Anyone else here playing the 100% self powered game?

[power.saver]

I am totally confused on this. It sounds like our systems are the same but behave differently. Maybe it's the mode we are in?

My charger is also on my main panel. I checked the graph and yes, both the grid and home show the charger load, but no energy is coming from the powerwalls. But in your case they are.

I was in Cost Savings mode. What mode were you in when you charged and took the snapshot?

Cost savings mode does not offset loads during off-peak. That is why you show no PW usage during your off-peak period.

If you go charge your car today for a little bit during part-peak or peak, you will see the PW discharge.

 

[Incredulocious]

I think this must be it.

The powerwalls must be monitoring both the costs and the energy required and is making the best decision based on cost. Since my off peak charging is a fraction of my regular rate, it would probably never use powerwall power to charge the car.

As @BrettS stated if I did the same test during peak like you did, I would probably see the powerwalls contributing since they would be cheaper than the grid costs. I will try that later today for a few minutes to confirm.

As I mentioned in my post above, I can confirm that the non-backed up car charging circuit will pull from the Powerwalls when the grid is available. This is true even overnight during off-peak hours. The "non-backed up aspect" of that circuit only matters when grid power disappears and the connection with the grid-side is shut off.

Here's an example of that where the car charging starting at 12:30 am proceeds to use the Powerwalls until they hit my minimum 20% reserve for backup and then it starts pulling from the grid:

 

[jboy210]

I think trying to be 100% self-powered is a good idea. Learning how and when to use electricity stored in PWs and blend it with solar is worthwhile. Because no matter where you are the power from the grid can go out. This can be due to weather (ice-storms, hurricanes, etc), natural disasters (fires, earthquakes), or humans (terrorist attack on grid, Public Safety Power Shutoffs - PSPS ). We got these PW systems to have power in these scenarios. We should learn how to use them now so we can be prepared with it hits the fan, and know what we can and cannot expect.

 

[aesculus]

Here's an example of that where the car charging starting at 12:30 am proceeds to use the Powerwalls until they hit my minimum 20% reserve for backup and then it starts pulling from the grid:

What mode is this in?

 

[Incredulocious]

What mode is this in?

I think that must be from when I was trying the "Balanced" time-of-use mode: in the graph you can see how it decided to immediately stop using the grid while charging the Powerwalls from solar at the morning switch from off-peak to partial-peak. And you can see it switch to grid power again in the evening exactly when the off-peak period started.

These peak/partial peak periods didn't actually apply in February; I was just testing to see what Balanced Time-of-Use mode would do. As I described earlier, I ended up switching to self-powered mode as it would let me run longer without using the grid and yet still tend to let me export to the grid in the late afternoons.

What I see with self-powered mode during the summer tends to look like this:

 

[aesculus]

OK. Mia Culpa. I stand corrected about where the charger is in relationship to if the PW will send energy to it.My charger is NOT backed up and sits in the meter panel.

My past charging experiences have always been during off peak where my energy charges are the cheapest, hence PW energy would only be equal but never cheaper, so it would make no sense to use that energy (AC-DC losses etc)

So a quick experiment today by charing at peak time showed the PW sending power to both the charger and the house while preserving the solar to go to the grid to get max dollars. I purposefully moved my amps down to stay below the 10 kW of the PWs so I did not incur grid charges. Confirmed that when I stopped charging the PW discharge went to 800 Watts, which is about normal.

 

[jjrandorin]

In my case, because I use "self powered" when I charge the car, the car will take all available solar + make up the difference with powerwall power. If solra + powerwall is not enough to charge at the rate my car charges at, the balance will pull from the grid. Since I dont have time of use, it doesnt matter WHEN I use my power, so its no cheaper for me to charge at night than during the day.

I just dont like to deplete my powerwalls charging the car, if they cant charge back up, so I will usually either throw it into backup only mode while I charge, or, leave it in self powered but set a reserve that I know the solar can fill back up from at the time my car finishes charging.

I have said it before, but I am way over micromanaging it, but am really enjoying doing so, so am not currently looking to automate anything in this workflow. At some point I will likely be forced onto a TOU plan, and I will revisit the advanced settings for myself then.

 

[NathanielHrnblwr]

Was it determined that if the PW's are in Cost Savings mode that they won't try to charge the car during off peak?

 

[cwied]

Yes, the Powerwalls always charge and never discharge during off-peak. Conversely, they always discharge and never charge during peak. It's harder to understand what they're doing during partial peak, but the basics seem to be they will try to be at a charge level at the end of partial peak that can last until the end of the next peak period. They do some obviously sub-optimal things during partial-peak, though, like discharging and then switching back to charging during the same period.

 

[aesculus]

Yes, the Powerwalls always charge and never discharge during off-peak. Conversely, they always discharge and never charge during peak. It's harder to understand what they're doing during partial peak, but the basics seem to be they will try to be at a charge level at the end of partial peak that can last until the end of the next peak period. They do some obviously sub-optimal things during partial-peak, though, like discharging and then switching back to charging during the same period.

Yeah. What they do during partial peak is a mystery to me, especially if your rate changes Sat. It seems like it tries to predict what your SoC will be by the end of peak on Friday night and only charges Thurs/Friday to that plus projected usage to minimize storing power that cannot be sold.

Occasionally this algorithm fails if cloud cover or extra heat requiring extra power during peak uses more energy. But if everything is fairly constant it seems to work OK for me.