Author Topic: High Voltage solar batteries  (Read 7870 times)

0 Members and 1 Guest are viewing this topic.

Offline maxx4096Topic starter

  • Newbie
  • Posts: 9
  • Country: be
High Voltage solar batteries
« on: February 13, 2018, 09:24:16 am »
Hello everyone,
I live in a net metered area but that is about to change, so while looking for a battery to add to my system (16 panels, 4.16kWp, Solaredge inverter, ~350VDC), I learned that there are basically low voltage ones (12 to 48VDC) and high voltage ones (350 to 450VDC).

I can picture how a low voltage system works, just plug the batteries in series and there you go (plus safety and charge control circuitry).
I just cannot imagine wiring 400V worth of lithium-ion battery packs in series and thinking that it is safe to do so.

How do these work? Do they bring down the voltage to charge lower voltage "banks" and then step that up while discharging; incurring losses on the way down and again on the way up?

Thanks
 

Offline ahbushnell

  • Frequent Contributor
  • **
  • Posts: 734
  • Country: us
Re: High Voltage solar batteries
« Reply #1 on: February 13, 2018, 12:13:41 pm »
Hello everyone,
I live in a net metered area but that is about to change, so while looking for a battery to add to my system (16 panels, 4.16kWp, Solaredge inverter, ~350VDC), I learned that there are basically low voltage ones (12 to 48VDC) and high voltage ones (350 to 450VDC).

I can picture how a low voltage system works, just plug the batteries in series and there you go (plus safety and charge control circuitry).
I just cannot imagine wiring 400V worth of lithium-ion battery packs in series and thinking that it is safe to do so.

How do these work? Do they bring down the voltage to charge lower voltage "banks" and then step that up while discharging; incurring losses on the way down and again on the way up?

Thanks
Higher voltage is probably more efficient.  300 to 400 volts is a convenient voltage to work with line voltages.  Higher voltage means lower currents at a given power.  So less copper is required and better efficiency.  Operating at 300 to 400 volts means there does not need to be a transformer between the batteries and the line.  Electric cars work in the 300 to 400 volt range. 
 

Offline maxx4096Topic starter

  • Newbie
  • Posts: 9
  • Country: be
Re: High Voltage solar batteries
« Reply #2 on: February 13, 2018, 01:00:41 pm »
Higher voltage is probably more efficient.  300 to 400 volts is a convenient voltage to work with line voltages.  Higher voltage means lower currents at a given power.  So less copper is required and better efficiency.  Operating at 300 to 400 volts means there does not need to be a transformer between the batteries and the line.  Electric cars work in the 300 to 400 volt range.
That's my understanding, right now my inverter sees 230VAC (grid voltage) or about 330Vpp. It makes a lot of sense to just chop the 350VDC from the string to a sinewave through a H-bridge or something like that.

I wasn't aware that EV were using that kind of voltage.

So theoretically, I can buy 10 3.7V batteries with built-in protection, wire them in series and have my own baby powerwall? (I have no plan to build it myself, I need something that comes with a warranty and all the required safeties and paperwork)

Thanks for your reply!

Edit: typo
 

Offline jolshefsky

  • Regular Contributor
  • *
  • Posts: 227
  • Country: us
    • Jason DoesItAll
Re: High Voltage solar batteries
« Reply #3 on: February 13, 2018, 02:28:45 pm »
I have a grid-tied solar system as well and want to get access to the solar energy separately—mostly for the case of a power failure, at which time my grid-tie system shuts down with the grid and 4000 watts of solar sits idle. Mine is 434VOC (volts open-circuit) which puts it out of reach for basically anything I can find. I think I can wire up a SMPS that's 240VAC rated, but that voltage is very close to absolute maximum ratings, and that makes me wary.

I found (but then lost and can't find again :rant:) a UPS that would emulate the grid for the grid-tie inverter I have when the power went out. I remember it was quite expensive, didn't work with all grid-tie inverters, and required a dummy load to dump excess solar energy since the grid-tie inverter expected to be able to do that.

I later decided that a better option would be a more traditional "generator" sub panel where I could switch in some local circuits to power. I found the Magnum Energy MS-4024-PAE Inverter as my preferred choice. Then I realized I might be able to use DIN-style DC-DC converters to boost power to a 24-volt battery system directly from the 434V-peak solar array. One example is the Mean Well WDR-480-24 from Mouser which supplies 24 volts at 20 amps. Plus, voltage can be adjusted up to 28V so it could actually float-charge a 24-volt pack. It isn't the prettiest solution, but it would make available some of the solar capacity when the power is out. I am skeptical that it could work because the WDR is probably too "smart" and would not be able to run off 400V DC.

I also found a non-helpful thread asking basically the same question. The problem is, if the grid is down, what the heck can you do with 400+VDC other than make yourself dead very very quickly with one tiny mistake?
May your deeds return to you tenfold.
 

Offline station240

  • Supporter
  • ****
  • Posts: 967
  • Country: au
Re: High Voltage solar batteries
« Reply #4 on: February 13, 2018, 02:34:29 pm »
At a basic level, the typical grid tie solar inverter takes the 150V-600V* from a string of solar panels, boosts it up to charge a bank of 350-450V capacitors. The inverter then takes the charge in the capacitors and uses it to output ~230V AC.

So adding a HV battery is simple, just provide an external connection into the capacitor bank, and a an external comms port (CANbus, RS422, etc). In theory at least, surprisingly few companies making the modified inverters.

For added strangeness, the Tesla Powerwall, despite having a HV DC version, is internally a 48V battery with a bi-directional DC-DC. Naturally the DC versions aren't being made any more |O due to lack of 3rd party inverters, the AC version with internal 48VDC to 240VAC inverter is all that is for sale.

* 600V is max string voltage for solar cells, typical max voltage actually produced is more like 440V.
 

Offline maxx4096Topic starter

  • Newbie
  • Posts: 9
  • Country: be
Re: High Voltage solar batteries
« Reply #5 on: February 13, 2018, 05:08:56 pm »
At a basic level, the typical grid tie solar inverter takes the 150V-600V* from a string of solar panels, boosts it up to charge a bank of 350-450V capacitors. The inverter then takes the charge in the capacitors and uses it to output ~230V AC.

So adding a HV battery is simple, just provide an external connection into the capacitor bank, and a an external comms port (CANbus, RS422, etc). In theory at least, surprisingly few companies making the modified inverters.

For added strangeness, the Tesla Powerwall, despite having a HV DC version, is internally a 48V battery with a bi-directional DC-DC. Naturally the DC versions aren't being made any more |O due to lack of 3rd party inverters, the AC version with internal 48VDC to 240VAC inverter is all that is for sale.
That's interesting, so, they are going for economy of scale (everything @48VDC) and taking the conversion losses...

My prime candidate* is a LG Chem "RESU7H For SolarEdge", it has all the ticks I need on the box (reputable brand, 350V-450VDC, RS485 interface to communicate with my inverter, warranty and all) but I wonder what is inside and how that works,... (Dave if you are reading this please do a teardown of a solar battery system or maybe a Fundamentals Friday on high voltage/high power DC-DC electronics)

@jolshefsky: I know that my inverter when coupled with a battery system can do backup power, but it requires a special component to safely disconnect the home supply from the grid. (Frying linemen is bad, really really bad). That option has to be approved in the country where the system is installed, only in AU, NZ, SA and US, if I read that correctly... As this is not available for me, my backup system consists of a camping stove, a BBQ, a gas genset and 2 extension cords so I can plug the fridges and central heating in case of an extended power cut. I leave the DC alone and curse loudly until the power comes back (can take some time, but it works!  ;) )...

* OFF TOPIC: The Tesla Powerwall was on my list until a car weight amount of space junk was added to the solar system (pun intended).
 

Offline Nusa

  • Super Contributor
  • ***
  • Posts: 2416
  • Country: us
Re: High Voltage solar batteries
« Reply #6 on: February 13, 2018, 06:34:57 pm »
The car was already in the solar system. It's just in a different orbit around the same sun now.
 
