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

Misleading Capabilities of Backup Mode?

This site may earn commission on affiliate links.
#41
BIC1 said:
Only been at this a week or so. House consumption, per Tesla app, seems to range from 1-4kW, I'd say typically about 2-3kW. You are correct, it seems to shut down when the solar production exceeds the needs of the house & battery together.

Tesla suggested increasing consumption in that case. Guess I could invite the neighborhood over to charge their phones.
Click to expand...
Yes, why not for free. :D
 
#42
pgrovetom1 said:
There is no technical reason the maximum charge rate is tied to the discharge rate. Discharge is the inverter power and should be limited by the heat dissipated by the PW2 inverter driving the home. That is why they say 7KW for 10 seconds as its not a hard limit. Its a heat thing. But when the heat gets too high, the PW2 should shut down its output.
Click to expand...

It's tied to the discharge rate because the batteries are DC and the grid side is AC. The batteries and internal circuitry can handle much more than 5kw or 7kw but since the inverter is a bottleneck charging and discharging is limited to the inverter spec, which is listed as 5kw continuous and 7kw for 10s.
 
#43
pgrovetom1 said:
I'm exploring designing a battery system to install in my garage that can soak up excess solar production while on backup and then discharge it at night reducing the PW2 discharge.

I'm trying to investigate the issues that are being discussed in this post since the system I want to build might solve the issues here.

Some questions:

What is the per-PW2 charging limit? Is it the 5KW output? If the PW2 has a low charging limit, it creates some problems. Solar Grid Tie inverters are current sources. That means they raises their output voltage until 100% of their capacity is being sunk by the grid. Since the grid is unlimited, a solar inverter can have all its power sunk so everything is ok. It never runs into the situation where the grid is disconnected and it is forced to push its output voltage into dangerous territory well above 240V. But when you add a PW2 into the picture ( or 2-4 of them), things get tricky. If the PW2s cannot sink 100% of the Solar Inverters power, the voltage will rise until the grid tie inverters are in over-voltage and will shut off. But the home usage sinks some of the power but it varies and may not work with all Solar Inverters. For example, Enphase M215 inverters are not split phase. They output 240V so 120V appliances cannot use M215 power directly. But many Solar Inverters do output split phase 120/120 so the home can soak up some power. That complicates the discussion in this thread.

Lets make some simple assumptions to illustrate. Lets say the maximum power a PW2 can soak up for charging is 5KW as in its output specification. Say an application has a real 6KW of Enphase M215 inverters on a single PW2. Lets say the home always draws at least 1KW. In this example, one might expect the PW2 to be able to accept 5KW always and since the home uses at least 1KW, it should be ok right? No because the Enphase inverters cannot power 120V home loads since its not split phase. That means the inverters will keep trying to raise their output voltage to force the grid to sink the power. But there is no grid in backup nor can the home plus PW2 soak it up. If the home plus the PW2 cannot sink 100% of the power produced, the Enphase inverters will shut down.

But if this example uses Solar Inverters that produce split phase, the home plus the PW2 will soak up the power and everything will work correctly. The PW2 will charge to 100% from the solar 6KW minus home 1KW = 5KW until its full and use the 60HZ frequency to shut the split phase capable Inverters down. Then when the PW2 is down to 96% or so it will return the 60HZ to exactly 60HZ and the inverters will come back on and the recharge will occur over and over.The PW2 will remain fully charged when it goes into backup so the user will get the full 13.5KWhr before full shutdown. If the user keeps the home usage under 13.5KWHr overnight, it should be ok.

Its gets more complicated with the new UL1740 inverters that have the capability to be partially throttled back but I doubt the PW2 can support the variable 60HZ or many people have the new spec inverters. So I believe the behavior of the PW system will vary depending on:

1) What can the PW2 sink while charging - 5KW or more?
2) Do the Grid Inverters on the solar have split phase or 240V only?
3) Their maximum solar power on the best sunny day.

I am having 3 PW2s installed:

1) PW2 #1 on one of my SMA Sunny Boy 7000 with about 6KW ( the SMA inverters support split phase)
2) PW2 #2 on another SMA Sunny Boy 7000 with another 6KW ( the SMA inverters support split phase)
3) PW2 #3 on a Enphase M215 system with 18 inverters at about 4KW ( the M215s don't support split phase)

My home always draws more than 1KW which can be powered by SMA split phase support.

Since the 60HZ from the PW2's is common, the PW2's cannot shut down one inverter at a time. If the 5KW per PW2 is a hard limit, I will have a problem if the sum of my 3 inverter outputs exceeds 5KW x 3 minus my home 1KW. So if I hit 6 + 6 + 4 = 16KW minus 1 KW for my home equal 15KW, it should be ok. I would be right on the limit limit if the 5KW charge limit is real. Fortunately my SMA inverters can power my homes 1KW split phase so my M215's shouldn't be a problem. It will be a bummer if on right on the edge and it shuts down at peak power in the summer.

This might explain some of the differences people see. Anybody considered adding say another 5-20KWHr of backup power to supplement the PW2's while in backup for EV charging or other high usage applications?

Thoughts
Click to expand...

pgrovetom1 said:
There is no technical reason the maximum charge rate is tied to the discharge rate. Discharge is the inverter power and should be limited by the heat dissipated by the PW2 inverter driving the home. That is why they say 7KW for 10 seconds as its not a hard limit. Its a heat thing. But when the heat gets too high, the PW2 should shut down its output.

But the maximum charge is based on the heat dissipated in the battery charger. Inverters are more complicated beasts than chargers and should have a little lower efficiency. I would expect the inverters in the PW2 to have an efficiency in the 95-96% range but the charger could more easily be 97-98%. So I would think the charger would be more tolerant. They may just spec them the same since its close. But what is the point where it shuts down on over-charge? If my 3 inverters put out another 1KW or 17KW at mid-day, would the charger overheat and shut down? I have spoken with 2 Tesla engineers neither mentioned this issue. But one put my Enphase inverters on the utility side of the PW2 Gateway and wanted me to use 2PW2s. I then asked to increase it to 3 PW2's and Tesla seems consternated - maybe this is why...
Click to expand...

You have a thread you started for technical discussions regarding adding on non powerwall batteries to a powerwall. These technical discussions on this topic should be in that thread, not hijaacking this OPs thread.

Tesla PowerWall DIY battery expansion capacity
 
#45
gpez said:
As @Ampster notes another solution to this problem is to have your inverters reduce their power during an outage through the microgrid frequency. During an outage your Powerwall will set the local frequency slightly higher which your inverters will notice and "ramp down" production. The Powerwall will dynamically set the frequency - higher than spec when the state of charge is higher or net (production - consumption) is higher, at spec when either are not. This allows your solar production to increase and decrease as the Powerwall charges and discharges with the house load.

@SteveInCal Enphase call it "curtailment" or "ramp down" (https://enphase.com/sites/default/f...Considerations-AC-Coupling-Micros-Battery.pdf, page 6). My IQ6 inverters support this.

SolarEdge inverters all it "P(F) Power Frequency" (https://www.solaredge.com/sites/default/files/application_note_power_control_configuration.pdf, page 6).

@BIC1 I know you said you're maxed on your budget - unfortunately your YC500As don't support this while the newer YC600 model does. It's called "ramp rate" - https://usa.apsystems.com/wp-conten...C600-For-USA-User-manual_Rev1.3_2018-3-15.pdf page 19.
Click to expand...

Thank you, @gpez, very informative post. Interestingly, I signed my PV system contract in October 2018 with installation in February 2019. Materials were delivered in January 2019. The new construction was occupied in late April 2019. Along with LG 320N panels, I was contracted for Enphase IQ7s. Somewhere along the way (probably Nov-Dec 2018), the installer said that Enphase was having supply issues and I would have to wait an additional six months for Enphase. He then suggested APsystems.

I specifically asked if I was giving up anything with AP YC500 vs Enphase IQ7. He said no, they were comparable. He said the cost was the same. I noted that AP claims they're less expensive because you only need half as many microinverters. The installer said that was not true after all costs are totaled.

Well, now it seems they're not comparable. Also, from what I can tell now, the YC600 was available at that time. In Sept 2017, AP announced the YC600 release around November 2017. The data sheet is dated June 2018 (could have been updated from an earlier version).

If I understand the ramping, it's analogous to an automated dimming night light, turning the light brighter and dimmer based on the ambient light. Does that sound right?

How high does the frequency go with ramping microinverters? Does it get up to the danger zone of 65Hz, or do they reasonably stay below 61Hz or so?

Does anybody know if Enphase was having supply issues in late 2018/early 2019? Anybody know when the YC600 was released? Guess I'm losing the confidence I had in my installer. Thanks.
 
#46
BIC1 said:
Thank you, @gpez, very informative post. Interestingly, I signed my PV system contract in October 2018 with installation in February 2019. Materials were delivered in January 2019. The new construction was occupied in late April 2019. Along with LG 320N panels, I was contracted for Enphase IQ7s. Somewhere along the way (probably Nov-Dec 2018), the installer said that Enphase was having supply issues and I would have to wait an additional six months for Enphase. He then suggested APsystems.

I specifically asked if I was giving up anything with AP YC500 vs Enphase IQ7. He said no, they were comparable. He said the cost was the same. I noted that AP claims they're less expensive because you only need half as many microinverters. The installer said that was not true after all costs are totaled.

