I don't like the fact that @Ulmo 's system is discharging the PowerWall during the Super Off-Peak rate period from 11pm-7am. I appreciate that it is maximizing self consumption, but I would like the ability to configure the system to NOT use that battery energy during the cheapest rate period and allow the battery to fill earlier in the morning and earn more NEM credits. This is especially important in the Winter when the solar is generating much less energy. It would really suck if the only way to manage this was to change the reserve setting every couple weeks according to how much your solar can generate per day. I suppose the other way to do it would be to set up a scheduled task to bump up the reserve at 11pm every night and lower it again every morning.
-
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: TOU/Rate Arbitrage
- Thread starter abasile
- Start date
-
- Tags
- Powerwall Tesla Energy
PlugNPlay85
Member
Sadly, you cannot change that fixed time period yet. However, Tesla says an app update is eminent which will allow you to tweak your time of use settings. For now, I use it only in backup mode, because I am unwilling to discharge at off-peak.
Ulmo
Active Member
I was thinking all the same things. The reserve capacity can be dynamically changed to effect the effect desired. That would mean some scripting that can both read and write into the API, something that seems quite plausible (anybody on that yet?). Also, I think I either asked Tesla for or they had already stated that they are working on those things as part of the control software updates they are programming.
You are right that it is worse in any situation in which this matters, otherwise it's just theoretical; I doubt any installation which is not meant to be off grid will ever make this simply theoretical; however, right now I'm living in the theoretical imaginary land in cloudless summer days (!) and temperate climate with a lack of EV: ever since I charged the PowerWalls fully, I have not taken any energy from the grid at all (except token amounts like 100watthours/day, maybe as a pulse or leak or monitoring level or something).
Let me repeat that: with sufficiently sized system (i.e., enough solar and enough PowerWalls), grid use can be nonexistent; the grid only exists as a place to sponge out excess solar.
Some people might prefer to set up dynamically operated cryptocurrency miners (and appropriate air conditioning for them) to use up the extra energy in a more profitable way. (It would have no net energy effect on Earth, because all that energy would eventually end up as heat on Earth anyway.) Before crypto miners, you could move some more energy intensive civil structure into your home, such as desalination, but that would require both a brine output and a saline water source; forward thinking utilities could offer that. But, both operations would essentially be making up for the low reimbursement rates from the utilities; doing such things would put an upward price pressure on the utilities paying for home generation, but with their regulated price controls, it might be decades for them to do that. In the near term, it seems crypto miners are the best sink.
Sinking excess energy only matters when there is more generation than use at the home. It's an ideal situation: full clean independence. "Waste" is not really waste in the full sense of the word. However, it seems like the utilities ought to be able to use the energy better by delivering them to dynamic processes such as water movement and refinement (treatment at all stages, etc.).
As soon as there's any need for grid electric use at all, then all of a sudden, offsetting the PG&E NEM 2.0 costs will matter. Then, using cheap rate from grid and being able to offload more solar at peak offset rates will help a lot more. Then, strategically planning how much to use at cheap rates will matter; basically, a daily prediction. Simply put, an AI; less simply done, a slightly complex prediction system, but it wouldn't be more difficult than a single smart person could throw together in a few months (say a programmer at Tesla who even knows what we're talking about).
Last edited:
PlugNPlay85
Member
Thanks for the reply, what is the fixed time period for the Powerwall to discharge on self powered mode?
Ulmo
Active Member
In honor of @miimura, I did this while drying my clothes today to optimize my financial position with PG&E (note that on weekends and particular holidays, the off peak time starts at 7PM, which is the same as the 11PM weekday rate @miimura mentioned above):
As you can see, I pulled 2.3kWh from the grid at the bottom rates (by forcing the flow logic to pull from grid with the high backup reserve setting), which will allow my battery to have more charge, allowing less solar electricity needed to recharge the battery and thus more solar electricity to pass to PG&E during top rates, getting my PG&E NEM2.0 in better position. This is the first time I've used grid power (in anything but trace amounts) since sunup July 21, the day after my PowerWall install.
Going over to CAISO:
7:45PM-8:45PM yesterday (closest CAISO has to today right now) I am using a mix of mostly coal ("imports") and methane ("thermal" (aka natural gas)), with some wind, hydro ("hydro" and a portion of "imports"), the end of the solar day and a sliver of nuclear and geothermal mixed in; @miimura, I'm not sure I like this rate shifting potential you call out as prudent, for only 4kWh/day.
As soon as I plug in an EV, though, it's game over: 40kWh to 60kWh in one sitting will blow through my measly 2 PowerWall 2's 13.5kWh*2=27kWh's and solar system's ~36kWh's worth of electricity in no time, so if I had that, I'd manually reset the PowerWall "minimum backup reserve" up past the current charge level every evening when I went to bed so that from then on all use would be from grid, and charge the car with the lower cost grid power. That is wrong of course, because we want to charge cars with ample solar panel power; what I should do is install another $20,000 worth of solar panels on our property (which begs the question, where? --- I've already been told I can't plant a pole with a swivling parabolic dish or flat array in the middle of the yard, and most of the roof has been taken), and PowerWalls to match (so, essentially, +50kWh/day of solar panels and +50kWh of (daily cycle) PowerWall 2's, which is 4 more PowerWall 2's for a total of 6 PowerWall 2's, so another $26,500 or so of PowerWalls). At WholeSaleSolar.Com, I get a basic price on solar panels that can +50kWh/day me (so, going from my current ~36kWh/day on a 5.6kW system, that's another 8kW of solar panels):
Our roof is basically maxed out on the easy part to fill; finding space for 30 panels is a lot harder than 24. Doing the 8.16kW system for about the same price we can save space with only 24 panels instead of 30. We could catch some morning sun on the garage with about six panels, and put the other 18 on the "wrong" side of the house. I thought of making a covered walkway to another building on the property, but then they'd starve the grass of sunlight. No carports make sense. We don't have room for a yard-mounted array given our architectural concerns. I was already told we can't put up a pole with a fully tracking array; I think it would look nice, but oh well. There is one place I might be able to still do it if I can hide it behind one of the buildings; I think 12 panels could fit there if I did that, and I think it would be fun to learn about 100% (dual axis) tracker systems. So, I suppose it could be done space-wise.
So, that's about $17,000 (including electrician, permits, and my labor) to put in the additional solar and about $26,000 for the additional 4 PowerWalls, for a total of $43,000, and then I could go 100% clean energy for my over-the-hill commute on my new EV.
Or, I can stop working in that field and start working from home, and just keep my current system and expand it about 40% to 100% (double my PowerWalls and add about 15 more solar panels), for a total of about:
Let's say I get one of the above, for about a max budget of $8,000 all installed, and 2 more PowerWalls for another $13,000, for a total of $21,000. That seems reachable, plus citing the space for 12 panels is a lot easier than 24 panels.
Ok, I finished drying my clothes, and my 3.7kWh of dirty energy saved me about 30 cents per kWh, so about $36/month or $13,000 for the life of the system, if I take this approach every time I dry clothes or some other heavy electrical use event (like my space heaters). But just for today, it's $1.20. I feel dirty. I could have used clean solar power instead to dry them.
Last edited:
I fully admit that my desires for the settings I mentioned in the PowerWall will decrease self-consumption and therefore the benefit to the environment. Well, hold on a minute.... My solar is going to generate the same amount of energy regardless of what I do with the battery, so that offset will always be there. I'm also not talking about changing my usage pattern because of the battery. I have no way of knowing if the extra mid-day feed-in at noon because I'm charging my car from the grid overnight will actually decrease the CO2 producing thermal generation or if it will cause the ISO to curtail some other renewable generator.
So, I will act in my own financial interest for this particular thing. I want the batteries to pay for themselves as much as possible. I still pay PG&E $1,000 per year at true-up, so I want to at least cut that in half if I get Powerwall(s). It just doesn't make sense to me to charge the battery from my own solar and forego $0.24/kWh Part-Peak credits and use that energy from the battery when I could get it from the grid overnight for $0.12/kWh. A more minor concern would be discharging during Part-Peak when you are just putting extra cycles on the battery and taking the efficiency hit from round trip charging and inverting losses. My solar is only 4.32kW DC and I can only add 1 kW if I want to stay on the NEM 1 Tariff. If you are on the NEM 2 tariff as I believe @Ulmo is, then you have more of a motivation for self-consumption because you will have to pay Non-Bypassable Charges for the kWh you pull from the grid. In that case, the NBC hit will cover at least the round trip efficiency losses.
Anyway, this is my way of saying more end-user control is better.
So, I will act in my own financial interest for this particular thing. I want the batteries to pay for themselves as much as possible. I still pay PG&E $1,000 per year at true-up, so I want to at least cut that in half if I get Powerwall(s). It just doesn't make sense to me to charge the battery from my own solar and forego $0.24/kWh Part-Peak credits and use that energy from the battery when I could get it from the grid overnight for $0.12/kWh. A more minor concern would be discharging during Part-Peak when you are just putting extra cycles on the battery and taking the efficiency hit from round trip charging and inverting losses. My solar is only 4.32kW DC and I can only add 1 kW if I want to stay on the NEM 1 Tariff. If you are on the NEM 2 tariff as I believe @Ulmo is, then you have more of a motivation for self-consumption because you will have to pay Non-Bypassable Charges for the kWh you pull from the grid. In that case, the NBC hit will cover at least the round trip efficiency losses.
Anyway, this is my way of saying more end-user control is better.
Ulmo
Active Member
Thank you, and I agree. I want to emphasize how correct, well written, and complete the above post is by @miimura. I encourage it as the first post to read of anybody who buys a PowerWall 2.
As an aside, my very latest update with my experience is I tried for a time to put my PowerWalls back into discharging mode, by lowering the backup reserve % to under the charge level %, but that did not do anything, even after waiting for half an hour. I will now attempt to mimic a power outage to see if that has an effect.
After 1 month of using the PW2 in Self-Powered mode here's what my PG&E usage looks like. (I started using it that way on June 27 after filling it that weekend from the grid) We are driving fewer local miles in the C-Max Energi since our gym commute is much shorter so our electric usage is way down from last year too. There was an early July vacation in there too. First month in a long time we were a net energy producer.
I'm using the battery in the spirit of how they're marketing it - collect solar during the day to use at night - and not really doing arbitrage in part because I'm concerned about our SGIP incentives. I really want it to look to PG&E and the SGIP incentive program like this is how people will use the battery. I may decide (especially in the winter) to do arbitrage but for now I'm happy being self-sufficient and selling back at peak times a few days of the week when the battery is at 100%.
I'm using the battery in the spirit of how they're marketing it - collect solar during the day to use at night - and not really doing arbitrage in part because I'm concerned about our SGIP incentives. I really want it to look to PG&E and the SGIP incentive program like this is how people will use the battery. I may decide (especially in the winter) to do arbitrage but for now I'm happy being self-sufficient and selling back at peak times a few days of the week when the battery is at 100%.
Attachments
Ulmo
Active Member
My PG&E usage detail from PG&E shows that ever since I got my PowerWalls, what I've been net sending to PG&E has gone down; the installation was the day before the blue bar and the blue bar was the result of me using a large amount of solar power to charge up the PowerWalls for the first time rather than offset PG&E use of the early morning.
Because of non-bypassable charges (in NEM2 but not in NEM1), I think I'm better off using my own solar power, but I'll check.
Because of non-bypassable charges (in NEM2 but not in NEM1), I think I'm better off using my own solar power, but I'll check.
bkp_duke
Well-Known Member
I was thinking the same thing, about non-bypassable charges, but for the past two months they have been PENNIES on my bill (was on NEM 1 in the old house, so only have 2 months data so far).
The numbers make me concerned that even with SGIP that the Powerwall may never pay for itself. I'm debating cancelling and waiting till the finances make sense. I'm all for the environment, but the finances have to line up as well.
The numbers make me concerned that even with SGIP that the Powerwall may never pay for itself. I'm debating cancelling and waiting till the finances make sense. I'm all for the environment, but the finances have to line up as well.
On NEM 1.0 in California, even with SGIP, it can be tough to justify the cost of a Powerwall 2 on the basis that it will reduce grid charges. However, the backup capability is certainly worth something. People spend thousands on whole-home backup generators (that use fossil fuels). With this factored in, there's enough value for us to move forward.
SoundDaTrumpet
Member
I was also borderline with cancelling until I was able to lower installation costs significantly. With a clear path forward (SGIP & ITC), the Powerwall pays for itself because my solar production is undersized.
Last edited:
Ulmo
Active Member
The sudden drop in energy being sent to PG&E couldn't be explained just by the PowerWalls, so I looked into it. According to my computer logs, I started using my heaters about the same time the power usage went up on the PG&E graphs. I live in Aptos which gets a lot of cool summer nights. My cryptocurrency miner was turned back on on July 23rd, which is the first heater I started using. There were some extra-cold nights (night of July 28 for a few days) which required my space heater in addition. As you can see in the PG&E graph, that explains most of the drop in energy being sent to PG&E from my solar system; I assume my cryptocurrency miner uses about 500 watts, more or less; I'm about to go to bed, otherwise I'd go test that assumption right now.
I'll try to run the numbers when my next PG&E bill comes in, and the following bill, too.
I've looked (admitted not that hard) for NEM 2 non-bypassable charges - how much does SDG&E charge for each net-metered kWh under NEM 2?
bkp_duke
Well-Known Member
Old house was NEM1. New is NEM2, and even on NEM2 and getting SGIP it does not appear that the powerwalls will ever pay for themselves.
bkp_duke
Well-Known Member
I have no idea how they are calculating NBC, because my two bills in the new house are:
June:
"Your non-bypassable charges for this month were based on usage of 13 kWh."
$0.22
July:
"Your non-bypassable charges for this month were based on usage of 13 kWh."
$0.22
The very odd thing is I KNOW our usage from the grid has been much higher than 13 kWh per month because the AC is running at night, almost every night, and we've let the cars charge after sunset at times.
I've overproduced like crazy these past two month because I was out of town, and the pool has not yet been installed.
June - 779 kWh back to grid
July - 1008 kWh back to grid
I'm not inclined to poke the bear and tell SDG&E that their calculations are way off (in my favor), but if this continues, the two powerwalls on order make zero sense financially.
I too would like to know how PG&E calculates the NBC. I also made a spreadsheet to track how much peak usage I could offset by time shifting with a Powerwall. Last month it was $8.62
Peak -224.7936000 kWh @ $.045389 -$102.01
Part Peak -146.452800 kWh @ $0.24986 -$36.59
Off Peak 239.541000 kWh @ $0.12225 $29.28
NBC Net Usage Adjustment $3.04
NBC $8.55
Peak -224.7936000 kWh @ $.045389 -$102.01
Part Peak -146.452800 kWh @ $0.24986 -$36.59
Off Peak 239.541000 kWh @ $0.12225 $29.28
NBC Net Usage Adjustment $3.04
NBC $8.55
Figured it out. According to this page: Net Energy Metering Program | San Diego Gas & Electric
NBCs are made up of the following (with the apparent rates at the end of each):
- Public Purpose Programs (PPP) charge - $0.01063
- Nuclear Decommissioning (ND) charge - $-0.00049
- Competition Transition Charge (CTC) - $0.00177
- Department of Water Resources Bond Charge (DWR-BC) - $0.00549
So in theory, if you import export 1000 kWh and also import 1000 kWh, instead of being NET-zero cost, you should have an additional non-bypassable charge of $17.40.
This blog post also explains this: The Ultimate Guide to NEM 2.0: Part 1, Non-Bypassable Charges
How SDG&E only came up with 13 kWh is a good question. They must be looking at interval data to figure this out and I'd have to guess 15 min intervals, but have to do more research.
As far as NBCs being used to help your financial return, they are so low that they don't even account for the overhead of the Powerwall 2.0 at 89%.
To maximize Powerwall return for those of us on NEM in California and claiming the SGIP, here's what I believe to be the optimal strategy:
1. Charge Powerwall using solar only during off-peak periods.
2. Use Powerwall for self-consumption during on-peak periods and to maximize solar exports.
3. Discharge Powerwall enough to meet annual cycling requirements without discharging below a minimum level to allow for backup use if you lose grid power.
In the end, though with current peak TOU periods still covering half of daylight, it's hard to make a financial case for storage and self consumption of solar.
Might look in to OhmConnect. In my limited experience it appears that it could provide good financial return if you can force discharge the Powerwall during OhmHours.
bkp_duke
Well-Known Member
Thanks for the info on how they calculate it. I have monitoring setup on my solar system to get a pretty good idea (+/- 10%) of what I pull from the grid and put back into it.
Example:
July I pulled approximately 1100kWh from the grid and put 2300kWh back in. I would expect a NBC of about $17-18 based upon that, not what I actually saw based upon 13kWh.
EDIT - Thanks for the OhmConnect info. Going to look into that further.
Last edited:
