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

My grid outage frequency issue is resolved!

This site may earn commission on affiliate links.
Unless Tesla decided to do something crazy... grid-tie is grid-tie and the grid makes the grid which is voltage and frequency. The battery backed inverter creates its own mini-grid for the grid-tie inverter to connect to so it can charge it's batteries and have another power source. The grid-tie inverter just follows along. Older grid-tie inverters were on or off but newer ones can actually decrease their power slowly as the batteries get full and the AC couple inverter slowing increases 'grid' frequency.
It does look like Tesla is violating UL1741SA + CA Rule 21, where the set frequency for turning off smart-inverters should be 63Hz for 10 seconds.

It doesn't matter if there's broken inverters out there that don't respect/follow UL1741SA + CA Rule 21 at 63Hz. There's an official CA database that lists which inverters are officially certified for it.

Telsa's installer interface has a list of inverters that suggests it recognizes some form of inverter database which should flag these frequency attributes. However, for example it only lists SunPower version C micro-inverters, instead of the later Rule 21 compliant D & E versions. This means to me Tesla's may be aware of this inverter "database", but is negligently out of date.

I'm wondering if reporting Tesla is being non-compliant will help speed up getting this fixed for everybody.
 
  • Informative
Reactions: MorrisonHiker
It does look like Tesla is violating UL1741SA + CA Rule 21, where the set frequency for turning off smart-inverters should be 63Hz for 10 seconds.

It doesn't matter if there's broken inverters out there that don't respect/follow UL1741SA + CA Rule 21 at 63Hz. There's an official CA database that lists which inverters are officially certified for it.

Telsa's installer interface has a list of inverters that suggests it recognizes some form of inverter database which should flag these frequency attributes. However, for example it only lists SunPower version C micro-inverters, instead of the later Rule 21 compliant D & E versions. This means to me Tesla's may be aware of this inverter "database", but is negligently out of date.

I'm wondering if reporting Tesla is being non-compliant will help speed up getting this fixed for everybody.

There is no solution to Tesla's issue outside of my 66 Hz ‘blast’ proposal. Since not all inverters provide digital information as to their make to programming within the PW2, it must assume the ‘worst’ and hit every inverter with 66 Hz.

However, a short blast of 66 Hz followed by steady state 62.5 Hz for the 5 minute cycle allows UPSs and home automation switches to work properly. Then when the cycle restarts, most inverters will remain off at the 62.5 Hz level. If the PW2 notes a return to power, it can ‘blast’ 66 Hz for 5-10 seconds again, then drop to 62.5 Hz, rinse and repeat.

It’s the legacy inverters that don’t provide info to the PW2 that are constraining Tesla right now.
 
The new firmware allowed them to tweak and they tweaked mine to match my inverters. There was no automatic discovery of my system, just noticed that the Gateway did know my inverters as part of my poking around.

gpez, the PW2 can query many inverters as to their make and model via software - but some old legacy inverters don’t respond AND take higher Hz values to turn off.

Thus the ‘one off’ approach they have chosen where they get you to vouch for the inverter type and then tweaked the value for you.

However, it’s not automated like the ‘blast’ approach I’ve suggested to them that would work for every system without one-off tweaking. The only downside is a single blast of 66 Hz for your type of system that lasts only for 5-10 seconds.
 
There is no solution to Tesla's issue outside of my 66 Hz ‘blast’ proposal. Since not all inverters provide digital information as to their make to programming within the PW2, it must assume the ‘worst’ and hit every inverter with 66 Hz.

However, a short blast of 66 Hz followed by steady state 62.5 Hz for the 5 minute cycle allows UPSs and home automation switches to work properly. Then when the cycle restarts, most inverters will remain off at the 62.5 Hz level. If the PW2 notes a return to power, it can ‘blast’ 66 Hz for 5-10 seconds again, then drop to 62.5 Hz, rinse and repeat.

It’s the legacy inverters that don’t provide info to the PW2 that are constraining Tesla right now.
I'm saying the whole point of the commissioning is to register the correct inverter information. There does not need to be any "provide digital information" probes.

Then, if the commissioning configuration is misconfigured, or wrong, with a known certified smart-inverter not responding properly to the protocol, THEN back off to failsafe blast with 66Hz for X seconds and then the rest of what they're doing now.

That should be the proper process to handle this instead of assuming all inverters do not follow protocol. What's the point of the certification and the protocol if Tesla just goes rogue and ignores it all?
 
I'm saying the whole point of the commissioning is to register the correct inverter information. There does not need to be any "provide digital information" probes.

Then, if the commissioning configuration is misconfigured, or wrong, with a known certified smart-inverter not responding properly to the protocol, THEN back off to failsafe blast with 66Hz for X seconds and then the rest of what they're doing now.

That should be the proper process to handle this instead of assuming all inverters do not follow protocol. What's the point of the certification and the protocol if Tesla just goes rogue and ignores it all?

I agree if the firmware can positively identify all inverters and have a database as to the Hz levels they will shut off with, it will automate this issue.

However, if that inverter malfunctions and doesn’t respond correctly, there has to be the 66 Hz failsafe.

So maybe the firmware first identifies the inverter (if it can) and then uses the requisite Hz level when the grid disconnects at full PW2 SoC. However, if that doesn’t immediately work within a second or two, it need to ‘blast’ the 66 Hz value to absolutely shut down a misbehaving (or unknown) inverter.

This issue is complicated...
 
Would it help anyone else wanting to have this same procedure done to reference any other specific tickets or just say we need the inverter shutdown set to 62.5 or the like?

I’ve not done a loss of power test under full PWs yet to see how my UPSs react but want to be ready to get with TE if things don’t respond correctly.
 
  • Helpful
Reactions: NuShrike
Would it help anyone else wanting to have this same procedure done to reference any other specific tickets or just say we need the inverter shutdown set to 62.5 or the like?

I’ve not done a loss of power test under full PWs yet to see how my UPSs react but want to be ready to get with TE if things don’t respond correctly.

I'd suggest you do it. It takes 5min.
 
What takes 5 minutes? The test or the fix? And the question was, what's the best way to get TE to do it? So far it took me 10 days to get a simple question about the app's stats answered. :D

A test " to see how my UPSs react but want to be ready to get with TE if things don’t respond correctly."

Best way is to get a Tesla employee to help you out and let him open a ticket.

The fix takes about an hour....Talking all time not just pushing the change.
 
  • Like
Reactions: boaterva
Do you have a phone number for Tesla support? My installer has certainly been able to get through to them, but when I was having issues with my installer and tried to reach Powerwall support as an end-user, I was placed in a circular loop with each Tesla rep telling me to call the number that referred me to them.
 
Last edited:
Do you have a phone number for Tesla support? My installer has certainly been able to get through to them, but when I was having issues with my installer and tried to reach Powerwall support as an end-user, I was placed in a circular loop with each Tesla rep telling me to call the number that referred me to them.

I don’t unfortunately. My Tesla contact did all that for me...
 
I think all iQ Enphase inverters also have a setting that will allow them to be more analog in the event of a power outage. Instead of solar production being ON/OFF based entirely on the capacity of your powerwall and solar production. My understanding is that you can configure the enphase micro inverters to gradually adjust the solar production based on load when there is a grid outage and there powerwall is fully charged.

That's called curtailing and it's a relative decrease on power output tied to the frequency on the micro-grid, see below, page 5

https://enphase.com/sites/default/f...Considerations-AC-Coupling-Micros-Battery.pdf
 
  • Helpful
Reactions: gpez
Yep. That's it. Unfortunately, I discovered that I don't have iQ microinverters, so it's not an option for me anyway.

What are you trying to accomplish? If your end goal is to be able to turn off your micros that are ac coupled when your micro grid frequency shifts high you can always use a contactor with controller to disconnect the micros from the grid (essentially telling them the grid is down and forcing a UL1741 rapid shutdown) then when the frequency goes back down close the contactor
 
One of the guys who helped install my 4 PW Founders opened a case with them. 1.37.1 I believe.
In case it's not obvious (it's way down there), your current firmware is at the bottom of the Gateway page in the app (where it says Powerwall and how many). Clicking on it also goes to the support website for (somewhat out of date on some topics) support docs.