High Voltage powerwalls?

[fafrd]

I don’t what to get hung up on the definition of ‘powerwall’ versus ‘battery’ but I’m interested to understand whether there are any budget high-voltage batteries out there that are a good value.

By High Voltage I’m taking about ~400V with absolute minimum of 250V.

A 10kW 400VDC battery from BYD can be had for $6000 ($0.60/Wh) while a 5kW 48VDC EG4 battery from Signature Solar can be had for $1500 ($0.30/Wh).

So are there any High Voltage batteries delivering the cost and quality of an EG4-like battery rather than the HV battery offerings of BYD and LG Chem costing ~twice as much?

Are there BMSes out there allowing DIYers to make LiFePO4 powerwalls in the 250-400V range? (100S to 136S)?

 

[Ampster]

There are BMSs used in EV conversions that can handle 400 Volts. Some names that I know of are Orion, EMUS and Elithion. I have used the first two. Any BMS that can handle a hundred or more cell groups is going to cost a lot more. All of them also use external Contactors and current shunts or CTs external to the BMS.

A DIY pack of a hundred cells has to be around $0.15-30 per Wh, plus the cost of a BMS and other parts.
What is the application or use case and size of the pack?

 

[Rocketman]

The Bms I use Batrium I believe can handle the higher voltage (if I remember correctly).
But the contactors are going to be more expensive. Then you will need the inverters and solar charge controllers.

Mass production is driving down the prices of the 48v stuff - I imagine there is not a lot of 200v+ DC stuff out there.

Safety becomes a much larger concern as the voltage grows.

 

[fafrd]

The Bms I use Batrium I believe can handle the higher voltage (if I remember correctly).
But the contactors are going to be more expensive. Then you will need the inverters and solar charge controllers.

I’m going at this on the opposite direction. I’ve found the AIO+EVC I want but it only supports batteries of 250-500V (400VDC nom): https://www.dcbel.energy/wp-content/themes/brook-child/assets/doc/r16_datasheet_FINAL_WEB.pdf

Like the new Solar Edge string inverter with EV charging, this new class of hybrid inverters designed to support fast DC charging directly from solar requires high string voltages to charge an EV and if they also support hybrid capability, the battery they will need will be an EV-like high voltage battery.

Mass production is driving down the prices of the 48v stuff - I imagine there is not a lot of 200v+ DC stuff out there.

Yeah, I built a DIY 14kW 8S 48V battery for $1500 all-in, but it doesn’t look like it’s going to be usable with the new breed of EV hybrid inverters…

Safety becomes a much larger concern as the voltage grows.

Agree. That’s another reason to choose OTS over DIY for an HV battery..,

 

[Ampster]

Like the new Solar Edge string inverter with EV charging, this new class of hybrid inverters designed to support fast DC charging

I was not aware that Solaredge had introduced a device capable of DC fast charging and as a hybrid inverter? The integrated J1772 EV charger has been around for a few years as part of the GT inverter. It has some advantages over separate components because it can program the EV charger to prioritize charging from solar and I believe modulate the EV charging to follow the solar production.

So far as I can tell there is no new class of hybrid inverters capable of DC fast charging but there is a new spin on Vaporware. We shall see if there is even a market for those kind of devices compared to the alternatives current available at significantly less cost.

I also do not think many EV owners feel the need for DC fast charging at home. DC fast charging is important for road trips at rates of 250 kWs. At home 30 or 40 Amps at 240 volts is the equivalent charging rate of 7200 to 9600 Watts per hour. That means for most EVs the equivalent of 28 to 36 miles for every hour of charging. Of course that assumes a solar array of 7 to 10 kW. If I need more I can always charge overnight at off peak rates.

 

[fafrd]

I was not aware that Solaredge had introduced a device capable of DC fast charging and as a hybrid inverter?

No, the Solar Edge inverter with EV charger is a string inverter (grid-tied-only, no battery).

The integrated J1772 EV charger has been around for a few years as part of the GT inverter. It has some advantages over separate components because it can program the EV charger to prioritize charging from solar and I believe modulate the EV charging to follow the solar production.

Yes, that’s what I was referring to. It also requires a high-voltage string…

So far as I can tell there is no new class of hybrid inverters capable of DC fast charging but there is a new spin on Vaporware. We shall see if there is even a market for those kind of devices compared to the alternatives current available at significantly less cost.

Of course it might end up being vaporware, but dcbel claims hybrid EV charger will be available in California next month…

And as far as cost, it’s priced below the equivalent Solark hybrid with all the capability of a Solark + the ability to charge an EV and use it as a second battery…

I also do not think many EV owners feel the need for DC fast charging at home. DC fast charging is important for road trips at rates of 250 kWs. At home 30 or 40 Amps at 240 volts is the equivalent charging rate of 7200 to 9600 Watts per hour. That means for most EVs the equivalent of 28 to 36 miles for every hour of charging. Of course that assumes a solar array of 7 to 10 kW. If I need more I can always charge overnight at off peak rates.

The DC charging and even the DC fast charging is actually secondary to the value proposition. You can also charge in AC and power loads from the EV battery in AC, just at modestly lower efficiency.

However you get the energy jammed into the EVs battery, you’ll end up using less of that energy if you have the battery powering the dcbel hybrid in DC like the house battery and allowing the hybrid inverter to perform the inversion.

It trivially easy to throttle energy in DC so having the EV power the hybrid in DC and the hybrid decide how much DC energy to invert to AC power is going to be much more simple (and also probably more efficient) than having the EV decide how much battery energy to invert and send back to the hybrid to offset loads.

My focus and interest is much less on the charging speed than it is in the use of an EV as an extended battery for overnight load offset as well as export reduction (with minimum wear on the EV’s battery).

 

[Ampster]

You can also charge in AC and power loads from the EV battery in AC,......

You can power AC loads from the EV battery, only if the specific vehicle has a charger that is bidirectional. So far I do not know any EVs that have an onboard charger that can do that. the Ford F150 has announced 240V output but the price tag is expensive and I do not know when they will be available. My definition of Vaporware is announcements without installed product. Pardon my scepticism but the F150 and dcbel fit that definition..

 

[fafrd]

You can power loads from the EV battery, only if the specific vehicle has a charger that is bidirectional.

dcbel claims to have DC charge and discharge capability to the EV. I’m not aware of any EVs supporting DC fast-charging that also support DC discharge, but that is the capability to keep an eye on.

Discharging a battery in DC is a far, far easier proposition than discharging in AC. The EV essentially becomes a high-voltage battery that offers it’s stored DC energy to the hybrid charger/inverter like any other battery. The hybrid does all the heavy lifting of deciding how much power is needed and inverting it to offset loads.

So far I do not know any EVs that have an onboard charger that can do that. the Ford F150 has announced 240V output but the price tag is expensive and I do not know when they will be available. My definition of Vaporware is announcements without installed product. Pardon my scepticism but the F150 and dcbel fit that definition..

You are right to be skeptical -we’ll believe it when we see it.

But AC out is already a common feature in many hybrids. They integrate inverters running off if the 12V battery (so very limited power and energy levels),

Changing that to an inverter running directly off of the EV battery for higher efficiency and greater output levels is a straightforward upgrade and that is all the Ioniq 5 and F-150 Lightening have claimed.

So simple V2L capability running off an islanded AC out port that has nothing to do with grid sync is easy-peazy, only costs a modest amount more than what the average hybrid already offers today, and I’ll be very surprised if we don’t see both vehicles launch with that capability when the hit the market later this year.

I’m assuming your skepticism is focused on the ‘V2G’ and home-backup capability meaning EV’s directly powering a critical loads panel.

You should certainly be able to peer a fridge through an extension chord. Powering a full backup loads panel means bushwhacking some wiring or getting an electrician involved who will insist on a transfer switch at a minimum and possibly additional safety features (so that’s a very different proposition than the ‘using an EV-like-a-generator’ approach that I’m discussing.

The dcbel hybrid inverter/EV charger may prove to be vaporware but the architecture is a natural and extension of the high-voltage hybrid inverters already in use and I believe the arrival of something like this is inevitable.

If the EV standard for DC charging does not also allow for DC discharge, then the problem is with the standard, not the concept.

Inverting power from DC to AC can be dangerous, principally because the AC power needs somewhere to go and so proper throttling must be performed by the inverter (which gets more complicated if that inverter is in the EV).

For an EV to throttle DC power costs nothing - just allow the EV charger access to the EV battery terminals the way it already has access for DC charging.

If DC charging requires a bunch of electronics within the EV between the charging port and the battery, the whole thing is more complicated than it had to be and it will take much more time for DC discharging capability to materialize than it could have.

I was all set to buy an Ioniq 5 and still may but I’m now going to hold off to understand more about the show DC charging and DC discharging landscape…

 

[Ampster]

Discharging a battery in DC is a far, far easier proposition than discharging in AC

I should have been more clear about my reference to discharging and the need for a bidirectional inverter/charger for AC loads. As you point out DC should be easier but I do not know about the protocol. Apparently Chademo has implemented or already had implemented some bidirectional control. I do not know what has been done with CCS. Of course J1772 is an AC protocol.

 

[fafrd]

I should have been more clear about my reference to discharging and the need for a bidirectional inverter/charger for AC loads. As you point out DC should be easier but I do not know about the protocol. Apparently Chademo has implemented or already had implemented some bidirectional control. I do not know what has been done with CCS. Of course J1772 is an AC protocol.

Yes, I’m with you - the key is what CCS has done for DC charging and whether it allows a simple DC discharge path from EV battery to charger or not.

From what the dcbel documentation and specifications indicate, they seem to believe they have access to that capability (EV-as-a-battery), but of course, talk is cheap and the devil is always in the details…

 

[fafrd]

Not as clear and divinities as I’d like, but this suggests that DC-bi-directional-charging’ is a thing: https://electrek.co/2020/12/11/vw-22-kw-bi-directional-dc-charging-station-electric-vehicles/amp/

‘For instance, as we learned with the Quasar, it basically turns your car into a Tesla Powerwall for power out situations.’

Can’t find the references to DC bi-directionality I would like, but the Quasar claims it is bidirectional and also claims it can offset loads when grid-tied: https://wallbox.com/en_us/quasar-dc-charger

‘Quasar is the first bidirectional charger for home use. It allows you to charge and discharge your electric vehicle so you can use your car battery to power your home or the grid.’

‘Add load balancing to allow Quasar to measure how much energy you are using in your home and automatically adjust the charge or discharge power, providing you with the most efficient energy management yet.’

So they sure make it sound as though the Quasar is a bi-directional charger/inverter that can act as a hybrid inverter offsetting loads in the home while grid-tied…

 

[Ampster]

Can’t find the references to DC bi-directionality

I have always assumed that a direct DC connection was capable of bidirectionality subject to protocols to control the load and or the charging current.
That is a different animal than an onboard charger being able to also invert the DC to the AC port. To do that means the onboard charger has to have the hardware to invert and the control protocol to avoid depleting the vehicle battery beyond some setting. If I recall correctly the J7112 protocol originates in the Charging station and tells the onboard charger the correct Amperage. The only thing the vehicle part of the communication does is it tells the Charging station to close the relay to power the high voltage pins on the plug.

 

[fafrd]

More on the Wallbox Quasar: https://support.wallbox.com/wp-content/uploads/ht_kb/2020/12/EN_Datasheet_Quasar.pdf

This states ‘Wallbox has created a direct current (DC) bi-directional electric charger for home.’

It also states: ‘Quasar charges through CCS[6] or CHAdeMO vehicle connectors and can be [b{discharged through a CHAdeMO connector allowing homeowners to take advantage of new vehicle to grid electricity tariffs.’

So the DC discharge is only through CHAdaMO and not CCS (at least this version).

7.6kW max charge power and 6.8kW max discharge power but since this version also states ‘50Hz’ as well as ‘230VAC’, I suspect this may be referring to a first product version released for the European market only…

 

[fafrd]

I have always assumed that a direct DC connection was capable of bidirectionality subject to protocols to control the load and or the charging current.
That is a different animal than an onboard charger being able to also invert the DC to the AC port. To do that means the onboard charger has to have the hardware to invert and the control protocol to avoid depleting the vehicle battery beyond some setting. If I recall correctly the J7112 protocol originates in the Charging station and tells the onboard charger the correct Amperage. The only thing the vehicle part of the communication does is it tells the Charging station to close the relay to power the pins on the plug.

Yeah, DC bidirectionality is far easier than AC bidirectionality.

At least the early version of the Wallbox Quasar appears to have been limited to CHAdeMO for discharge (which supports DC discharge) so that appears to be at least one option for both-directional DC charging but hopefully CCS offers something similar…

A modest PSW (off-grid) inverter for direct V2L power (islanded) with bidirectional DC charging capability to deliver grid-tied or even backup power when connected to a hybrid bidirectional charger seems to be the winning combination.

 

[fafrd]

Found this article from dcbel: https://www.dcbel.energy/2022/01/12/new-year-new-bidirectional-cars/

‘The cars listed above use the CCS charging protocol, which does not yet support CHAdeMO-style bidirectional power flow. So how can automakers promise these features in 2022? It’s because they included additional power inverters under the hood and are outputting alternating current (AC) to satisfy the load through special adapters or standard 120V power outlets.’

‘Once the CCS protocol is formally revised, the majority of EVs in North America will be able to send high-voltage DC power straight from the car’s battery to a Home Energy Station via a proper charging cable, at which point it will be converted to AC that can be used to power the home and potentially be sold back to the grid.’

So seems as though CCS does not yet support DC discharge but is hoped to soon…

 

[fafrd]

Looks as though CCS is planning to support but-directional charging in 3 years (2025): https://www.electrive.com/2019/01/24/charin-bidirectional-charging-with-ccs-standard-by-2025/

 

[Alber C]

High voltage battery system required BMS and BMU. its more complicated then the 48v system. while you use the batteries, the communication is important with inverter. in terms of protection voltage, over charge, and over discharge details . here is a small system based on 512v https://www.coremax-tech.com/produc...v-600v-li-ion-lifepo4-512v-hv-battery-system/

 

[Warren93nature]

High voltage battery system required BMS and BMU. its more complicated then the 48v system. while you use the batteries, the communication is important with inverter. in terms of protection voltage, over charge, and over discharge details . here is a small system based on 512v https://www.coremax-tech.com/produc...v-600v-li-ion-lifepo4-512v-hv-battery-system/

Thanks for this information

 

[bogi]

Apart from comm protocol (BMS <-> inverter), safety and HV equipment something that push me off from HV ideas is the fact that once you build your HV battery you can not grow it easily. Not as easy as in 48 V system. The need for battery<->inverter communication makes things so unDIY friendly for now. I planned to use 96s or 192s battery build based on Chevy Volt modules but the next step would be power capacity increase. Inverter is talking to one battery set (i do not know HV inverters capable to talk with more than one battery pack). Then there is charging issues - imagine you have 100kWh (my goal for the beginning is 60kWh) - HV inverters i found are able to use up to two PV strings only. That means those string must be quite big in terms of PV panels count. In my case i can't place so many PV panels in one sopt. I have more smaller spots, different angles, different parts on the property. Yes, i can use AC coupling but for now i'm at the stage that is it much easier to hook up any 48V MPPT charger to the battery bus bar if i find another place on the roof to install another 4 PV panels.
Both systems have their own pros and cons. I like small currents with high power in HV systems and all the advantages that come from it though.

 

[loki429]

Hello
Great discussion so far and some interesting points raised. For me the decision came down to high current vs high voltage and the efficiencies of powering a large load. Cabling becomes a lot easier at 200A than 1000A!
As for the dangers of high voltage? Well 60V is already considered lethal so is it any more dangerous to use 384V? Don't forget you're also working with 110 or 240V AC so no matter what DC voltage you use there is lethal voltage somewhere on the system. Best to consider all parts of the system to be potentially lethal until confirmed otherwise.

I'm building a 384V battery to power a high DC load (electric motor). As the EV industry has discovered, to power high load situations low voltage motors are not suitable. Instead they have opted for higher voltages in motors and batteries. And HV for large electric motors is not something new; it's been used for 20 years or more in industrial applications.
I found a bi-directional onboard charger from an EV supplier that supports the HV battery with charging from AC, with the added benefit of a 12V DC output plus 240V AC output for vehicle to load situations. Coupled with a HV MPPT solar controller I will have the ability to charge the battery from solar or charge from AC supply. Then the reverse - power an AC load from the battery.
I have the charger and cells already (128s1p) but am waiting on some parts for electrical connections. For BMS I have an Orion. I'm aiming to start some testing in a couple of weeks - I'll let you know how it goes!

Cheers,
Jon