Aggregated throughput means discharged kWh per footnote 5 (Measured at the battery AC output), after 2 years you should be at no more than 9.855MWh with a full 13.5 kWh discharge every day. The 37.8 MWh over 10 years works out to an average of 76.7% discharge or 10.36 kWh/day which I think in order to exceed you would need to be doing grid charge for most of the year or between the Puerto RIco longitude and the equator.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Tesla Virtual Power Plant in CA
- Thread starter power.saver
- Start date
A few years ago I posted the ITP research where they cycled a bunch of batteries and measured their SoC degredation. Their latest findings are updated here.
You need to keep in mind that the 11.8kW site limit that you have includes the solar generation. So your Powerwall max discharge will be 11.8 - (Solar - House), For my system my Powerwalls are discharging at 4.9 kW = (7.8kW - (3.4kwW - 0.5W)). If you increase your house load, by say charging an EV, then you can increase your discharge rate.
What matters is only what the Powerwall discharges (during the VPP and the baseline for days prior to the VPP) , it doesn't matter if goes to the grid or to your house.
So if I’m understanding right…I get more VPP $ if I actually use more energy, at least while the sun is up and my solar is generating?
Pretty sure Tesla made a big deal when promoting VPP that THEIR VPP wouldn't void the unlimited cycle warranty but other VPP programs would. My average aggregate AC output per powerwall for my last 12 month period was 1.989 MWh, so it would take me 19 years at that rate to exceed the 38.8 MWh, long after the 10 year warranty is up.
Ok, so this one is interesting...since this VPP is starting at 4pm while the solar is still producing, I'm experimenting with some things and found some interesting results.
At the beginning of the event (4pm), I turned off as much as I could at the house (well, mostly just shut off the AC), and here's what happened:
- The house was using 0.2kW
- Solar was producing 5.7kW
- The powerwalls were discharging at 2.2kW (total between 2 PWs)
- The grid was getting 7.7kW
Since 7.7kW is the most the grid will allow me to export (that's 32 amps) the powerwalls decreased their output so that their output plus solar minus house = 7.7kW. At only 2.2kW, that's not enough to export my 27ish kWh over the course of 5 hours (although it would probably speed up as the solar waned throughout the evening).
Then, I turned the AC back ON to suck some power from the solar and here's what I got:
- The house was using 3.1kW
- Solar was producing 5.7kW
- The powerwalls were discharging at 5.1kW (total between 2 PWs)
- The grid was getting 7.7kW
So here's my theory...with a longer VPP window (which includes time that solar is producing well), the most profitable route would be to set the backup reserve to 0% and get the house consumption to match the solar output to minimize the amount of excess solar going to the grid and maximize the amount of powerwall power going to the grid. Theoretically I could get the PW up to 7.7kW export. Do that for an hour (from 4pm to 5pm), which would give you 7.7kWh of VPP credit (about 30% of the total 2xPW capacity). Then, change the backup reserve to 100% from, say 5pm to 6:30pm to export nothing and charge the PW back up to full. Then change the backup reserve back to 0% and let it rip for the rest of the VPP.
If I'm looking at this right, that means I could theoretically get 130% of my power wall capacity for each VPP even that starts early enough in the day to take advantage of solar to re-charge the PWs
...even better results could be reached if they started at say 2pm at some point 
Does that seem right?
Edit: looks like @Redhill_qik started talking about this a little above while I was in the middle of typing this up.
At the beginning of the event (4pm), I turned off as much as I could at the house (well, mostly just shut off the AC), and here's what happened:
- The house was using 0.2kW
- Solar was producing 5.7kW
- The powerwalls were discharging at 2.2kW (total between 2 PWs)
- The grid was getting 7.7kW
Since 7.7kW is the most the grid will allow me to export (that's 32 amps) the powerwalls decreased their output so that their output plus solar minus house = 7.7kW. At only 2.2kW, that's not enough to export my 27ish kWh over the course of 5 hours (although it would probably speed up as the solar waned throughout the evening).
Then, I turned the AC back ON to suck some power from the solar and here's what I got:
- The house was using 3.1kW
- Solar was producing 5.7kW
- The powerwalls were discharging at 5.1kW (total between 2 PWs)
- The grid was getting 7.7kW
So here's my theory...with a longer VPP window (which includes time that solar is producing well), the most profitable route would be to set the backup reserve to 0% and get the house consumption to match the solar output to minimize the amount of excess solar going to the grid and maximize the amount of powerwall power going to the grid. Theoretically I could get the PW up to 7.7kW export. Do that for an hour (from 4pm to 5pm), which would give you 7.7kWh of VPP credit (about 30% of the total 2xPW capacity). Then, change the backup reserve to 100% from, say 5pm to 6:30pm to export nothing and charge the PW back up to full. Then change the backup reserve back to 0% and let it rip for the rest of the VPP.
If I'm looking at this right, that means I could theoretically get 130% of my power wall capacity for each VPP even that starts early enough in the day to take advantage of solar to re-charge the PWs
...even better results could be reached if they started at say 2pm at some point Does that seem right?
Edit: looks like @Redhill_qik started talking about this a little above while I was in the middle of typing this up.
Last edited:
Really bad UI on that site, I think I clicked through the same page cycle three times before realizing that I needed to click on the cover image to get the report PDF. This is the direct link for anyone that wants it https://itpau.com.au/documents/32/Battlab_Report_12_Final.pdfA few years ago I posted the ITP research where they cycled a bunch of batteries and measured their SoC degredation. Their latest findings are updated here.
