Elon, I love you... but the PowerWall isn't that great...... yet.

[schueppert]

The main problem I see with other types of energy storage that involve a mechanical aspect is the huge round trip efficiency loss. Such storage mechanisms could never compete with a more solid state approach such as batteries.

Why would you think other systems would have worse efficiency? Large scale hydro and compressed air systems in place already have better round trip efficiency than batteries, and improved systems are coming along. Oh, and batteries aren't sold state: they're chemistry. But just to be clear, I certainly think there is an opportunity for batteries at utility scale if the price is right.

 

[jgs]

JGS, I respectfully but vehemently disagree. In states like mine, the public service commission (PSC) is chaired by politically appointed members by the governor. Such that when a change in government happens, saying going from Democratic to Republican or vise versa, the public commission is going to represent the interests of the party in power. By that statement all that I mean is that what your calculated ROI is today with a 6 year return on investment may no longer be valid after the next election. You may say "yeah free market wins" but not necessarily. Such as in the case of Wisconsin, PSC specifically both raised the grid connection fee (the amount you pay regardless of how much power you use) and lowered the delta between on and off peak kWh usage. This drastically changed the ROI on solar. They did this specifically to make sure that solar is non-competitive, they even said as much. And the net result is they raised the fees ON EVERYONE. The only way in which this made solar more competitive is if you were completely off grid. Which is very expensive to do and in no way competitive to a standard grid connection when you look at todays cost of a complete off grid solution (the amount of solar and storage need for year round consumption in a northern state). Thus in one fell swoop they turned a 6 year ROI into a 20+ year ROI AND RAISED FEES ON EVERYONE ELSE!

Oh. Wisconsin. Sincere condolences and best wishes for your ability to bring your state government under control again. Although here in Michigan we don't exactly have anything to brag about. Anyway…

I take your point, and it's true that especially in the north where a grid connection is almost mandatory, the utility still wields a lot of anti-competitive power. But is it the case that the utility is able to prevent a customer from installing a PV-and-battery system that never ever sells power back to the grid, that is, where all power produced in a given home remains in that home? Insofar as the system would still be physically coupled to the interior wiring of the home, which would also be physically coupled to the grid, and therefore transitively the PV system would be in some sense coupled to the grid, I can see how electrical codes could be used to prevent this, but my expertise in these matters is almost nil.

 

[glhs272]

I believe net metering is still legal. However solar is just not competitive. Personally I am going to do it anyway, ROI be damned. But I believe I am the very much in the minority. And that's the problem, if we really want to move the needle on solar adoption, we have a long way to go and many battles to fight.

 

[stopcrazypp]

Why would you think other systems would have worse efficiency? Large scale hydro and compressed air systems in place already have better round trip efficiency than batteries, and improved systems are coming along. Oh, and batteries aren't sold state: they're chemistry. But just to be clear, I certainly think there is an opportunity for batteries at utility scale if the price is right.

I thought pumped storage is 70-85% efficient round trip? Compressed air is also ~70% efficient. Tesla quotes this battery at 92% round trip (DC-DC I presume, so throw in inverter at ~95% efficient and you get 88% overall). So batteries still have a slight edge (although I suppose not necessarily true of all battery types).

However, in a large scale, the cost of pumped storage and compressed air per kWh is so much lower, the lower round trip efficiency doesn't matter much.

 

[Johan]

However, in a large scale, the cost of pumped storage and compressed air per kWh is so much lower, the lower round trip efficiency doesn't matter much.

But even in massive applications where the properties of a battery (for example can go from charge to discharge in the blink of an eye) there will be a time when the higher price will cancel out by higher efficiency. This of course depends on variables such as total power handled, life time of the system and price/cost of electricity (being wasted). And massive storage applications will handle massive amounts of MWhs for years and years...

 

[wk057]

"Never compete"? I would imagine that in many deployments a more important metric than round-trip efficiency would be cost per kilowatt hour, something it's easy to imagine a mechanical storage system doing well at.

Why would you think other systems would have worse efficiency? Large scale hydro and compressed air systems in place already have better round trip efficiency than batteries, and improved systems are coming along. Oh, and batteries aren't sold state: they're chemistry. But just to be clear, I certainly think there is an opportunity for batteries at utility scale if the price is right.

So far in my testing of the Model S cells I've been able to achieve right around 99% charge/discharge round trip efficiency with the cells (1X Ah in, 0.99X Ah out). So efficient in fact that it might as well be ignored for my personal use case. Pretty impossible to beat that with something mechanical. By solid state I meant no moving parts.

 

[schueppert]

Batteries don't get 99% round trip efficiency. Tesla isn't even claiming anything close to that. I suspect you are evaluating Model S based on information you are getting from the user display - you really need to measure the battery directly if you want to know the real story. I've seen posts around here before where people have done that and found numbers more like the 92% referred to above. And that's DC to DC, so 85 to 90% for AC to AC seems realistic. An that doesn't include losses due to battery thermal management, which is essential for Tesla's claimed cycle life and may well be quite high if the temperature is very hot or very cold.

