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

PowerWall technical discussion

This site may earn commission on affiliate links.
I think that is a good idea but the low hanging fruits are the markets where there are arbitrage opportunities and/or resistance by public utilities for net metering.

Also I should point out that in Wisconsin the public works commission recently approved several anti-solar/net metering policies. They raised the grid connection fees and significantly reduced the TOU price delta. Irks me so much that I am now more motivated than ever to go solar. Unfortunately my budget for solar isn't that significant. I have about $7K saved up for the project, not sure how much I can accomplish with that.

- - - Updated - - -

If an MS 'goes to Zero' you might be able to lift your PowerWall unit off its wall and along with a Wall/Mobile Charger go and charge the MS enough to drive it home.
--

Possible maybe, but until we can get a DC based EVSE it would require an inverter as well (and all the losses associated with it). That's why I am still so facisnated with the DC charging setup. I am just a teeny bit disappointed that Tesla didn't provide this solution with the introduction of the power wall. I understand why they didn't, but right now it sort of up to the community to figure this out. I am curious if the progress that Tony Williams and company have made with Chademo (both setting up the cars to charge with and the chargers themselves) might be helpful.
 
Also I should point out that in Wisconsin the public works commission recently approved several anti-solar/net metering policies. They raised the grid connection fees and significantly reduced the TOU price delta. Irks me so much that I am now more motivated than ever to go solar. Unfortunately my budget for solar isn't that significant. I have about $7K saved up for the project, not sure how much I can accomplish with that.

- - - Updated - - -



Possible maybe, but until we can get a DC based EVSE it would require an inverter as well (and all the losses associated with it). That's why I am still so facisnated with the DC charging setup. I am just a teeny bit disappointed that Tesla didn't provide this solution with the introduction of the power wall. I understand why they didn't, but right now it sort of up to the community to figure this out. I am curious if the progress that Tony Williams and company have made with Chademo (both setting up the cars to charge with and the chargers themselves) might be helpful.

For God's sake let's not incorporate Chademo in to any direct DC charging solution. It's an abomination of a standard.
 
For God's sake let's not incorporate Chademo in to any direct DC charging solution. It's an abomination of a standard.

What it has going for it is that it's an open standard. Other than that not much else. But since Tesla released the adapter it's possible to incorporate it in a DIY scenario. Would I rather use a native supercharger protocol? Obviously yes.
 
What it has going for it is that it's an open standard. Other than that not much else. But since Tesla released the adapter it's possible to incorporate it in a DIY scenario. Would I rather use a native supercharger protocol? Obviously yes.

Off topic, but au contraire mon frère: Chademo is not an open standard. Native SC protocol or SAE Combo connector DC protocol (according to earlier discussion on this forum highly likely very similar) would be a better choice.
 
Chademo standard can be obtained here: http://www.webstore.jsa.or.jp/webst...TS+D+0007:2012&dantaiCd=JIS&status=1&pageNo=0

I haven't purchased it, but the document pre-view gives you an idea of what is available.

Point me to the CCS protocol and I will take a look. My google-fu hasn't found what I am looking for yet...

Pheonix Contact sells some charge controllers that may be able to implement this. Looks interesting. But the Tesla hasn't officially supported CCS yet, so still not sure it would work even if it were implemented properly.
 
Last edited:
Like you I am very interested in knowing more about the PowerWall & PowerPack technical side and how it interfaces with and existing Solar install, and listened twice the full Q1 2015 Earnings call to learn more:


Here is the post I wrote this week end about it:


"Elon Musk & JB Straubel gave more details about the Tesla Energy PowerWall & PowerPack during the Q1 2015 Shareholders Call !


Elon: "The response has been overwhelming, like crazy !" ... "Its has gone super viral"


- 60,000 PW reserved (38,000 customers ordering 1,5-2 per installation)
- 25,000 PP reserved (2,500 customers ordering 10 PP at once !)
=> they are sold out as of now through mid 2016, until GigaFactory can supply their first cells


- 2,500 solar companies requests to install the PowerWall, they can not even answer them for now


- 18650 cells used are generic, but some are optimized for the Daily Cycling PP and some for the Backup version, but a different number of cycles :


10 kWh Backup Power is for High Power/Energy application and cells are similar to the ones used in car (Nickel Cobalt Aluninum) : 255 cycles / year for 1000-1500 cycles
7 kWh Daily Cycling: cells are Nickel Manganese Cobalt : 60-70 cycles / year for 15 years / 5000 cycles


- Daily Cycling PowerWall (several units) is for going Off Grid


- Elon: a DC-DC converter is already included in the PowerWall, and wired to the existing PV it will charge the battery pack directly, the existing DC-AC inverter is used to push to power back into the grid
JB: the PowerWall will act like a solar panel in a way


- GigaFactory could be dedicated to Battery Manufacturing for Stationary Storage




=> My thoughts on this:
So, from what they said, I understand that the PowerPack will charge directly from the solar PV array, which is a very good thing because it means less loss in converting power from DC to AC with the existing inverter and then AC to DC by the Charger that would be required to charge the battery pack
But, once fully charged, and when it's getting dark, will the PowerWall have the ability to push the right amount of voltage & current in order the satisfy the house loads only ? Or will it push full power into the existing inverter and generate full 2kW if it's a 2kW inverter ? That question remains and I would like to have even more details on that ..."

- - - Updated - - -

Sorry I meant: I understand that the PowerWALL will charge directly from the solar PV array

- - - Updated - - -

https://youtu.be/EGHsDqwSY28
 
10 kWh Backup Power is for High Power/Energy application and cells are similar to the ones used in car (Nickel Cobalt Aluninum) : 255 cycles / year for 1000-1500 cycles
7 kWh Daily Cycling: cells are Nickel Manganese Cobalt : 60-70 cycles / year for 15 years / 5000 cycles

I think you got it backwards. the 10kW is for backup and is the 50 cycles / year while the 7kW for daily cycling is the 365 cycles / year
 
- Elon: a DC-DC converter is already included in the PowerWall, and wired to the existing PV it will charge the battery pack directly, the existing DC-AC inverter is used to push to power back into the grid
JB: the PowerWall will act like a solar panel in a way

=> My thoughts on this:
So, from what they said, I understand that the PowerPack will charge directly from the solar PV array, which is a very good thing because it means less loss in converting power from DC to AC with the existing inverter and then AC to DC by the Charger that would be required to charge the battery pack
But, once fully charged, and when it's getting dark, will the PowerWall have the ability to push the right amount of voltage & current in order the satisfy the house loads only ? Or will it push full power into the existing inverter and generate full 2kW if it's a 2kW inverter ? That question remains and I would like to have even more details on that ..."

Sorry I meant: I understand that the PowerWALL will charge directly from the solar PV array
If you look at how most off-grid systems work, they go direct from solar panels to the battery, all in the DC domain. The PowerWall is different because it does so with DC-DC Converters instead of MPPT Charge Controllers. I would be very interested to see a given application size specified with a "traditional" off-grid system and also with a PowerWall based system. By specified, I mean a complete competitive bid with BOM and labor clearly and completely spelled out.
 
just spoke with a SolarCity representative who called me about my PowerWall reservation. His understanding of what he's been informed is that both the 7kW and 10kW battery packs are for backup only and will ONLY WORK when the grid is down. Yes thats right, the 7kW will only work when the grid is down. When I asked about then how could people 'buy cheap at night and use that energy during the day to keep costs down just like Elon stated on multiple occasions' he said that basically everyone else is wrong and that's not how it works and you can't do that with it and you'd only be able to use the either of the packs during the day if the grid goes down.

I have a hard time believing that given all the public statements from Elon himself and all the techincal discussions on the differences between the 7kW and 10kW packs.

He also said that the PowerWall will only be available through SolarCity until end of 2016 which is when it will be made available to other (non-SolarCity) vendors.

sigh. I still think HE was wrong though. on pretty much every account. he did say he could be wrong though and to not quote him on it haha. well I dont know what the point of that phone call was when he can't explain the technical details of it. they shouldn't be calling anyone if they don't understand it.

Anywho, regardless of how the system works or doesn't work, I canceled my PowerWall reservation. The ROI is just not there. And even if it was there, lets say ROI was 7-10 years, in 7-10 years from now how much kWh in batteries do you think you'll get for $3k? probably triple if not more. I'll wait until the price per kWh is below $100 before I buy.
 
He is incorrect on all counts.
just spoke with a SolarCity representative who called me about my PowerWall reservation. His understanding of what he's been informed is that both the 7kW and 10kW battery packs are for backup only and will ONLY WORK when the grid is down. Yes thats right, the 7kW will only work when the grid is down. When I asked about then how could people 'buy cheap at night and use that energy during the day to keep costs down just like Elon stated on multiple occasions' he said that basically everyone else is wrong and that's not how it works and you can't do that with it and you'd only be able to use the either of the packs during the day if the grid goes down.
He also said that the PowerWall will only be available through SolarCity until end of 2016 which is when it will be made available to other (non-SolarCity) vendors.
 
Here is my GUESS....
The DC/DC is used during charging to buck/boost the PV output to a current at the correct voltage for the battery. If the panel series is slightly lower than the pack, it boosts the voltage so that pack will charge. If the PV string has a higher output voltage, the converter will buck it. This, of course, raises the question "does the DC/DC have an MPPT capability for the best PV harvest?".

