Tesla Battery

SolarPowered

PowerWalls voltage parameter is unorthodox and actually requires more solar panels as an investment.

The string size for panels would require 400vmp per string, to account for voltage drop and losses.
This also means even though power wall is within a C10 charge rate it is more susceptible for the system to stay in bulk/absorb, if the system doesn't have enough amps to drive the charge on cloud cover days.

Unlike smaller voltage systems (under 48v) you could use a single GT panel at 60v, down convert voltage, which actually maintains a better harvest (per individual panel), and can drive a system out of bulk absorb better even at a C20 rate, VS the higher voltage lithium system at a C10 rate.

Essentially Power Wall because of it's reversed engineering will have elevated initial system costs 75% greater than a standard AGM system. During system life there is no guarantee that even though the batteries charge rate is more efficient that the system will "harvest" better and be optimized compared to lower voltage systems.

Ampster

What is the "reverse engineering" that you mention and how does that affect system cost. There has to be some savings of wire gauge because they operate at lower amps. From what I have seen from the SolarEdge promotionnal videos they say it means less costly hookups for the installer.

The PowerWall/SolarEdge combo just looks like a DC coupled system with no need for a charge controller because the communication between the BMS and the Power optimizers takes care of setting the voltage. If the voltage is higher than the Powerwall current will flow proportionate to the energy produced by the panels.With Lithium chemistry they don't have to worry about "out of bulk absorb" If the Powerwall doesn't need a charge then that energy will be inverted and sold to the grid or run the load. I assume the Power Optimmizers take care of maximmizing the harvest. Are you saying there is an inherent flaw in the SolarEdge inverter/ Optimizer design and they are less efficient inverting string voltage DC to AC?

SolarPowered

Ampster wrote: »

What is the "reverse engineering" that you mention and how does that affect system cost. There has to be some savings of wire gauge because they operate at lower amps.

The PowerWall/SolarEdge combo just looks like a DC coupled system with no need for a charge controller because the communication between the BMS and the Power optimizers takes care of setting the voltage. If the voltage is higher than the Powerwall current will flow proportionate to the energy produced by the panels.With Lithium chemistry they don't have to worry about "out of bulk absorb" If the Powerwall doesn't need a charge then that energy will be inverted and sold to the grid or run the load. I assume the Power Optimmizers take care of maximmizing the harvest. Are you now saying there is an inherent flaw in the SolarEdge inverter/ Optimizer design and they are less efficient inverting string voltage DC to AC?

Reverse Engineering(I.E) referring to TESLA's terms using kWh in regards to Ah (battery capacity), or using higher voltages than common off grid low voltage systems under 48V, (TESLA 380V).

It's in the initial solar array design.

Lower voltage systems require less panels per sub array/less surface area for "harvest". Each panel works more independently for harvesting power then plugging (x) number of panels into a series string for a higher voltage.

Lower voltage systems under 48V will convert left over volts to amps, where as with solar edge TESLA more of the focus is to increase voltage, while minimally regarding amperage.


As an example since I compare a 10kWh powerwall, to a 12V AGM bank here is how the scenario plays out.
(I.E) if you use a 2000watt array using 60V (33.5 amp parallel) GT panel(s) on a 12V AGM battery system, it will harvest more power efficiently, VS a 2000watt DC optimized (400V series of panels), 5amp TESLA 10kWh battery system. The AGM system is more efficient because amps are not vulnerable, where as voltage is susceptible to voltage drop, and temperature rise. The lower voltage system requiring less volts for a a 12V system would also convert left over volts to amps, driving a charge even on cloud cover days where as the the 400V tesla system would just lose amps (optimizers do not help in this scenario).

I don't know how exactly solar edge sets it's charge settings but typically for outback you can set modes for bulk,absorb, silent, equalize, and float for lithium batteries, when used with lithium home board, and BMS.

Or disregard the homeboard and use a schnieder electric lithium/nihm solar charge controller, which offers preset's.

PNjunction

The thing I'm just pointing out is to remember the context. That is, the powerwall is a hands-off system, whereas one can make their own system out of lifepo4 prismatics following the KISS principle with the proper knowledge.

With that in mind, I guess I'm guilty of thread-drift. The thing is, we can get so lost in the content and bench-racing of specs, that newcomers generally tend to think only in terms of small-cylindrical cells, usually used / gray market to save on cost, and put all together with duct tape, unless they just want to buy a properly engineered system and walk away from it.

This would be akin to someone wondering why their $4 little agm from a computer ups doesn't seem to act the same as a Rolls-Surrette.

Speaking of which, I could have helped you in the agm department with the ebike. Too late now.