The conclusion is 79% remaining after 2,520 cycles which is a 6.9 years. This is a single Powerwall2 that isn't being used in the typical solar based installation that would have lower charge and discharge rates. Definitely interesting information, but not what I would call full characterization. so YMMV.
It's Australian, so you have to turn your laptop upside down...
Do you have a link or reference for that? My output total shows 10.2MWh for 2 year old system from the App
Well, you get more VPP $ for a large difference between your baseline discharge and the discharge during the VPP. Yes, you can game the system to do this when you are still generating from solar and limited by your site export limit. Should you do this? Probably not, but I have previously admitted to charging my EV during a grid outage during a VPP last year, this got me a few more $, but I was also able to use solar that would have otherwise been disabled due to a full Powerwall.
Can't find it now. It was one of the things that convinced me to join the program otherwise I would have declined.
Now that you mentioned the app, I went ahead and checked that and for the last 12 months from today it says 4.16 MWh combined discharge for my 3 powerwalls, but when you add up each 5 minute segment from the beginning of August 2022 to end of July 2023, it says 5.97 MWh combined. It's a huge discrepancy. Not sure how the app is calculating it but the in my spreadsheet, it's the combined sum of kw output from the PWs for each 5 minute interval / 12 / 1000 to get total kWh hours (5,969).
The app is messed up. If I add up each month discharged by month in the app instead of the 1 year total, it comes out to 6,025 kwh, so the API is correct and the app month to month is correct but the year view is totally screwed.
Can’t you use ChargeHQ to basically take the solar and put it into an EV, basically matching PV outputOk, so this one is interesting...since this VPP is starting at 4pm while the solar is still producing, I'm experimenting with some things and found some interesting results.
At the beginning of the event (4pm), I turned off as much as I could at the house (well, mostly just shut off the AC), and here's what happened:
- The house was using 0.2kW
- Solar was producing 5.7kW
- The powerwalls were discharging at 2.2kW (total between 2 PWs)
- The grid was getting 7.7kW
Since 7.7kW is the most the grid will allow me to export (that's 32 amps) the powerwalls decreased their output so that their output plus solar minus house = 7.7kW. At only 2.2kW, that's not enough to export my 27ish kWh over the course of 5 hours (although it would probably speed up as the solar waned throughout the evening).
Then, I turned the AC back ON to suck some power from the solar and here's what I got:
- The house was using 3.1kW
- Solar was producing 5.7kW
- The powerwalls were discharging at 5.1kW (total between 2 PWs)
- The grid was getting 7.7kW
So here's my theory...with a longer VPP window (which includes time that solar is producing well), the most profitable route would be to set the backup reserve to 0% and get the house consumption to match the solar output to minimize the amount of excess solar going to the grid and maximize the amount of powerwall power going to the grid. Theoretically I could get the PW up to 7.7kW export. Do that for an hour (from 4pm to 5pm), which would give you 7.7kWh of VPP credit (about 30% of the total 2xPW capacity). Then, change the backup reserve to 100% from, say 5pm to 6:30pm to export nothing and charge the PW back up to full. Then change the backup reserve back to 0% and let it rip for the rest of the VPP.
If I'm looking at this right, that means I could theoretically get 130% of my power wall capacity for each VPP even that starts early enough in the day to take advantage of solar to re-charge the PWs
Does that seem right?
Edit: looks like @Redhill_qik started talking about this a little above while I was in the middle of typing this up.
Select EV instead of Battery?
Not sure I understand why you shouldn’t do this? If you’re charging your EV, energy is still going to something productive. It’s not like you are using it for resistance heating of the air or something…
Since my VPP event wasn’t correctly discharging batteries to grid I decided to take matters into my own hands to increase my baseline usage… your move Tesla
New one scheduled for tomorrow for 6pm-9pm.
Not that it matters for compensation but it should be a little cooler here so we don't have to use AC. For a few minutes during tonight's event we were actually pulling a small amount from the grid with the AC running and the new algorithm to end the event at reserve
Not that it matters for compensation but it should be a little cooler here so we don't have to use AC. For a few minutes during tonight's event we were actually pulling a small amount from the grid with the AC running and the new algorithm to end the event at reserve
Someone posted recently in this thread with CAISO alerts, for yesterday, today and tomorrow…that so far has a 1:1 correlation. That said…correlation does not prove causation
Last edited:
Was able to manual eyeball close enough. The 30min delay seemed to work, and it kept up 11.8kW right until 9p with a bit of a dip just before.
Still trying to figure out how to not have solar go to grid from 4p-630p. Maybe use one of the EV’s with PV priority Solar tracking with ChargeHQ?
This could all be moot, if PGE/Tesla render me a “typical” powerwall exporter at 11.8kW from 4p onwards each day. Maybe I’ll get credit for the 8-9p hour since I’m normally done exporting each day about 8p.
I just hope Tesla fixes my clipping inverter on Thurs when they show up for their appointment. Looks like those book 1.5mo out now, so really hope they fix this in one go!
The program is an Emergency Load Reduction Program intended to remove load from the grid either by lowering usage and increasing generation (Powerwall discharge). Charging an EV is adding load which may get you more $, but is the opposite intent of the program.
Similar threads
