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

App Config Option

Customize

  • Backup-Only

  • Self-Powered

  • Advanced - Cost Saving

  • Advanced - Balanced


Results are only viewable after voting.
This site may earn commission on affiliate links.
I'm in the UK and have cheap electricity overnight. until about 3 weeks ago I was running Advanced - Cost Saving to fill the battery with cheap electricity to be used during the day. However, the algorithm has failed me recently by only filling up based on what appears to be the last few days worth of solar generation which left me short of stored electricity a couple of times and using expensive electricity. Since then I have changed to Self-Powered and have set up some automation on a Raspberry Pi which sets the backup percentage to be the percent I want to have in the battery at the end of cheap electricity and then sets the backup percentage back to 5% at the point cheap electricity ends. I'm currently setting that percentage based on looking at the next days weather forecast and "making a call". Ideally I'd work out a way to automate it but so far it's based on manual intervention. I'm expecting in a few weeks time we will have enough solar generation to just leave the unit on Self-Powered.
 
The size of my solar system, 1x PW and typical household consumption makes Self-Powered make the most sense to me (and most of the time at 0% Reserve, with Storm Watch disabled). I have Net Metering and TOU in my area (which presumably is the same as the OP's), so it definitely makes sense to bank some of the solar to keep me off the grid during the peak hours (as opposed to Backup Only). Most of the year I can run sundown to sunup without touching the grid (though sometimes I'm within a few % left when that happens), so any Reserve would only cause me to draw from the grid more without any real benefit, though I am trying to work out how to programmatically predict the rare (for me) really bad production days where I may not get to fully-charge my PW, or an intense overnight storm, to keep some Reserve on those days in case there is a storm-related outage. This also generally makes Storm Watch useless to me, since it would only cause me to draw power from the grid when I generally don't need to do that to have a fully-charged battery by sundown.

And since there's generally enough battery capacity to last overnight, I don't see any benefit to either of the Advanced modes, in the end I believe they'd only cause me to incur more Non-Bypassable Charges without earning me any extra credit (I already have much more NEM credit than I need, so there's no value in any more of those).

But the "right" answer depends on a lot of factors, such as how billing works, how big the solar system is, and how much overnight household consumption compares to the storage capacity of the PW(s).
 
We have 15.4 KW of solar panels and 4 PowerWalls, currently on a $.091/KWh electricity plan. We're using self-powered mode with 20% reserve percentage. Excluding charging of our Tesla S & X, the rest of the house has been 61% self-powered in 2020. Including EV charging, we've reduced our electricity use by 55% and with the low electric rate we've reduced our electricity bill by 65%.

When our current electricity plan expires in a few months, we'll likely shift to a "free nights" plan that charges almost 2X for electricity during the day and provides free electricity from 9PM to 9AM. Based on the estimates for 2020 so far (and 2019 smart meter data), it appears we could cut our electric bill by an additional 50% by changing the TEG setting to time-based control.

When we planned our system, we made projections on how long it would take for us to get an ROI on the solar/PowerWall investment, assuming we'd cut our electric bills by 50%. If we can get the savings to 75%, we'll do even better.

For homes with smart meters, it should be possible to download actual usage data (in Texas, we can get it the power usage in 15 minute intervals) and make a pretty good estimate of how much the various electricity plans should cost, based on actual data.
 
With simple net metering and no Time of Use differentials in cost of electricity, it really doesn't matter. I set my PWs at 50% just to be sure they are working properly. It also gives me a good idea as to how long I can last without the grid.
 
I am still trying to decide. Our install should be complete in the next few days. We are in a PG&E service area with a Time of Use plan. Rate is 0.40 in summer from 4-9PM M-F, 0.20 otherwise. During the rest of the year the rate is 0.26 peak, 0.22 off-peak. 2 Powerwalls. We have 2 Teslas, but don't drive a lot. Any suggestions?
 
I use TBC-Balanced. I am in PG&E territory and have been pushed from the EV-A rate to the EV2-A rate schedule. This means that my generation has mostly moved from Part-Peak to Off-Peak. I also have a relatively small solar system compared to my two Powerwall system. I have estimated that I should save about $700 per year due to Powerwall time shifting. The system did not work very well in Cost Saving mode on EV-A because the generation was during Part-Peak. Now, on EV2-A, the behavior should be more similar between Balanced and Cost Saving since the morning generation is during Off-Peak.

Since the rate schedule change in November, I have filled the Powerwalls to 100% and exported solar only twice, and the export was inconsequential at less than 1kWh each time. As the weather improves and solar generation improves, I should be able to limit my grid draws to only the Off-Peak period. The Powerwalls have enough storage to completely power the house from 3pm to midnight.
 
I am still trying to decide. Our install should be complete in the next few days. We are in a PG&E service area with a Time of Use plan. Rate is 0.40 in summer from 4-9PM M-F, 0.20 otherwise. During the rest of the year the rate is 0.26 peak, 0.22 off-peak. 2 Powerwalls. We have 2 Teslas, but don't drive a lot. Any suggestions?
Set it so you charge the PWs before 4pm, and use their energy starting at 4. Set the reserve very low; start at 20%, and see if you can recharge them fully before 4pm the next day. Start recharging the Teslas at 9pm.
 
Net metering 1.0 and on a standard tiered rate plan (have not been forced into time of use). In my situation, the tiered rate plan works best for me because the size of my solar system means for I just about break even on the year true up, even with my model 3.

The less electricity I use "overall" from the grid, the cheaper it is for me as in my plan the NBCs are fairly low (most can be offset by solar production).

I use self powered, and normally set my reserve to 45% for my 2 powerwalls, as that allows me to run almost 100% off the grid starting in about mid february, on most sunny days. If I see its going to rain, I move my reserve higher ( to a higher number, like 50 or 60%). My goal is to ensure that I will ALWAYS be able to run off my powerwalls in a loss of power situation, until the sun comes up and starts filling them again. If I see there is going to be less sun, I make sure there is more reserve in my PWs to cover more time.

I dont want to go without power in my home ever again, and wont, if I can help it (so always have "some reserve"

Since I dont have TOU, self powered works best for me.
 
8.55kw PV, 1x Powerwall, 100% backup. No ToU, full net metering, electricity is cheap here. I expect one if not more than one of those to change over the next 10 years. ROI for me today was comparing to home standby generators for outages which the PW handily won.
 
Advanced - Cost Saving with 2X Powerwall and 9.5 kW-DC of solar.

1) Recoup as much of the TOU cost with PG&E under NEM 2.0. Currently under EV-A and eventually will transition to EV2-A.
2) Provide backup power as we are in a neighborhood with overhead power lines. If the cost of an equivalent natural gas system (installation and maintenance) is factored with the TOU cost saving, I believe we actually will have a reasonable rate of return.

As a side note, To mitigate for (save on) $0.02-0.03/kwh of
non bypassable charges (NBC) we try to use heavy loads when solar production coincides with our off peak rate. By doing this we are directly using our solar production and mitigating the "penalty" for using the grid "as a battery". When practical, We charge our cars and dry our laundry at this time. Under EV-A this is during the weekend till 3:00 p.m. and with EV2-A this would be till 3:00 p.m. everyday.

The use of Powerwalls to mitigate NBCs are a wash. The cost of the in-out energy loss (10%) is almost equivalent to the NBC saved. I rather save on the wear on the battery.
 
Last edited:
  • Informative
Reactions: Sonus
I'm in the UK and have cheap electricity overnight. until about 3 weeks ago I was running Advanced - Cost Saving to fill the battery with cheap electricity to be used during the day. However, the algorithm has failed me recently by only filling up based on what appears to be the last few days worth of solar generation which left me short of stored electricity a couple of times and using expensive electricity. Since then I have changed to Self-Powered and have set up some automation on a Raspberry Pi which sets the backup percentage to be the percent I want to have in the battery at the end of cheap electricity and then sets the backup percentage back to 5% at the point cheap electricity ends. I'm currently setting that percentage based on looking at the next days weather forecast and "making a call". Ideally I'd work out a way to automate it but so far it's based on manual intervention. I'm expecting in a few weeks time we will have enough solar generation to just leave the unit on Self-Powered.
Please tell us more about how you use Raspberry Pi to automate your powerwall backup percentage.
 
Please tell us more about how you use Raspberry Pi to automate your powerwall backup percentage.
I'm not who you asked, but there are two ways to do this:
1. Use the API on the gateway over the LAN that's used by the setup wizard: vloschiavo/powerwall2
2. Use the Tesla server API that's used by the mobile app: TOU api

#2 seems to be more reliable than #1 now and has more control (the local API doesn't do TOU settings). The drawback is it will not work if your internet connection is down.