DIY conversion of 15s to 16s

[piomark]

Hi All,

I have 2x Pylontech US5000 15s batteries + 1x custom built 15s batt (I had to order 16 cells anyway so I have 1 spare).
As I was starting to plan to expand to second custom battery, I was wondering if I could "easily upgrade" my 15s pylontechs into 16s battery.
I can see 2 options here:
Disassemble existing batteries, change BMS'es and create 2 x 16s batteries out of it => really costly and a lot of work,
or
Add 1 single cell 200Ah+ (probably 280Ah to be on the safe side?) in front of the 2 paralleled pylontechs.

2nd option is really simple on paper, although ofc I would be loosing some of the BMS functions of pylontechs (which I don't use anyway as I have additional custom battery). The BMS would still deliver the overcurrent/overtemperature protection (at least of the existing 15s pack). The additional cell would never be balanced to the rest of the battery as there would be no balancer (might be a new project for winter for me ), and would need some additional monitoring (like temp, voltage).

Did anyone tried that? Anything that I am clearly overlooking from security/maintainability perspective?

Thanks,
Piotr

 

[robbob2112]

Running a LFP cell without a BMS connection to prevent overvolt/overcharge, low voltage cutoff is a bad idea.

Also the extending of the wire outside the unit would add resistance between the one cell and the rest and probably result in it being charged more/used more or charged less/used less..... also bad and would require the BMS to keep from having a melt down

So I would recommend you do option A - add cell 16 to the existing string and get a 16s BMS if the existing BMS can't be programed to do 16s and has an extra lead not used.

Also before I did either I would probably get a YR1035+ meter ($40usd off alliexpress) and measure the impedance of the cells in the unit and the extra cell and make sure they are closely matched. To ensure they are reading correctly make sure batteries and that single cell are at the same SOC -- fully charged. If the impeadance is way off you will have the same problems as if you extended it outside the case with a wire.

 

[profesor79]

I tell you what I did with my 3 x 5kW pylontech us5000... sold them and got cells on alibaba.
I was thinking about that idea, but after a research and reading posts where mixing 15s with 16s was discouraged.
There a few issues, like getting a new bms (and loosing the good stuff from pylontech), the extra cabling needed for a custom cell added to the set.
so keep it simple - get rid of pylontech or love it

 

[piomark]

Well it's not that I hate them, but I'm not a fan of 15s battery after 1y+ of use. Also when buying cells they usually come in packs of 4...

I was trying to sell my 2 pylons no luck so far that's why I was investigating other options.

Regarding running the pack without BMS, actually I can create a simple Bms based on esp32 for this additional cell and use pylon remote port for disconnect (in case of over/under voltage or temperature issue) this module could also have a simple passive balancer.

I'm more concerned about resistance difference, probably I should try it out (I have the meter).

If I manage to sell the pylons I would buy more eve cels and create my own packs.

 

[frankrudis65]

A 15s would be slightly out of date these days,I would think. I'm running 17s's happily....Running a cell without bms is asking for trouble imho...Is it worth the risk ?Do you want to put anyone under risk?Or do you prefer a healthy system for years to come.....Best regards Frank