diy solar

diy solar

Failed TinyBMS and recovery of dead cells

theta

New Member
Joined
Dec 9, 2021
Messages
6
Hi all,
I woke this morning to no power after a week of clouds. Turns out Energus tinybms decided to stop working and failed on, and my 8s lifepo4 pack is now sitting at 9 volts. I think even the balance system of the BMS also seems to have crashed, I have 3 cells sitting in between 0.2 and 1.7; the rest are right above 2v, hard to say though given the fast voltage drop to SoC. Further, It seems it might not have done too well with over-charge either; a few of the paralleled cells are sightly bloated cell together. It even Insulted me with the low voltage alarm once I cycled power to the BMS.

A couple of questions:
Safety: I've disconnected the pack, but is there any reason for immediate concern? These cells are under my bed, and I don't have a place to dispose or store them outside my van for the next few days.
Salvage: Is it possible to salvage the cells sitting at 2.0V? Since I have a 2P8S pack I was hoping to maybe turn it into a 1P8S or a 12v so I can limp along till I get new batteries. I also assume that the whole pack is now considered trash; I shouldn't just replace the cells under 2.0v with new cells.
Disposal: Do recycling centers take LiFePo4 batteries, I see there's a mail service to dispose of them, but it seems costly.
BMS: I though the TinyBMS would be a good product considering it has a number of EV customers, is European made and was pricey. I really only went with it as it supported external current monitoring and relay cutoffs, which was important as I pull over 100Amps when I run the inverter. Any suggestions for a replacement?
Other Batteries: I'm tempted to get Battleborn after this catastrophe for the peace of mind, but for price and the space I have I would only get 1/3 of the capacity in comparison to raw cells.
 
Last edited:
1st, I will not address the BMS, that's another thing altogether.
The cells are NOT necessarily dead or damaged. Harmed maybe. Unless you severely overcharged (above 4.5Vpc) or totally flattened them and froze them. 2p8s = 24V so does 1P8S. 8 Cells in series = 24V, 4 cells in series = 12V.

First, you need to examine each cell, do a voltage test, IR test is possible (a YR-1035+ Battery Cell tester is great for this).

Next charge each cell to at least 3.600, allow the amps taken to drop to <1.5A, then let rest for 1 hour and retest Voltage & IR, they will settle to close to around 3.500 +/- a bit within an hour (normal behaviour). Cells that behave weirdly should be watched, in many instances sometimes a full cycle (2.50 to 3.60 and 3.60 to 2.50 discharge cycles) may be needed to recover the cells. DO NOT Overdrive the cells with High Amps, best if below 0.25 C-Rate charge.

Once you have identified the good "recovered" cells then it is time to make some decisions.
Unless you are using Genuine Matched, Batched & Binned cells it is never good to put cells in parallel, a weakling can cause havoc within the paired set. Many argue this point, it is a Dead Horse Topic and I won't go through that again. Simple Straight Series Packs are more amenable in general, especially with Bulk & Commodity cells or those of dubious provenance.

Bloat is relative to many things. Appreciate that the blue prismatic cells commonly used for ESS do & can expand & contract by 2-4mm depending on size/capacity & temperature factors. Cylindrical cells are "bound" by the very nature of their construction unlike the big block cells. Even cells with Plastic/ABS housings can expand & contract but it is less noticeable. Not many notice 1mm or 2 mm change generally.

When working with cells, charging/discharging & testing it is highly recommended you do so at "Room Temp" or roughly 25C +/- a few degrees but not much more and make sure the cells are at fully "room temp".

TinyBMS has a good rep overall and I would be sus[icious that the BMS itself failed, a possible component or sensor failed OR its programming was incorrect or got changed somehow. This would require a far more in-depth examination to determine what happened and how. Swapping it out without knowing what went wrong may result in a similar issue coming back at you again. I would say, that if that happened you would be most Irate I am sure.

For better information & troubleshooting please provide the following.
Brand & Model of cells used, link to product is good to add.
Full detail on the Battery Pack Assembly, photo's, schematics if possible.
The TinyBMS model, and what components are used, along with settings etc as used.
General Solar Info: Make/Model of Solar Charge Controller, Inverter/Charger and solar panel info as well.

The more complete the information is the better responses you will get. NONE of us can guess or read minds and divine what you have in any way.

Hope it Helps, Good Luck.
Steve
 
The cells are NOT necessarily dead or damaged. Harmed maybe. Unless you severely overcharged (above 4.5Vpc) or totally flattened them and froze them.
Nothing frozen in the van, but it was -1C outside last night and the batteries are not in a warmest spot. And the heater didn't come on due to lack of power. Its questionable.

2p8s = 24V so does 1P8S. 8 Cells in series = 24V, 4 cells in series = 12V.
Yep, My system is 24v 2PS8. My 12v comment is about converting the pack to 1P4S just to get 12V lights working.
Unless you are using Genuine Matched, Batched & Binned cells
They were cells recommended by Will Prowse from BLS and stated to be matched
TinyBMS has a good rep overall and I would be sus[icious that the BMS itself failed, a possible component or sensor failed OR its programming was incorrect or got changed somehow.
All I can say is the heartbeat LED turns off after a period of time and there is no communication through the serial port and it left the relays powered. Cycling the power revives it and everything works again for a bit, the alarm came on, relays were left open, and then it stops again. It's worked correctly once before after I accidentally left the circuit breaker for the solar off for a week, so I don't think is programming. And I intently put in conservative thresholds when I programmed it.
For better information & troubleshooting please provide the following.
Brand & Model of cells used, link to product is good to add.
120ah cells
Full detail on the Battery Pack Assembly, photo's, schematics if possible.
The TinyBMS model, and what components are used, along with settings etc as used.
Will get back to you tonight.
General Solar Info: Make/Model of Solar Charge Controller, Inverter/Charger and solar panel info as well.
Victron 100|30 MPPT in LiFePo4 mode with 2x sunpower X21-335 , Victron 500|24 inverter, Sterling Pro 12 to 24v 60A with LiFePo4 settings
 
Last edited:
The BLS cells are Voltage & IR Matched at storage voltage only.

These are NOT properly Matched & Batched which is a lengt6hy & costly test process of charging discharging cells at Hi & Lo C-Rates and testing at threshold points through the cycle, typically they will run. The IR is recorded at the voltage points for consistency (matched) through their cycles, cells that deviate outside of spec goto Bulk Sale. Matched & Batched cells will show an identical (very close) IR whether at 2.50, 3.0, 3.2 or 3.4 V, unmatched will float all over the place and most especially outside of the "Working Voltage" of 3.000-3.400.

This is an example of a Match/Batch report for 280AH cells. (VERY FEW VENDORS DO THIS) it is more commonly done by the Production Factory for Large Batches or an external test service which is more costly.
1639078206700.png

This is extracted from a Resource I prepared for everyone that I gave to Luyan Tech. This will help you somewhat.
Source: Luyuan Tech Basic Lifepo4 Assembly Guide
1639078520306.png
 
Until you figure out why your downstream load (inverter) continued to pull current even when the battery bank was low, I don't think buying off-the-shelf batteries is going to solve your problem.
 
Until you figure out why your downstream load (inverter) continued to pull current even when the battery bank was low, I don't think buying off-the-shelf batteries is going to solve your problem.
Wasn't just the inverter. I have a lot of DC loads, the larger of which have voltage protection. But, lights, fans, and phone chargers aren't that smart. And at least off-the-shelf should have a working BMS and are easier to get a hold of without a shipping address if these cells are damaged.
 
Update:
The cells are pretty swollen, I would say the worst are 5-7 mm on each side or 1.5cm overall. I had a hell of a day trying to extract them from their box. Ultimately the trick was to slip a 3ft zip tie behind the cell to pull on. Also turns out they are 150ah cells

Most of the cells recovered without charging from 2.0v to 2.6v overnight, and have charged up to 3.0v in no time. However, the one set of cells that was sitting a 0.2 hasn't recovered at all. I've been charging it a 3V/3A for over 1 hour and it is now at 1.4V. Considering all the cells that recovered by themselves were +/- 0.1v from one another I think this one cell set was drained by the BMS. The tinyBMS balance system is software controlled, you can set its parameters, given when the BMS failed it left the relay in its powered state; I think the balance system failed in a similar manner, leaving this cell connected to the bleed resistor for months before the pack drained. This has been my experience with microcontrollers, the outputs of the chip are latched and require a signal to change state, thus when the software crashed it left that pin in its current state.

How bad is it to replace cells in a pack? I am getting the feeling that the 0.2v cells won't make a recovery. And I definitely won't be able to fit the pack back in its box with the bloated cells.

Also the tinyBMS is hardware v2.1. I've been trying to get it connected, but after I power it operates for about 5-10s before crashing, so I don't have enough time to get it connected to the software and pull the config. It is weird because for those 5 seconds it does work, the serial message looks all good, it recognizes the pack was screwed and cut the relays. And then it just stops, I can't even reset it with a manual hex command. I really want it to work as I will have to remake bus bars if I have to pull the relays out of the system.
 
The 0.2v cell has been charging for an hour and a half; it hit 1.5v and has dropped back down to 1.1 over the last 5 min while still under CC charging at 3A. Has anyone seen this behavior before, I am stopping charging for now
 
Last edited:
The 0.2v cell has been charging for an hour and a half; it hit 1.5v and has dropped back down to 1.1 over the last 5 min while still under CC charging at 3A. Has anyone seen this behavior before, I am stopping charging for now
In such an instance the really low cells will be Very Slow to get back up into the curve. Believe it not it's a good sign because if it shot up fast then you are shorted & it's pooched OR if it tool no charge then likely it partly delaminated internally (I won't go tech on this, too much and not important). IF they have bloated and are beyond 5mm @ room temp without compression they ARE suspect and caution IS WARRANTED.

There are things that "can" be done with staged charging / discharging with compression being applied but it is a Long Ardous process and risky if you do not know what you are doing exactly - not typically for DIY folk, too impatient, sorry but it is what it is. Even then there is no guarantee that things can be resolved. There is NO SIMPLE FIX. Whatever you do DO NOT HIT THESE CELLS WITH HIGH AMPS ! If it is a delamination issue they would Puff out like balloons and vent off (very very bad), IF they are Shorted the results would be even less pleasant BUT you'd have seen that rear it's head when applying a charge of any amperage quite quickly.
 
So those 2 cells have been charging for 48h and their voltage continues to drop. And the open circuit voltage is now < 0.1v. They are luke warm to the touch, enough that I think its pure heat from the 3 watts going into the cell. I think these might have shorted.

I was looking for new cells to replace these but apparently, CATL doesn't make the 150ah anymore? The listing I used now has changed its title to "not 150ah" and edited the capacity in the photo, and is now selling 120ah. I don't have the equipment to do a capacity test but my cells do match the listed size for the 120ah(H163mm/W173mm/W46mm). IAre their different grades with different capacities? I can't find any other 150ah cells on aliexpress.
 
Back
Top