I'll shut up, since this IS about the TESLA battery and needs no more drift.

Ampster

PNjunction wrote: »

The thing I'm just pointing out is to remember the context. That is, the powerwall is a hands-off system, whereas one can make their own system out of lifepo4 prismatics following the KISS principle with the proper knowledge.

With that in mind, I guess I'm guilty of thread-drift. The thing is, we can get so lost in the content and bench-racing of specs, that newcomers generally tend to think only in terms of small-cylindrical cells, usually used / gray market to save on cost, and put all together with duct tape, unless they just want to buy a properly engineered system and walk away from it.

This would be akin to someone wondering why their $4 little agm from a computer ups doesn't seem to act the same as a Rolls-Surrette.

Speaking of which, I could have helped you in the agm department with the ebike. Too late now.

I'll shut up, since this IS about the TESLA battery and needs no more drift.

I like your description of the Powerwall as a hands off system. I think that will have appeal to some segment of the market but not to many people on this forum. It will have an impact on the market just by virtue of its $350/kWhr price. As an example I was assisting friend who lives in Hermosa Beach and desperately wants of get off the grid or at least independant of Southern California Edison. He has a Toyota RAV4 EV and previously had an EV1 so he is comfortable with newer technology unorthodox as that might be. LOL We calculated the cost per kWhr of storage for two bids he got and they were around $800/kWhr. He decided to wait and see under the expectation that prices will come down. The relevance to this thread is the effect the Tesla battery has on prices for storage.
I am with you on making my own. I was fortunate to find some used Voltronix (relabled Winstons) and they are serving me well. I should start a thread because I learned a lot from that experience. BTW, My AGM story is almost 4 years ago and I know what you mean by the duct tape E bike crowd.

Ampster

SolarPowered wrote: »

Reverse Engineering(I.E) referring to TESLA's terms using kWh in regards to Ah (battery capacity), or using higher voltages than common off grid low voltage systems under 48V, (TESLA 380V).

It's in the initial solar array design.

Lower voltage systems require less panels per sub array/less surface area for "harvest". Each panel works more independently for harvesting power then plugging (x) number of panels into a series string for a higher voltage.

Lower voltage systems under 48V will convert left over volts to amps, where as with solar edge TESLA more of the focus is to increase voltage, while minimally regarding amperage.
.............
.

Okay I understand how you mean the term "reverse engineered" As far as using kWhrs vs. Ahrs is concerned maybe its about what we have become accustomed to thinking about. I know when I think about most batteries I do think in terms of Ahrs because that is how they are labeled. When I think about my electric bill I think in terms of kWhrs. . You mentioned a panel as having so many watts because panel voltage can vary so we would have a hard time if we had to multiply voltage times amps every time we tried to figure out how much power it put out. Voltage and Amps are what we can measure, but it is Watts that give us the answer to compare panels. I don't think the Tesla PowerWall people invented the term kWhrs to confuse any body. I do think they chose to do a high voltage battery pack for a number of reasons that make sense to an engineer. They already had expertise in high voltage with their cars. They had done some grid level battery systems that were perhaps in the kilovolt range. I have also heard that inverting high voltage DC to 240 or 120 AC is more efficient than inverting 48v DC. Those are just guesses on my part as I try to understand what they were thinking.

I think I understanding what you are saying about keeping the panel voltage low so you get more amps. I believe that is essentially what MPPT is all about, ie. finding that sweet spot of voltage and amperage that produces the most watts. That is what micro inverters and as I understand them what the SolarEdge Optimizers do. Microinverters do it by outputting parallel AC devices at the same voltage so the amperage adds. Each micro inverter finds that sweet spot of amps vs volts to output the most power. Since the output is a fixed voltage the amperage adds to produce optimum power (Watts). I don't completely understand the optimizers but I have heard it explained that they also output a fixed voltage but do so in series by using powerline communications to keep the string at a fixed DC voltage, but each panel is operating at a low voltage. In the end all we really care about is how much power a system produces and the only way to measure that is multiplying Amps times Volts. If anyone has an explanation that is more correct or more clear please chime in. I am not an Electrical Engineer but am trying to understand this subject from a layman's standpoint.

I do not understand what you mean when you say that "systems under 48V will convert left over volts to amps" I don't remember anything like that from physics class. How can there be left over volts when the system is at 48 volts? Where do I attach my VOM to measure such a thing?. If this leftover voltage thing does exist, why does it exist in low voltage systems and not in high voltage systems?

PNjunction

Ampster wrote: »

I like your description of the Powerwall as a hands off system. I think that will have appeal to some segment of the market but not to many people on this forum.

Guess I get to put a dollar in the broken promises jar.

The appeal may be for this thread topic, but not to the forum in general. Here, much of the discussion revolves around those piecing together batteries, controllers, panels, chargers, gen sets etc, based upon their needs and helped by those who have gone before. It is not really a hands-off kind of thing like the powerwall is.

And please take this the right way - I'm trying to be very nice about it...

I mentioned the outside advice coming from those primarily in the motive-power application field, be it commercial EV, E-bike, rc-modeling - what have you. Those needs, and the batteries we choose to support a long-term solar users needs that come here differ greatly from a hands-off system for the average consumer who doesn't care what is screwed into the wall. That is, we are also calculating solar-insolation hours, autonomy, and even things like the fact that most don't ever want to go past 50% DOD no matter what chemistry they use.

The basic analogy is like having an auto-parts salesman who deals primarily with SLI starting batteries, giving advice to off-gridders who run flooded forklift batteries along with extensive power-budget modeling, to create on their own - a finely crafted system to give them a week or so of autonomy. Different needs, different techniques, and different battery setups.

My preoccupation in regards to small cylindricals is nearly the same. On a Pb level, I see no need to use hundreds of 2v/5a Hawker Cyclon Agm's to build a huge solar house bank instead of using large lead-prismatics typically. Yes, it CAN be done, but that would be a specialist "one off" kind of system done just for the kicks of it.

Anyway, that's my concern. I see a lot of advice from those who have NEVER built a properly functioning solar system, even from lead, to those who are already doing it with lifepo4. That, and the dismay that some solar forums that used to cater to the DIY element made from *commonly available* items, going the way of EV / Ebike forums that already exist.

Ok, that's TWO dollars in the broken promise jar!

SolarPowered

I would just end on this note. Power Wall is a rich mans "toy".

The majority of the people here are looking for the most practical, affordable, and efficient means of storing power, or distributing power at lower costs than what can be provided by a third party.

TESLA essentially has to do something phenomenal to attract buyers, and they aren't achieving that phenomenon by making things reverse engineered, or proprietary as being different from the norm.

Any grid system at this point that costs $4.00 a watt would rather stay paying the utility company.

While those off grid are looking for that practical under $5.50 watt, or consider going grid.

Grid parity has everything to do with economics, a way of life and deviating from that fundamental ideology, makes it the top 1% rich mans toy.

Ampster

SolarPowered wrote: »

I would just end on this note. Power Wall is a rich mans "toy".
..............
Any grid system at this point that costs $4.00 a watt would rather stay paying the utility company.

While those off grid are looking for that practical under $5.50 watt, or consider going grid.

Grid parity has everything to do with economics, a way of life and deviating from that fundamental ideology, makes it the top 1% rich mans toy.

Okay, I just touched a sensitive nerve, and perhaps I haven't shown enough respect for the cost effective systems you may be running..

In order to understand your per watt figures, I need to ask a question. I know the $4.00 per watt number for grid tie is based on the panel capacity. For the $5.50 per watt figure are you using per watt of storage or per watt that can be delivered over a fixed number of hours?

For example, 9 months ago I paid an installer $17,000 for a 4kw grid tie system ( $4.25 per watt)
A few months ago I bought an inverter and some batteries for $7,000) and tied them together with the grid tie system. That system has a battery capacity of 24kWhs. and can deliver 4kW continuous for 3 hours (50% of battery capacity) I have a total system cost of $24,000 at a cost of $1.00 per watt of storage or $6.00 per output watts. What is your benchmark per watt figure base on, capacity or output.?

westbranch

Ampster, it is the WATT you use that counts! sitting in storage has no value till it is put into use...

ie I have 2 battery banks, on gets a tickle once a year for a test load...otherwise I have 'sunk costs of $XX' but I need redundancy (100% OG) so my per watt (overall cost has to include the redundant equipment I have 'sitting' in wait to be put into action...

inetdog

For a GTI system, the single figure of merit is the wattage. How many kWh that translates to will depend on local solar conditions.
For a unit with storage both kW and kWh are important. All the storage in the world with low output will not drive your loads, but all the instantaneous output in the world will not run things at night.

SolarPowered

Ampster wrote: »

Okay, I just touched a sensitive nerve, and perhaps I haven't shown enough respect for the cost effective systems you may be running..

In order to understand your per watt figures, I need to ask a question. I know the $4.00 per watt number for grid tie is based on the panel capacity. For the $5.50 per watt figure are you using per watt of storage or per watt that can be delivered over a fixed number of hours?

