Thanks for the contribution. That's the first time I've heard that partial curtailment of solar production is working in a Powerwall setup. With my system (SolarEdge inverter) it appears to be all or nothing.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Powerwall/Solar Edge behavior in power outage
- Thread starter blulegend
- Start date
Thanks for the contribution. That's the first time I've heard that partial curtailment of solar production is working in a Powerwall setup. With my system (SolarEdge inverter) it appears to be all or nothing.
This is a good point. If the temperature of the Powerwall is an issue for charging during grid outages, then the software should apply the same principle - allow as much solar production as the Powerwall can take. I'm expecting a Tesla site visit in the next couple of weeks. We'll see what they say about our Powerwalls not being able to charge during grid outages.
The problem is that your typical inverter is either on or off and nothing in between. Unless you have a dump load you can control, you're stuck. A few inverters will taper their output gradually as frequency climbs, but not many.
If your PW2 is connected to your home LAN, you can check yourself what frequency the PW2 is seeing/pushing.
https://(PW2 IP address)/api/meters/aggregates will give you:
site
last_communication_time "2019-03-04T13:16:18.279886213+11:00"
instant_power -2172.479960680008
instant_reactive_power -473.01000022888184
instant_apparent_power 2223.377529766984
frequency 49.95000076293945
energy_exported 2067486.7526071235
energy_imported 4121237.4017737904
instant_average_voltage 719.4400024414062
instant_total_current 0
i_a_current 0
i_b_current 0
i_c_current 0
timeout 60000000000
last_communication_time "2019-03-04T13:16:18.279886213+11:00"
instant_power -2172.479960680008
instant_reactive_power -473.01000022888184
instant_apparent_power 2223.377529766984
frequency 49.95000076293945
energy_exported 2067486.7526071235
energy_imported 4121237.4017737904
instant_average_voltage 719.4400024414062
instant_total_current 0
i_a_current 0
i_b_current 0
i_c_current 0
timeout 60000000000
and same for the solar, house load lines.
Thanks. I checked this out. During a simulated outage while producing solar with a battery charge of 78%, my frequency was 65.94. When I have grid power, it's 60. I chatted again with my installer and they said Tesla will fix this issue in their next software release. Currently I'm on 1.32.0
dortor4ev
Member
1.33 is out - and i’d be surprised if your installer knew what was or was not fixed in any upcoming updates - i do however believe it could be fixed in a future update - but claiming knowledge of a specific fix in a specific update is in consistent with my experience with Tesla and the tech industry in general..
My installer is working directly with Tesla support to try and resolve the issue since they have several customers who complained during a recent outage. Tesla mentioned the fix in a future update...which tells me they are at least aware of the issue and working to resolve. Once my PW updates to 1.33, I'll test again and report my results. If it's not fixed, I try again with 1.34.
How long did you leave the system with the simulated outage? It seems regardless of the charge state, the PW2 will raise the frequency to get the inverters to stop - but if the charge is less than ~96%, then it should ramp the frequency down over the next 5 - 10 minutes or so and the inverter should re-start. Sometimes an inverter enforces a 5 - 10 minute timeout before it will restart, just to let things settle down.
If you had the breakers off for at least 15 minutes and nothing changed, then yes theres a problem. But if you flipped the breakers, saw the frequency rise, and flipped the breakers back within a minute or two (I was guilty of premature reconnection recently
), perhaps try again, and leave it for half an hour before reconnecting.How long did you leave the system with the simulated outage? It seems regardless of the charge state, the PW2 will raise the frequency to get the inverters to stop - but if the charge is less than ~96%, then it should ramp the frequency down over the next 5 - 10 minutes or so and the inverter should re-start. Sometimes an inverter enforces a 5 - 10 minute timeout before it will restart, just to let things settle down.
If you had the breakers off for at least 15 minutes and nothing changed, then yes theres a problem. But if you flipped the breakers, saw the frequency rise, and flipped the breakers back within a minute or two (I was guilty of premature reconnection recently
My issue is that even if this is true and it will settle down after 10 minutes, there seem to be a lot of electronics that don't do well with a frequency as high as 66 Hz. My fireplace fan stops, my LED lights blink or don't work at all, UPSs beep, dogs go crazy, and who knows how other motor-based appliances like refrigerators and freezers will react to that high of a frequency. I have seen frequencies at 66 Hz with the SOC less than 50% multiple times or even at 8%. I understand at a SOC of 97% or if the solar is putting out more than the PW2 can take, but the current situation seems just not correct. I haven't tested the system yet with 1.33. Perhaps it is better. I certainly hope so.
How long did you leave the system with the simulated outage? It seems regardless of the charge state, the PW2 will raise the frequency to get the inverters to stop - but if the charge is less than ~96%, then it should ramp the frequency down over the next 5 - 10 minutes or so and the inverter should re-start. Sometimes an inverter enforces a 5 - 10 minute timeout before it will restart, just to let things settle down.
If you had the breakers off for at least 15 minutes and nothing changed, then yes theres a problem. But if you flipped the breakers, saw the frequency rise, and flipped the breakers back within a minute or two (I was guilty of premature reconnection recently
I did two simulated outages. One for 1hr 33 minutes .The other for 25 minutes.
Interestingly, Tesla called yesterday to reschedule our on-site Powerwall service for a later date. The technician says that he wants to allocate a full eight hours to look into our issue! Glad to see Tesla taking this seriously. Even when I've had to bug them a bit, Tesla has always come through for us in the end, and I expect this issue to be no different.
This morning, a Tesla Energy technician came to our home to investigate the issue of our Powerwalls not charging from solar during grid outages. I purposely discharged the Powerwalls to about 55% prior to his arrival, by setting the backup reserve level down to 10%. With high clouds, we may have been getting about 1 kW of power from the solar panels.
The technician simulated a grid outage (by flipping the main breaker) and confirmed that our solar inverters were not turning back on. He upgraded the firmware from 1.32.0 to 1.34.2, but this did not seem to make a difference. I suggested that he use his multimeter to check our line frequency during a simulated outage, and sure enough, he measured 63 Hz which is too how for the solar inverters. As expected, this also explains why our induction stove won't function on backup power.
The technician spoke with Tesla's diagnostics team and was told that there's a known firmware issue with line frequency during outages. It appears that we'll have to wait for software fixes. He encouraged me to continue to follow up with Tesla, using my case number and job number, and expressed optimism that there will be a resolution. I emphasized that being able to charge from solar during outages is important to us, given our semi-rural mountain location.
The technician simulated a grid outage (by flipping the main breaker) and confirmed that our solar inverters were not turning back on. He upgraded the firmware from 1.32.0 to 1.34.2, but this did not seem to make a difference. I suggested that he use his multimeter to check our line frequency during a simulated outage, and sure enough, he measured 63 Hz which is too how for the solar inverters. As expected, this also explains why our induction stove won't function on backup power.
The technician spoke with Tesla's diagnostics team and was told that there's a known firmware issue with line frequency during outages. It appears that we'll have to wait for software fixes. He encouraged me to continue to follow up with Tesla, using my case number and job number, and expressed optimism that there will be a resolution. I emphasized that being able to charge from solar during outages is important to us, given our semi-rural mountain location.
SSonnentag
埃隆•馬斯克
So when the power is out, you can draw power from the Powerwalls, but the solar panels are worthless as the inverters the panels are connected to are shut down due to incorrect AC frequency? This is very bad. This also means that your power will be limited to what the Powerwalls can provide instead of the combination of what your solar panels are producing and what the batteries can provide. I suspect my system performs the same way. I have 12.5 kW of solar panels, but when I had the power cut for an hour to test the Powerwalls, I was not getting any power from the panels, only the Powerwalls. Unacceptable.
SSonnentag
埃隆•馬斯克
This would also explain the errors our induction stove was throwing when testing our system as well. Please keep us posted if you receive any further updates. I'll keep watching this thread.
Yes, I agree that this is unacceptable. If you haven't done so already, I would recommend contacting Tesla. While they are aware of the issue, it can't hurt for them to continue to hear from Powerwall owners that we expect a resolution. Being able to power our home during extended grid outages was a key Powerwall selling point.
It is too bad that you had to wait for a service person to show up and tell you what you and a number of others on this forum already knew for some time now. My installed tells me that my report of this problem through the installer was the first they had heard of it, but there were others now reporting the same thing. How to you say BS politely to the installer you need on your side?
SSonnentag
埃隆•馬斯克
I sent an e-mail to [email protected] right after posting here. Hopefully they'll act sooner with more reports. This was the first I'd heard of the problem, so I tested our system out and got the same results as posted here.
woferry
Member
I'll file a report on mine as soon as I can demonstrate it. Unfortunately my final inspection was today and it failed for one issue, so it'll be another 2 weeks to the next city inspector appointment, then however long it takes for PG&E PTO.
I had the 63Hz issue when I had a real outage in November. I reported it to Tesla by phone and opened a case. While the outage was happening, they were able to remotely force it back to 60Hz and my Enphase inverters came back online. I have not had a significant outage since then. However, yesterday I went home for lunch and did a test by turning off my main breaker. During the 1hr 5 minutes I was home, my system was stable at 60.0Hz and 75%-83% SOC and the solar generated continuously except the 1 minute immediately after I shut off the main breaker. Tesla has never updated me on my open case.
Similar threads
