CPUC NEM 3.0 discussion

[holeydonut]

Ohmman, what you are writing is perfectly clear and I agree with you. In fact, if the CPUC adopts NEM 3, that's what I plan to do when my NEM contract expires in two years. I'll put an export limiter on my inverter to stop sending my surplus PV output to PG&E. Instead, I'll add batteries to my system and charge them with my surplus PV production. I'll select a new rate plan, probably E-1, to replace my expired NEM contract. No more $8/kW/installed capacity/month NEM 3 charge. Since my PV array and batteries will provide most of the electricity I need my PG&E bills won't amount to much more than the $10/month cover charge that everyone pays. And with my battery backup I won't lose power every time PG&E sets fire to a nearby forest.

Hi Sky Blue; welcome to TMC!

I wish things were as simple as the layout you propose, but it won't be so easy unless vehicle-to-home bidirectional power flow becomes a reality. There aren't many reasonably affordable solutions in place today that can house daily energy production in the summer months without exporting to the grid.

For example, I have a very small solar array (6.7 kWp AC) compared to the amount of Powerwall energy capacity 40.5 kWh (3x Powerwall 2). Normally, Tesla would only recommend 1 or 2 Powerwalls for a small solar system like mine. But even with this small-ish solar system, over-sized battery, and my AC's running hard, I was still exporting about half of my daily solar production under NEM 2.0 in the Summer months.

But get this, my ratio of large storage to tiny production is so abnormal that PG&E fought me on my system. They felt I was either lying to them about the size of my solar array, or being conned by Sunrun. PG&E sent two "independent assessors" to my house to take pictures of my batteries and solar generation panel to prove I hadn't gone off the script defined on my permits.

Anyway, my point to have ESS sized so large as to avoid exporting to the grid requires an obscene amount of ESS. By my estimation I'd need another 3x Powerewalls (40 kWh) of capacity to completely avoid exporting. At that point, I think Vehicle to Home (V2H) solutions where an EV acts as the ESS is a key enabler to your plan.

Also, keep in mind my small solar array is sized to 100% of my annual consumption excluding EV charging. I have natural gas for heating (furnace and water). If you actually want to go "completely zero carbon" and use solar to generate electricity to generate home heating in the Wintertime, things become completely untenable. NorCal just doesn't generate enough daily solar in the Winter months to keep your house warm. You'll still need things like a wood fireplace or some wonky wind-setup to stay warm without relying on PG&E as your primary source when Solar isn't generating enough.

 

[gene]

Thursday's demonstration at the Los Angeles CPUC brought 1800 people, the one in San Francisco was 1200 people strong.

 

[ohmman]

I wish things were as simple as the layout you propose, but it won't be so easy unless vehicle-to-home bidirectional power flow becomes a reality. There aren't many reasonably affordable solutions in place today that can house daily energy production in the summer months without exporting to the grid.

So when I'm off-grid on my PWs, they curtail my excess solar once they are fully charged. Why can't they do the same thing in a non-export situation?

 

[Vines]

So when I'm off-grid on my PWs, they curtail my excess solar once they are fully charged. Why can't they do the same thing in a non-export situation?

In the microgrid situation you described most PV systems (besides Powerwall +) will get a frequency shift signal from the Gateway to your microgrid. When on grid, there is no possibility to shift the grid frequency in the same way to shut down any solar system using the curtailment parameters of Rule 21 SA.

Certain inverters like the PW+ and Solar Edge configured with CT's designed for non-export situations curtail production internally with the inverter, by reading a set of CT meters around the main incoming utility feed and never pushing more PV than the home load will consume right now.

 

[h2ofun]

In the microgrid situation you described most PV systems (besides Powerwall +) will get a frequency shift signal from the Gateway to your microgrid. When on grid, there is no possibility to shift the grid frequency in the same way to shut down any solar system using the curtailment parameters of Rule 21 SA.

Certain inverters like the PW+ and Solar Edge configured with CT's designed for non-export situations curtail production internally with the inverter, by reading a set of CT meters around the main incoming utility feed and never pushing more PV than the home load will consume right now.

So, dumb question. When the grid is off, the gateways seems to have the smarts to be able to shut down the solar if the batteries are full and no need for solar. So, why could the fw not be changed to do this non export 100% of the time?

 

[holeydonut]

So when I'm off-grid on my PWs, they curtail my excess solar once they are fully charged. Why can't they do the same thing in a non-export situation?

If a homeowner curtails solar production in the summertime, they may solve a "daily NEM cycle", but that means they'd be hosed on their annual cycle. Imagine a home needing 12,000 kWh per year (excluding EV charging). In the NEM 1.0 and NEM 2.0 days, it was somewhat reasonable to put enough solar on a normal rooftop (I am excluding outlier monster sized arrays like H2ofun) for a homeowner to generate 12,000 kWh in a given year. Maybe 8,000 kWh would be generated in the Spring/Summer, but only 4,000 kWh in the Fall/Winter.

Since the energy generation and use wouldn't net out each month, it required an annual true up where Summertime banking was cashed in during the Winter.

I think a home that still relies on natural gas for cooking and heat may care about this issue less simply because that home's electricity use also decreases in the winter time. But homes that converted to all electric because they were trying to be good Californians would get hosed hard if they simply stopped generating in the Summer rather than bank useful NEM credits.

 

[gene]

But homes that converted to all electric because they were trying to be good Californians would get hosed hard if they simply stopped generating in the Summer rather than bank useful NEM credits.

Under NEM3 the true-up would be monthly, the summer over production will no longer offset winter use of power from the grid.

 

[holeydonut]

Under NEM3 the true-up would be monthly, the summer over production will no longer offset winter use of power from the grid.

That... really sucks for homes that have electric heating. Since homes will be on time of use plans (excluding Medical Baseline and CARE), they'll get absolutely destroyed by winter time heating bills that don't conform well to the normal peak/non-peak curve. RIP.

Time to put in those wood fireplace inserts...

Edit, so yeah, even if you do export to the grid in the summertime, you're probably only getting $0.00 to $0.03 per kWh under NEM 3.0. So you're basically losing nothing by just turning off your solar array once your ESS is full and try to avoid that $8 per month per kW fee. NEM 3.0 is total trash.

 

[wwhitney]

So, dumb question. When the grid is off, the gateways seems to have the smarts to be able to shut down the solar if the batteries are full and no need for solar. So, why could the fw not be changed to do this non export 100% of the time?

Vines addressed this in his answer, but let me expand on it:

It's a question of how the Gateway/Poweralls can communicate with the solar inverters. If there's an out-of-band (separate communications channel) way for them to speak to each other, that both support, then absolutely that would be possible. I believe that's the case with the PW+, for example, where the Gateway and the solar inverter are integrated in the same (upper) box.

Failing that, the only way for the Gateway to communicate with the solar inverters is in band, via the AC waveform. When the whole system is on the grid, the Gateway/Powerwalls have no control over that, the Powerwalls are just grid-following inverters. While when the Gateway disconnects from the grid, the Powerwalls now run in grid forming mode, and have control over the AC waveform. They can force the solar inverters to shutoff by shifting the frequency of the AC waveform they are generating.

Cheers, Wayne

 

[RKCRLR]

Vines addressed this in his answer, but let me expand on it:

It's a question of how the Gateway/Poweralls can communicate with the solar inverters. If there's an out-of-band (separate communications channel) way for them to speak to each other, that both support, then absolutely that would be possible. I believe that's the case with the PW+, for example, where the Gateway and the solar inverter are integrated in the same (upper) box.

Failing that, the only way for the Gateway to communicate with the solar inverters is in band, via the AC waveform. When the whole system is on the grid, the Gateway/Powerwalls have no control over that, the Powerwalls are just grid-following inverters. While when the Gateway disconnects from the grid, the Powerwalls now run in grid forming mode, and have control over the AC waveform. They can force the solar inverters to shutoff by shifting the frequency of the AC waveform they are generating.

Cheers, Wayne

What they should do (assuming NEM 3 goes through similar to what is proposed) is allow a homeowner the option of selecting between no solar connection fee with no compensation for excess generation or connection fee with generation compensation. That would all ow an individual to select the best option for their particular situation. That's sort of a win-win.

But the best solution is a standard connection fee for everyone. Then figure out how to incentivize solar.

 

[RKCRLR]

Vines addressed this in his answer, but let me expand on it:

It's a question of how the Gateway/Poweralls can communicate with the solar inverters. If there's an out-of-band (separate communications channel) way for them to speak to each other, that both support, then absolutely that would be possible. I believe that's the case with the PW+, for example, where the Gateway and the solar inverter are integrated in the same (upper) box.

