or another gateway where the meter disconnect isn’t allowed.
If I upgrade to a 400A service, my understanding is a GW can only cover 200A of load, so I’d have to have a least one 200A sub-panel behind the main. If I had two 200A subs behind the main, and wanted whole home backup…that would mean 2x GW, right? And if so, would that mean two PW+ and one PW2? Can you “stack” two PW+?
 
Regardless of what that marketing material shows, I think this configuration would be disallowed due to a violation of NEC 110.26.A.3 required working space. This picture also does not align with what I have heard of the Powerwall+restrictions. In fact it doesn't even seem to adhere to the clearances required between stacked units to dissipate the heat.
Could you elaborate on what you have heard about PW+ restrictions? And when you say adhere to clearances…you mean laterally?

For workspace, clearly not to scale..but looks like 6ft above, right? And no wall or electrical in front.
 
And when you say adhere to clearances…you mean laterally?
Yes, in the illustration the exhaust from the stack on one side would be blowing into the cooling intake on the adjacent PWs, not good. [Edit: Plus the wiring compartment on the righthand stack in the illustration would be inaccessible.] There's a specified minimum clear space between non-stacked PWs on the same wall.

For workspace, clearly not to scale..but looks like 6ft above, right? And no wall or electrical in front.
If the inverter section is "likely to require examination, adjustment, servicing, or maintenance while energized" (which seems quite plausible), then the issue is that other related equipment above or below it can only stick out 6" past the face of the inverter enclosure. So you could stack 1 PW in front of a PW+ (I think), certainly not two. However, if an extension enclosure were available or fabricated, to simply expand the depth of the inverter enclosure, then you could stack 2.

Cheers, Wayne
 
  • Like
Reactions: YRide
If I upgrade to a 400A service, my understanding is a GW can only cover 200A of load, so I’d have to have a least one 200A sub-panel behind the main. If I had two 200A subs behind the main, and wanted whole home backup…that would mean 2x GW, right?
2x GW is the simplest solution, but with a tricky transfer switch setup, you could get away with just one GW. The idea is during grid operation, one 200A panel gets its power through the GW, the other doesn't; but when off grid, the second 200A panel switches to getting its power from behind the GW. Not a solution that Tesla would support, I believe.

Can you “stack” two PW+?
Definitely not, as the inverter section needs access.

Cheers, Wayne
 
  • Like
Reactions: YRide
2x GW is the simplest solution, but with a tricky transfer switch setup, you could get away with just one GW. The idea is during grid operation, one 200A panel gets its power through the GW, the other doesn't; but when off grid, the second 200A panel switches to getting its power from behind the GW. Not a solution that Tesla would support, I believe.


Definitely not, as the inverter section needs access.

Cheers, Wayne
I think it will be sufficient to have only one 200A backed up. Not looking for a complex Xfer switch…though if I could have one fit a *possible* temp generator that would be nice. Not expecting to lose power too much, more for TOU mgmt, especially now that one can dump onto grid (at peak rates) from wha is charged via solar.

So 2x GW is what it sounds like. But keep it at one sub 200A off the wall, the rest can come off main.

I mean if you lose power, you’re not gonna need 2x wall chargers live…
 
Then you'd only need 1 GW.

Without much complexity at all, you could add a manual transfer arrangement to allow the other panel to also be powered during an extended grid outage.


Cheers, Wayne
Ok…so then one GW (PW+), to the sole 200A sub, all three PW’s (PW+ and 2x PW2) on that single gateway. All 3 can export to grid. Most daily stuff is backed up via 200A sub.

And then I could put the manual Xfer onto the 400A main…and everything behind it is backed up, assuming I gave a generator that could take 400A load?
 
And then I could put the manual Xfer onto the 400A main…and everything behind it is backed up, assuming I gave a generator that could take 400A load?
My comments have not included consideration of any mechanical generator. I was referring to temporarily powering the 200A panel that is not behind the GW from behind the GW, i.e. from the PWs and PV. You'd need to manually ensure you don't turn on more equipment than your inverters can support, but that's true even with a single 200A panel, as an upper bound on the power output from PW+ and 2 X PW is roughly the sum of the breaker sizes, 50A + 30A + 30A = 110A.

Throwing a mechanical generator into the mix complicates things quite a bit, and I've not considered the options for that. Because hydrocarbons.

Cheers, Wayne
 
My comments have not included consideration of any mechanical generator. I was referring to temporarily powering the 200A panel that is not behind the GW from behind the GW, i.e. from the PWs and PV. You'd need to manually ensure you don't turn on more equipment than your inverters can support, but that's true even with a single 200A panel, as an upper bound on the power output from PW+ and 2 X PW is roughly the sum of the breaker sizes, 50A + 30A + 30A = 110A.

Throwing a mechanical generator into the mix complicates things quite a bit, and I've not considered the options for that. Because hydrocarbons.

Cheers, Wayne
PW+ is 50A, and not 30A too?
 
PW+ is 50A, and not 30A too?
Yes, PW+ connects with a 50A breaker, and its controller limits the joint inverter current to 40A while on grid to facilitate its use in a 200A main panel under the 120% rule, but allows up to 50A when off grid.

So off grid, up to 110A when the sun is up and the PV inverter can contribute, or 90A when the sun is down and you just have the (3) PW inverters (one of which is part of the PW+).

I'm avoiding the details of peak vs continuous power, etc, to keep this simple, and just going by breaker size.

Cheers, Wayne