When in discharge mode, the PW must be using the DC/DC or some other method to regulate current draw from the pack. Current inverters will consume everything the DC side can supply up to the power limit of the inverter so, with today's hardware, that function must be provided by the PW.

Next generation inverters should be capable of managing battery state by properly apportioning available PV power between AC generation and battery charging. At night, that same inverter should be capable of supplying the house (or a portion of the house's needs) from the PW(s). This will require a grid isolation function or a current clamp on the grid in lines to hold grid usage to near zero when living within the battery's ability to source power to your house.

Interesting times we live in. Obviously Tesla sees the writing on the wall WRT grid tie and net metering. Utilities have way too much say with what we do to generate our own power and they will use this say to stifle the PV competition. They have already started in some locations. Given this, my guess is there will be an emphasis on grid isolation while your house consumes your stored power. Having your house off grid would seem to be the only way to combat the utilities argument that your equipment exists on their grid thus they have control. Regretfully, this will require either incredibly fine granularity in the switching or over sizing of both the PV and Storage elements to catch the boundary conditions.
 
Here is my GUESS....
The DC/DC is used during charging to buck/boost the PV output to a current at the correct voltage for the battery. If the panel series is slightly lower than the pack, it boosts the voltage so that pack will charge. If the PV string has a higher output voltage, the converter will buck it. This, of course, raises the question "does the DC/DC have an MPPT capability for the best PV harvest?".

When in discharge mode, the PW must be using the DC/DC or some other method to regulate current draw from the pack. Current inverters will consume everything the DC side can supply up to the power limit of the inverter so, with today's hardware, that function must be provided by the PW.

Next generation inverters should be capable of managing battery state by properly apportioning available PV power between AC generation and battery charging. At night, that same inverter should be capable of supplying the house (or a portion of the house's needs) from the PW(s). This will require a grid isolation function or a current clamp on the grid in lines to hold grid usage to near zero when living within the battery's ability to source power to your house.

Interesting times we live in. Obviously Tesla sees the writing on the wall WRT grid tie and net metering. Utilities have way too much say with what we do to generate our own power and they will use this say to stifle the PV competition. They have already started in some locations. Given this, my guess is there will be an emphasis on grid isolation while your house consumes your stored power. Having your house off grid would seem to be the only way to combat the utilities argument that your equipment exists on their grid thus they have control. Regretfully, this will require either incredibly fine granularity in the switching or over sizing of both the PV and Storage elements to catch the boundary conditions.
Yes, you are right, there must be MPPT function + Boost/Reduct
The PW will have priority on the Inveter during the day to use PV generation (first come first serve), once full or near full (reduced Amps going into Battery Pack) the Inverter will use the remaining
To use the exact amount of energy stored once its dark or a high usage (> PV generation) is required (dryer, A/C, hoven), if there is not AC clamp / sensor, on both generation + usage wires, how will the PW know how much current to give back to the inveter; Here I don't agree: grid isolation contactor can not work because we are limited to 2kW of output by the PW, and if your loads are higher ... you need the grid for the rest

About the SolarCity phone call, no comment ... he should educate himself before anything, and that's what we are trying to do here ;)
 
In Quebec, 99% of all electricity produced is from renewables - Hydro and wind.
Presently there is not Time Of Use (TOU) pricing. Cost of electricity is the same during all 24 hours.
This makes the PW return on investment very far in the futur.
Solar grid-tied PV systems are the economic way to go here, and even there, it takes years to meet the
break even point.

The only viable business case here is with off grid houses or cottages, located far in the woods...

That said, the technology is interesting, but it needs more integration with the house or grid, with a
technical electrical plan to manage the loads with the PW.

Francois
 
If you look at how most off-grid systems work, they go direct from solar panels to the battery, all in the DC domain. The PowerWall is different because it does so with DC-DC Converters instead of MPPT Charge Controllers. I would be very interested to see a given application size specified with a "traditional" off-grid system and also with a PowerWall based system. By specified, I mean a complete competitive bid with BOM and labor clearly and completely spelled out.

My impression is that the SolarEdge Power Optimizers move the MPPT function to a junction box such that each panel has an individual MPPT function that works independent of other panels. Then the amperage is aggregated and one of the big advantages of micro-inverters over string inverters is mitigated - where the different panels have different output due to shading or whatever. The PowerWall system then plugs in after that and sees constant fixed voltage.

Power Optimizer | SolarEdge

The reason I really was looking forward to micro-inverters is the possibility of adding increments without penalty. However, the more I researched, the more I wasn't sure this was really the right approach. There is a worry about micro-inverter reliability as well as cost. Finally, adding any means for using your solar array during a power outage was complicated and expensive. That leads me to the next question... are there DC to AC inverters that are designed to be daisy chained and can be expanded in a modular fashion? Maybe that can work in a N+1 or just N redundancy arrangement? Does that even make any electrical sense?

