I think the point was that if the PWs are say 50% full when power goes out, in principle, if the PV generation is less than current home use plus PW charging capacity, generation could potentially continue uninterrupted, even when the grid is lost. However, it sounds like Tesla is apparently shutting down PV production even when these conditions are met (this is not something I have tested personally.) PV will resume 5 minutes later, so it is admittedly not the end of the world, but it is perhaps unexpected.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Powerwall 2 - Outage “Switchover” Delays?
- Thread starter Freakyguy666
- Start date
I was not aware that PV generation was shutdown unless all of the places that the PV power could go (i.e. the Powerwalls since the other sink, the grid, is gone) were full. But, it is something I will test when Tesla lowers the shutdown frequency to support my UPS units.
So, this was bugging me as I "thought" I had tested this for myself but I wanted to test it again on my own system to see what result I got.
I tested this on my own system this morning, and will need to retract what I was saying previously, as my PV definitely did remain producing through me throwing the main breaker.
I tested this at 7:45am my time. My 2 powerwalls are at 30% charge, and my PV was producing 2.1kW while the home was consuming approximately 1kW. I threw the main breaker, and my PV stayed on and continued to produce. Im actually "off grid" right now, while I send this email.
Also, my inverters are old ones from 2015 that do not have any sort of "ramp" feature to them, they are either on or off (they are from a company called ABB, which solar city installed in 2015 on my home). My inverters dont even have ethernet available on them.
Anyway, this may be like many other things with powerwall / electrical stuff in that "it depends", but I thought I would circle back for OP and say I tested it on my own system, and it indeed kept producing PV as my powerwalls + home load were absorbing the PV.
Interesting, I wonder if there is a way to define the conditions where the PV wont turn off?
I just got my own order in, so sometime next year I will be able to test my own installed Powerwalls!
I had the same experience. If my powerwalls are fully charged and I turn off the service disconnect to simulate an outage, then the inverters go off, as expected. If my powerwalls are not fully charged and I turn off the service disconnect to simulate an outage then the inverters stay on.
Other people have reported that their inverters will always shut off in an outage. My personal theory on this is that in a simulated outage it’s a pretty clean break and the inverters can probably easily ride through it. Obviously when we test for this all we can do is turn off the service disconnect. But in a real outage it’s much more likely that there may be a period of dirty power as the outage happens. I suspect that this is why sometimes the inverters will turn off during an outage even when the powerwalls aren’t fully charged. It’s not that the system is telling the inverters to turn off, but the dirty power that caused the outage also caused the inverters to shut down.
Congrats. That’s great news
. Are you doing the install yourself?They are certainly designed to be able to do that. However I think we’re talking about two separate things here.
-When you are off grid and the powerwalls are fully charged the inverters have to shut off. It would be dangerous for them to continue operating with nowhere for the excess power to go. If the grid is down and it’s a sunny day then likely you will get into a cycle with them going on and off, but this is pretty unavoidable. However, even this isn’t going to be an all day every day type thing. It will only happen when the grid is down, and only when the grid is down during the middle of the day when the powerwalls are charged and there is a lot of solar production.
-But the second thing, and the thing I was referring to, is the inverters shutting off when an outage happens. This may be avoidable in some cases, but unless you are getting outages all the time it’s not going to cause the inverters to continually cycle on and off over the duration of a day. It’s going to be a one time event that could potentially happen at the start of an outage, but it will not be happening frequently and I don’t believe it’s worth being concerned about.
This shouldn’t happen continuously. As I mentioned recently in another thread, when you are off grid and you get into the cycle where there is excess solar production so the inverters are being turned on and off the powerwalls will slowly raise the frequency as their state of charge increases. It will start at 60.1, then 60.2, etc. Modern inverters shut off at 60.5Hz, so 60.5Hz is the highest the frequency will get during that cycle. That should be low enough not to bother UPS’s or other devices.
The only time you will hit that max frequency is if there is a power outage when the powerwalls are fully charged. At that point the frequency will be raised to the max (65Hz by default, maybe 62.5Hz if tesla has lowered it). This could upset the UPS’s and make them beep. But then once the SOC lowers and it gets into the cycle the frequency will only hit a max of 60.5 and the UPS’s won’t have a problem anymore.
Congrats. That’s great news
Thanks! Thank SGIP for that win too!
I will have the company I work for do it, so yes sort of. I typically don't do field installs but in this case I have to get my hands dirty.
I will need to do a bit of trenching and preparing a concrete pad. Also, some of the prep work of installing the HPWC, and the new 400A service panel.
Just because something is “able” to do something doesn’t mean it is ideal.
Well, then you’ll be shocked to know that this was an all day issue as the 8 hour outage caused the PV to go ON & OFF at least 10 Times during the course of the day.
Again, it happens. Frequently.
Multiple outages in my area resulting in all day events causing the PV to cycle On/Off a dozen times.
Not sure if this is correct.
I believe 65Hz is the default unless you call Tesla and request it be lowered to resolve UPS issue.
Again, based on numerous posts in these forums, the frequency is 65Hz unless requested to be lowered.
If what you wrote were true, then why are there hundreds of people who have the issue with their UPS continuously during an outage when PW shuts down the PV (due to excess energy having no place to go) ?
Thanks for admitting being wrong. It helps establish the baseline.
I’m curious if the PV production would be REDUCED via the Gateway/PW system immediately following the outage.
Perhaps you could throw the breaker during a period when PV is producing closer to peak power (with PWs at 30%) and see if the output from the PV takes a sudden dip as others have suggested?
I’m still not sure that we’re talking about the same thing here. If what you’re referring to is the PV going on and off as the powerwalls get charged, then no, that’s not shocking and that’s expected behavior.
If you’re saying that the PV turned on and off at least 10 times during an 8 hour outage when the powerwalls were not fully charged, then that is very unusual and definitely not expected behavior.
The PV maybe have shut off once at the start of the outage, for reasons discussed above, but then should have stayed on until the powerwalls were fully charged. If that is not what happened for you, then something is wrong.
Again, if the PV cycling is due to the powerwalls being fully charged, then this is expected behavior and necessary. If you have multiple outages in the same day, then the PV may cycle at the beginning of each outage. But frankly, if you’re having multiple outages that frequently then you need to work with your power company to figure out what’s wrong.
This may need a little more investigation. I posted based on my experience, but someone else mentioned that he saw different results. Certainly from my understanding of how the system operates, and my own observations of my system, during off grid operation I don’t believe that the frequency should get any higher than 60.5 (or whatever the cut off frequency that the inverter is set to) no matter what the maximum frequency is set to on the powerwalls. But it’s possible that other systems behave differently than mine or I am misunderstanding how it works.
I’d love to see others do some tests with their systems to see what results they get.
————
Just to be clear though, I would absolutely expect your PV to cycle multiple times during an extended outage if your powerwalls are fully charged or become fully charged during the outage.
Other than that, and perhaps once at the start of the outage your PV should not cycle at all.
I was simply saying that it hardly seems worth worrying about that one potential cycle that may or may not happen at the start of the outage. But maybe I misunderstood and that’s not the only cycle that you were concerned about.
Recall that within your post that I was replying to, you wrote in effect that frequent on/off of the PV is not an issue. My point was that it is not ideal to have the PV continually doing this ON/OFF dance and your response was that it is not something that should happen frequently.
In fact, as I wrote above, it DOES happen VERY frequently. And you yourself seemed to agree that this COULD occur frequently, however, for whatever reason you feel that it will have no long term detrimental effect on the longevity of the PV. I disagree.
The reasons for WHY it occurs are a SEPARATE MATTER:
Reason 1: Outage occurs and despite the low (~30%) soc in the PWs the PV resets (this is something I’m trying to correct)
Reason 2: During outage, PWs allow the PV to charge the PWs once loc is below ~97%. This occurs very rapidly and once the PWs reach 97%+, the PV shuts off (this is normal behavior). I’m not complaining that this occurs. My issue with this specific scenario is that I can’t set the % at which the PV comes on in order to minimize the number of times this occurs. For example, set the % to 50% before the PWs kick the PV back on unless the PWs simply don’t have enough output by themselves to satisfy the home load.
Actually, you yourself wrote that when the outage first occurs (IF PWs are at a “high” stage of charge) the frequency goes up to whatever MAX Hz Tesla set (i.e. 62.5Hz) to eliminate the known UPS issue. Others have written that even after that initial spike in Hz the PW system may raise the Hz to above 60.5Hz to turn off PV production during off grid operation.
I too would be curious to discover the reason for the varying experiences.
I think we might be splitting hairs by now, but what I was trying to convey way way way up there is that I’m not sure it’s worth trying to correct this one particular issue. My thought was that the PV will cycle multiple times during an extended outage anyway, due to the powerwall state of charge, so I’m not sure it’s worth spending a lot of time and effort to fix one more cycle that might happen at the start of the outage. Perhaps I didn’t really communicate my thoughts well.
It would be interesting to be able to set this, but I think the risk is that it could potentially leave your powerwalls at a lower state of charge at the end of the day. If, for example, you had it set for 90% instead of 97%, then it would be possible that your powerwalls could be in the draining mode with the PV shut off during the last bit of strong sunlight for the day. Once they hit 90% and turn the PV back on there could be no light left to charge them and that would leave them at 90% capacity to make it through the night. Whereas with the current method the lowest they could be at the end of the light is 97%.
But that said, if this was a configurable parameter then you could make that decision for yourself.
Yes, again perhaps I was a bit unclear, but if the outage happens when the powerwalls are fully charged then it definitely will initially raise the line frequency to the max value. But then once it drops, I believe that during the rest of that outage it will only get to 60.5. However, I think this may require some additional investigation.
It sounds like you might be experiencing the same issue we've had for the past 3+ months. Our electric company is providing dirty power and causing up to three dozen "outages" per day. Each outage is resulting in 5 minutes where our PV system shuts down so some days we are losing out on nearly 3 hours of PV production. The app now reports 580 outages with probably ~530 of those happening since mid-May.
We found that we could reduce the number of outages by extending the peak hours. Normally, the Powerwalls would cover our peak period from 2 pm until 6 pm. We've scheduled "peak" to start at noon or even 10 am and that greatly reduces the number of outages but sometimes doesn't allow the Powerwalls to charge back up to 100%.
We know the line voltage provided by our closest transformer drops very low and believe the line frequency might be an issue as well. We've escalated the issue with Tesla and our utility company and have been waiting months for the utility company to install a new transformer.
It's not an issue, per se, just not efficient use of power or battery cycles. Some are concerned that on/off cycles cause excess wear on the inverters though I have not read anything yet to indicate that's an actual problem.
Off grid the most efficient use of your PV production is immediate consumption due to the Powerwall's 90% roundtrip efficiency. Back when I first did my off-grid test the Powerwall had to get below 95% for the frequency to get low enough to signal the inverters to turn back on (I believe that's been changed to 97% though I have not confirmed). Since I only have 1 Powerwall I really wanted to eek out those few extra %s to avoid the scenario where the battery is charged, the PV is turned off, then the clouds roll in and I miss out on those few precious watt hours.
I wonder if this is an effect of you having only one powerwall, so there is a higher load to powerwall capacity ratio. Once the powerwall drops the line frequency the inverters won’t actually restart for 5 minutes. I wonder if your powerwall dropped the frequency at 97% but that 5 minute delay was enough for your powerwall to drop to 95% before the inverters actually came back on and started producing.
These would seem to have some bearing, but I am just now how much of a concern it really is. We have around 99% grid uptime. And we are affected by 3-4 days of Public Safety Power Shutoffs. So a grid outage is hopefully a pretty rare event.
And don't inverters stop production every night and renew production in the morning?
Just trying to see if there is anything to be concerned about.
Morning and evening transitions are gradual as the sun rises and sets. The abrupt cutoff and restart could be harder on the equipment.These would seem to have some bearing, but I am just now how much of a concern it really is. We have around 99% grid uptime. And we are affected by 3-4 days of Public Safety Power Shutoffs. So a grid outage is hopefully a pretty rare event.
And don't inverters stop production every night and renew production in the morning?
Just trying to see if there is anything to be concerned about.
However, I think some people on this thread are over-reacting to this issue. As long as you have the max frequency lowered to below 63Hz, there shouldn't be any significant problem with the way the Powerwalls modulate the solar production. We can all dream up ways to make the system better, but part of the appeal of Powerwalls is the near universal compatibility with solar grid-tied inverters. Sure, it would be great if there was a way to have the Powerwalls proportionally curtail solar so that the consumption and solar production are completely balanced and the SOC is held at exactly 95%. That would be fantastic. However, there is probably a direct conflict in solar inverter settings between optimal off-grid proportional curtailment and grid ride-through.
My inverters do not have any ramp feature at all. They are older ones, so solar is either on or off on mine, there isnt any ramp. That came with later inverters.
Similar threads