The following users thanked this post: fourtytwo42

Offline GerryBags

  • Frequent Contributor
  • **
  • Posts: 334
  • Country: gb
Re: High Voltage solar batteries
« Reply #7 on: February 13, 2018, 07:21:16 pm »
Not sure if this helps, Jolshefsky, but many industries in the UK use 400v, and isolation transformers are available (like the Scneider ABL6TS160U - 230v-400v/230v 1.6kVA) which could be useful.
 

Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 8973
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: High Voltage solar batteries
« Reply #8 on: February 14, 2018, 01:49:53 am »
If you have a battery pack that's more or less the same voltage as rectified mains, a common VFD will work nicely from it. EV battery packs tend to be just above the voltage of rectified 240V which is helpful.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Offline MarkL

  • Supporter
  • ****
  • Posts: 2120
  • Country: us
Re: High Voltage solar batteries
« Reply #9 on: February 14, 2018, 03:56:33 pm »
...
So theoretically, I can buy 10 3.7V batteries with built-in protection, wire them in series and have my own baby powerwall? (I have no plan to build it myself, I need something that comes with a warranty and all the required safeties and paperwork)
I realize your scenario is theoretical, but one thing to watch out for is if you're putting a bunch of individually protected cells in series, many integrated protection schemes (such as a Polyfuse) will go open on fault.  If that happens, the protection device will have the remaining series voltage across it from the other batteries and it may not be rated for that voltage.
 

Offline maxx4096Topic starter

  • Newbie
  • Posts: 9
  • Country: be
Re: High Voltage solar batteries
« Reply #10 on: February 14, 2018, 08:44:39 pm »
I realize your scenario is theoretical, but one thing to watch out for is if you're putting a bunch of individually protected cells in series, many integrated protection schemes (such as a Polyfuse) will go open on fault.  If that happens, the protection device will have the remaining series voltage across it from the other batteries and it may not be rated for that voltage.
That's why I want to learn more about how these things are actually made.
Seen under that light, the observation made by @station240 earlier in this thread, makes a lot of sense, keeping the battery bank voltage low (48V) eases the requirements for the safety device(s), and may even offset some of the costs of stepping the voltage down then up, by using more mundane, off the shelf parts... (speculation here).
Thanks!
 

Offline John Heath

  • Frequent Contributor
  • **
  • Posts: 444
  • Country: ca
  • 2B or not 2B
Re: High Voltage solar batteries
« Reply #11 on: February 15, 2018, 12:20:44 am »
An area I do not see being explored is high high voltage. Doubling the voltage stored on a condenser is the square of the stored energy. that energy curve is going up rather fast as the voltage goes up. At some voltage it will be better than a lithium battery and the energy density of oil. For a quick example 1 u f condenser charge to 1 K volt is 1 M watt 1 u second or 1 watt for 1 second , 1 Joule. At 10 K volt it is 100 joules , 100 K volt 10,000 joules and 1 M volt 1 M joules of stored energy. That is a lot of stored energy for a 1 u f condenser. A lightening bolt in a vacuum jar 1 u f condenser to the tune of 1 M joule of energy. In hours it is 280 watts for 1 hour or equivalent to a 23 amp hour 12 volt battery. At 10 M volt it is 100 23 amp hour 12 volt batteries. Now we are talking. 1 u f condenser charged to 10 M volt = 1 hundred 23 amp hour 12 volt batteries. That should keep the porch light running for a while. One minor problem , it is little dangerous to deal with a 1 u f condenser that is charged to 10 M volts. You would not want to lick your finger and touch it to see if is charged or not.
 

Offline maxx4096Topic starter

  • Newbie
  • Posts: 9
  • Country: be
Re: High Voltage solar batteries
« Reply #12 on: February 15, 2018, 04:36:44 pm »
An area I do not see being explored is high high voltage. <snip>
 1 u f condenser charged to 10 M volt = 1 hundred 23 amp hour 12 volt batteries. That should keep the porch light running for a while. One minor problem , it is little dangerous to deal with a 1 u f condenser that is charged to 10 M volts. You would not want to lick your finger and touch it to see if is charged or not.
That is properly scary! 10MV would probably arc to anything coming within meters of it, and I am not sure you can even find anything to bring that kind of voltage, in a controlled manner, to something more "reasonable" like 220V.

As a home defense device, I can see some potential... Just imagine:
Quote
Hi ! Willy Mays here with the MightyCap!
Scared of thieves? Annoyed by mail, rodent, wildlife or stepmother? Get rid of all that with our killer 10MV capacitor, the MightyCap! Powered by mother nature wrath, it will get rid of every annoyances in your life! Watch it convert them into a bright blue flash and enjoy the sweet smell of burned flesh with a nice refreshing touch of ozone... All that for only 19.99M$! Call NOW and receive this amazing "Danger High Voltage" sticker absolutely free !

In all seriousness I think a HV capacitor based system will simply be way bigger and more expensive than a comparable battery based system. Wider separation between component, better insulation, and so on. Not ruling it out in the future but it doesn't seems practical now.
 

Offline John Heath

  • Frequent Contributor
  • **
  • Posts: 444
  • Country: ca
  • 2B or not 2B
Re: High Voltage solar batteries
« Reply #13 on: February 16, 2018, 04:10:32 am »
I love the smell of ozone in the morning. Yes , need to stew on the finer details of a practical version of the mighty 10 MV cap.
 

Offline f4eru

  • Super Contributor
  • ***
  • Posts: 1086
  • Country: 00
    • Chargehanger
Re: High Voltage solar batteries
« Reply #14 on: February 19, 2018, 09:43:11 pm »
Yeah, basically you want to optimize the voltage to have a "reasonnably low" current in order to avoid too much copper.

- 12V system, typically up to max 1-2 kW
- 24V system, typically up to 4-5 kW
- 48V system, typically 10 kW -> this is the maximum nominal battery voltage that is usually considered safe to touch, as it goes close to 60V when fully charged.

- 300-600V system : typical for up to a100-200 kilowatts, like big UPS, powerful EVs,

It's simply limited by physics. 100-300A continuous is a current that is quite OK to handle with serious copper, over that, the switching, protection, and cabling elements become too expensive, and you should go up in voltage.

Also, requirements differ wildly. An EV for example, has to provide short peaks at maximum power, solar storage has rather to be efficient for continuous power, very very different designs !

Usually, battery systems don't go to kilovolts, that's too dangerous because High voltage DC loves to creep and spark, while batteries love to burn down to the ground in a smelly firework when in presence of sparks. Not a good combination.
you rather parallelize smaller <1kV systems. Not sure how Telsa does it in Australia.

There's also not much systems inbetween 48V and 300V, as you have the hassle of protecting against those crazy humans coming in contact with the high voltage, you may as well go up in voltage. Some electric forklifts run off 96V perhaps.



10 Megavolt storage caps?
Forget it.
Look page 22, Fig. 5.4.1-2, that's a microfoarad range smoothing cap, good for 50ms.
https://www.energy.siemens.com/ru/pool/hq/power-transmission/HVDC/HVDC_Proven_Technology.pdf
And that's only a ridiculously small 0,5 MV !
« Last Edit: February 19, 2018, 10:24:10 pm by f4eru »
 

Offline f4eru

  • Super Contributor
  • ***
  • Posts: 1086
  • Country: 00
    • Chargehanger
Re: High Voltage solar batteries
« Reply #15 on: February 19, 2018, 10:35:29 pm »
Quote
Then I realized I might be able to use DIN-style DC-DC converters to boost power to a 24-volt battery system directly from the 434V-peak solar array. One example is the Mean Well WDR-480-24 from Mouser which supplies 24 volts at 20 amps.
Not a good idea.
No, really.
Seriously.

I saw one guy doing that with a similar converter, and wondering why these poor supplies go up in smoke after a while.

But wonder no more :

These converters are designed for a constant voltage input, and have a regulated constant voltage output, without regulated current(but OL protection) !
That's not what you want to connect to it.
With a PV panel, you have a slowly varying constant current source at the input -> not good !
With a battery, you absolutely need to have a combined constant current/constant voltage (like a lab PSU), or else your PSU will overload trying to push too much current in a discharged battery.

Your converter will hiccup, be unstable in regulation loop,oscillate and struggle between a badly behaving load, and an even more badly beaving source, switch on and off, make strange noises and smells, and most likely, ultimately, find the working point in which it will die in fire or smoke.