More importantly, for most any time shifting application I can think of the cost of the input electricity is very low (<5c per kWh and quite possibly free) versus the cost of storage (12c per kWh or more) so efficiency doesn't contribute much to overall solution economics.

 

[wk057]

Batteries don't get 99% round trip efficiency. Tesla isn't even claiming anything close to that. I suspect you are evaluating Model S based on information you are getting from the user display - you really need to measure the battery directly if you want to know the real story. I've seen posts around here before where people have done that and found numbers more like the 92% referred to above. And that's DC to DC, so 85 to 90% for AC to AC seems realistic. An that doesn't include losses due to battery thermal management, which is essential for Tesla's claimed cycle life and may well be quite high if the temperature is very hot or very cold.

More importantly, for most any time shifting application I can think of the cost of the input electricity is very low (<5c per kWh and quite possibly free) versus the cost of storage (12c per kWh or more) so efficiency doesn't contribute much to overall solution economics.

I'm talking about individual 18650 cells from the Model S as used in my off-grid solar project.

I'm also talking about the battery's efficiency, nothing else. Not a DC-DC converter, DC-AC inverter, etc. Generally, in mechanical based storage such as a flywheel, compressed air, liquid salt, whatever there is a huge efficiency loss just inherit in the core process, where the battery cells (core process here) are much more efficient.

 

[schueppert]

Cool. DC-DC roundtrip efficiency just means measure the DC power in when charging and DC power out when discharging. If that isn't what your figures are based on, what are you measuring?

I agree with you that moving parts means friction, heat, and reduced efficiency. But the chemical processes in batteries are exothermic which means heat and reduced efficiency. So you have unavoidable problems either way.:smile:

 

[stopcrazypp]

Cool. DC-DC roundtrip efficiency just means measure the DC power in when charging and DC power out when discharging. If that isn't what your figures are based on, what are you measuring?

I think what DC-DC means is any DC-DC conversion like with the Powerwall (which would step up the DC voltage of your panels to charge the batteries and step down the voltage of the battery to supply the inverter to your equipment). The 92% roundtrip figure for the Powerwall includes such DC-DC conversion losses (but not DC-AC inverter losses).

 

[wk057]

Cool. DC-DC roundtrip efficiency just means measure the DC power in when charging and DC power out when discharging. If that isn't what your figures are based on, what are you measuring?

I agree with you that moving parts means friction, heat, and reduced efficiency. But the chemical processes in batteries are exothermic which means heat and reduced efficiency. So you have unavoidable problems either way.:smile:

I meant to have the word converter/inverter in the previous post. (corrected)

The internal resistance of the cell is a huge factor when figuring what is lost during charge cycles, and for these cells it is incredibly low (lower is better).

For any system there will need to be a charging mechanism and a discharging mechanism, the efficiency of which will vary. For batteries, there are also super efficient chargers available (upper 90s%). Depending on the load, there are also super efficient inverters/converters available also (upper 90s%). So even an AC->DC->AC round trip efficiency is going to be better with batteries than any known mechanical system.

For these batteries:
Efficient storage: check.
Efficient charging: check.
Efficient utilization: check.

 

[JeffreyY]

Ok, I went back and did a lot of data lookups. For this analysis I crunched 2 months of power usage data (Juli and January) and I looked at our daily usage pattern. The bottom line is that the Powerwall would allow us to go off-grid for about 8-9 months of the year resulting in significant cost savings.

I also looked at our daytime vs. nighttime consumption and found that we never exceed 7 kWh of nighttime usage, typical number is about 4 kWh. So the Powerwall has the capacity to shift enough solar power from midday to the night. Basically we can cover all of our power needs with solar because of the Powerwall.

And now for the numbers. First up daytime power consumption vs nighttime. In this context daytime means sunrise to sunset.

Representing summer:

5th of May: daytime: 6.48 kWh, nighttime: 3.18 kWh.

And in the winter corner:

5th of January: daytime: 6.37 kWh, nighttime: 6.28 kWh.

I chose these days as examples because they were worst-case within that particular month. It is clear that we would never exceed the amount of power that the Powerwall can provide. Also I was unable to find a single instance of our house drawing more than 1,05 kWh of power at any one hour. So the 2 kW specification + 3kW peak is well within what would be needed.

I used a small 4.32 kWp solar-array as an example. This is the kind of array that would fit on our small roof. We could probably fit a 6 kWp array but that would just make all the numbers even better .

Next up is savings pr. month. In the summer would be around $160 in electricity, 30-40% of that comes from the powerwall shifting the power resulting in powerwall savings of about $56 pr month in summer. In the wintertime the savings are $8 pr month. On average we are looking at around $33 pr month in savings from powershifting. Multiply up to a year and we have about $400 pr year in savings from the powerwall alone and it pays for itself in ~8 years. Basically it will be under warranty the whole time and if you buy the extended warranty you should be good for another 10 years.

This is all based on current prices for electricity and the price is expected to go up over the next 10 years and the payout from selling power to the grid is expected to fall about 50-60%. At that point the powerwall becomes a complete no-brainer here.

