Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Ugh, these two-lead LFP batteries without any sort of BMS communication are fairly nasty devices. The inverter/charger does not know the maximum safe charge or discharge current, the cell temperature, the cell balance state, or really anything else except the voltage. If the actual BMS in the battery (assuming there’s one in there at all) wants the charger to slow down, it has no way to tell it to do so. The charger has no way to know what it needs to do to get the cell balancing circuit (if any) to work. And the BMS (again, assuming it exists) can’t even communicate the state of charge to the inverter/charger.

At least this particular setup uses a somewhat dignified 24-ish volt setup instead of the usual awful “12V” that is often seen in this genre of battery.



BMS <-> "solar inverter" communication is not required. All LFP batteries that aren't raw cells have a BMS, which have over/under voltage protection. A bad charger isn't going to do anything.

There's also nothing wrong with 12V setups (see all the RVs that are out there).


> BMS <-> "solar inverter" communication is not required.

It’s “not required” in the sense that it works. Poorly. Have you contemplated how balancing between paralleled “12V” LFP packs works? LFP has a voltage vs SoC curve that is very unpleasantly flat for purposes of cell balancing without intelligence, and almost no intelligence is available in these arrangements.

> Theres nothing wrong with 12V setups (see all the RVs that are out there).

There is so much wrong with them that it’s hard to even know where to start.

Let’s suppose you have an RV-like setup that is designed to power one single 15A 120V receptacle. That’s 1800W, which is pretty small by the standards of modern inverters. It’s also 150A at 12V. Want decent efficiency? That means you want to lose less than, say, 2V on the wires, so you need to carry those 150A over a round trip resistance below 13.3mOhm. If you use the ABYC ampacity table, that’s 2AWG or larger wire. Fortunately 2AWG will carry 150A a respectable distance without excessive voltage drop.

Now, 150A is a lot of current. 150A will make massive sparks, light things on fire, and weld sizable pieces of metal even if 12V is unlikely to electrocute you. And you’re feeding that wire or bus in multiple places (because you may have multiple batteries and a beefy charger), so you remember to put giant 150A fuses or breakers at all the feed locations, right?

But wait, 12V needs below 0.08 ohms to produce 150A. Want to comfortably blow a 150A fuse? You need well below 0.08 ohms total resistance in the entire circuit, including the fault. If you have a high resistance fault (and “high” could be 70 mOhms!), you will never blow the fuse and you can produce literally kilowatts of heat. Crispy! (Your house likely has circuits bigger than 150A, but it’s at a voltage that can much more easily trip the breaker, and in civilized countries there will be some sort of RCD as well.)

Oh, and that 2AWG cable is expensive, hard to bend, and requires actual skill to terminate well. And modern systems often target powers well above 1800W.

At least in a car, the high current portion of the 12V system is made and tested in a factory. And it still makes amazing showers of sparks if anyone messes up a jumper cable. There’s a reason that car makers want to move away from it.

So I stand by my claim: 12V sucks.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: