Energy Products

Feature request: during outage, Powerwall reserve 5% to jumpstart solar

All 3rd party solar systems (like the one that came with my house) are engineered to not generate electricity if there is no line power detected. This is to protect backfeed to the grid which could harm linesman trying to repair the grid.

When you have a Tesla Powerwall and Gateway, you can operate solar off the grid... but if your powerwall fully discharges, the solar is engineered not to turn on and charge the battery.

So my feature request is for Tesla to implement a setting where you can specify “when battery gets to X%, shut off powerwall discharge. Then when the sun is up again, turn on powerwall discharge to allow solar panels to generate and charge the battery. If charging does not commence after 10 minutes (too cloudy a day to charge) then shutdown powerwall discharge again until for an hour and try again... or get really smart by tying into weather predictions to see if cloudy is expected for a week. And don’t bother.

Either way, there needs to be more control over this Backup mode.


  • Have you actually seen a Powerwall behave in this way (i.e. not come back up after full discharge at night)? This would be a serious functionality issue, as you're supposed to be able to run completely off-grid, charging and discharging the PWs daily.

    Mine works just fine over several days, when tested with the main breaker opened.
  • How does this actually behave in real life? I have never discharged my Powerwalls to 0% before.

    In a power outage situation, if the Powerwalls get to 0% at night, does the solar no longer turn on the next day?

    Do the solar inverters need a certain minimum percentage in the Powerwall before it will allow the solar panels to turn on?

    I understand why solar is turned off during an outage when there is no Powerwall present. Avoid solar energy feedback injuring electrical workers in the area.

    But how does the SolarEdge inverter behave when Powerwalls are at 0% and the sun come up? The energy demand from the house might be uneven initially with appliances trying to pull right away while solar is producing only 0.1 kW at sunrise. With Powerwall at 0%, I could imagine all sorts of things tripping breakers.
  • In my experience, PW behavior during grid outages is exactly as designed. PV runs the house and charges the PW during the day, the PW runs the house after the sun goes down (and stops when it's depleted), and in the morning the PV recharges the PW until there's enough charge to turn on the house loads again. It's all controlled by the TEG.

    The OP and PapaSmurf are introducing FUD, by discussing "what if" scenarios that have no basis in how these things actually work.

    Note that I asked the OP if they actually saw the PWs behave in the way they described, i.e. "but if your powerwall fully discharges, the solar is engineered not to turn on and charge the battery." I got no response, because they aren't designed to work that way, and they don't. Period.

    PWs are fully capable of running off grid indefinitely, and the TEG will control power flows to insure that it's done seamlessly.
  • I had this exact same concern and was told by my installer that this degenerate scenario is in fact possible albeit very unlikely. I would have hoped the PWs would hold a reserve used ONLY for the purpose of providing AC to the inverters, thus allowing the inverters to not shut down the PV when the grid is out. But the installer said nope, the worst case is possible. I also cannot prove it one way or the other -- so I also hope this is just wrong. It would truly be ironic (to put it nicely) if I couldn't get my inverters to power on and charge the batteries because the batteries were dead. So I have to have a generator to get out of this worst apocalypse case?

    But frankly, this wouldn't be the only design flaw I've seen so far in the way this stuff works. The other one that really drives me nuts (and this one I have directly observed): grid out, batteries full, PV exceeds load --> load center shuts off PV by telling PWs to shift frequency to 62.5 Hz, which causes inverters to assume bad things and shut down. This 62.5 Hz causes various devices in the home to be very unhappy (like clocks, some consumer grade UPSes). Why would I have a UPS in a house with solar and batteries? Great question -- because the brief voltage drop when the grid goes out or comes back causes sensitive devices (computers) to reboot.

    So yeah, there are definitely some not so great design flaws in Tesla's system. And it wouldn't shock me if the OP's doomsday scenario is among them.
  • I forgot to mention that my installer is not 3rd party -- works for Tesla directly and very knowledgeable.
  • A newer firmware update tightened up the frequency delta to reduce the prevalence of unhappy UPSs. My UPS wouldn't come back online during full battery PV shutdown frequency shifts before the firmware update. They work fine now, after the update. Insure that you've got the latest firmware.

    I've been on these fora for over five years, and there's not been a single instance of this "doomsday scenario" actually happening reported.
  • I agree with @gregbrew's summary, which is exactly the way our two PW2's have behaved over past 2+ years operating in self powered mode. We operate independent of grid for ~9 months of the year with no grid draws, and actually export ~50 percent of our excess solar energy back to grid. We have never had a problem charging up the PW2's with our solar PV during the Spring, Summer, Fall. You obviously need to be careful about what house loads are on when operating after the sun goes down. We always limit out EV charging to coincide with the maximum solar production, to stay within the operating limits of the PW2's, and do not charge at night, which would quickly drain the PW2s. No doomsday scenario for me either.
  • my installer explained the powerwalls maintained a reserve for startup purposes. he said that why the specs state 14kwh total energy & only 13.5kwh usable.

    supposedly there is some depleted scenario you can get yourself into where you need a 12v battery to jumpstart the unit from some terminals inside the batteries
  • > @gregbrew_98470014 said:
    > A newer firmware update tightened up the frequency delta to reduce the prevalence of unhappy UPSs.

    I'm on 1.48.1 -- what version do you have? According to the release notes, even that is newer than the latest documented, which is 1.48 (although I saw a rumored 1.49 in another thread).

    I saw the frequency jump from 60 Hz to 64.5 Hz in this scenario just today. My UPS specs says it will tolerate 57-63 Hz.

    I still stand my belief that the load center should be providing smooth 120VAC / 60Hz to my loads regardless of what's going on behind the green curtain. UPSes should not even be necessary.

    And re the doomsday scenario -- yes, I agree it's VERY unlikely to happen. But the point is that even an actual Tesla pro said the possibility exists. I think that's just silly engineering again. But if a 12V jumpstart could mitigate (along with supported instructions for doing so), I guess that would be acceptable. I was just worried 120VAC from a bonafide generator would be required.
  • Ah, @gregbrew_98470014 , I see from another thread you are running 1.48.1 as well. So your UPSes must be more tolerant than mine. I have a commercial grade APC Smart-UPS 1000XL which has no issue -- but my consumer grade APC Back-UPS 850 gets heartburn.
  • IIRC, the "out of spec" frequency should be around 62Hz. Your 64.5Hz sounds too high. You might give Tesla a call (shudder) to see if it's possible to tighten yours up.
Sign In or Register to comment.