This is based on the SYSTEMS TOTAL NAME PLATE DC WATT

Ampster wrote: »

For example, 9 months ago I paid an installer $17,000 for a 4kw grid tie system ( $4.25 per watt)

That is fighting a losing battle to the (tier 1), kWh price.
Lets say your energy in tier 1 is $.15 per kWh, however your bills sky rocket because you are either on TOU, which is the silly mans game, or you deal with the "conservative" tier structure up to tier 4.
Lets say you are that guy that gets into tier 4, your monthly bill is $300 AVG, per month, (these are my typical clients).
(I.E) If we take tier 1 rate @ $.15kWh, and tier 4 rate @ $.36kWh, our means AVG cost of that kWh would be $.255kWh.

Last system(s) I design built for 5 clients on a group on buy, was the LG300N1c-B3 with the enphase micro inverter. Between the 5 clients in the group on the AVG cost to build all 5 roughly came to $3.60 per watt for (39kWh systems), one system required a 200amp servuice main upgrade and teh final cost on that system was $3.79 per watt.
Prior to the fed tax credit the calculation of production equated that the value harvested from the system would generate power @ $.14kWh, after fed tax credit it was estimated to accelerate the investment production at under $.11kWh.
This means teh systems beat the consumer cost of grid parity by 1 penny before fed tax credit, and at fed tax credit the production is below whole sale which beats grid parity.

The LG systems are expected to see the ROI of the system break even on year 8 before fed tax credit, and on year 5.5 after fed tax credit. Any thing after that isn't relative and is considered free power against the utility provider.

Ampster wrote: »

A few months ago I bought an inverter and some batteries for $7,000) and tied them together with the grid tie system. That system has a battery capacity of 24kWhs. and can deliver 4kW continuous for 3 hours (50% of battery capacity) I have a total system cost of $24,000 at a cost of $1.00 per watt of storage or $6.00 per output watts. What is your benchmark per watt figure base on, capacity or output.?

To me it makes no sense why some people go hybrid. GTI is as efficient as the system can be at a 95.5% thresh old and better, so if the system is in NEM as it should be in the grid, that is your maximized value.
the moment battery "hybrid" systems are installed into the grid, there is a 15% efficiency loss in the transmission of charging the battery (for lithium based systems, the number is worse with AGM/lLead Acid). Even though TESLA claims the battery is 90% efficient there are still transmission losses of 5% between the CC/battery/inverter, so any over production harvest is "lost", there is no added value in the over production and maximum summer insolation.
This means people are taking systems for GTI that were optimized at 95.5%, and then decide to have a system that is "hybrid and optimal at a maximum of 80.5%. It makes absolutely no sense even if the guy that invested into hybrid thinks he can sell the power back to the utility at a higher peak TOU, (It doesn't work that way).

SolarPowered

At $6 a watt your return wont happen until year 17 of a system that see's a 20 year warranty, and that's if you built the system accordingly to 110% of the last years annualized consumption. If the system is not built to 110% then you will NEVER see the return against grid parity, and the rate that utility companies inflate energy cost from year to year.

Ampster

SolarPowered wrote: »

At $6 a watt your return wont happen until year 17 of a system that see's a 20 year warranty, and that's if you built the system accordingly to 110% of the last years annualized consumption. If the system is not built to 110% then you will NEVER see the return against grid parity, and the rate that utility companies inflate energy cost from year to year.

I will start a new thread in a while so we can discuss this at more length without hijacking the Tesla Battery thread. I'm on TOU and not tiers, but you are correct the paybacks are long.

lkruper

westbranch wrote: »

Ampster, it is the WATT you use that counts! sitting in storage has no value till it is put into use...

ie I have 2 battery banks, on gets a tickle once a year for a test load...otherwise I have 'sunk costs of $XX' but I need redundancy (100% OG) so my per watt (overall cost has to include the redundant equipment I have 'sitting' in wait to be put into action...

I am thinking of adding battery backup to my existing generator/transfer switch so that I can have quiet evenings when the power goes out. I am grid-connected. I know that it is not recommended to add batteries to an existing bank because they would not be the same age and the older batteries would limit the utility of the new batteries. But what if I start with two 6v T105's and later decide I need the power to last longer? I know I could have two battery banks, but how easy is it to switch the inverter to use the secondary bank? I am thinking of a manual switch-over, but are there connectors that are robust and allow for a quick cut-over to the second bank?

I address this to you because I see you mention you have two battery banks.

westbranch

I use a marine battery selection switch right now, tested only, but I plan to add a second inverter soon ,so I will shift some of the loads from inverter 1 to #2.... you want the switch to be a 'break before make 'not the reverse so it guarantees the 2 banks will not feed simultaneously

lkruper

westbranch wrote: »

I use a marine battery selection switch right now, tested only, but I plan to add a second inverter soon ,so I will shift some of the loads from inverter 1 to #2.... you want the switch to be a 'break before make 'not the reverse so it guarantees the 2 banks will not feed simultaneously

Thanks. What happens if both batteries are momentarily connected to the inverter at the same time? One advantage would appear to be no loss of power to load.

westbranch

yes that would be a benefit, I just have a good tummy feeling when there is no chance of an 'ooops' occurring.. personal preference

inetdog

westbranch wrote: »

yes that would be a benefit, I just have a good tummy feeling when there is no chance of an 'ooops' occurring.. personal preference

The biggest potential problem is that if the two battery banks are charged to different voltages (State Of Charge, SOC) then there can be a really large (damaging) current flowing from the higher voltage battery bank into the lower voltage bank during the time that both are connected.
A brief connection during switching will not necessarily be bad, but if the marine battery switch has a 1 + 2 position, you need to make sure you never leave the switch there unless you have matched the bank voltages first.

lkruper

inetdog wrote: »

The biggest potential problem is that if the two battery banks are charged to different voltages (State Of Charge, SOC) then there can be a really large (damaging) current flowing from the higher voltage battery bank into the lower voltage bank during the time that both are connected.
A brief connection during switching will not necessarily be bad, but if the marine battery switch has a 1 + 2 position, you need to make sure you never leave the switch there unless you have matched the bank voltages first.

Thanks, this is helping me with a potential strategy to increase battery capacity should I start small. I was considering a charger (this is grid-connected battery backup) that charges two different banks. That should isolate the two banks during charging, correct?

Ampster

lkruper wrote: »

Thanks, this is helping me with a potential strategy to increase battery capacity should I start small. I was considering a charger (this is grid-connected battery backup) that charges two different banks. That should isolate the two banks during charging, correct?

I used a similair strategy with my 48v Lithium pack to increase capacity. I started with 32 cells and paralleled two together in 16 pairs. That way you only need one inverter and one charger/charge controller. Later I added another 16 for three parallel sets and 16 sets (48V). I think the hobbyist Lipo guys call this 3P16S. As long as the Ahr rating of the cells is the same it should work fine. I have seen some comments that this is not a good idea with Lead Acid. I don't know all the reasons for for not doing this in Lead Acid. I do know when I added the third battery to the buddy pairs they were all very close to the same voltage because I paralleled those cells for a few days to get them all to the same voltage and SOC. The charging/discharging curves are very flat so it is hard to do with a VOM and charger/discharger.. I have 4 spare cells in case one of mine goes bad.

As far as your question about charging is concerned this eliminates the need to isolate strings for charging. Just remember if the batteries are paralleled to the inverter, isolating the chargers will have no effect because the chargers will be in parallel by virtue of their inverter connection. If you use separate strings (which I also could have done with Lithium) the system gets much more complicated, but an AB switch, like another poster suggested, is a simple solution.

Since this thread is about the Tesla battery I will describe how I think this is handled by the PowerWall. To add PowerWalls together they are connected in parallel and the internal BMS and DC to DC converter take care of balancing the cells within the PowerWall and the SOC of each Powerwall with respect to the other Powerwalls. I have read that those devices communicate via DC Powerline but there may be additional low voltage communications available.

bebop89

I like the Aquion battery system better because it can be easily integrated to existing solar components like charge controllers and inverters.

Deveak

Honestly, beyond high efficiency and current rating I see no advantage to lithium. High charge and discharge is not exactly super important in off grid power systems or back up systems. Price, lifetime and capacity is. Lithium fails on price, the lifetime is good but no better than a fork truck cell with proper care (cycle life is also hotly debated and still to early to verify, I've seen 2000 cycles listed at 50%, I've seen 6000, I know I get 3250 on my fork truck cells at 50%.) They also age like lead acid, dont know if its faster or slower. They have sever balance and resistance issues so a bad cell can be a death sentence. They require constant monitoring by a bms that can fail horribly, expensively and quickly lead to damage. The power wall is a high voltage battery with next to no inverter that will work with it (100% sure, Tesla or the sister solar company his brother owns will sell one though), its high voltage so in most areas you wont be able to install it yourself. Weight is irrelevant in a stationary home application. Capacity is related to price. Which it fails at. Most home systems already have high capacity banks for autonomy. The low charge and discharge rate of other batteries is less important when you have larger batteries. Given a choice I would choose a quality fork truck battery, cycle it shallow or a NIFE battery bank which can be charged at better rates that lead with no sulfation (prices do suck though).