Well, now it seems they're not comparable. Also, from what I can tell now, the YC600 was available at that time. In Sept 2017, AP announced the YC600 release around November 2017. The data sheet is dated June 2018 (could have been updated from an earlier version).

If I understand the ramping, it's analogous to an automated dimming night light, turning the light brighter and dimmer based on the ambient light. Does that sound right?

How high does the frequency go with ramping microinverters? Does it get up to the danger zone of 65Hz, or do they reasonably stay below 61Hz or so?

Does anybody know if Enphase was having supply issues in late 2018/early 2019? Anybody know when the YC600 was released? Guess I'm losing the confidence I had in my installer. Thanks.
Click to expand...

Welcome!

My IQ6s can be configured in many possible ways. I currently have applied the "PREPA off grid frequency-watt 83/hz" profile, which is a standard profile used in Puerto Rico (the "PR" in "PREPA"). Because my PV system is entirely behind the Powerwall it's opaque to my utility and I'm mostly free to use whatever PV profile I'd like as long as the Powerwall is compliant with the local utility's interconnection agreement.

The PREPA 83 profile has these properties set:
- Ramp Down Start Frequency: 60.2hz
- Ramp Down Rate: 83.0%/hz
- Over Frequency Disconnect: 61.8hz

The Powerwall is smart. During off grid operations I found that the frequency that it sets the local microgrid to will vary over time as the house load changes and the battery's state of charge (SoC) changes. Utility outage simulation data dump

My Powerwall is configured to set the local grid frequency to 62.0hz when it wants to fully disable the PV, which is higher than the Over Frequency Disconnect of 61.8hz so that works great.

As an example:
High SoC, low load = 62.0hz -> PV off
Low SoC, low load = 60.0hz -> PV 100%
High SoC, high load = 61.0hz -> PV curtailed to 33.6% (100% - [61.0hz - 60.2hz] * 83.0%/hz = 33.6%)
Low SoC, high load = 60.0hz -> PV 100%

Of course there are infinite "in between" scenarios but generally as the battery charges from the sun the frequency increases, curtailing solar production. If the house is consuming more and the battery starts to drain the frequency decreases, curtailment lowers or stops which allows more solar production.

You have to remember that there is no way for any part of the system to be able to predict how much power the PV will produce next - once the sun's rays hit the panel and generate a current that current has to go somewhere right that instant. When the grid is down and the Powerwall is 100% full there is literally nowhere else for the electrons to go except "out" in the form of sparks and fire. It's very important to ensure all of your settings are correct to prevent that from happening.
​
 
  • Informative
  • Like
Reactions: GenSao, jjrandorin and brianman
#47
BIC1 said:
Thank you, @gpez, very informative post. Interestingly, I signed my PV system contract in October 2018 with installation in February 2019. Materials were delivered in January 2019. The new construction was occupied in late April 2019. Along with LG 320N panels, I was contracted for Enphase IQ7s. Somewhere along the way (probably Nov-Dec 2018), the installer said that Enphase was having supply issues and I would have to wait an additional six months for Enphase. He then suggested APsystems.

.
Click to expand...

I think your have to choices to make your system work better.

1. Try what gpez is suggesting. I don't know if your microinverters support this function. If your system was installed recently, maybe your installed will agree to swap them out for Enphase?

2. The other option is to take one of the branches and move it outside of the PW gateway.
 
#49
So my takeaway on this issue is that the problem isn't really a Tesla Powerwall problem. Perhaps if the Tesla Powerwall installer was better informed he could have known that those microinverters were not capable of ramping down like the more modern Enphase inverters.
In the long run the impact on system performance is insignificant since in one scenerio the GT solar is cycling on and off to keep the Powerwall charged. In the other scenerio the GT solar would be modulated and the the PowerWall would not cycle as much. As others have said, that amount of cycling is probably not affecting the life of the PowerWall. At least from my perspective the title of this thread is misleading.
 
#52
BIC1 said:
That is my question; is there any way to divert/dissipate the excess solar while off-grid in backup mode so that the PV doesn't shut down? Thanks.
Click to expand...

I'm not sure I understand the problem.
If you are off-grid (the grid has been disconnected/failed) and the battery is full, then the solar inverters can't produce any more energy than your house is using at that instant. Sure, if you turn on a pile of things and make more load, your solar can then supply all that extra too - but it can't be making more than your house is using, generation and consumption must balance out in a zero-sum-game when the grid is disconnected.
Now in practice, the system will cycle - turn off the solar and run the house load from the battery until the battery is down to 95% or so, then the solar will ramp up, supply the house AND the battery to charge it back up again, until the battery tells the solar to pause for a while. Repeat.
On average, the solar generation will match the house load as designed, plus a little bit to account for the 10% round-trip loss in the battery.

So what exactly is the issue - is it when the grid is out, you want to run more than 5kW of load, and the battery won't supply more than 5kW, even though the solar panels could and the generator you didn't buy could?
 
