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

Powerwall 2 + UPS Connundrum - and solution

This site may earn commission on affiliate links.
Navigate to https://<powerwallip>/api/site_info while logged in to your TEG locally you'll get a reply that should include "grid_code_overrides" with "soc_freq_droop_config_df_max". The value there is the hz increase above baseline.

Code:
grid_code_overrides":[{"name":"soc_freq_droop_config_df_max","value":2}]}}

Is what I get, which means my max frequency is 62hz.
 
  • Helpful
Reactions: aesculus
Good Morning guys, I've been reading this and the other thread.

It seems i have a different issue (or at least didn't find many with mine).

When the PW goes into backup (no matter the charge %) all my UPS shut down immediately. I initially thought it was the ups that was too old but i recently bought another cheap one and they all behave the same. Power Outage? Instant UPS shutdown.

If i manually re-arm the UPSes everything goes fine and the UPS does use the PW supplied energy.

The problem is that i need to manually re-arm the ups, that's why i'd like to understand the reason why my ups just instantly shut down without providing any current to the attached devices. I always thought that in case of power surge or issues with the input, the ups would disconnect itself from the mainline and start providing the current from the internal batteries until Vin, Hertz etc was fine again.

Can some of you shed some light to my situation?
 
Good Morning guys, I've been reading this and the other thread.

It seems i have a different issue (or at least didn't find many with mine).

When the PW goes into backup (no matter the charge %) all my UPS shut down immediately. I initially thought it was the ups that was too old but i recently bought another cheap one and they all behave the same. Power Outage? Instant UPS shutdown.

If i manually re-arm the UPSes everything goes fine and the UPS does use the PW supplied energy.

The problem is that i need to manually re-arm the ups, that's why i'd like to understand the reason why my ups just instantly shut down without providing any current to the attached devices. I always thought that in case of power surge or issues with the input, the ups would disconnect itself from the mainline and start providing the current from the internal batteries until Vin, Hertz etc was fine again.

Can some of you shed some light to my situation?
I've not heard of this before. What brand and model UPS are you using? What is the frequency that the Powerwalls are supplying to the house?
You are correct that if the grid voltage or frequency is out of bounds, the UPS should go to battery, not shut down the load.

The only time I have seen this behavior is when the battery cannot support the cutover load. I couldn't figure out why my server was randomly shutting down. It turns out that the self-test was not failing with an alarm to indicate the failure, it was failing so bad that it was shutting down the load and not showing any error afterwards. Replacing the battery restored normal operation. I realize that this is not likely to be your problem since you have the same symptoms across multiple units.
 
I've not heard of this before. What brand and model UPS are you using? What is the frequency that the Powerwalls are supplying to the house?
You are correct that if the grid voltage or frequency is out of bounds, the UPS should go to battery, not shut down the load.

The only time I have seen this behavior is when the battery cannot support the cutover load. I couldn't figure out why my server was randomly shutting down. It turns out that the self-test was not failing with an alarm to indicate the failure, it was failing so bad that it was shutting down the load and not showing any error afterwards. Replacing the battery restored normal operation. I realize that this is not likely to be your problem since you have the same symptoms across multiple units.
So, my UPS is a Legrand -> Keor PDU single-phase VFD UPS - 800VA 480W - 8 sockets DE/IT for FR/DE/IT sockets - autonomy up to 15 min - 3 103 32 - Legrand

it should support my load as the management interface shows 36% load attached to the ups. Yesterday I tried to pull the plug of the ups to check once again if the "instant shutdown" was due to powerwall or something else. It turns out that it instantly shat down.

So I guess the battery is also defective as the the ups linked was bought a week ago.

The other ups was a refurbished one and could very well have a died battery.

I just ordered another ups as a replacement for the legrand to check if it works out of the box and will test if it will keep working tthrough the next few days before trying a "powerwall backup test"
 
Finally, after 7 weeks and 3 phone calls, I got the frequency shift overridden. Yesterday I got a lady on the phone who contacted a tech ("Robert Jackson") live, and got it done as we were talking. New entry in the api/site_info log:

grid_code_overrides:[{"name":"soc_freq_droop_config_df_max","value":2.5}]

Only glitch was that I had the grid disconnected to show them the freq when he reset the Powerwall without notifying me, so I lost power to the house for a couple minutes.
 
I recently got 2 APC BVN650M1 UPSs. They are going into backup mode and refusing to charge the battery when operating off the Powerwalls and eventually go dead when operating off the Powerwalls. They are passing power through the surge outlet. My solar panels are covered with snow and the Powerwalls are down to about 80% so frequency shift shouldn't be a problem. They are supposed to allow 50/60 Hz +/- 5Hz and 92 Vac to 139 Vac before going into backup mode. I've tried resetting them but they still won't charge. My LED lights are sensitive to power/frequency fluctuations and are fine right now.

Has anyone else encountered this?
 
The UPSs may be defective. OTOH, they may sense the "unstable" frequency when the Powerwalls take over, and properly switch over. From what I've seen on my system, they shift frequency immediately when the grid goes down, and then slowly reduce to 60 Hz. They are small units, so with any significant load they will run down quickly.
 
The UPSs may be defective. OTOH, they may sense the "unstable" frequency when the Powerwalls take over, and properly switch over. From what I've seen on my system, they shift frequency immediately when the grid goes down, and then slowly reduce to 60 Hz. They are small units, so with any significant load they will run down quickly.
I called Tesla. They said my Powerwalls were putting out 65 Hz and I can expect the Powerwalls to continue to do so until they reach about 80% (I was at 83% when I called). They have submitted a Tier 2 support request to lower the frequency.
 
Last edited:
  • Informative
Reactions: navguy12
Just ended a 2 hour outage in SoCal (Edison). PW was at 100% when it started, then both APC UPS's kicked in, but died within 15 and 20 minutes respectively. I do feel a lot more informed after reading this thread, because I was really pissed before wondering why these two devices stopped pulling from the wall that was powered by the Tesla battery.

Anyways, suggestions for 2 Eatons to replace my APC units would be appreciated - one unit to power a router and NAS, the other a desktop. Thanks!
 
Just ended a 2 hour outage in SoCal (Edison). PW was at 100% when it started, then both APC UPS's kicked in, but died within 15 and 20 minutes respectively. I do feel a lot more informed after reading this thread, because I was really pissed before wondering why these two devices stopped pulling from the wall that was powered by the Tesla battery.

Anyways, suggestions for 2 Eatons to replace my APC units would be appreciated - one unit to power a router and NAS, the other a desktop. Thanks!
I got the

CyberPower CP1500PFCLCD PFC Sinewave UPS System, 1500VA/1000W, 12 Outlets, AVR, Mini Tower​

But it was only $150 on a flash sale. A smaller one would probably be OK. You just need to figure out how long you need them to run all your devices. It also will step aside at 62.5 hz which if you have your PWs throttled back should last forever.
 
  • Like
Reactions: navguy12
Just ended a 2 hour outage in SoCal (Edison). PW was at 100% when it started, then both APC UPS's kicked in, but died within 15 and 20 minutes respectively. I do feel a lot more informed after reading this thread, because I was really pissed before wondering why these two devices stopped pulling from the wall that was powered by the Tesla battery.

Anyways, suggestions for 2 Eatons to replace my APC units would be appreciated - one unit to power a router and NAS, the other a desktop. Thanks!
To me, the main take-away from this thread is that you should call Tesla to have them adjust your system frequency down to 62.5Hz. Then you won't need to replace your APC UPS units. That is a far better solution than allowing your micro-grid to run at a high frequency and buy a new UPS that is tolerant of that extreme frequency.
 
To me, the main take-away from this thread is that you should call Tesla to have them adjust your system frequency down to 62.5Hz. Then you won't need to replace your APC UPS units. That is a far better solution than allowing your micro-grid to run at a high frequency and buy a new UPS that is tolerant of that extreme frequency.
I wonder why Tesla doesn't just put the default at 62.5Hz, given the number of cases they probably get on it. I sent an email to have mine changed, but probably going to take a few weeks before anyone actually gets to it. Especially for new PW+ systems installed with no other solar systems, not sure why they couldn't just change the default.
 
  • Like
Reactions: navguy12
I wonder why Tesla doesn't just put the default at 62.5Hz, given the number of cases they probably get on it. I sent an email to have mine changed, but probably going to take a few weeks before anyone actually gets to it. Especially for new PW+ systems installed with no other solar systems, not sure why they couldn't just change the default.
I agree. The number of solar inverters that won't shut down at 62.5Hz is probably very small. They should just make the default 62.5 and make the installers test to make sure the solar goes offline at that frequency. Only if it's a 3rd party inverter that doesn't shut down would the value be increased.
 
I agree. The number of solar inverters that won't shut down at 62.5Hz is probably very small. They should just make the default 62.5 and make the installers test to make sure the solar goes offline at that frequency. Only if it's a 3rd party inverter that doesn't shut down would the value be increased.
Are there any inverters that won't shut down at 62.5 Hz? I know my string inverter will shut down at 60 +/- 0.5 Hz and I thought that smart inverters were less than 62 Hz. I thought these were compliance requirements.
 
  • Like
Reactions: navguy12