The Tts (time to smoke) is inversely proportionnal to LC_elko, which is the level of "Chineseness" of the electrolythic capacitors.

Use a proper MPPT/charger instead.
« Last Edit: February 19, 2018, 11:14:08 pm by f4eru »
 

Offline digsys

  • Supporter
  • ****
  • Posts: 2209
  • Country: au
    • DIGSYS
Re: High Voltage solar batteries
« Reply #16 on: February 19, 2018, 11:27:59 pm »
IF you have money to throw away, or your premises travels at greater than 5km/hr :-) , go with the LiIons, otherwise AGM, SLAs etc (lead-acid) are
a heck of a lot cheaper, easier to keep balanced and for some chemistries, better life. We often have members putting in between 12-24X series
systems. Often, members pick up whole sets from truck companies etc for incredibly cheap prices, as they are often swapped at ~ 80% capacity.
Sometimes, they can even be re-conditioned to get a few more %. Just a thought
Hello <tap> <tap> .. is this thing on?
 

Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 8973
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: High Voltage solar batteries
« Reply #17 on: February 20, 2018, 01:16:59 am »
Lithium is already cheaper than SLA. Conventional deep cycle is still cheaper but not by much and won't last as long for regular cycling use.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Offline f4eru

  • Super Contributor
  • ***
  • Posts: 1086
  • Country: 00
    • Chargehanger
Re: High Voltage solar batteries
« Reply #18 on: February 20, 2018, 05:47:14 am »
Yep exactly !
Lithium got cheaper, except for wet lead acid :
Pb Battery Wet   0,13   €/Wh   2000 cycles
Lithium battery    0,26   €/Wh   6000 cycles
Pb Battery Gel   0,27   €/Wh   1000 cycles
I wonder, do you have wildly different numbers ?

« Last Edit: February 20, 2018, 05:50:07 am by f4eru »
 

Offline NiHaoMike

  • Super Contributor
  • ***
  • Posts: 8973
  • Country: us
  • "Don't turn it on - Take it apart!"
    • Facebook Page
Re: High Voltage solar batteries
« Reply #19 on: February 20, 2018, 06:35:02 am »
Nissan Leaf modules regularly go for $200/kWh, even less in bulk. I have also used 32650 LiFePO4 batteries that are about half that cost, but the time to assemble a sizable pack out of those cannot be ignored. The latter ends up cheaper than good lead acid (and is close to a drop in replacement with the addition of a BMS) and hence why I went for that to replace a deep cycle that was over 5 years old at the time.
Cryptocurrency has taught me to love math and at the same time be baffled by it.

Cryptocurrency lesson 0: Altcoins and Bitcoin are not the same thing.
 

Offline digsys

  • Supporter
  • ****
  • Posts: 2209
  • Country: au
    • DIGSYS
Re: High Voltage solar batteries
« Reply #20 on: February 20, 2018, 07:42:54 am »
Comparing like for like.
Yes, many of our members (and others), are scoring huge bargains on ex-Teslas, Leafs, iMievs etc packs, but bargains on Lead acid are far greater
right now. One even scored a couple 100 Yuasas for ~$100 (transport).
You need nearly 4x the number of series cells for the same voltage, and cell matching / management is a HECK of a lot more complex and dearer.
(one of my main roles). They also go out of match a lot easier. Plus, many of the Lithium BMS systems I see are pretty crappy.
The #cycles is also questionable. Lithium types can easily have reduced life IF they're treated badly (which I often see when coming from ex-EVs).
In a fixed installation, where space is not a concern, the weight / volume is usually not a consideration.
I would always go for Lithium, but that's what I like to play with, but even now I've scored a few 120AH AGMs for free, hope I can recondition them.
Hello <tap> <tap> .. is this thing on?
 

