PowerWall Cold Start without Grid Power

[Story]

I was wondering about this note from (insert name) above:

"Since the Powerwall inverter's max sustained value is 5.0kw on a sunny day where the solar microinverters are maxed the Powerwall will shut them down to reduce the input below the inverter maximum to protect it, even if the battery needs to charge."

Where is this documented? I'm curious because I wonder if my solar array will overwhelm my new 2 PW2 system. Currently, my aging solar system seems to peak at 5.3KW, but I'm about to replace over half the panels, and expect to be able to generate 10KW with the new panels. (52 x 165w array originally, now 44x165 + 8x265, going to 12x165 + 24x305 + 8x265 soon.)

If I generate 10KW from solar, and have 2 PW2, I should never face the PW inverters being overloaded, yes, because I have 2x5-7KW inverters running in parallel?

 

[shs1]

I was wondering about this note from (insert name) above:

"Since the Powerwall inverter's max sustained value is 5.0kw on a sunny day where the solar microinverters are maxed the Powerwall will shut them down to reduce the input below the inverter maximum to protect it, even if the battery needs to charge."

Where is this documented? I'm curious because I wonder if my solar array will overwhelm my new 2 PW2 system. Currently, my aging solar system seems to peak at 5.3KW, but I'm about to replace over half the panels, and expect to be able to generate 10KW with the new panels. (52 x 165w array originally, now 44x165 + 8x265, going to 12x165 + 24x305 + 8x265 soon.)

If I generate 10KW from solar, and have 2 PW2, I should never face the PW inverters being overloaded, yes, because I have 2x5-7KW inverters running in parallel?

I am not sure where it is documented, but when connected from the grid, my single PW2 will typically limit its charging (from solar) to 3.4 kW and only occasionally goes higher. That is why with my test today I was impressed that it charged at a full 5 kW for an extended time. My two solar arrays can put out 11 kW but in the summer because of temperature derating, typically only put out about 9 at noon. I have seen 10-11 occasionally when it is colder and sunny. A sunny February morning is a good time to blow your line fuses!

The other factor to consider is that the house is always drawing some power and so if it is drawing 1 kW that is 1 kW less that the PW2 has to take. Once I get a second PW2, I am not at all concerned that my 11 kW solar will cause over charging rate troubles, especially during the summer, as my house seems to always draw about a kW or sometimes a lot more!

 

[mongo]

I was wondering about this note from (insert name) above:

"Since the Powerwall inverter's max sustained value is 5.0kw on a sunny day where the solar microinverters are maxed the Powerwall will shut them down to reduce the input below the inverter maximum to protect it, even if the battery needs to charge."

Where is this documented? I'm curious because I wonder if my solar array will overwhelm my new 2 PW2 system. Currently, my aging solar system seems to peak at 5.3KW, but I'm about to replace over half the panels, and expect to be able to generate 10KW with the new panels. (52 x 165w array originally, now 44x165 + 8x265, going to 12x165 + 24x305 + 8x265 soon.)

If I generate 10KW from solar, and have 2 PW2, I should never face the PW inverters being overloaded, yes, because I have 2x5-7KW inverters running in parallel?

It's not really overloaded in the sense th hst the PW will be damaged. The PW shifts the frequency(or allows the frequency to shift) when it is converting the maximum it can. Even if your solar is less than PW inverter capacity, it will curtail production when it is full.

This is similar to the way the grid works, overproduction increases frequency and voltage, underproduction reduces them. This regulation comes about due to the way AC induction motors, and generators, operate.

What will be interesting to see what happens to the grid stability as more devices have power converters on their inputs which pull more current at lower voltages.

 

[gpez]

I am not sure where it is documented, but when connected from the grid, my single PW2 will typically limit its charging (from solar) to 3.4 kW and only occasionally goes higher. That is why with my test today I was impressed that it charged at a full 5 kW for an extended time. My two solar arrays can put out 11 kW but in the summer because of temperature derating, typically only put out about 9 at noon. I have seen 10-11 occasionally when it is colder and sunny. A sunny February morning is a good time to blow your line fuses!

