Understanding PowerWall Roundtrip Efficiency

[jsurpless]

I am that the PowerWall has a 90% roundtrip efficiency but am having a little trouble understanding how the PowerWall is operating.

For clarity, we have two PWs so 27 kWh is the available energy - I assume that this means that more than that is used to fill it up (say 5%?) and that we then only get 95% out or similar?

Am I understanding this correctly?

In addition, we are participating in Connected Solutions and we're seeing more variability in the energy discharged during an event that I'd expected.

In most cases, the batteries are 100% charged at the start of the event but the energy discharged ranges from as low as 19.5 kWh to as high as 20.9 kWh.

Today, the battery discharged from 100% to ~20% over 3hrs, yielding only 19.5 kWh. Since events won't go below 20%, I do not understand how 20.9 kWh was ever discharged during a 3-hr event.

 

[power.saver]

Efficiency is calculated based on how much energy comes out versus how much goes in. The amount shown coming out is the reference.

So starting at 100% if you discharge 20kWh from your PW, you would need to put 22.2kWh back in to get to 100% again. (20/0.9 = 22.2)

As for the amount you can discharge during your events, the app reserve setting of 20% is actually translated into a internal value of 24%. That means only 76% of your 27kWh is available, or 20.5kWh which is very close to the maximum you see.

Here is the chart that shows the conversion from the app setting reserve to the actual (api) value:

 

[jsurpless]

Efficiency is calculated based on how much energy comes out versus how much goes in. The amount shown coming out is the reference.

So starting at 100% if you discharge 20kWh from your PW, you would need to put 22.2kWh back in to get to 100% again. (20/0.9 = 22.2)

Right, I follow this - makes perfect sense but not sure about the following

As for the amount you can discharge during your events, the app reserve setting of 20% is actually translated into a internal value of 24%. That means only 76% of your 27kWh is available, or 20.5kWh which is very close to the maximum you see.

What is the function of the y = 0.95x + 0.05 to translate from app % to API %? Is the 0.05 some minimum reserve that Tesla has restricted use of?

Any thoughts on why the kWh discharged varies?

Thanks for the explanation!

 

[power.saver]

We don't know exactly why Tesla implemented the percentage conversion, but it appears to be so that if a user set their app to 0% it would still hold 5% in the battery. If the battery were to reach 0% then the system could not restart if the grid failed as it takes some energy to activate the solar inverters and get charging started.

I think the variation in discharge might be because you said that in most cases the batteries were 100% charged. If they were not, and let's say you started at 97%, then the max discharge would be 27kWh * 0.97 * 0.76 = 19.9kWh. A lower starting point will reduce the energy output.

Another possibility is variations in efficiency. It has been reported, and I've seen it on my system, that the efficiency changes slightly. While the spec is 90%, I have seen variations from 88% to 92% over a period of time. It is not clear what causes this. It could be temperature, discharge pattern, battery recalibration or something else. So there is always going to be some variation in available energy.

 

[jsurpless]

We don't know exactly why Tesla implemented the percentage conversion, but it appears to be so that if a user set their app to 0% it would still hold 5% in the battery. If the battery were to reach 0% then the system could not restart if the grid failed as it takes some energy to activate the solar inverters and get charging started.

I think the variation in discharge might be because you said that in most cases the batteries were 100% charged. If they were not, and let's say you started at 97%, then the max discharge would be 27kWh * 0.97 * 0.76 = 19.9kWh. A lower starting point will reduce the energy output.

Another possibility is variations in efficiency. It has been reported, and I've seen it on my system, that the efficiency changes slightly. While the spec is 90%, I have seen variations from 88% to 92% over a period of time. It is not clear what causes this. It could be temperature, discharge pattern, battery recalibration or something else. So there is always going to be some variation in available energy.

I was thinking that it might have had something to variations in efficiency.

I had considered that perhaps the battery wasn't at the same charge level but I'm pretty sure that it was. It's still new to us so I've been keeping a somewhat close eye on the events, and so I've seen the app say 100% before each event.

 

[jsurpless]

Hi

So I came up with a 'discharge efficiency' modification to my tracking spreadsheet, as follows:

Do you think that this is a meaningful way of looking at it? Basically, I was interested in calculating the actual energy discharged vs how much energy I "could have" discharged for the observed change in charge capacity.

Not sure about most of the End Capacity values but I assumed that it was at the level Tesla says is the cutoff.

Confusing to me as to how I could have discharged greater than the ~20.5 kWh for 100% -> 20% but apparently, it does happen.

 

[lensovet]

Digging up this ancient thread as I'm basically never seeing the promised 90%.

For example, two days ago PW shows being charged 3.8 and discharging 3.2 (85%). On another day, I have 3.6 charged, 3.0 discharged (80%). On most days, I'm seeing 85%.

Is there some way to force grid export before the battery reaches 100%? If I don't have an outage, I would prefer to keep the battery at 75% and just use net metering instead of losing 15% of my solar production to roundtrip inefficiencies (which comes out as actual dollars on my electric bill). Plus, batteries age better when they are kept between 20-80%.

 

[JohnRatsey]

My Powerwall has averaged 85% over nearly 5 years. I suspect that Tesla's efficiency figure excludes the power overhead consumption with operation of the battery and the gateway. The lower the power flow into / out of the battery then the bigger the proportional effect of this power consumption. My battery has a fairly easy life for half the year.

 

[Redhill_qik]

Digging up this ancient thread as I'm basically never seeing the promised 90%.

For example, two days ago PW shows being charged 3.8 and discharging 3.2 (85%). On another day, I have 3.6 charged, 3.0 discharged (80%). On most days, I'm seeing 85%.

Is there some way to force grid export before the battery reaches 100%? If I don't have an outage, I would prefer to keep the battery at 75% and just use net metering instead of losing 15% of my solar production to roundtrip inefficiencies (which comes out as actual dollars on my electric bill). Plus, batteries age better when they are kept between 20-80%.

The Powerwalls lose 200-250Wh per day, so the 3.2kWh was really about 3.4kWh that needs to be recharged and 3.4/3.8 = 89.47%.