Failing that, the only way for the Gateway to communicate with the solar inverters is in band, via the AC waveform. When the whole system is on the grid, the Gateway/Powerwalls have no control over that, the Powerwalls are just grid-following inverters. While when the Gateway disconnects from the grid, the Powerwalls now run in grid forming mode, and have control over the AC waveform. They can force the solar inverters to shutoff by shifting the frequency of the AC waveform they are generating.

Cheers, Wayne

But if you have solar plus Powerwalls couldn't you electronically go off-grid when the Powerwalls are full and run off them and solar, then go back on grid when the Powerwalls get to a certain point?

 

[ohmman]

In the microgrid situation you described most PV systems (besides Powerwall +) will get a frequency shift signal from the Gateway to your microgrid. When on grid, there is no possibility to shift the grid frequency in the same way to shut down any solar system using the curtailment parameters of Rule 21 SA.

Thanks, I was thinking about the capacity of the system to do a thing, but not the facility with which it did that thing. Much clearer now.

If there's an out-of-band (separate communications channel) way for them to speak to each other, that both support, then absolutely that would be possible. I believe that's the case with the PW+, for example, where the Gateway and the solar inverter are integrated in the same (upper) box.

I'm guessing that if NEM 3.0 is adopted as proposed, it will create a market for add-on hardware that provides out-of-band communication so that this is possible. Interesting stuff, though. Thanks both of you for the responses.

Edit, so yeah, even if you do export to the grid in the summertime, you're probably only getting $0.00 to $0.03 per kWh under NEM 3.0. So you're basically losing nothing by just turning off your solar array once your ESS is full and try to avoid that $8 per month per kW fee. NEM 3.0 is total trash.

Exactly what I was trying to say above. I'm not amenable to wholesaling my electricity back to the grid so they can arbitrage it to my neighbors, especially when they're charging me fixed connection fees for my PV. While I haven't done the math to see how it works out, it's probably a very close call and I am not going to incentivize them to continue scapegoating solar customers for their IOUs.

 

[wwhitney]

But if you have solar plus Powerwalls couldn't you electronically go off-grid when the Powerwalls are full and run off them and solar, then go back on grid when the Powerwalls get to a certain point? The main problem I see with this is when the solar production is exceeding home consumption and maximum Powerwall charge rate but it seems like that could be managed.

Yes if the software supported that, it would be possible, there's no technical barrier. But for any scheme that involves frequent switching between off grid and on grid, it would be nice to have the transition seamless, which I don't think is always the case with the current Gateway. Although I've not tested the "go-off-grid" button, perhaps they've added some functionality to make that transition seamless when it is planned, rather than due to a grid outage?

[BTW, are the transitions from off-grid to on-grid ever seamless? With the topology of the GW and PW inverters, I don't think it's possible for the PW inverters to presynchronize their AC waveform with the grid before switching back. So at present do you necessarily get a partial cycle of zero power?]

Another thing the PWs could conceivably support with the proper software would be a "run off grid normally mode" where the GW only switches to on-grid when the PW batteries are below a certain threshold, then switches back to off-grid when they are sufficiently charged. I guess with a little embedded computer to poll the PW battery charge status, and a transfer switch that could be controlled by the computer, you could implement this now: the PV and PWs are always a little microgrid, and the critical loads are switched from the utility grid to the microgrid according to battery state. I don't think switching could be made seamless with a bolt-on solution like that.

Cheers, Wayne

 

[yblaser]

What they should do (assuming NEM 3 goes through similar to what is proposed) is allow a homeowner the option of selecting between no solar connection fee with no compensation for excess generation or connection fee with generation compensation. That would all ow an individual to select the best option for their particular situation. That's sort of a win-win.

Well that would be nice I don't think there is anyway that would happen. I don't think anyone would pay the connection fee as even exporting 100% of your solar would not make up for the fee. At my location 1 kW of south facing solar panels will generate 1660 kWh of electricity and with estimated $0.05/kWh credit under NEM 3.0 yields $83/year credit compared to the fee of $96/year.

The $8/kW fee is really going after behind the meter generation and consumption.

But the best solution is a standard connection fee for everyone. Then figure out how to incentivize solar.

 

[holeydonut]

Exactly what I was trying to say above. I'm not amenable to wholesaling my electricity back to the grid so they can arbitrage it to my neighbors, especially when they're charging me fixed connection fees for my PV. While I haven't done the math to see how it works out, it's probably a very close call and I am not going to incentivize them to continue scapegoating solar customers for their IOUs.

Yeah, I feel like the only way the 1,000,000+ residential solar generators can get their point across that their combined investments is actually worth more than the $0.03 per kWH avoided cost calculator (and stupid fixed costs) is to wait until the hottest day in Summer. And they collectively turn off their solar arrays at 4pm.

Of course the SEIA and CALSSA cannot coordinate such an event since it's probably illegal. And whatever grassroots way homeowners could plan this would also give the IOUs and CAISO enough lead time to just get standby peaker plant natural gas generation and some pricey imports teed up from neighboring states. They'll pass this emergency scramble to procure energy back to rate payers anyway.

 

[holeydonut]

Thursday's demonstration at the Los Angeles CPUC brought 1800 people, the one in San Francisco was 1200 people strong.

I'm kind of disappointed in the SF Bay Area news. None of the local outlets in the Bay Area picked up this event for coverage. That means collectively, Fox, ABC, NBC, and CBS all felt this wasn't worth their time to even speak to on the news telecasts. They had some correspondents tweet about it, but it didn't have anywhere near the normal news penetration.

On the other-hand, Fox 2 up here did a segment about some dude fixing potholes in front of his house without a permit.

 

[skepticcyclist]

Thursday's demonstration at the Los Angeles CPUC brought 1800 people, the one in San Francisco was 1200 people strong.
View attachment 755630

Even more interesting is the fact that NEM 3.0 won't even affect LA. They're under LADWP.

 

[ohmman]

Yeah, I feel like the only way the 1,000,000+ residential solar generators can get their point across that their combined investments is actually worth more than the $0.03 per kWH avoided cost calculator (and stupid fixed costs) is to wait until the hottest day in Summer. And they collectively turn off their solar arrays at 4pm.

Of course the SEIA and CALSSA cannot coordinate such an event since it's probably illegal. And whatever grassroots way homeowners could plan this would also give the IOUs and CAISO enough lead time to just get standby peaker plant natural gas generation and some pricey imports teed up from neighboring states. They'll pass this emergency scramble to procure energy back to rate payers anyway.

The other problem with this is that it gives the utilities a convenient villain - those “wealthy solar brats”. It would have the possibility of publicly validating the policy, especially if it causes brownouts.

 

[miimura]

Yes if the software supported that, it would be possible, there's no technical barrier. But for any scheme that involves frequent switching between off grid and on grid, it would be nice to have the transition seamless, which I don't think is always the case with the current Gateway. Although I've not tested the "go-off-grid" button, perhaps they've added some functionality to make that transition seamless when it is planned, rather than due to a grid outage?

[BTW, are the transitions from off-grid to on-grid ever seamless? With the topology of the GW and PW inverters, I don't think it's possible for the PW inverters to presynchronize their AC waveform with the grid before switching back. So at present do you necessarily get a partial cycle of zero power?]

Another thing the PWs could conceivably support with the proper software would be a "run off grid normally mode" where the GW only switches to on-grid when the PW batteries are below a certain threshold, then switches back to off-grid when they are sufficiently charged. I guess with a little embedded computer to poll the PW battery charge status, and a transfer switch that could be controlled by the computer, you could implement this now: the PV and PWs are always a little microgrid, and the critical loads are switched from the utility grid to the microgrid according to battery state. I don't think switching could be made seamless with a bolt-on solution like that.

Cheers, Wayne

Wayne,

I believe the transition back to on-grid is supposed to be seamless. The Gateway can sense the waveform on the grid input and when I have done tests by manually switching my main breaker, the time to close the Gateway switch after grid restoration has always been significant - like 30 seconds to a minute before re-engaging. I assume that is to allow time to re-synchronize the Powerwall output to the grid but I haven't hooked up a scope or anything to verify it.

 

[wwhitney]

I believe the transition back to on-grid is supposed to be seamless. The Gateway can sense the waveform on the grid input and when I have done tests by manually switching my main breaker, the time to close the Gateway switch after grid restoration has always been significant - like 30 seconds to a minute before re-engaging. I assume that is to allow time to re-synchronize the Powerwall output to the grid but I haven't hooked up a scope or anything to verify it.

Ah, OK, then the GW must sense the phase angle and communicate that to the PWs, as the inverters in the PWs can't see the grid. It would be interesting to see oscilloscope traces of the AC voltage waveform during various transitions.

Cheers, Wayne