a request for knowledge on the demise of my 30Q battery

[Minexplorer]

Hi ,my question is pretty pointless now really, but i still irk for answers and would appreciate some understanding of what went wrong.Especially as it totally ruined my e biking experience. Much thanks in advance,if anyone can shed any light.

It started with my BBSHD build.my samsung 30Q (sdi 136 cells) 5P 52 volt battery started playing up 3 years ago with just 12 months of use on the BBSHD. At the BMS i found 6 of the 14 cell groups were at lower voltages. manually charging them to close resolved the issue of the battery hitting the LVC of 42 v when it still had say 47 volts remaining.It began holding right to the LVC even on bursts of 30 amps.Though they were still not fully 4,2 v, it didnt last long tho and got worse,i just could not get them to fully reach the same voltage of the other 8 groups. i tried adjusting the charger to only charge to the lowest group voltage. Again it worked for a while then the battery would sag and hit the LVC under sudden high load ,like hills and hitting the throttle, with as high as 50 volts remaining on the display.

People advised me it was trash and i had some bad cells. Inquiries for repair got quotes of £400 .i read the 30Q was a crap cell prone to discharge issues. It got to the point i couldnt trust going out on it. The bike would go at any time under a higher load demand once the voltage was down to 50/52 volts. The battery got chucked under the bed a year or more ago charged up. Today those 8 good cell groups are reading 3.84 volts,the bad groups 3.68 ,3.65 ,3.52 ,3.47, 3.25 ,3.00.

So ive been stripping it down and found it seems to have ultimately 14 sets of 5 cells.Which the 14 wires at the BMS obviously refer to.Im not going to attempt to understand the complicated way these are joined into 7 sets of 10 then to other sets,i was expecting to see the cells joined somehow in 14s. However what has surprised me is i haven't found a handful of random bad cells. All 5 cells in a 'bad' group have the same voltage as each other,after the nickle strip joining them is removed ???

Please someone tell me it wasn't the BMS and ive just trashed a £450 battery for a lifetime supply of vape batteries.I did try many times manually charging the lower cell groups and they never fully came up to the good ones ,getting slowly worse over time.
I havent yet gotten to trying to recharge the low cells. The better cell groups so far are reaching 4.08v-4.12v. I admit i did hammer that battery with the BBSHD tons of full 30 amp whizzing around and constantly running it down to the LVC and having to walk the last cpl uphill miles home. A 5P 30q battery is supposed to be able to handle 37.5 amps continuous or 75 amps for bursts i thought . Was there anything else i could have done im wondering.thanks

 

[pwd]

I don't blame the 30Q cell; perhaps it not the best but it is still an OK cell from my experience. My guess would be too much discharging below the LVC and too much charging to 100% as opposed to 90% for better cycle life.

Cell ratings are always embellished compared to real world test. I usually cut the rating in half. IE: If a cell is rated for 15A of current; I'd recommend only pushing it to 7.5A for less voltage sag and cycle life.

To get a more accurate idea of the cell's health I'd recommend use a four wire internal resistance measurement tool like the YR1035 or at the very least; one of those cell chargers that does a rough measurement of cell capacity.

 

[Minexplorer]

I don't blame the 30Q cell; perhaps it not the best but it is still an OK cell from my experience. My guess would be too much discharging below the LVC and too much charging to 100% as opposed to 90% for better cycle life.

thanks
hmm so i hammered it then and damaged it. Was advertised as the ideal choice for BBSHD. Yes i was discharging it repeatedly and re charging to 100% maybe 90 times in the year it lasted. i read here i think the 30Q had been found to suffer 1 in 10 cells failing after as few as 50 recharge cycles.

Ive charged up a few of the lowest cells now and they are 4.10v slightly higher than the ones that didnt discharge much.and higher than when they were in the complete battery.

If only a case of some self discharging for whatever reason why the catastrophic sagging under load.i can only imagine the lowest cells were hitting the 42v LVC being why the battery was shutting off. Thats quite a sag from perhaps 50 volts capacity showing on the display. The lowest cell groups were at that time ,even topping up manually ,refusing to charge above 4.05v.

 

[The Battery Doctor]

Unbalance like this most common when pulling more amps then the cells can handle. Yes the 30Q can handle 15A, but when you read the datashet that will leave the cells at 60% at 250 cycles. What BMS did you use? Where did you buy the cells, new or used?

 

[eMark]

It started with my BBSHD build.my samsung 30Q (sdi 136 cells) 5P 52 volt battery started playing up 3 years ago with just 12 months of use on the BBSHD.

My guess is that your 30Q "136" 14s5p battery was close to 5yrs old and although unused the battery may have been storage in less than cool dry storage condition. Even new 30Q "136" cells are more prone to high self-discharge than other name brand high energy cells such as Sony VTC6 or LG HG2.

Basically (IMO) it's a manufacturing QC problem where some of the underperforming "136" cells in 6 of your p-groups were actually closer to B-grade cells (that were passed as A-grade cells). IMO, the same was true (hit and miss) with some of the 5 different manufacturing runs of the 30 "141" cells used in my 30Q 10s3p DIY experimental build using new cells.

Have had to replace 9 of the 30 "141" cells after 125 cycles (1yr) with 9 30Q6-T (newer version) and then 4 more "141" cells after another 125 cycles (after 2yrs) that were again suffering from higher than normal self-discharge. They were replaced with previous "141" regen cells. However, they now need to again be removed and replaced plus another 3 "141" cells with all 7 being in just one of the 10s strings.

Please someone tell me it wasn't the BMS and ive just trashed a £450 battery for a lifetime supply of vape batteries. At the BMS i found 6 of the 14 cell groups were at lower voltages.

The battery got chucked under the bed a year or more ago charged up. Today those 8 [5p] good cell groups are reading 3.84 volts, the bad [5p] groups 3.68 ,3.65 ,3.52 ,3.47, 3.25 ,3.00. All 5 cells in a 'bad' group have the same voltage as each other after the nickle strip joining them is removed ???

But that's only temporary until separated into individual cells. After a weeks rest you can tell which are still good cells that may have actually slightly increased in voltage ... meaning they're possibly still useable. So, 8 of the "good" 5p groups with 'minimal?' self-discharge from say 4.10V to 3.84V over 12 months may still be useable (IMO?) for 30Q high energy cells using questionable A-grade older "136" cells. Most (all) new 30Q cells for sale are now 30Q6-T or 30Q6-150-T. There still may be some "141" cells for sale, but doubtful any "136" cells as more likely questionable counterfeit cells.

You need to separate the six "bad" 5p groups of your 14s5p battery into sixty individual cells. Then charge balance all 30 cells to the same voltage (say 3.850V). Then let rest for one week and check the individual voltages again to see if some are still good enuf (at least 3.844V). Then let all rest for another week to see how many are still good enuf to use again (say at least 3.838V). If possible use a DMM that reads to the nearest 0.001V (1mV) ... https://www.amazon.com/GUOSHUCHE-Multimeter-Capacitance-Temperature-Transistor/dp/B09B3Q2WBJ/ref=sr_1_10?crid=35UD9CVW7IQ58&keywords=19999+Counts+True-RMS+Auto+Range+Digital+Multimeter+Volt&qid=1670881272&sprefix=19999+counts+true-rms+auto+range+digital+multimeter+volt%2Caps%2C429&sr=8-10&ufe=app_do%3Aamzn1.fos.006c50ae-5d4c-4777-9bc0-4513d670b6bc

The better cell groups so far are reaching 4.08v-4.12v. I admit i did hammer that battery with the BBSHD tons of full 30 amp whizzing around and constantly running it down to the LVC and having to walk the last cpl uphill miles home. A 5P 30q battery is supposed to be able to handle 37.5 amps continuous or 75 amps for bursts i thought . Was there anything else i could have done im wondering. thanks

IMO, 18650 cell manufacturers overrate "MCD" (and peak burst) just as do RC Lipo manufacturers. That's why a DIY builder's preference when "hammering" a 14s battery would do everything possible to make room for an 84 or 98 cell DIY battery ... preferrably a 14s7p. Hopefully, you can salvaged some of the cells in the six "bad" 5p groups.

Have to replace 3 more "141" cells (after 3yrs and 330 cycles). All of the 30Q6-T cells are still OK. Have ordered ten new 30Q6-150-T cells (newer Samsung version made in China). Present plan is to replace one of the three 10s "141" series runs of ten cells with ten of the new 30Q6-150-T cells as seven of the "141" cells in that 10s string need to be replaced.

Then next winter replace any additional "141" cells (then suffering from unacceptable self-discharge) with 3 good "141" cells in the 10s "141" string having 7 "bad" cells now suffering from various degrees of unacceptable self-discharge. My 10s3p was purposely an experimental 30Q battery from the Get-Go in 2020. The Vruzend V2.1 Kit is easily disassembled and reassembled (compared to spot-welding) for testing individual 30Q "141" cells suffering from unacceptable self-discharge and replacing with good 30Q cells.

The following post in my 1/21 30Q thread "Self-Discharge UPDATE On 30Q 141 Cells" ... is not unusual about DIY builds using 30Q cells ... https://endless-sphere.com/forums/viewtopic.php?t=110052#p1612408. Could be wrong, but doubt you'll find one DIY builder that is bullish about the lasting performance (at least 250 performnce cycles) of his Samsung 30Q DIY build.

LunaCycle has used Samsung 30Q cells in the past for some of their battery packs. It's very possible that those Luna Cycle 30Q packs have been good performers for up to 250 cycles with no complaints.

 

[neptronix]

30Q is a cell with a kinda short lifespan to begin with:
https://eu.nkon.nl/sk/k/30q.pdf

Another thing is that if you've gone from nice temperatures to freezing temps, you can expect the cell's internal resistance to go up 4x when comparing 70f weather to 30f weather. This means it will poop out quite quick during the winter.

Whether your BMS is bad or your cells weren't matched well from the factory is up for debate and hard to figure at this point since you don't have baseline measurements from when you first bought the battery.

I can tell you right now that that 3v cell is your limiting factor. If a cell goes below LVC, the pack will shut off, and that one is responsible.. replacing that bank would help but there's a few other cells in line to become the weakest link..


My suggestion ( for the future )

If you want to avoid mismatched packs and bad BMSes in the future, only buy a battery from a top brand that's devoted to making good batteries and can prove it. ( or build it yourself )

The top shelf stuff is more expensive but that extra 20 min of work spent matching cells together before assembling is totally worth the extra coin.
One example is em3ev, they send a battery calibration sheet with batteries and that's the reason i bought most of mine from them.

>95% of battery pack sellers don't take those extra steps into account and sell stuff that won't last.
Most ebike part consumers see a lower price and don't know the difference..

..even worse is a low quality BMS that drains exclusively from 1 to 4 cells.. which is sadly the most common thing in cheap packs.

 

[mxlemming]

As others have said but in fewer more concise words...

30Q is garbage. They die by becoming leaky by no fault of the owner. This happens to cells from a wide span of ages and batches.

It wrecked my first battery, I spent too much time fixing it only to find other cell groups started discharging a few weeks later. Scrap it/give it away for people to use as vape cells.

Mine lasted about 6 months before it first became a problem.

 

[docw009]

My ebiking is slow riding on urban paths, probably pulling under 200 watts on average,so life is pretty easy for my batteries, I've got two Luna batteries that use 30Q's that I purchased in 2015 and 2016. Using a load tester with a constant 3A DC load, one pack currently delivers 2.6Ah/cell and the other delivers 2.7Ah/cell. I do not think that's bad at all.

By the way, a rare chance for some science here. I have cells stored outside in a BBQ at ambient, It's currently 3F. I fetched a few indoors to check the AC resistance using a YR-1035. I had checked them earlier when it was warm.

Brand____________ 20C ___________ -16C
30Q (new)_______12.7-13 mΩ __ 18-19 mΩ
Eve 25P (new)___ 11-12 mΩ ___ 18-19 mΩ
LG MH1 (used)___ 20-32 mΩ ___ 38-40 mΩ
Molicel 18650A___11.7 mΩ _____ 15.7 mΩ
Lishen LR2170-SA 15.3 mΩ _____ 19.5 mΩ

On a percentage basis, the 30Q and the Eve's have the highest increase when cold, so their high current numbers probably fall off the most.

 

[batteryGOLD]

Brand____________ 20C ___________ -16C
30Q (new)_______12.7-13 mΩ __ 18-19 mΩ
Eve 25P (new)___ 11-12 mΩ ___ 18-19 mΩ
LG MH1 (used)___ 20-32 mΩ ___ 38-40 mΩ
Molicel 18650A___11.7 mΩ _____ 15.7 mΩ
Lishen LR2170-SA 15.3 mΩ _____ 19.5 mΩ

Thanks for all details.

There is an issue with 30Q cells, those with time tend to acquire voltage leak issue.
Means if you have example one 10P block at battery with just one 30Q leak, than all other 9cells keep dischargin.
Have seen voltage leak at 30Q many times. soo seems a issue at those..

Panasonic 18650B green(I think those used at initial teslas battery, no wrap) Those, have high capacity and about 30mOhm , but with time those increment up to 45mohm and keep same capacity. this leads to high voltage drop and soon bms cut. But if you test those at 1C discharge rate it puts out almost full capacity. the only issue comes to high current demand, so if U use less throttle acceleration it should put out most of its capacity/range.. :thumb:

 

[docw009]

Good info about some 30Q's becoming leakers.

 

[Tony01]

The battery got chucked under the bed a year or more ago charged up. Today those 8 good cell groups are reading 3.84 volts,the bad groups 3.68 ,3.65 ,3.52 ,3.47, 3.25 ,3.00.

You are an extremely lucky guy, I suggest you buy a lottery ticket while your luck hasn’t yet run out. Put that POS outside away from your house ASAP! Trust me you do NOT want to breathe whatever it is leaking and if it catches fire under the bed... well...

 

[eMark]

My ebiking is slow riding on urban paths, probably pulling under 200 watts on average,so life is pretty easy for my batteries, I've got two Luna batteries that use 30Q's that I pu rchased in 2015 and 2016. Using a load tester with a constant 3A DC load, one pack currently delivers 2.6Ah/cell and the other delivers 2.7Ah/cell. I do not think that's bad at all.

Have always thought that Luna 30Q batteries use all A grade cells (same manufacturing run?). Whereas 30Qs purchased by DIYers from suppliers (my thirty 30Q 141 cells) may be a mixture of A grade and borderline A/B grade cells (from different manufacturing runs).

Nine of my thirty 30Q 141 cells had to be replaced after 125 cycles with nine newer 30Q 6-KH1T cells (same manufacturing run). Then another three 141 cells had to be replaced the next year and now another three 141 cells (of the original 30) need to be replaced again due to higher than normal self-discharge. All nine of the 30Q 6-KH1T cells are still AOK (no leakage).

Currently in the process of pre-testing ten newer 30Q 6-150-T cells (made in China). Self-discharge after 10 days ranges from 10mV to 16mV. Charged from 3.45V resting volts (as received) to 3.85V. Current plan for 2023 is for my 10s3p 30Q experimental pack to have: One series string of ten 141 cells; Second series string of nine 6-KH1T cells and one 141 cell; and Third series string of ten 6-150-T cells (different manufactering runs).

My 10s3p is an experimental Vruzend V2.1 battery pack to evaluate what is generally believed to be a higher than normal self-discharge tendency of 30Q cells compared to other high energy dense 18650 cells (e.g. VTC6, HG2). Then replace leaking cells as needed for optimum cycle life.

Brand____________ 20C ___________ -16C
30Q (new)_______12.7-13 mΩ __ 18-19 mΩ
Eve 25P (new)___ 11-12 mΩ ___ 18-19 mΩ
LG MH1 (used)___ 20-32 mΩ ___ 38-40 mΩ
Molicel 18650A___11.7 mΩ _____ 15.7 mΩ
Lishen LR2170-SA 15.3 mΩ _____ 19.5 mΩ

What's the takeaway? So, who in their right mind stores a battery pack at -16°C (3.2°F)? Or are you suggesting that sub-freezing storage of 1865/2170 cells (or battery pack) for a few months over winter is OK? Even with temps as low as 3.2°F assuming casual mild-mannered urban use once again come warm weather ?

"Storing lithium-ion batteries at below-freezing temperatures can crack some parts of the battery and separate them from surrounding materials, reducing their electric storage capacity"... https://www6.slac.stanford.edu/news/2021-08-24-how-extreme-cold-can-crack-lithium-ion-battery-materials-degrading-performance#:~:text=Now%2C%20researchers%20at%20the%20Department,reducing%20their%20electric%20storage%20capacity

"To improve electrical performance in the extreme cold, researchers reporting in ACS Central Science have replaced the traditional graphite anode in a lithium-ion battery with a bumpy carbon-based material, which maintains its rechargeable storage capacity down to -31°F" ... https://www.acs.org/pressroom/presspacs/2022/acs-presspac-june-8-2022/lithium-ion-batteries-that-last-longer-in-extreme-cold.html

 

[leffex]

Hi ,my question is pretty pointless now really, but i still irk for answers and would appreciate some understanding of what went wrong.Especially as it totally ruined my e biking experience. Much thanks in advance,if anyone can shed any light.

...

Learning money.

Don't worry about the 30Q cells. I would consider them worthless or you can use them for light use like vape as you said or flashlights and small powerbanks. If you can't use em for that then they have no value I would say.

This came to light a year or two ago little after some guys in this forum started testing so I said to myself never to use those cells as I haven't. However this can happen other brands or noname brands as well. Usually if the cell is bad it can happen after 1 cycle. Worlthess crap. Or after a 2-3 hundred cycles which I consider very bad.

What went wrong may be that internal components in the cell from the beginning was flawed from manufacturing making the cell to self-discharge very rapidly. If you have one cell that is ligthly or very bad it will kill a cell group very fast making the BMS system to activate and turn off your battery more and more often untill your full battery is almost useless for normal use.

 

[eMark]

I would consider them worthless or you can use them for light use like vape as you said or flashlights and small powerbanks. If you can't use em for that then they have no value I would say.

This came to light a year or two ago little after some guys in this forum started testing so I said to myself never to use those cells as I haven't.

That seemed to be a consensus for "136, 138 and 141" 30Q cells. Most reputable suppliers now stock the 30Q 6T and newer 30Q 6 150 T (Made in China). I have been using both the 6T and 6 150 T without any significant self discharge. IMR still shows the "141" cell - Samsung 30Q 18650 3000mAh 15A Battery - but shipped me the10 of the 6T in '21 and 10 of the 6 150 T in '22.

The manufacturing runs of "136", "138" and even "141" 30Q cells i hopefully no more. Even if you can still find some i wouldn't buy them even if only $2; even moreso when building a spot-welded battery pack. Maybe worth a gamble for a N.E.S.E. battery pack as easy to replace a bad 136, 138 or 141 cell.

 

[tomjasz]

Most reputable supplier

Who are they?

a consensus

who is the consensus?

 

[richj8990]

How many miles did they actually get out of the battery? One year could be 500 miles or 10,000 miles. 10K miles for a 450 pound sterling battery cost is not exactly optimal but not horrible either.

 

[eMark]

How many miles did they actually get out of the battery? One year could be 500 miles or 10,000 miles. 10K miles for a 450 pound sterling battery cost is not exactly optimal but not horrible either.every

Depends on the number and degree of unacceptable self discharge in a DIY 30Q build. Storing a 30Q battery for several days / weeks between a charge/discharge cycle degrades any parallel group sooner with an unacceptable 30Q cell. 30Q DIY builders should pretest their 30Q cells so as to hopefully remove any cells with evidence of unacceptable self discharge.

High energy dense cells(e.g. LG, Sony/Murata, Samsung) are generally more prone to self discharge. That said the ES "consensus" since the advent of the 30Q "136" cell is that 30Q is more problematic than other name brand high energy performance cells.This 30Q problem (e.g. "136" cell) occurs when a 30Q cell with unacceptable self-discharge starts to gradually degrade the other good 30Q cells in its parallel group until it becames quite quite noticeable even as soon as 50 cycles. When recharging a 30Q DIY pack (e.g. with "136" cells) every day or every 2-3 days you should be able to get at least 150 cycles before a questionable parallel group ever becomes so problematic as to require replacing an entire parallel group.

The general ES DIY "consensus" since the advent of the 30Q "136" cell is that 30Q is more problematic than other high energy dense name brand cells.The 30Q self discharge problem becomes more evident when storing a 30Q pack for a couple weeks, months, voer winter, as a bad cell more adversely affects (drains) the other good 30Q cells in its parallel group. The amount of adverse cell drain in any particular parallel group will depend on its variance of unacceptable self discharge.

The recent 30Q manufacturing runs of the newer 30Q6T and 30Q6 150 T seems less problematic (iimproved QC manufacturing runs). Previously with my experimental 10s3p 30Q pack i used a DMM to measure the mV (0.000v) of each 30Q "141" parallel group after resting discharge. Then 1.5amp cbottom charged any problematic parallel group to within +2-3 mV of the other good parallel groups. Then charging the 10s3p experimental pack to 41.0 volts (90%). The two strings of 30Q6T and 30Q6 150 T cells do not show any significant self discharge as was apparent with my "141" cells from 2020 thru 2022. The remaining string of 10 "141" cells (350 cycles) are not showing any indication of unacceptable self discharge in 2023 (so far). Won't know for sure untill disassembling pack this winter to do a thorough test of self discharge (ten "141" cells, ten "6T" cells, ten "6 150 T" cells.

 

[docw009]

So, who in their right mind stores a battery pack at -16°C (3.2°F)? Or are you suggesting that sub-freezing storage of 1865/2170 cells (or battery pack) for a few months over winter is OK? Even with temps as low as 3.2°F assuming casual mild-mannered urban use once again come warm weather ?

Published storage specs for cells often state -20C to +40C. We're in Northern Illinois, so typical outdoor temperatures are down to -4F a few nights in January. My garage temperature will freeze beer and pop in winter, whatever that is but I think it doesn't go much under 20F. I have kept ebike packs in the garage at 30% SOC for two winters now. Two packs did have cell groups drain in 2021, but these were made with cheap chinese cells. I scrapped all of the cells and no longer have any packs with unknown cells.

Yep, I keep several hundred bare cells outside in the BBQ. They're not new M50LT's, 50E's, or GA's. When I get those, I build them right away into battery packs. These are cells I bought on clearance from Battery Clearance House or Battery Hookup, That include 30Q's from 2019, new Eve 25P's. old LG MH1's. I don't have much use for them because if I build a battery now, it's with 5Ah 21700 or 3,5AH 18650. WHen my Luna Dolphin packs finally wear out, maybe I will recell them with the Eve cells. I've sampled these BBQ cells on occasion. The IR is going up, but I don't see much change in capacity. In any case, I think they're safer outside than on my desk or in the garage.




.



.

 

[eMark]

The [30Q 14s] battery got chucked under the bed a year or more ago charged up. Today those 8 good cell groups are reading 3.84 volts, the bad groups 3.68v ,3.65v ,3.52v ,3.47v, 3.25v, 3.00v.

Hopefully that degree of self discharge would never happen with the newer 30Q6T and 30Q 6 150 T cells.
Store DIY battery in cool/dry safe place in a protective container (rather than under the bed). Pretest 30Q cells.

Hookup all 30Q cells in parallel charging to around 4.100 resting volts. Then disassemble and store cells for 2 weeks before testing each individual 30Q cells' voltage. Don't use cells with noticeable self discharge (e.g. 5 to 7mv) below the norm of the good 30Q cells for your DIY build. If this is too much of a bother than disregard and take your chances that all your 30Q cells are A grade and good to go.

This is one way of pretesting 30Q cells ... you may have another method.