• Have you tried out dark mode?! Scroll to the bottom of any page to find a sun or moon icon to turn dark mode on or off!

diy solar

diy solar

400 volt batteries, the good, the bad, the ugly

SnickeringBear

Solar Addict
Joined
Jul 3, 2025
Messages
647
Location
Northwest Alabama
I'm looking into the logistics of 400 volt batteries. Advantages include higher voltage which means they can use smaller cable sizes. Inverters are available that support the voltage range with Solis 11.4 an example. The disadvantages I'm seeing look like there is a disparity with cell sizes where for example 314 AH cells are 1 kWh each so 128 of them would be 128 kWh of storage. There is a wide range of cell configurations that should be viable. The high end is 450 volts and the low end is as low as 100 volts. As an example, 128 LiFePo4 cells would deliver 435.2 volts at full charge and drop to 320 volts at 0% charge. String sizes could range from 116 at the low end up to 128 at the high end. Perhaps most intriguing, a string of 128 MB56 cells would deliver 256 kWh of storage in a 128 cell string weighing about 3500 pounds and 1175 amp hour cells would deliver 481 kWh of storage with a weight of about 6800 pounds..

I looked into BMS and found a couple that would work. It helps that there are already several commercial 400 volt batteries on the market including the much maligned LG. From a charging perspective, there are chargers, converters, and MPPT's that work with 400 volts. Compatible inverters are available from Solaredge, SMA, Goodwe, Growatt, LG, Delta, and Solis. I have not delved very deep so there are probably several more that will work. Keep in mind that even with an inverter that can handle 400 volts, the BMS still has to be compatible.

So a few questions for discussion, would you consider a 400 volt battery and if so why or why not? There is a slight advantage when charging EV's if using 400 volt batteries as a source. It should be fairly easy to use a buck up MPPT to feed from a 400 volt battery into an EV up to 800 volts.
 
String sizes could range from 116 at the low end up to 128 at the high end. Perhaps most intriguing, a string of 128 MB56 cells would deliver 256 kWh of storage in a 128 cell string weighing about 3500 pounds and 1175 amp hour cells would deliver 481 kWh of storage with a weight of about 6800 pounds..
That's a big a meatball!

So a few questions for discussion, would you consider a 400 volt battery and if so why or why not?

I would, but I'm not very smart.
 
I have always thought the battery pack should be a generic voltage so its easy to invert to AC. So 120vDC or 240vDC
Or a bit higher to make the AC peaks
 
I have and do, here is my system

HV rocks, no fans or noise, battery wiring cost savings over 48V high power hybrids alone covers most of the inverter cost and the S6 is less than half the cost of SolArk / EG4 stuff.

Get the Seplos Hiten ( single 20kWh 4 stack ) for the required canbus communications and then add generic strings of (8) 48V batteries and parallel to the Hiten.

FYI, 128S is considered 409.6V and battery charging with good cells would be ~458V and the Seplos Hiten has RS485 communications and everything is configurable in the BMS.

The "dangers" are no more or less than a 48V system, actually its less due to the 1/8th current, most that poo poo HV have no idea what they are talking about.
 
The "dangers" are no more or less than a 48V system,
Well higher voltage is inherently more dangerous (higher potential of arcing for example, no pun intended), but yes the lower currents would be very nice and reduce other types of danger (thermal for example).
 
I have and do, here is my system

HV rocks, no fans or noise, battery wiring cost savings over 48V high power hybrids alone covers most of the inverter cost and the S6 is less than half the cost of SolArk / EG4 stuff.

Get the Seplos Hiten ( single 20kWh 4 stack ) for the required canbus communications and then add generic strings of (8) 48V batteries and parallel to the Hiten.

FYI, 128S is considered 409.6V and battery charging with good cells would be ~458V and the Seplos Hiten has RS485 communications and everything is configurable in the BMS.

The "dangers" are no more or less than a 48V system, actually its less due to the 1/8th current, most that poo poo HV have no idea what they are talking about.
If you can’t connect the S6 to grid for grid-tie due to no interconnect allowed, how could you charge from grid if needed, just use second S6 for charging, no loads on it?
 
High voltage is deadly. Which takes a simple oopsie and turns it into funeral arrangements. Can it be handled safely? It sure can.
But if the question is, is it more dangerous? Absolutely yes.
Higher voltage means more cells in series and less in parallel. Which also means less redundancy. And redundancy is a very important part of my system.
I'm not for, or against, high voltage. I'm sure that it's where the market is going. But, there's always a tradeoff to everything.
Each person must decide what will work best for their situation.
 
If you can’t connect the S6 to grid for grid-tie due to no interconnect allowed, how could you charge from grid if needed, just use second S6 for charging, no loads on it?
A second one would still fall under the same rules.
 
Same rules but no chance of sell back if there are no loads connected to it and quickly stop. Forgot you and I already had a similar exchange on this topic. lol.
An interconnect agreement has nothing to do with loads.
It is about the type of equipment connected.
Now, if you are talking about breaking the rules anyway, and hopefully not getting caught. Then not having loads, could help. But equipment capable of exporting, can export from the battery.
 
Well higher voltage is inherently more dangerous (higher potential of arcing for example, no pun intended), but yes the lower currents would be very nice and reduce other types of danger (thermal for example).

The lower currents have less ability to keep the ionized path, so the 400 vs 48 V isn't that different. All typical wiring and breakers are already 600V rated and 240Vac is really ~340-380Vdc when rectified, no one is saying the sky will fall over that ...

For the breakers, one should use multi-poll, simple and helps address the arcing concern. On the battery, the S6 has both a contactor and fuse builtin and since communications are required, it shuts down in less than a second the HV DC port for a fault, so no load, no arc.

Most welders are actually under 48V, its the current that does the work.

If you can’t connect the S6 to grid for grid-tie due to no interconnect allowed, how could you charge from grid if needed, just use second S6 for charging, no loads on it?

No different than any other hybrid, you use and external charger. This would work for $250 but need manual on/off, primary charging would be typically solar, or a generator which would from the S6
 
The lower currents have less ability to keep the ionized path, so the 400 vs 48 V isn't that different. All typical wiring and breakers are already 600V rated and 240Vac is really ~340-380Vdc when rectified, no one is saying the sky will fall over that ...

For the breakers, one should use multi-poll, simple and helps address the arcing concern. On the battery, the S6 has both a contactor and fuse builtin and since communications are required, it shuts down in less than a second the HV DC port for a fault, so no load, no arc.

Most welders are actually under 48V, its the current that does the work.



No different than any other hybrid, you use and external charger. This would work for $250 but need manual on/off, primary charging would be typically solar, or a generator which would from the S6
Well ok. A HV chargeverter. This may have answered the mail on how to do this and not grid connect at all, just double conversion. Can this hook into the Hiten or LG batteries easily? Understand I need an on/off or controlled contactor.
 
Last edited:
The lower currents have less ability to keep the ionized path, so the 400 vs 48 V isn't that different. All typical wiring and breakers are already 600V rated and 240Vac is really ~340-380Vdc when rectified, no one is saying the sky will fall over that ...

For the breakers, one should use multi-poll, simple and helps address the arcing concern. On the battery, the S6 has both a contactor and fuse builtin and since communications are required, it shuts down in less than a second the HV DC port for a fault, so no load, no arc.

Most welders are actually under 48V, its the current that does the work.



No different than any other hybrid, you use and external charger. This would work for $250 but need manual on/off, primary charging would be typically solar, or a generator which would from the S6
You can get an ionized path way easier at higher voltages than lower voltages. Lighting is an extreme example, but it is one of very high voltage with very low current.

240VAC is far different than DC because it is AC crossing the zero line.

I am not against using HVDC but you have to treat it with more respect than LVDC. HV is always inherently more dangerous than LV.

Go grab your car battery terminals, then grab the terminals of a HVDC battery, and get back to me. Note: I've been hit by 300+VDC before, it is NOT fun.
 
You can get an ionized path way easier at higher voltages than lower voltages. Lighting is an extreme example, but it is one of very high voltage with very low current.

240VAC is far different than DC because it is AC crossing the zero line.

I am not against using HVDC but you have to treat it with more respect than LVDC. HV is always inherently more dangerous than LV.

Go grab your car battery terminals, then grab the terminals of a HVDC battery, and get back to me. Note: I've been hit by 300+VDC before, it is NOT fun.

An ionized path is a negative resistance characteristic, meaning Increased current leads to lower voltage for an arc to happen ( Google it ).

Its really a moot point, with the S6 there is no way to cause a continuous arc ( which is the concern for fire ), that is the benefit on UL9540 which the S6 complies with. You need a load for any arc, the S6 stops it with 1 second thanks to canbus communications, voltage monitoring ( battery low voltage disconnect due to the arc voltage drop as seen by the inverter ) and or communications stopping or commanded by the battery.

Anything with home power systems deserve respect and education, its the BS about 48 vs 400V that I'm addressing. Not long ago the move from 12->24->48V had the same conversations ....
 
I've been hit by 480v a few times. And while that did suck, the worst was a neutral that fell out of a bad connection and against my hand. This was the neutral of a 277v lighting multi wire branch circuit. I really thought that it was my last day above ground.
But this was AC. And I would not have survived if it were DC.
 
The biggest danger of DC is if the voltage makes your muscles contract, you have no way to let go, and your skin resistance would be the key to what voltage is going to cause that.

Bottom line is wear protective gloves and long sleeve shirts
 
The biggest danger of DC is if the voltage makes your muscles contract, you have no way to let go, and your skin resistance would be the key to what voltage is going to cause that.

Bottom line is wear protective gloves and long sleeve shirts
Which just shows that it is indeed more dangerous.
 
I have and do, here is my system

HV rocks, no fans or noise, battery wiring cost savings over 48V high power hybrids alone covers most of the inverter cost and the S6 is less than half the cost of SolArk / EG4 stuff.

Get the Seplos Hiten ( single 20kWh 4 stack ) for the required canbus communications and then add generic strings of (8) 48V batteries and parallel to the Hiten.

FYI, 128S is considered 409.6V and battery charging with good cells would be ~458V and the Seplos Hiten has RS485 communications and everything is configurable in the BMS.

The "dangers" are no more or less than a 48V system, actually its less due to the 1/8th current, most that poo poo HV have no idea what they are talking about.
how 'batteries' are arranged? For example, the 'top' cells in 128 cell battery would be at 400+ potential relatively to the '-' of the battery if that's what is connected to the ground. Wouldn't it create difficulties for measuring/communicating those cell's individual voltages? Is active balancing even possible?
 
how 'batteries' are arranged? For example, the 'top' cells in 128 cell battery would be at 400+ potential relatively to the '-' of the battery if that's what is connected to the ground. Wouldn't it create difficulties for measuring/communicating those cell's individual voltages? Is active balancing even possible?
You would never connect the battery positive or negative to ground.
And balancing is done at cell voltages, the total pack voltage has nothing to do with balancing.
 
You would never connect the battery positive or negative to ground.
Any distribution system that relies on getting power to a distant location in one way or another has to rely on an earth ground. Look at AC distribution which at high voltage has a single lead that is hot and a single lead that is the return. The return lead is bonded to ground in any MGN system. It is the only way to protect the overall system from lightning.
 

diy solar

diy solar
Back
Top