  • Like
Reactions: Ampster
#53
PBBear said:
So what exactly is the issue - is it when the grid is out, you want to run more than 5kW of load, and the battery won't supply more than 5kW, even though the solar panels could and the generator you didn't buy could?
Click to expand...

That basically sums up my concern at first. I've come to realize it's not as big a deal as I initially thought. But, guess I should see what kind of load I need to support. My passive load seems to range up to 3.5k. Guess I should see what turning on the oven or induction cooktop does.
 
#55
BIC1 said:
That basically sums up my concern at first. I've come to realize it's not as big a deal as I initially thought. But, guess I should see what kind of load I need to support. My passive load seems to range up to 3.5k. Guess I should see what turning on the oven or induction cooktop does.
Click to expand...

I have a electric Double Wall oven that requires a 40 amp 240v connection. If I turn on both Ovens, the usage can spike up to 10k, although its not a constant usage at that rate (ovens pulse on and off like some other items). While my oven is backed up, if I am in a true outage situation I would only use my oven if I wanted to drain power from the powerwalls for some reason.

I have 3 BBQ grills (1 natural gas, 1 Kamado style, and 1 pellet style... I enjoy BBQ'ing lol), and can use each of them as an oven if I choose to. I make Biscuits, Bacon, etc out in my grills, so would use them as ovens. My cooktop is a gas 6 burner so I would use that in an outage. OP especially since I believe you only have 1 powerwall, I would actually be surprised if your cooktop and oven (if double wall oven) are backed up... but you did say your installer is not super powerwall savvy. It would appear to me that 1 powerwall would not be enough to run the startup or loads from those two devices in an outage, but I dont know which ones you have.

Given the size of your PV / Solar installation, if I was in your shoes, I would have at least 2 powerwalls. Might be something to consider.

I feel this is an issue where your installer did not provide you with adequate information, not because they did it on purpose, but because they are new as a tesla installer per what you mention. If I was in your shoes, I would (very politely) lean on them a bit to install a second powerwall for you, at the cost of the powerwall. The powerwall physical cost should be 6500 (they pay the same for powerwalls that we do, I am fairly sure).

I would ask them to install the 2nd powerwall for cost of the powerwall, and see where that goes. They might agree if they are embarrassed by not giving you the proper recommendation to get at least 2 powerwalls for a system that size. They might push back for a small labor charge, they might say no.... but I suspect they would really work with you to get that 2nd powerwall installed at a cost not too much more than the device itself (and both would qualify for the 26% rebate).

I believe you would be much happier with 2 powerwalls.
 
Last edited:
  • Like
Reactions: GenSao
#58
PBBear said:
So what exactly is the issue - is it when the grid is out, you want to run more than 5kW of load, and the battery won't supply more than 5kW, even though the solar panels could and the generator you didn't buy could?
Click to expand...
The other problem with having too much solar generation on the backup side of the Gateway is that when the solar comes back online it will immediately trip and solar will be forced offline again. Even if your consumption is moderate, the solar will never charge the batteries while the instantaneous generation is above the Powerwall capabilities. The simplest solution to this problem is to merely turn off the breaker to one or more solar inverters so that it won't overwhelm the Powerwalls when the sun is high in the sky. You may want the additional solar online in the morning and evening when the production per panel is lower.
When the batteries are full, they will force the solar offline. That is not a problem. When the solar is too big, it won't charge the batteries at all until the solar output naturally drops low enough in the afternoon. Failing to charge when the batteries are low is definitely a problem.
 
#59
Earlier in this discussion the issue was that the GT solar was cycling on and off. The GT solar has a capacity of 12 kWs and when the sun is shining and the grid is down that should be able to power his household. If he needs to power his AC and oven at night during a power outage, then he probably would need another PowerWall. He would need to clarify his goals, before I would recommend another PowerWall.
 
You must log in or register to reply here.

Similar threads

X
Delaying/Limiting Powerwall Charging Until Midday Peak Solar Production
Replies
7
Views
413
Tesla Energy
Mark Purcell
Mark Purcell
NathanielHrnblwr
SDG&E, NBC's, EV-TOU2, and Optimizing Powerwall modes on NEM 2.0
Replies
3
Views
1K
Tesla Energy
NathanielHrnblwr
NathanielHrnblwr
RedModel3
Battery backup without solar?
Replies
2
Views
240
Tesla Energy
arnolddeleon
arnolddeleon
B
Emergency Power PWs Shutdown Due to PV Overproduction
Replies
11
Views
580
Tesla Energy
BIC1
B
J
Tesla Powerwall2 on Two Phase installation
Replies
7
Views
762
Tesla Energy
BGbreeder
B
Top
Back
Blog
New
Forum list
Marketplace
Menu