I haven't got that information from Tesla yet. I have a 100 amp panel that is maxed out. The upgrade is probably to a larger 200 amp panel with more space. Also, I read in the Powerwall 2 specifications for the Energy Gateway that it has disconnect current of 200A. So, I may need a 200 amp panel regardless.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Powerwall 2: Installation
- Thread starter JeffreyY
- Start date
-
- Tags
- Powerwall Tesla Energy
Great system diagrams provided by jeep1979! It seems to me that it'd be hard to justify the added cost of a subpanel to back up only a subset of one's loads. For a "normal" home with EVs, could there be a good reason to go with anything other than a whole house backup?
Even with only a single PowerWall, in the event of a power outage (we had one that lasted 26 hours during a heavy snowstorm this past January) I'd prefer to have the flexibility of using any circuit, including our EV charging station. Of course, we'd have to be careful with our power and energy usage; I wouldn't want to be adding too many miles to an EV, particularly with no sunlight on the PV panels. This could mean going outside in the middle of the night to unplug an EV if the power goes out - the only downside I can think of.
I'm assuming that if, in an outage scenario, we were to turn on too many electric appliances simultaneously, the PowerWall would protect itself from overloading via an internal breaker or a similar mechanism - the same as if it's "empty".
Any thoughts?
Even with only a single PowerWall, in the event of a power outage (we had one that lasted 26 hours during a heavy snowstorm this past January) I'd prefer to have the flexibility of using any circuit, including our EV charging station. Of course, we'd have to be careful with our power and energy usage; I wouldn't want to be adding too many miles to an EV, particularly with no sunlight on the PV panels. This could mean going outside in the middle of the night to unplug an EV if the power goes out - the only downside I can think of.
I'm assuming that if, in an outage scenario, we were to turn on too many electric appliances simultaneously, the PowerWall would protect itself from overloading via an internal breaker or a similar mechanism - the same as if it's "empty".
Any thoughts?
My main panel has the utility meter integrated, so there is no good way put a transfer switch before the main panel. All the circuits I'm interested in backing up are already on a 125 amp sub-panel in the middle of the house, so it will be easy to put the transfer switch between the main panel and the sub-panel. However, in order to move the solar from the main panel to the backup side of the transfer switch, it will probably be necessary to put a 200 amp panel next to the 400 amp main panel for those extra breakers. In theory, any additional circuits I would want to back up from the main panel could also be moved over there. I see no reason to provide EV charging from a single PowerWall with my small 4.32kW solar. However, if you have a larger solar system, you could use the EV to soak up some of the excess production when the grid is down.
The issue of how to shed load automatically is a significant one and I don't think Tesla has addressed this at all. Since I was planning to install a backup generator when I built my house, I already installed a APC UTS10BI Universal Transfer Switch. It can be configured for UPS size and Generator size as well as required running minutes per hour for each connected circuit. It will automatically do round-robin load shedding on a circuit by circuit basis. Every connected circuit basically has a three-way switch inline so that the circuit can be fed by utility power, UPS power, or generator power as appropriate. Since I never got around to pulling the special permit and buying the NatGas generator, I am basically using it to provide UPS coverage to my structured wiring cabinet (cable modem, network switch, etc.) and my home office through the house wiring. Given the way the PowerWall works and the desire to have the solar charge the PowerWall during a grid outage, there is no good way to use this for load shedding with its simplistic internal logic. Also, to make it the most useful, it would have to know the SOC of the PowerWall to prioritize loads when the battery gets low. If this transfer switch could be upgraded with some communication capability, it could be really cool to integrate with the Tesla Gateway.
I'm debating this myself (critical loads sub-panel or whole home). This is where I'm at:
Critical loads sub-panel:
Pros:
1) Only important loads would be enabled
2) Would use less powerwall overall since a lot of power-wasting devices would be disconnected
Cons:
1) Additional cost for sub-panel
2) Wouldn't have the freedom to use whatever we want
Whole home:
Pros:
1) Cheaper (no sub-panel required)
2) Would have the freedom to use whatever we want
Cons:
1) Power-wasting devices would drain the batteries faster. Sure, I could turn disconnect these circuits but what if I'm not home, etc.
Lastly, one of my major concerns with the whole home solution is getting my wife to use electricity more carefully when on the batteries during a grid failure. I have a big clock that has indicators for our on-peak and off-peak TOU hours. However, a couple of weeks ago she was off on a weekday and I came home to her baking cake layers in both ovens, running the clothes dryer with the AC cranked down and half the lights in the house on.
Critical loads sub-panel:
Pros:
1) Only important loads would be enabled
2) Would use less powerwall overall since a lot of power-wasting devices would be disconnected
Cons:
1) Additional cost for sub-panel
2) Wouldn't have the freedom to use whatever we want
Whole home:
Pros:
1) Cheaper (no sub-panel required)
2) Would have the freedom to use whatever we want
Cons:
1) Power-wasting devices would drain the batteries faster. Sure, I could turn disconnect these circuits but what if I'm not home, etc.
Lastly, one of my major concerns with the whole home solution is getting my wife to use electricity more carefully when on the batteries during a grid failure. I have a big clock that has indicators for our on-peak and off-peak TOU hours. However, a couple of weeks ago she was off on a weekday and I came home to her baking cake layers in both ovens, running the clothes dryer with the AC cranked down and half the lights in the house on.
Very good point. I hadn't considered this, as our main panel also has the utility meter integrated. I'm not sure if it would be practical for us to relocate our utility meter.
We don't tend to have frequent outages, but in the event of an unusual, extended outage, I'd want to be able to use some of our PV production to charge EVs for local or emergency driving. Perhaps it would make more sense, in that scenario, to use a mobile connector and charge from a 120V outlet at 16A. That would be less convenient, but it would avoid inadvertently draining the PowerWall if an EV is connected to the EVSE when the power goes out. Even with a draw of only 16A at 120V, a fully charged PowerWall 2 would be drained in seven hours (until it can be recharged via PV).
So maybe it wouldn't be a big deal to move the most important circuits into a subpanel.
Yes, it would be nice to be able to prioritize the backed up circuits, etc. It would be good to see Tesla address load shedding in an intelligent manner. However, in our case, I don't think we don't have enough outages to make this a big concern. So I'd probably opt to keep the complexity and cost down, and not worry about automatic load shedding.
It would probably be better to use a low amperage 240V circuit if you want to charge your car from the PowerWall when the grid is down. Ideally you would be taking the power directly from the solar and they normally rely on the utility transformer to provide the 120V current because they don't connect to the neutral. The DC PowerWall with SolarEdge inverter requires an external auto-transformer to balance the neutral and supply 120V loads when the system is islanding.
I agree that automatic load shedding is not a major concern. However, it will be interesting to see how Tesla handles overload situations. Maybe send out a notification when it does overload and shut down, with a request to turn off higher power loads before restarting.
Just received a call from Solar City requesting a recent copy of my SCE bill to start the paperwork for my rebate. Since I require a main panel upgrade (100 amp to 200 amp), it appears it will be a 2-step process. Do the main panel upgrade first, then install the Tesla batteries.
Tesla says it was needed. In looking at the specs for the energy gateway, it has a disconnect current of 200A and an overcurrent breaker 100-200A. It is probably the reason. It may be that high because the energy switch can support up to ten Powerwall 2s.
My PowerWall estimate came in about a week ago. While I calculated I'd only need 2 Powerwalls to cover my critical loads, my service is distributed on my property (400A service comes into a meter on the far side and splits across the property to two different structures). They prefer installation as close to service as possible and have suggested four Powerwalls. I'm not terribly opposed to more storage, except that I really don't have room to mount these on the far side of my property. I barely have room for two on a service board.
I've been sitting on the bid while I decide. Going with four Powerwalls will start to complicate installation because of the mounting situation. I foresee some runaway costs that I'd prefer to avoid. On the other hand, having more storage would mean I could cover my entire usage over more of the year during an outage...
I'm undecided on whether to even try for SGIP.
I've been sitting on the bid while I decide. Going with four Powerwalls will start to complicate installation because of the mounting situation. I foresee some runaway costs that I'd prefer to avoid. On the other hand, having more storage would mean I could cover my entire usage over more of the year during an outage...
I'm undecided on whether to even try for SGIP.
Ulmo
Active Member
Don't take my advice: I too often muck things up when I give advice to single entities. But if you throw caution to the wind, here's some of my thoughts:
- What do they mean by "as close to service as possible"? You and they might interpret that differently. If you consider a looser interpretation, and push to the limits, this may open up a wider array of positioning options that you might have not already considered, and might become more natural, both from the perspective of access (and installation) and from the perspective of beauty and safety. That's what happened to me, at least in theory, when SolarCity site surveyor said "gee, this 4th possible location you told me is by far the best", which was an old shed concrete pad (shed long since rusted and removed) on the far side of the garage outside with plenty of access, but was also 20 to 35 feet away from the main service panel. None of the other locations were sufficient for more than one PowerWall without significant rework and/or compromises. In SolarCity's initial 4 PowerWall proposal, I did not see confirmation of this (they didn't have the site layout detail), but when I asked them to downsize the proposal to 2 PowerWalls, they did include the site layout detail and did confirm this location. Either way, I think the only sensible location with room for 4 or 2 PowerWalls was the further location, so the fact they included that location in the reduced footprint pretty much confirms they were thinking of that in the increased footprint. The details look hand-drawn in a CAD program from both satellite imagery (or jurisdiction data (county docs)) and the site survey they did; they're spending a lot of money on this proposal.
- Avoid runaway costs. There's often an easier solution that is not too bad for everyone once you know more of the facts. Not always!
- SGIP doesn't seem like that much of a headache considering SolarCity is helping us with the paperwork. It's our tax money; we might as well get some benefit from it. It takes 5 minutes to fill out the rather easy and simple paperwork and click send (it's all computerized -- I'm sure you've seen your email for this (DocuSign link (a different doc from the PowerWall proposal DocuSign link -- don't confuse the two)). (If you do apply for SGIP, do it very soon, since apparently there's a first come first served aspect to this with limited funding.) Also, don't forget to send them your latest PG&E bill pdf (from pge.com user portal). The SolarCity/Tesla rep can send you a demand email for the utility bill if you ask them, and you can reply to that, or you can submit to the email address listed in the SGIP docusign doc, at least in my particular case. (I did both.) The whole process was easier than pie so far. Not bad from my point of view.
Last edited:
K-MTG
Sunshade Captain of TMC
Are you gonna put your power inside or outside? Even though it will function in both places, I think this can be compared to keeping your car outside vs inside
Thanks.
What they mean by "close to service" is that they prefer it to be near the meter. My meter is mounted on a service panel out on the far side of a vineyard. I think your advice is good - I should work with them on panel placement. I don't think that location is ideal from a mounting perspective, and I think we can find other locations that will require minimal (or no) trenching.
Likewise with SGIP. I suppose I'll submit for it. I'm generally averse to applying for rebates that are intended to incentivize something that I would have purchased anyway. I can take the rebate and donate it to another sustainable energy cause, though.
