So, I joined the BURN club.

[doing it different]

scratched case from the poor welding in the back. the case is a negative charge.

Not to dunk on you, but as a guy that builds stuff and welds etc. There probably was some splatter or something that didn't get cleaned or ground down... it's easy to miss,

As someone that has rebuilt rebuilt engines. Even if the machine shop has done everything. It's the assemblers job to make sure everything has been machined correctly, measurements etc. and one thing that often gets forgotten proper cleaning. making sure there is no machining chips caught in any oil passages etc.

Whenever custom things are built, it's the responsibility of the builder to get it right.

Chalk this up to learning curve, even though it's hard to swallow.

You can't break things, if your not doing things

 

[Brucey]

can you explain that? how would a not tightened terminal allow overcharging? Isn't the voltage what the voltage is? say worst case 58v. how would it be more? I could see maybe some micro arching, causing warmth. but dang... it was 100% for 20 minutes, and it was already on it's way down as the deye shows it detected 100% voltage, so it was throttling down the solar power, as can be seen on the solarman screenshots. yeah, the DIY is more painful. Our original intent, 2 years ago, was to have custom built racks that are easily cell-swappable in a large EV boat. That fell apart due to lack of technical expertise here, so we got railroaded here. Now, with the advent of cheap cells and cases, it seems obvious to buy a case, but, it wasn't 2 years ago. quite frustrating to try to be a leader in any field.

Speculation is that without comms the deye simply delivered max charging current which was feeding into the case short, looks like you had fr4 between the cells but not on the outside edges contacting the shelves.

 

[Mattb4]

Thanks for posting about your equipment failure. It is not always an easy thing to do this in a Public Forum for others to learn by since you also have to put up with potential criticism. I am glad to see your batteries, while damaged, are not burned to cinders. It furthers my trust in LiFePO4 as being somewhat fire safe.

To my old eyes it looks to be an issue with lugs and terminal connections that were not perfect and heat eventually created a cascade effect. With 48vDC the voltage is high enough that anything a little off reveals itself versus the same arrangement at lower DC voltages.

 

[SomethingAppropriate]

The fact that your inverter said full charge and all your cells are at 3.2 ish means they either let all their energy out. Or never got much.

 

[400bird]

If you look at pics 1, 7, and 8 in the first post, are those all of the same pack?

And the rub through shown in post #11 is also the same pack? Looks like positive and of that pack. Any signs of a rub through on the negative end?

Anyways pic 1 looks to me like the damage centers around the positive terminal on the left (negative) end of that pack. Pics 7/8 look like the damage comes from the positive terminal also.
The cell case has continuity to the positive terminals. So, if you didn't insulate the cells from the steel mounting frame, then when the useless blue cell liner rubbed through, you put current through the case of both those cells causing the heat and damage to the end cells.

Also, if you've never charged that BMS up to your high voltage limit, the SOC is useless because yo haven't given it a reference for what 100% actually is. Until it is calibrated, the SOC is just gibberish.

 

[XO4001]

If they aren't communicating how is the deyw supposed to know it's at 100%

Dunno how it works but I have no comms and the inverter (Deye) say batteries are 100% and the BMS (Seplos) say it's less and still charging batteries until set voltage is reached.

 

[Daddy Tanuki]

I think you can see the battery breakers, green ones, in one of the pics. Suntree i think, 125a (5x). Ok, didn't know Daly was so bad. but arguably.. maybe there was nothing for them to protect if the short was external? and yes LITHIUM mode implies that, but, it still has protections.

Daly is not any worse then any of the other offerings from China. a quick search of this forum proves that as there are more JK/JBD threads currently then there are Dalya threads. it boils down to what is being used the most at the moment in time. if more Daly's are being used, more complaints, if more JK then more JK complaints. etc. etc.

I have used daly form the start and other than the ones damaged by a lightening strike I never had any issues after the first one I got. (i damaged it out of stupidity.)

if set properly they are just as good as any other. you have a hole in the case of one cell, probably from expansion against the weld as you suggested. clean up your racks, rebuild and this time set the BMS properly as well as the charging source. the BMS should be a last resort cut off, not the go to option.