I simulated expected changes to power-pricing and the monthly saving went up to $50 per month for the powerwall , now it would pay for itself in 5 years.

In closing here are some stats:

January:
Total power used: 362 kWh.
Solar power generated: 72.9 kWh.
Max daily usage: 17
Min daily usage: 7

July:
Total power used: 265 kWh
Solar power generated: 576 kWh
Max daily usage: 14 kWh
Min daily usage: 4 kWh.

Now I'm off to figure what is so power hungry in our house, 17 kWh is just embarrassing. Watch out, something is about to get turned off for good.

PS. If anybody wants a copy of my spreadsheet let me know. Most of it is in danish, you have been warned

I'd love the spreadsheet

 

[JeffreyY]

Now I'm off to figure what is so power hungry in our house, 17 kWh is just embarrassing. Watch out, something is about to get turned off for good.

PS. If anybody wants a copy of my spreadsheet let me know. Most of it is in danish, you have been warned

Dude, don't feel bad. My house (with cars) pushes 80 KWH/Day in the worst part of the year.

 

[scaesare]

Dude, don't feel bad. My house (with cars) pushes 80 KWH/Day in the worst part of the year.

Ditto. Except it's >100kW/day

 

[Xminus6]

Ditto. Except it's >100kW/day

Yup. Same here. I've yet to isolate what causes all my houses (not just this current one) to use so much energy. Obviously the one constant is me and my family but we've always been highly energy consumptive for some reason and it's never been really obvious to me why.

 

[macpacheco]

PowerWall/PowerPack 1.0 was the lowest hanging fruit type of product. All it needed was 0.01% of total potential energy storage customers (which is at least a US$ 1 trillion/yr market worldwide).
2.0 version is good for ~5% of the total market, but still not the right solution for most.
3.0 will be the real mainstream product.
But the fundamental problem with doing your spreadsheets is if you have net metering you can't assign an ROI on purchasing a PowerWall. But once the PowerWall costs about 1/4 of your total solar installation and includes the inverter it just becomes an affordable luxury that future proofs your system so that when net metering is gone, your solar panels are still economical. And then you have a house wide UPS.
I hope the PowerWall 3.0 will have two sizes, one <10kWh really cheap such that everybody can have some storage (doubling as a basic home wide UPS) and a >20kWh for people that want to live off grid without compromise.
And the PowerPack 3.0 should have at least 300kWh and double the charge/discharge power vs 2.0 with at least 1/3 lower price/kWh.
I predict there will still be a 4.0 version by the time Tesla has several 250GWh/yr GFs producing batteries by the boatloads.
The world needs in the order of 100TWh worth of batteries in stationary storage+another 100TWh (one billion 100kWh average packs) in vehicles to go 100% sustainable energy. Its a huge market.
I wouldn't recommend people try to find a reason to get a PowerWall just to be trendy. There are millions of customers that have real economic need today. Like Australia with its unbalanced metering rules (buy full price, sell at 2/3 discount) or Hawaii.
Its much better to have one PowerPack in the grid than 10 PowerWalls in the same area.

 

[scaesare]

Yup. Same here. I've yet to isolate what causes all my houses (not just this current one) to use so much energy. Obviously the one constant is me and my family but we've always been highly energy consumptive for some reason and it's never been really obvious to me why.

I have a rack of servers/storage/switchgear in my basement that's undoubtedly a good chunk of it...

 

[jbcarioca]

I have a rack of servers/storage/switchgear in my basement that's undoubtedly a good chunk of it...

I have just completed a three month long energy usage audit of my house in Rio de Janeiro. I do not use air conditioning (possibly the only house I know of in Rio that does not) so my total use excludes that giant category.
40% of my total use is in voltage stabilizers/battery backups for computers and other electronic gear. If I can go to a complete off-grid solution I'll eliminate all those losses. Even with grid-tie that could happen, but grid-tie here in Rio is still grossly expensive.
The problem is battery storage, since every type is excessively expensive here. Powerwall would be great but not available here, and no other options is currently economic. A crashed Tesla 85 would be more than enough, if I had the expertise of wk057 or several others.
So, in the context fo the thread, Powerwall seems to be an excellent solution when more traditional solutions are not feasible.

 

[scaesare]

That's the overhead for the backup units themselves and not them and their loads?

 

[jbcarioca]

That's the overhead for the backup units themselves and not them and their loads?

Yes, it excludes the loads. We have some dramatic voltage swings (107-133 v in one twelve hour period) so end out using both conventional battery backup and power regeneration. The latter because I have some serious audiophile equipment and use the same backups for computers. I'd have vastly lower usage if I'd drop battery regeneration. Before I did that I blew the power transformers in a $9,000 power amplifier because of a sudden voltage drop that the battery backup could not cope with. Thus my interim solution.
FWIW, I formerly had a photovoltaic system on an island, where I also had 'pure-sine-wave' inverters. That was impeccably clean power, another advantage of going off grid.