The other factor to consider is that the house is always drawing some power and so if it is drawing 1 kW that is 1 kW less that the PW2 has to take. Once I get a second PW2, I am not at all concerned that my 11 kW solar will cause over charging rate troubles, especially during the summer, as my house seems to always draw about a kW or sometimes a lot more!

In case folks are looking the inverter limits are documented as part of the Powerwall 2 specs, check the Real Power max continuous and peak values. @shs1 look at the (1) note: "1 Values provided for 25°C (77°F), 3.3 kW charge/discharge power." My interpretation of that is that 3.3kw is the optimal charge/discharge rate under normal circumstances.

If the grid is unavailable to dump excess power and the PV output minus the house consumption exceeds the capabilities of the Powerwall(s) inverter(s) then they will work to either shut down the PV or taper it down to manageable levels. When the grid is up though the Powerwall can charge at the inverter max and allow the grid to absorb the rest.

 

[Story]

In case folks are looking the inverter limits are documented as part of the Powerwall 2 specs, check the Real Power max continuous and peak values. @shs1 look at the (1) note: "1 Values provided for 25°C (77°F), 3.3 kW charge/discharge power." My interpretation of that is that 3.3kw is the optimal charge/discharge rate under normal circumstances.

If the grid is unavailable to dump excess power and the PV output minus the house consumption exceeds the capabilities of the Powerwall(s) inverter(s) then they will work to either shut down the PV or taper it down to manageable levels. When the grid is up though the Powerwall can charge at the inverter max and allow the grid to absorb the rest.

Thanks for the pointer @gpez -- so as I read the specs, when off the grid (no Utility power) I can only expect 3.3KW max charging per Powerwall, or 6.6KW for two powerwalls, before the PW's start reducing solar input.

But when I'm on the grid, I can expect the PW2's to charge at "real power (max continuous)" rates? Thus it's 5KW * 2 = 10KW?

Thanks!

 

[Story]

My two solar arrays can put out 11 kW but in the summer because of temperature derating, typically only put out about 9 at noon. I have seen 10-11 occasionally when it is colder and sunny. A sunny February morning is a good time to blow your line fuses!

@shs1 - have you actually blown your line fuses?

 

[shs1]

@shs1 - have you actually blown your line fuses?

Yes, a couple of times and always when it is sunny and very cold, i.e. a morning in February. I monitor solar output from both inverters now as I didn't detect the blown fuse for a couple of months the first time and lost some valuable solar output.

 

[cwied]

Thanks for the pointer @gpez -- so as I read the specs, when off the grid (no Utility power) I can only expect 3.3KW max charging per Powerwall, or 6.6KW for two powerwalls, before the PW's start reducing solar input.

But when I'm on the grid, I can expect the PW2's to charge at "real power (max continuous)" rates? Thus it's 5KW * 2 = 10KW?

Thanks!

Actually you get 10 kW both on and off-grid. The footnote just refers to the efficiency number. Two Powerwalls are capable of absorbing 10 kW of excess power from solar and can provide 10 kW continuous, 14 kW peak when off grid. I haven't seen my Powerwalls go over 10 kW when on grid, presumably because there's no need to stretch to cover the load if the grid can just make up the difference.

 

[gpez]

Thanks for the pointer @gpez -- so as I read the specs, when off the grid (no Utility power) I can only expect 3.3KW max charging per Powerwall, or 6.6KW for two powerwalls, before the PW's start reducing solar input.

But when I'm on the grid, I can expect the PW2's to charge at "real power (max continuous)" rates? Thus it's 5KW * 2 = 10KW?

Thanks!

Not quite how I read the spec. The (1) note about the 3.3kw charge/discharge is attached to the "Round Trip Efficiency1,3: 90%" line while the Real Power Max and Real Power Continuous are 5kw and 7kw, respectively. I'm interpreting that to mean that the optimal inverter power level is 3.3kw but it can go up to 7kw for brief moments and 5kw forever. Likely at those higher power levels the inverter efficiency drops so the Powerwall is configured to charge/discharge at 3.3kw when possible, such as during normal grid operation. Of course how it determines the best charge/discharge rates is only known by Tesla and they don't publish it best to my knowledge.

 

[Kacey Green]

Yes, a couple of times and always when it is sunny and very cold, i.e. a morning in February. I monitor solar output from both inverters now as I didn't detect the blown fuse for a couple of months the first time and lost some valuable solar output.

Same blew it on the first full sunny day after getting more panels (someone overlooked the fuse)

 

[Musterion]

Hi all, I’m currently under yet another planned PG&E all-day power outage and have gateway software 1.37.1. Although previous power outages were handled fine, this one started with Powerwalls at 50%, but Gateway immediately started outputting 61.3Hz (at 240Vac) and thus never enabled Solar to come back online. 4x Powerwalls can easily charge at even my max solar output (and now is not near max anyway) and in past this scenario triggered solar to come online immediately after grid failure and begin to charge powerwalls. I called Tesla and they said all their agents are busy and best they can promise is someone gets back to me within 24 hours (or to try back when the night team arrives to support the Aussies...). But since this outage will extend through the night based on previous experience, I would like to get the solar kicked on during the day so we can make it through the night.

Back to the point of this thread. This problem mirrors some of the problems detailed by @MorrisonHiker but I did not see that a cold-start method was identified there in the end. I am tempted to reboot the gateway but Tesla said that system would probably not restart if grid is out. I have read through this whole thread but nobody has given a tested “cold-start” method. There are a lot of questions on why one would want to do this. Well, this is one example. I don’t want to reboot if I can’t restart the Powerwalls, because they are for time being powering the house, but Gateway seems “stuck” at outputting exactly 61.3 Hz and I’d like to kick it back to how it operated before which was to work nicely with the solar system during grid failure. So, can anyone confirm a “cold-start” will work when rebooting gateway during grid outage, having 50% charge, and during day when solar can be enabled and powerwall capable of accepting all solar power?

 

[MorrisonHiker]

Hi all, I’m currently under yet another planned PG&E all-day power outage and have gateway software 1.37.1. Although previous power outages were handled fine, this one started with Powerwalls at 50%, but Gateway immediately started outputting 61.3Hz (at 240Vac) and thus never enabled Solar to come back online. 4x Powerwalls can easily charge at even my max solar output (and now is not near max anyway) and in past this scenario triggered solar to come online immediately after grid failure and begin to charge powerwalls. I called Tesla and they said all their agents are busy and best they can promise is someone gets back to me within 24 hours (or to try back when the night team arrives to support the Aussies...). But since this outage will extend through the night based on previous experience, I would like to get the solar kicked on during the day so we can make it through the night.

Back to the point of this thread. This problem mirrors some of the problems detailed by @MorrisonHiker but I did not see that a cold-start method was identified there in the end. I am tempted to reboot the gateway but Tesla said that system would probably not restart if grid is out. I have read through this whole thread but nobody has given a tested “cold-start” method. There are a lot of questions on why one would want to do this. Well, this is one example. I don’t want to reboot if I can’t restart the Powerwalls, because they are for time being powering the house, but Gateway seems “stuck” at outputting exactly 61.3 Hz and I’d like to kick it back to how it operated before which was to work nicely with the solar system during grid failure. So, can anyone confirm a “cold-start” will work when rebooting gateway during grid outage, having 50% charge, and during day when solar can be enabled and powerwall capable of accepting all solar power?

Our system didn't come back up (and stay up) until the grid came back after being out for 46 hours. They did come out and adjust the inverters but I think they might've made them too sensitive as I've seen them restarting when hitting just 60.5 Hz. We need to do another simulated outage and confirm things will start back up correctly.

I think you should be able to get the system going again if the Powerwalls output the correct frequency. If the frequency is too high, then you'll experience the off/on loop that I had all day long.

 

[Dan123]

So, can anyone confirm a “cold-start” will work when rebooting gateway during grid outage, having 50% charge, and during day when solar can be enabled and powerwall capable of accepting all solar power?

I ran the powerwalls down to below 10%. They turned off and stayed off for 30 minutes. Then they tried to restart, but shut down immediately again (probably because of some large load).

I did the following:
1. Turned off all of the load circuit breakers, so there is nothing drawing power from the PWs. The only breaker that was on - solar.
2. Turned off the circuit breakers for the PWs.
3. Turned off the 1/0 switch on the Powerwall. Turned on back the switch.
4. Turned on the circuit breakers for the PWs.
5. Waited 5 minutes, and the solar came back on.

You solar inverter should have a setting (grid profile) in which you can set it to restart in 5 seconds instead of waiting for 300 seconds. That will improve PW restarting capability.

 

[Musterion]

@MorrisonHiker @Dan123
Thanks guys — you beat Powerwall support response time by a long shot!! Telsa NA telling me to wait for the Australian support team to wake up, seriously??

In the meantime I tried to change various settings to force charge, and also cut and reset solar breakers, etc. with no success. Everything short of rebooting Gateway which is the last resort. I’ve cut most demand in case I can ride out the solar outage and hope PG&E restores power before I run out of juice. (In past few outages, unfortunately PG&E went ~8-9 hours overtime, and restored at ~2am instead of 5pm). I’ll track usage and if I’m feeling lucky, I might try the protocol of @Dan123 before sun goes down. Thanks!

 

[joebob73]

In reading this thread, should my single Powerwall (which is hooked up to solar) be charging at 3.3kw or 5kw? Currently it is charging from solar at 3.3kw and I plan to do an off grid test to see if it will bump to 5kw when 100% off grid.

 

[Ampster]

Currently it is charging from solar at 3.3kw and I plan to do an off grid test to see if it will bump to 5kw when 100% off grid.

It would depend on your loads and solar output. Going off grid most likely wont change that unless your CT also reads your non essential loads. In that case the non essential loads wont be powered by the Powerwall, leaving more PV output available to charge the Powerwall.

 

[woferry]

My single PW is typically fully-charged before my solar is producing >5kW, but if the morning gets a cloudy start then suddenly clears-up, my PW sits at 5.0kW consumption until it gets around 90-95%, then starts to ramp-down. Here's an example.

 

[mongo]

In reading this thread, should my single Powerwall (which is hooked up to solar) be charging at 3.3kw or 5kw? Currently it is charging from solar at 3.3kw and I plan to do an off grid test to see if it will bump to 5kw when 100% off grid.

Does you solar setup have 5kW of extra generation after subtracting all the loads on the backed up side of the Gateway?

 

[joebob73]

It would depend on your loads and solar output. Going off grid most likely wont change that unless your CT also reads your non essential loads. In that case the non essential loads wont be powered by the Powerwall, leaving more PV output available to charge the Powerwall.

My solar roof generates 10kw peak and I understand that the maximum the Powerwall will be able to receive is 5kw. I ask about the 3.3kw vs the 5kw because some posts have suggested that the Powerwall is optimized for 3.3kw and thus does so under normal on-grid operation. However, it is spec'd to absorb 5kw when offgrid. I wanted to confirm if that is true and if anyone has observed this in operation. Alternatively, should it be absorbing 5kw both on/off grid?

 

[joebob73]

Does you solar setup have 5kW of extra generation after subtracting all the loads on the backed up side of the Gateway?

Yes, 10kW and my normal usage is about 0.7kW. @woferry's post seems to indicate that 5kw while on grid should be normal.