- - - Updated - - -

just spoke with a SolarCity representative who called me about my PowerWall reservation. His understanding of what he's been informed is that both the 7kW and 10kW battery packs are for backup only and will ONLY WORK when the grid is down. Yes thats right, the 7kW will only work when the grid is down. When I asked about then how could people 'buy cheap at night and use that energy during the day to keep costs down just like Elon stated on multiple occasions' he said that basically everyone else is wrong and that's not how it works and you can't do that with it and you'd only be able to use the either of the packs during the day if the grid goes down.

So, I don't believe the question of backup or daily cycling algorithms are contained in the PowerWall itself. The PowerWall is merely a battery with built in charge controller, discharge protection, and liquid thermal management. The actual power apportion functionality is handled by the inverter. In other words, if the inverter they are using and the way they are wiring up will only use DC power from the PowerWall when the inverter has detected a grid power outage, then it doesn't matter which PowerWall is hooked up, it will only use the PowerWall as backup. If the PowerWall is wired into a solar panel string, how would the inverter know when the PowerWall is operating versus the solar panel string? In that case, it would daily cycle no matter which PowerWall you plug in. You'll just get capacity degradation of the 10 kWh backup version of the PowerWall to 80% or 60% or whatever they rate "end of life" in 2.75 to 4 years or so. A smarter inverter could take into account the cost from the grid at time, the solar power output, and apportion power appropriately.

The PowerWall product does seem to be a side show to the real meat right now, which is Power Pack.
 
(snip)
The reason I really was looking forward to micro-inverters is the possibility of adding increments without penalty. However, the more I researched, the more I wasn't sure this was really the right approach. There is a worry about micro-inverter reliability as well as cost. Finally, adding any means for using your solar array during a power outage was complicated and expensive. That leads me to the next question... are there DC to AC inverters that are designed to be daisy chained and can be expanded in a modular fashion? Maybe that can work in a N+1 or just N redundancy arrangement? Does that even make any electrical sense?

Sure they can. You have to use inverters of the same brand (sometimes also same model) and wire all of them in parallel. Plus they need some sort of communication network between them, so only the high end models are capable of this (but they are also the ones you would want in a typical home application.)

When you do so, you can add up the power outputs (ie. 3 1kW inverters in parallel will give a total 3kW output.)

Look at wk057's thread about his off-grid system. I think he has something like 10 inverter in parallel. :D

So, I don't believe the question of backup or daily cycling algorithms are contained in the PowerWall itself. The PowerWall is merely a battery with built in charge controller, discharge protection, and liquid thermal management. The actual power apportion functionality is handled by the inverter. In other words, if the inverter they are using and the way they are wiring up will only use DC power from the PowerWall when the inverter has detected a grid power outage, then it doesn't matter which PowerWall is hooked up, it will only use the PowerWall as backup. If the PowerWall is wired into a solar panel string, how would the inverter know when the PowerWall is operating versus the solar panel string? In that case, it would daily cycle no matter which PowerWall you plug in. You'll just get capacity degradation of the 10 kWh backup version of the PowerWall to 80% or 60% or whatever they rate "end of life" in 2.75 to 4 years or so. A smarter inverter could take into account the cost from the grid at time, the solar power output, and apportion power appropriately.

The PowerWall product does seem to be a side show to the real meat right now, which is Power Pack.

With the 10kW PW, you just have to setup your inverter to only use the PW when the grid is down. Even if you don't size the PW correctly (ie. it is empty after 5hours and the grid still isn't back up) I sure hope you don't have power outages more than once a week !
 
Last edited:
Micro-inverters were a cool idea, but they don't make much sense once you add a battery into the equation because you really want to keep everything DC at least until you reach the battery. That's why SolarEdge is splitting the micro-inverter functionality into two parts: a DC-DC converter (the Power Optimizer) that gives you the per panel power optimization you want, then a single DC-AC inverter after the battery. You certainly could build a modular DC-AC inverter but it doesn't make much sense if you aren't combining the power optimization function - you're better just having people over provision the inverter to begin with (which won't add much cost) and then have them use up the excess capacity as they add panels. If PW is a big hit, Enphase is basically screwed.

I suspect you are right that PowerWall is a side show for Tesla compared to PowerPack. Thermal management is Tesla's special sauce for long battery life and it doesn't scale down very well to a 10 kWh pack. I suspect that is why Tesla is not really pushing the small daily cycle pack other than in a very few markets (e.g. Germany) where the economics are compelling for local reasons.
 
I agree, but the 10kW version for power outage backup is compelling any where. Yeah, it's more expensive than a combustion generator, but so what? Not having a noisy belching generator during power outages is worth a lot, as is the complete lack of maintenance needed.