Offline maxx4096Topic starter

  • Newbie
  • Posts: 9
  • Country: be
Re: High Voltage solar batteries
« Reply #21 on: February 20, 2018, 01:13:15 pm »
Yep exactly !
Lithium got cheaper, except for wet lead acid :
Pb Battery Wet   0,13   €/Wh   2000 cycles
Lithium battery    0,26   €/Wh   6000 cycles
Pb Battery Gel   0,27   €/Wh   1000 cycles
I wonder, do you have wildly different numbers ?
I'm down with the flu with a good fever, and my brain isn't exactly working right, but on a "per cycle basis", lithium appears as the ultimate winner.
Pb wet 0,39€/Wh as I'll have to replace them 3 times to get 6000 cycles
Li-ion   0,26€/Wh 6000 cycles (no replacement)
Pb Gel  1,62/Wh   6000 cycles (6 replacements)
Not taking into account the rate at which the batteries degrade (say instant death on the 6000th cycle vs slow performance decrease) nor if the price for a given tech goes down with time.

Am I missing something or is it just a question of maintaining the wet batteries to keep them healthy and then cheaper as a consequence?

My use case is for a domestic install, the Li-ion system I have in mind is more or less a plug and play deal. The only Pb based system (an UPS) I ever came across was a custom job for a company data center.

Thanks all for your posts I'll revisit them after a good nap an post something if I have anymore questions...
 

Offline jolshefsky

  • Regular Contributor
  • *
  • Posts: 227
  • Country: us
    • Jason DoesItAll
Re: High Voltage solar batteries
« Reply #22 on: February 20, 2018, 03:25:30 pm »
Quote
Then I realized I might be able to use DIN-style DC-DC converters to boost power to a 24-volt battery system directly from the 434V-peak solar array. One example is the Mean Well WDR-480-24 from Mouser which supplies 24 volts at 20 amps.
Not a good idea. [...] Use a proper MPPT/charger instead.

So far I've had no luck finding a charger that can handle an input of over 400V. I had considered 250VAC rated devices, but they only expect a peak voltage of 350V or so.
May your deeds return to you tenfold.
 

Offline mtdoc

  • Super Contributor
  • ***
  • Posts: 3575
  • Country: us
Re: High Voltage solar batteries
« Reply #23 on: February 20, 2018, 03:51:33 pm »
So far I've had no luck finding a charger that can handle an input of over 400V. I had considered 250VAC rated devices, but they only expect a peak voltage of 350V or so.

They exist.

But - better to run a string voltage with a Vmp of 80-100 V and use a standard MPPT controller and 48V nominal battery bankk.
 

Offline f4eru

  • Super Contributor
  • ***
  • Posts: 1086
  • Country: 00
    • Chargehanger
Re: High Voltage solar batteries
« Reply #24 on: February 20, 2018, 06:19:29 pm »
Am I missing something or is it just a question of maintaining the wet batteries to keep them healthy and then cheaper as a consequence?
You're not missing something. Lithium is cheaper on the long run if you want a lot of cycles.
But you mentionned Li-ion, this ain't li-ion, it's lifepo4. Li-ion probably doesn't have much cycles.
My price was last year (might have gone down 10-20% since then), and for big square cells 60/80 Ah, so the price of the cell is higher than the round cells, but you have much less effort grouping them :)

NiHaoMike : are your prices for used or new LEAF batteries ?
« Last Edit: February 20, 2018, 06:42:50 pm by f4eru »
 

Offline maxx4096Topic starter

  • Newbie
  • Posts: 9
  • Country: be
Re: High Voltage solar batteries
« Reply #25 on: February 20, 2018, 08:20:07 pm »
You're not missing something. Lithium is cheaper on the long run if you want a lot of cycles.
That's good to know, I'm trying to learn about battery technology and cutting my way trough all the marketing stuff..

But you mentionned Li-ion, this ain't li-ion, it's lifepo4. Li-ion probably doesn't have much cycles.
You're right, it is hard to get rid of old bad habits... I read lifepo4 as 'Life-Pee-Oh-four', which sounds like marketing speak, instead of the proper LiFePO4 (lithium ferrophosphate). While it is still a lithium-ion battery (according to Wikipedia) it is no longer the Li-ion of old...

On another subject, that Siemens HVDC stuff you linked earlier already answered a few questions I had yet to ask, as a new 1000KW HVDC line is being built just a few kilometers from where I live...

Thanks again!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf