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

Keeping PW+ at fixed charge percentage

This site may earn commission on affiliate links.
Is there some way I can keep PW charged at a single percentage while in grid connected mode?

E.g. set it to 75% so that any excess is immediately exported into the grid, rather than charging to 100% first, and kept as a reserve for power outages?

I tried tricking it by creating my own rate plan in which daytime hours are more expensive and choosing the time shift (rather than self powered) option. Somehow that resulted in all solar going into PW, rather than being exported.
 
Is there some way I can keep PW charged at a single percentage while in grid connected mode?

E.g. set it to 75% so that any excess is immediately exported into the grid, rather than charging to 100% first, and kept as a reserve for power outages?

I tried tricking it by creating my own rate plan in which daytime hours are more expensive and choosing the time shift (rather than self powered) option. Somehow that resulted in all solar going into PW, rather than being exported.

I can't guarantee these settings will work for you...I think the AI may play a role in it not always sending to the grid...

Using "Peak", I use buy=.26 and sell=.03 from 7am to 9pm. When I switch to time-base mode, it will send to the grid instead of the pw.
 
Is there some way I can keep PW charged at a single percentage while in grid connected mode?

E.g. set it to 75% so that any excess is immediately exported into the grid, rather than charging to 100% first, and kept as a reserve for power outages?

I tried tricking it by creating my own rate plan in which daytime hours are more expensive and choosing the time shift (rather than self powered) option. Somehow that resulted in all solar going into PW, rather than being exported.
what would be the purpose of doing this?
 
what would be the purpose of doing this?
It's pretty simple really.

From a financial standpoint, charging the PW and then discharging it will always result in net losses as neither process is 100% efficient. OTOH, exporting to the grid will always be 100% efficient. In addition, from a battery longevity standpoint, not having the battery spend any time in the top 20% (or bottom 20%) is better to maximize its life.

My primary motivation to have PW is to have power during outages. For that, I want to keep it at 75% during the winter (when the sun won't be out every day) and around 35% in the summer (when solar production will suffice).

If you want to maximize the amount of solar power your house uses, the mode that Tesla provides is useful, but only to a point, and imho only if you can go mostly off-grid. My array and my single battery are not sufficient to power the house entirely, especially when you factor in car charging, so I'd prefer to minimize my conversion losses.
 
It's pretty simple really.

From a financial standpoint, charging the PW and then discharging it will always result in net losses as neither process is 100% efficient. OTOH, exporting to the grid will always be 100% efficient. In addition, from a battery longevity standpoint, not having the battery spend any time in the top 20% (or bottom 20%) is better to maximize its life.

My primary motivation to have PW is to have power during outages. For that, I want to keep it at 75% during the winter (when the sun won't be out every day) and around 35% in the summer (when solar production will suffice).

If you want to maximize the amount of solar power your house uses, the mode that Tesla provides is useful, but only to a point, and imho only if you can go mostly off-grid. My array and my single battery are not sufficient to power the house entirely, especially when you factor in car charging, so I'd prefer to minimize my conversion losses.

I think the only way to achieve this is some sort of automation, where a process checks (maybe every 15 min) for the 75% and when it hits this it will probably have to go time-based mode and then excess will flow to the grid. I was playing around with doing something like this but I finally got grid charging so it's been easier to hit the 100% and sending back to the grid.
 
If you only want to use PW as a backup, perhaps you could just set the reserve at 100%.

I'm not sure that your concern about full charging is well founded. Especially if you don't discharge and recharge daily, any degradation may not be all that significant. We know that there is a 5% reserve below 0%, but there could also be a hidden reserve above 100% as well. Tesla used to do this on their early cars to mask the range loss due to early battery degradation.

Before Tesla gave us a grid charging option, I once hacked together some software on a home computer to automate changes to my PW settings, so this is a possibility. But I am not aware of a turn-key program to do this. I also have heard of third party services which can do so as well.

One last idea would be to have your solar re-wired to the meter side of the gateway, so the PW is unaware of your solar. I don't know how PW behaves in this non-solar configuration, but a reserve setting of 75% might stop charging at that level.
 
I'm not sure that your concern about full charging is well founded. Especially if you don't discharge and recharge daily, any degradation may not be all that significant. We know that there is a 5% reserve below 0%, but there could also be a hidden reserve above 100% as well. Tesla used to do this on their early cars to mask the range loss due to early battery degradation.
This has been discussed many times here with any real conclusion as no one outside of Tesla has access to the reliability data. Tesla guarantees at least 70% of capacity after 10 years when used to power the home "normally" which appears to be one full discharge/recharge cycle per day based on the other spec for kWhs. The 5% reserve at the bottom is with the app reporting, but if you are in using the REST APIs and looking at the nominal_energy_remaining values then those are not adjusted as far as I know.

Some Powerwalls have see rapid declines in capacity during the first year and then slow down and others don't experience this at all. My Powerwall has been near 100% for most of it's life and using the REST APIs both are still above the rated 13.500 kWh after 28 months. I thought that this might be an issue with the battery management system not having enough data across the full range, but during the VPP events I discharged down to 5% and then back to 100% multiple times with no significant change in the reported capacity.
 
If you only want to use PW as a backup, perhaps you could just set the reserve at 100%.
This is literally not recommended for any traditional (non-LFP) battery so I'm definitely not going to do that.

I tried time-based but this resulted in PW charging above its reserve and the house using entirely grid power, which is obviously not desired and frankly made no sense. But perhaps I'll try it again.
 
This is literally not recommended for any traditional (non-LFP) battery so I'm definitely not going to do that.
My PW is nearly 3 years old, cycled every day from 100% to 20% and back. Prompted by your statement about "literally not recommended" charging, today I queried my PW API via the local internet interface (https://powerwall/api/system_status), and among the returned info is the current battery capacity:

nominal_full_pack_energy":13245

That is 98% of the spec capacity after 3 years of 100%, not very significant degradation.

Tesla's PW warranty includes covering excess degradation. They designed PW to regularly charge to 100%. So apparently Tesla is not swayed by whatever recommendations you are seeing. Perhaps those recommendations you saw apply to, as you say, traditional batteries, which do suffer from heating during rapid (1C) charging. PW has thermal management and charges and discharges at a rate of less than .5C.

The Time-Based Control behavior you describe is not what mine does. I'll attach yesterday's battery info below. TBC does rely on several settings, including time zone, utility rate periods and buy and sell pricing, as well as a few days of usage and solar history. So double check your settings and watch it for a few days. Without Export Everything, it would cover the entire partial and peak periods, and end at around 50% state of charge.

IMG_8091.jpg


The PW charged from solar reaching 100% around noon. When the rate went up at 4 PW supplied power to the house. Because we set it to Export Everything it also dumped what it thought it would not need onto the grid. Around 11 pm it was down to the 20% reserve and it stopped discharging. Our off peak rate doesn't start till midnight, but PW calculated that exporting a little too much at our 4pm peak rate was a better deal than saving it to avoid consuming at the partial peak rate from 11 till 12.
 
This is literally not recommended for any traditional (non-LFP) battery so I'm definitely not going to do that.

I tried time-based but this resulted in PW charging above its reserve and the house using entirely grid power, which is obviously not desired and frankly made no sense. But perhaps I'll try it again.
This is literally recommended by Tesla as they designed the Powerwall control system to specifically charge to 100% without any option to set the maximum charge to lower percentage. There is no recommendation to attempt to do anything different than this. Tesla is on the hook for meeting their performance guarantees and if this was going to result in more warranty claims being filed then they wouldn't be allowing the system to do this as it would cost them money. You think that is a problem to have the Powerwall at 100% for any length of time, but Telsa that has money on the line and the reliability data doesn't.

If you want to go to extremes in an attempt to fix a problem that does not appear to exist for Powerwalls have with much lower charge/discharge rates than car battery packs experience then that is your prerogative, but realize that your attempt to limit the max charge to 70-80% is essentially accepting the warrantied 10 year number as your capacity starting on day one.
 
Last edited:
It's pretty simple really.

From a financial standpoint, charging the PW and then discharging it will always result in net losses as neither process is 100% efficient. OTOH, exporting to the grid will always be 100% efficient. In addition, from a battery longevity standpoint, not having the battery spend any time in the top 20% (or bottom 20%) is better to maximize its life.

My primary motivation to have PW is to have power during outages. For that, I want to keep it at 75% during the winter (when the sun won't be out every day) and around 35% in the summer (when solar production will suffice).

If you want to maximize the amount of solar power your house uses, the mode that Tesla provides is useful, but only to a point, and imho only if you can go mostly off-grid. My array and my single battery are not sufficient to power the house entirely, especially when you factor in car charging, so I'd prefer to minimize my conversion losses.

Yea, I can't understand why Tesla doesn't have a max charge SOC setting. In the winter, even with 100% peak offset usage, I cycle between 75% and 100% every day that it is sunny. I'd much rather cycle between 25% and 75%. The batteries would last TWICE as long easily.
 
Yea, I can't understand why Tesla doesn't have a max charge SOC setting. In the winter, even with 100% peak offset usage, I cycle between 75% and 100% every day that it is sunny. I'd much rather cycle between 25% and 75%. The batteries would last TWICE as long easily.
I cycle between 95-100% during the winter, just to exercise them a little. Who says that the first thing to go after 10 years is the battery capacity, maybe it is the electronics, the inverter or controller most likely, that goes bad. Consumer spec for ICs is 20 years with a lower junction temp of 85C and Industrial spec is generally 10 years due to a higher junction temp of 100C. So, it becomes a numbers game at that point.
 
I cycle between 95-100% during the winter, just to exercise them a little. Who says that the first thing to go after 10 years is the battery capacity, maybe it is the electronics, the inverter or controller most likely, that goes bad. Consumer spec for ICs is 20 years with a lower junction temp of 85C and Industrial spec is generally 10 years due to a higher junction temp of 100C. So, it becomes a numbers game at that point.

Maybe, but the electronics lifetime is unlikely to be changed by cycling 37.5% to 62.5% vs 75%-100%. (previous post should have states 37.5 to 62.5, not 25 to 75), yet we know for sure this is way better for the batteries.
 
cycled every day from 100% to 20% and back
I’m not sure how that’s relevant in response to a suggestion to keep the battery pegged at 100% 24/7.

There’s nothing wrong with charging these batteries to 100%. However, they don’t defy the laws of physics, and keeping them at 100% for days on end is actually a great way to shorten their lifespan.

we set it to Export Everything
That option isn’t even available for me, but it sounds like exactly what I’d want in this case.
 
I’m not sure how that’s relevant in response to a suggestion to keep the battery pegged at 100% 24/7.

There’s nothing wrong with charging these batteries to 100%. However, they don’t defy the laws of physics, and keeping them at 100% for days on end is actually a great way to shorten their lifespan.


That option isn’t even available for me, but it sounds like exactly what I’d want in this case.
lensovet,

I was only sharing my experience with degradation, which appears to be minimal. Mine sits at 100 only from around noon till 3 when it starts discharging to cover the house load, so is not the long term 100% which you want to avoid.

I gather that you don't have a time-of-use rate plan, so the only utility of the battery for you is to cover grid grid outages. I'm not sure Export Everything would help you either. Yours is quite different from my case, because our peak rate is 2 to 3 times the off-peak rate. So for us, time shifting consumption and solar export give extra financial leverage on our solar production, as well as reducing our grid draw during times of high loads. In fact, under the Virtual Power Plant program our utility offers to pay us (through Tesla) $2.00 per kWh we export during occasional high stress "grid emergencies".

PW used to have a mode called Backup Only, but I don't think even that would limit the charging to the reserve setting. Now they say to accomplish the Backup Only functionality, use Self Powered and 100% reserve. Sadly that won't do what you want either.

Before Tesla added the option of grid charging last year, I did some experimentation with my PW which did allow me to charge from the grid, as I reported in an old thread. But a modification of what I did might, just possibly, be a way to keep your PW from charging from solar. I have not tried this exact set up, so I'm not sure how it would behave, but it might just solve your dilemma. The jist is to re-configure the PW to think you have no solar, so it will let your solar run the house and export the excess rather than charging. Setting this up involves going into the installer menus, which TMC moderators prefer not to see posted in these forums, but if you send me a private message I could fill you in on this potential work-around.

SW
 
  • Like
Reactions: lensovet
Here's another data point. My total pack energy is at 73.7% after 4 years. Two Powerwalls in the basement and solar on roof. For the first year or so, it was set to backup only at 100% reserve. Then, I switched it to self powered mode with a reserve of 20%. Now, I mostly use autonomous mode (unless it's being dumb) with time of use and a reserve of 10%. The batteries do about a full cycle per day. I'm keeping an eye on it to see if it drops below 70%. If I were to do anything differently, I probably would not have kept it on backup mode at 100% reserve earlier on.

f942f5b5-8b5a-4dee-b339-57e32e587d70.png
 
  • Like
Reactions: lensovet and RHinCT
Here's another data point. My total pack energy is at 73.7% after 4 years. Two Powerwalls in the basement and solar on roof. For the first year or so, it was set to backup only at 100% reserve. Then, I switched it to self powered mode with a reserve of 20%. Now, I mostly use autonomous mode (unless it's being dumb) with time of use and a reserve of 10%. The batteries do about a full cycle per day. I'm keeping an eye on it to see if it drops below 70%. If I were to do anything differently, I probably would not have kept it on backup mode at 100% reserve earlier on.
That's an interesting data point. Have you checked to see whether both batteries show the same degradation (e.g., using the local /api/system_status)?