goatman
10 MW
youre going to use your balance charger to desulfate the 5s2p pack you made of the 10 self discharging cells
Self-Discharge UPDATE On 30Q 141 Cells
- Thread starter eMark
- Start date
eMark
100 kW
Balanced the five 5S2P p-groups within 1v of each other, then straight charged at 0.33C (1 amp) to 3.96v with further top balancing to achieve 3.96 resting volts of the five p-groups (∓ 0.005v). Storing this experimental 5S2P pack in cool, dry place until end of month. Then will record each of the five p-group voltages before removing all bus bars and recording voltages of the ten individual cells.goatman said:
Hopefully, will PM you on 3/2 with recorded voltages and we can brainstorm any further ideas (if still interested). Appreciate any suggestions ... two inquisitive minds are better than one.
Have been asked by a third party not to post photos with further unwarranted info (e.g. 1/31/chart) of said cells. Bottomline is that the 1/31 chart isn't really necessary as it further expresses the 1/10 and 1/15 charts. Best if we PM going forward. Will PM you with the 2/28 p-group voltages and individual cell voltages on 3/2.
Happy Trails,
Mark
PS: I have two fire extinguishers :thumb:
goatman
10 MW
im not a big fan of not showing data for everyone else to see
ive had a couple PMs where people ask me a question so ill do a test and post the result without anyone else understanding why i did something but i put the data out for all to see
it just rubs me wrong to hide data from everyone especially when the data isnt available anywhere else
im not a battery guy but i had questions and noone was doing the tests or atleast sharing the data if they were
thats why i bought a tester, to do the tests, share the data
heres an interview of Justin, a 2 minute listen about why he bought endless sphere
https://youtu.be/IxB2j-egWcQ?t=791
ive had a couple PMs where people ask me a question so ill do a test and post the result without anyone else understanding why i did something but i put the data out for all to see
it just rubs me wrong to hide data from everyone especially when the data isnt available anywhere else
im not a battery guy but i had questions and noone was doing the tests or atleast sharing the data if they were
thats why i bought a tester, to do the tests, share the data
heres an interview of Justin, a 2 minute listen about why he bought endless sphere
https://youtu.be/IxB2j-egWcQ?t=791
eMark
100 kW
Does resistance discharging by some BMSs continue even during storage if p-group voltage variance is more than 30mV from highest to lowest p-group voltage? Have asked this more than once, but seems no one really knows. There are also those that believe the charger has to be left plugged in after green light comes on for a BMS to continue balancing p-group voltages when more than 30mV variance among p-groups. Again, no one seems to know as BMSs are not all the same. Then add "high" self-discharge to the equation and even Einstein might not have an answer :wink: Thus one reason why i decided to bottom balance (and top balance if needed) and not to use a BMS.
Wasn't there already enough data presented(1/5, 1/10, 1/15 charts) echoing other self-discharge reports over the years without belaboring it any further with this particular high energy cell. Is Li-ion "desulfating" even partially fixable ??goatman said:
I'm willing to continue with minimizing "high" self-discharge experimenting if a partial *fix* is even possible. I'm re-posting the following again as one would think the following data isn't characteristic of that one cell with the "highest" self-discharge out of nine cells with higher discharge than normal. Does the following strike you as not what one would have expected (3, 394mAh capacity at FULL and low IR at only 26mΩ from 98% to FULL) ...
Is this what you would have expected from a high energy cell with "high" self-discharge (3.75v on 1/2 to 3.44v on 1/15). Some may be thinking that this cell has an internal short that's dangerous with a fire just a matter of time. Is it not just as likely that this cell could just keep discharging until it's dead at 0.00V?
On average 95% of charge/discharge pack cycles (41V to 35-34V) were every 1.4 days over 170 cycles in 2020, with bottom balancing before straight charging at 0.5C with top p-group variance averaging 10-15 mV with pack voltage of 41V. Only discharged the 10S3P pack at Controller cut-off (32V) three times.BlueSeas said:
So, this one high energy "high" self-discharge cell may have sufficed for another 100 cycles considering it's current low IR and available capacity even after 170 charge/discharge cycles (every 1.4 days on avg) according to BlueSeas
Will replace the nine cells with higher than normal self-discharge. Will store them in fire proof container to see if it any eventually burst into fire or just drain to 0.000V ... assuming there isn't a "fix" for high self-discharge.madin88 said:
BlueSeas
100 W
There is a near zero chance of needing the fire proof container. Left to their own demise, no more charging, no more physical trauma, they will just die. We have "stored" to death many LFP cells, and while they show a nominal 3.2V, they have near zero stored energy. In the one case where I tried, slow cycling did rejuvenate them to 1/2 capacity, but usable only at low discharge rates, basically making them paperweights.
madin88 is correct though. There would be some risk in continuing to use the cells. However, if monitored as closely as you do, a few more cycles with the cell remaining within the normal envelope probably not a huge risk. It's certain to say there are lots of abused battery packs operating with even worse cells, and few fires. But the best decision is replacement.
madin88 is correct though. There would be some risk in continuing to use the cells. However, if monitored as closely as you do, a few more cycles with the cell remaining within the normal envelope probably not a huge risk. It's certain to say there are lots of abused battery packs operating with even worse cells, and few fires. But the best decision is replacement.
999zip999
100 TW
I got a Samsung pack 40t 20s6p still testing and a little worried as these thread has changed my mind. I brought a 8p 14s nese kit and was going to buy the Samsung 30q. But will not know. My be second choice is Samsung 25r. But be is Samsung any good. Maybe a should do a new thread ? Confused.
eMark
100 kW
Excellent post, THANKS! Rarely, if ever does someone post that a p-group with a weak/bad cell destroyedBlueSeas said:
their DIY battery pack ... thanks to BMS protection circuit.The self-discharge voltage of the five p-groups charged to 3.96V on 2/2 (see previous post) of my experimental 5S2P with ten cells having higher self-discharge than normal are today at: 3.95V, 3.95V, 3.95V, 3.94V, 3.93V. Will record p-group voltages on 2/7, 2/14 and 2/28. Will report back with p-group voltages and individual cell voltages on 3/2. Hoping by then some of you may have some thoughts on another test that might minimize or even reverse the self-discharge although the consensus seems to be less than encouraging ...
Should have posted the following earlier as intro to this thread for those new to Li-ion batteries; especially high energy cells ...
goatman
10 MW
im just wondering if desulfating is doing something to to the SEI layer
https://circuitdigest.com/article/what-is-solid-electrolyte-interface-sei-to-improve-lithium-ion-battery-performance
and if desulfating a pack before putting it into storage or every 75 cycles solves the self discharging
i dont have any self discharging cells to test with, nice that youve taken this on
https://circuitdigest.com/article/what-is-solid-electrolyte-interface-sei-to-improve-lithium-ion-battery-performance
and if desulfating a pack before putting it into storage or every 75 cycles solves the self discharging
i dont have any self discharging cells to test with, nice that youve taken this on
eMark
100 kW
That suggests that you believe "desulfating" is possible with Li-ion cells to minimize or eliminate "higher" self-discharge than normal self-discharge. Did you mean to say "sulfating" instead of "desulfating"?goatman said:
What leads you to believe "desulfating" is even possible with Li-ion cells suffering from a "higher" than normal self-discharge? It would be helpful if you could post any links that support the belief that "desulfating" is possible with Li-ion cells suffering from higher than what BattU considered at one time was the normal rate of Li-ion self-discharge.
By protection circuit are they referring to the cell's protection circuit or the BMS? Figuring five percent a month: 3.75V - 0.1875 = 3.56V. On 1/15 (2 weeks) only the bottom cell shown below was suffering from "high" self-discharge according to the above BattU quote, and on 1/31/21 only the bottom three cells had lost enough voltage to be considered suffering from "high" self-discharge according to the above BattU quote...
All 30 cells in my 10S3P were at 3.75V on 1/2/21. Here are the voltages of these cells on 1/15/21 ....
5 cells at 3.75V
15 cells at 3.74V
1 cell at 3.73V
2 cells at 3.72V
1 cell at 3.71V
1 cell at 3.70V
1 cell at 3.68V
1 cell at 3.66V
1 cell at 3.62V
1 cell at 3.57V
1 cell at 3.43V (mAh capacity of 3,394mAh from 2.5 resting volts to 4.2V (FULL) at 0.60 amps (0.2C) on 2/1/21.
Up until now the thinking seemed to be that a high energy Li-ion cell suffering from “high” self-discharge also suffered from a loss of diminishing ability of storage capacity. That theory was disproved with the one 30Q cell having the highest self-discharge from 3.75V to 3.43V over 14 days (Jan 2nd to January 15th) ...
eMark
100 kW
Another thread is needed. Entitled something like "Do I Really Need A High Energy Cell". Explain in intro that you have a 14S8P NESE kit and your ebiking application. Be as specific as possible will help determine if you really need a high energy cell as they generally have a shorter cycle life. With 8 p-groups "raw performance" may still be possible without using a high energy cell like 30Q or 25R? Defining "raw performance" as it relates to your own ebiking application at the very beginning of the thread is essential to determining whether or not you really need a high energy cell :thumb:999zip999 said:
It would be great if Justin (ES Host and Grin Tech) would offer his take should you decide to start a new thread. Do you remember that Jan-Erik-86 thread with his Vruzend 14S7P build in which the consensus by ES members after some lengthy discussion was to use the M36 cell ? As for me i'm still a newbie on the learning curve so will bow out should you start a new thread, but will follow it closely.
goatman
10 MW
to answer youre quote from above
"That suggests that you believe "desulfating" is possible with Li-ion cells to minimize or eliminate "higher" self-discharge than normal self-discharge. Did you mean to say "sulfating" instead of "desulfating"?
What leads you to believe "desulfating" is even possible with Li-ion cells suffering from a "higher" than normal self-discharge? It would be helpful if you could post any links that support the belief that "desulfating" is possible with Li-ion cells suffering from higher than what BattU considered at one time was the normal rate of Li-ion self-discharge."
i use desulfate, because i dont know what to call it, "desulfate" is used for lead batteries
what do you call what is happening to that 30Q self discharging cell once you "desulfate" it?
it doesnt work on all batteries
im beginning to think its rebuilding the SEI layer
SEI layer
SEI layer is the most important and less understood component in the electrolyte. Though the discovery of the SEI layer is accidental, but an effective SEI layer is important for the long life, good cycling ability, high performance, safety and stability of a battery. The formation of the SEI layer is one of the important considerations in the designing of batteries for better performance. Well adhered SEI on electrodes maintains good cycling ability by preventing further consumption of the electrolyte. The proper tuning of porosity and thickness of the SEI layer improves the lithium ions conductivity through it, results in improved battery operation.
During the irreversible formation of the SEI layer, a certain amount of electrolyte and lithium ions are permanently consumed. Thus the consumption of lithium ions during the formation of SEI results in a permanent loss of capacity. There will be SEI growth with the many repeated charges and discharge cycles, which causes the increment in battery impedance, temperature rise, and poor power density.
"That suggests that you believe "desulfating" is possible with Li-ion cells to minimize or eliminate "higher" self-discharge than normal self-discharge. Did you mean to say "sulfating" instead of "desulfating"?
What leads you to believe "desulfating" is even possible with Li-ion cells suffering from a "higher" than normal self-discharge? It would be helpful if you could post any links that support the belief that "desulfating" is possible with Li-ion cells suffering from higher than what BattU considered at one time was the normal rate of Li-ion self-discharge."
i use desulfate, because i dont know what to call it, "desulfate" is used for lead batteries
what do you call what is happening to that 30Q self discharging cell once you "desulfate" it?
it doesnt work on all batteries
im beginning to think its rebuilding the SEI layer
SEI layer
SEI layer is the most important and less understood component in the electrolyte. Though the discovery of the SEI layer is accidental, but an effective SEI layer is important for the long life, good cycling ability, high performance, safety and stability of a battery. The formation of the SEI layer is one of the important considerations in the designing of batteries for better performance. Well adhered SEI on electrodes maintains good cycling ability by preventing further consumption of the electrolyte. The proper tuning of porosity and thickness of the SEI layer improves the lithium ions conductivity through it, results in improved battery operation.
During the irreversible formation of the SEI layer, a certain amount of electrolyte and lithium ions are permanently consumed. Thus the consumption of lithium ions during the formation of SEI results in a permanent loss of capacity. There will be SEI growth with the many repeated charges and discharge cycles, which causes the increment in battery impedance, temperature rise, and poor power density.
eMark
100 kW
Neither you or I have ever indicated or provided any evidence (link) that a Li-ion cell once sulfated could be desulfated? On the contrary my position has been that desulfating a cell with higher than normal self-discharge is unlikely (but willing to run some tests with your suggestions).goatman said:
Is there a particular link that you've found supporting a belief that once a Li-ion cell is sulfated to an extent resulting in a higher degree of self-discharge than normal that a Li-ion cell can be desulfated (cured/fixed) or whatever word you'd prefer to use instead of desulfate?
So you seem to imply that my above test of that one 30Q cell with highest discharge is bogus? Will post it again in case you find it near impossible to believe that a 3000mAh rated cell with "high" self-discharge could be charged from 2.5V to 4.2V resulting in a capacity of 3,394mAh after 170 charge/discharge cycles during 2020 ...goatman said:
So, is it your belief that the above test putting 3,394mAh into a 3,000mAh capacity cell that suffers from "high" self-discharge after 170 charge/discharge cycles is not possible?
I don't mean to be picky, but do we know for certain that a cure (fix), if one is possible, is best accomplished by what you refer to as "desulfating" with the cause/effect possibly the result of a lack of Quality Control during production and inspection by not weeding out inferior cells ...
goatman
10 MW
im not disputing any numbers youre putting out
lets do 1 question at a time
you desulfated the self discharging cell
is it still losing capacity everyday?
or holding steady?
lets do 1 question at a time
you desulfated the self discharging cell
is it still losing capacity everyday?
or holding steady?
goatman said:
If those cells which have a problem with self discharge could be healed by discharging to 2.5V, they all would repair themselve because those cells (or groups) would, so to say, go down to that voltage each cycle as they reach LVC of BMS before the controller cut power.
From my experience it does not heal them but the time i spent measuring voltage drop over the days was not that much and the only thing i can tell for sure is that self discharge was dependent on actual voltage and the higher the SOC the bigger the issue.
Fact is that 30Q behaves different than most other cells without ever beeing abused or mechanically stressed, and the best is to not use them as long as we know the issue hasn't been fixed.
eMark
100 kW
Not so as the 9 cells with higher self-discharge than the other 21 cells of the 10S3P 30Q 141 battery were not healed as they still suffer from a higher rate of self-discharge over 14 days than the other 21 cells with no noticeable change (+/- 0.05V) over 14 days (2/2 thru 2/15).goatman said:
Desulfation has to do with lead acid batteries not LI-ion batteries. We need to stop using the term "desulfated" with respect to Li-ion cell chemistry and higher than normal self-discharge. Here's a definition of desulfation ...
It would be helpful if you stop using the term "desulfated" when referring to high energy Lithium-ion cells and Li-ion chemistry. Is there another term that is more appropriate?
With self-discharge a Li-ion cell looses capacity. It just happens that 30Q seems to be the one high energy cell most prone to higher than normal self-discharge, but that doesn't necessarily mean it can't be filled again closer to its original capacity than one would think. Thus, part of the motivation for starting this UPDATE thread with respect to the newer "141" 30Q cell and recent 2020 addition of the "6 KH1T" 30Q cell by a major US supplier.goatman said:
The thinking is that a Li-ion cell suffering from higher than normal self-discharge can't be charged as close to it's original mAh capacity as previously thought. My recent test seems to disprove that thinking and so will continue onward to see if it was a fluke or a uniqueness of a 30Q cell. My 30Q 10S3P battery cells with 170 charge/discharge cycles last year with 9 cells having noticeably higher self-discharge than the other 20 cells.
The 30Q cell (3000mAh rating) with the highest self-discharge of the thirty 10S3P battery cells (3.75v to 3.43V = 0.31V in 14 days) was still capable of receiving 3,394mAh when charged from 2.51V to 4.2V FULL. Am going to discharge one of the unused 30Q 141 cells i've had in storage (going on 14 months) having very little self-discharge. Will discharge it to 2.5V resting and charge at 0.2C to 4.2V FULL. Am going to do the same to one of the 3Q "6 KH1T" cells recently purchased. Will report back on their FULL capacity and mΩ IR. Suspect the mAh capacity of both cells (141 & 6 KH1T will also be around 3,400mAh). Will report back with results Monday afternoon.
I do appreciate your input suggestions and am willing to do other tests you may suggest that could reduce high self-discharge closer to what is considered normal Li-ion self-discharge by BattU.
So far the 30Q test with the highest self discharge doesn't support the belief that it has lost the ability to store capacity. It just leaks capacity ... kind of like a leaking water bottle that can still be refilled to capacity. Kind of like the 30Q cell with the highest self-discharge that can be refilled close to if not exceeding manufacturers mAh rated capacity.
____________ edit typo of 0.2 amps corrected to 0.2C
goatman
10 MW
you just said this
The 30Q cell (3000mAh rating) with the highest self-discharge of the thirty 10S3P battery cells (3.75v to 3.43V = 0.31V in 14 days) was still capable of receiving 3,394mAh when charged from 2.51V to 4.2V FULL. Am going to discharge one of the unused 30Q 141 cells i've had in storage (going on 14 months) having very little self-discharge. Will discharge it to 2.5V resting and charge at 0.2 amps to 4.2V FULL. Am going to do the same to one of the 3Q "6 KH1T" cells recently purchased. Will report back on their FULL capacity and mΩ IR. Suspect the mAh capacity of both cells (141 & 6 KH1T will also be around 3,400mAh). Will report back with results Monday afternoon.
its still self discharging?
The 30Q cell (3000mAh rating) with the highest self-discharge of the thirty 10S3P battery cells (3.75v to 3.43V = 0.31V in 14 days) was still capable of receiving 3,394mAh when charged from 2.51V to 4.2V FULL. Am going to discharge one of the unused 30Q 141 cells i've had in storage (going on 14 months) having very little self-discharge. Will discharge it to 2.5V resting and charge at 0.2 amps to 4.2V FULL. Am going to do the same to one of the 3Q "6 KH1T" cells recently purchased. Will report back on their FULL capacity and mΩ IR. Suspect the mAh capacity of both cells (141 & 6 KH1T will also be around 3,400mAh). Will report back with results Monday afternoon.
its still self discharging?
eMark
100 kW
That's been a given from the GetGo and reason for starting this thread. What we were discussing (questioning) is the 30Q "141" cell with the highest self-discharge having an apparent ability to store more capacity than originally thought from 2.5V (resting) to FULL (4.2V). Normal procedure is 2.5V cut-off and then immediately charging knowing that there is also bounce back voltage (say 3.2V and higher) when immediately charging again.goatman said:
Instead my method was intentionally to keep reducing cell voltage below 2.5V until the bounce back voltage after 2 hours was no higher than 2.51V. In order to do this I had to nearly drain the cell in order to only achieve a bounce back no higher than 2.51V (after 2 hour rest). Then begin charging at 0.60 amps (0.2C) until 100% FULL (4.2V) that resulted in 3,394mAh capacity as previously posted. FWIW, it's OK to nearly drain cell as long as it's soon charged again for the purpose of this test.
Thus, proving that a 30Q "141" cell with "high" self-discharge can still store more capacity than we originally thought was possible.
________________
Here's the charged capacity of one of the six unused 30Q "141" cells in storage (going on 14 months) charged at 0.60 amps (0.2C) until FULL (4.2V). Then 1.25 amp discharge until nearly drained in order that bounce back voltage after 2 hour rest was only 2.50V before charging. Then (and only then) charged at 0.60 amps (0.2C) resulting in capacity of 3,479mAh at FULL (4.2V). IR of 58mΩ during final charge from 95% to FULL. Surprisingly this "141" unused cell had a higher IR of 58mΩ than the "141" high self-discharge cell with 170 cycles in 2020 having an IR of only 26mΩ during its final charging from 93% to FULL (4.2V).
Here's the charged capacity of another of the six unused 30Q "141" cells in storage (going on 14 months) charged at 0.60 amps (0.2C). Then immediately charged again after 1.25A discharge cut-off of 2.50 volts ... resulting in capacity of 3,161mAh at FULL (4.2V). IR of 52mΩ from 95% to FULL. Surprisingly lower IR (only 26mΩ) with the "141" cell with 170 cycles having the highest self-discharge during final charging from 95% to FULL (4.2V).
Here's the charged capacity of one of the ten recently received "6 KH1T" 30Q cells charged at 0.60 amps (0.2C). Then discharge of 1.25A to 2.50V cut-off and further discharged until bounce back voltage is only 2.50V after 2 hr rest. Then immediately charged at 0.60 amps (0.2C) resulting in 3,420mAh capacity at FULL (4.2V).
Here's the charged capacity of another of the ten recently received "6 KH1T" 30Q cells charged at 1.5 amps (0.5C). Then discharge of 1.25A to 2.50V cut-off and immediately charged at 1.5 amps(0.5C) resulting in 2,720mAh capacity at FULL (4.2V).
__________________
What we're all wondering is: "What's the deal with the "6 KH1T" code on the 30Q pink tube wrap?" Is it a "cull" ? The supplier would only say that the WARNING label was printed by Samsung. Further phone calls and emails were unanswered as to why i didn't receive 30Q "141" cells as shown on their web page OR Do you no longer have "141" cells in inventory? OR What is the minimum order required to ensure delivery of 30Q "141" cells instead of "6 KH1T" cells?
Maybe, one or more of you could call IMR to find out if they are as tight-lipped with you when asking them if they are now only able to ship the 30Q cell with the "6 KH1T" pink tube label instead of the "141" pink tube label shown on their 30Q web page.
Attachments
goatman
10 MW
eMark said:
when you built this 5s2p pack, did you discharge all the cells to 2.5v and charge to 4.2v before building it?
im pulling these quotes below out of your/this post
https://endless-sphere.com/forums/viewtopic.php?f=14&t=110052#p1615369
"This is the pairing of the ten cells in the 5S2P experimental pack ...
1S2P... 2S2P.. 3S2P.. 4S2P.. 5S2P
3.65v, 3.62v, 3.53v, 3.53v, 3.56v
3.44v, 3.73v, 3.70v, 3,70v, 3.70v"
did you discharge that cell? or did it self discharge to 3.44v?
eMark
100 kW
Just to clarify these were the ten cells in the 10S3P 30Q 141 with self-discharge after two weeks (1/2-1/15). The other twenty cells remained at 3.75V to 3.74V ... the same voltage on 1/2 as on 1/15.eMark said:
Then i removed the 3.44V cell from the pack and continued to experiment with this 3.44V cell suffering from "high" self-discharge with further charging and discharging (at different rates) as suggested by you to see if a "fix" were possible, but not the case so far. Then after these tests it was returned to 3.44 resting volts and again assembled with the other nine cells suffering from higher than normal self-discharge in this 5S2P experimental pack. The 3.73V cell in the above diagram isn't typical of high self-discharge, but i needed ten cells to make up this 5S2P 3Q 141 experimental pack. The other twenty cells (10S3P 30Q 141) were still at 3.75V to 3.74V after two weeks (1/2 to 1/15).goatman said:
The past few days i've been experimenting with these ten cells (out of thirty) with self-discharge assembled instead as a 2S5P pack (see photo) instead of the previous 5S2P pack. Close to concluding that self-discharge cannot be reversed, healed or fixed with the experiments run so far. That isn't to say it's not possible to lessen high self discharge, but doubt high self-discharge can be eliminated or even lessened at this experimental stage.
Drained the newly assembled 2S5P pack below 2.50V to 1.00V (more than once) with the bounce back voltage after 24 hrs at 2.58V. Then bottom balanced the 5 p-groups within 10mV of each other. Then balance charged this 2S5P pack at 0.60 amps (0.2C) for twelve hours with the pack voltage only 3.74V after all that time. All of my previous tests were with bottom balancing the p-groups as close as possible followed by straight charging with top balancing not necessary as p-groups were within 10-15mV.
The obvious reason for 12 hrs with pack voltage only 3.74V was that there was at least one cell in this 2S5P pack suffering from voltage well below 3.74V. The result (12 hrs balance charging the 2S5P pack from 2.58V to 3.74V) was that balance charging (resistance discharge) was constantly discharging the higher voltage cells to bring them in line with the lowest cell(s). Because it took 12 hours (2.58V to 3.73V) gives us a pretty good idea that the resistance discharge balancing voltage was less than 0.60 amps, but not by much ... probably close to 0.40 amps.
What has become apparent is that even though the p-groups can be bottom (or top) balanced within 10mV of each other by no way means that the individual cells are within 10mV of each other ... quite the contrary. The reason is that the interconnecting of the series and parallel bus bars is very useful in balancing the pack p-group voltages within 10-15mV (my goal), but gives a false impression when the variance between the ten cells in the above 5S2P pack diagram were as much as 0.29 volts (3.73V to 3.44V) disparity with the other twenty cells in the 10S3P 30Q 141 pack still at 3.75V to 3.74V after two weeks.
The culprit in the 2S5P pack with the same self-discharging cells as in the 5S2P pack was not the 3.44V cell (as one might think), but rather it was the 4S2P 3.53V cell. Apparently the 3.44V cell with the highest self-discharge could withstand being drained to as low as 1.00V (more than once) than could one of the 3.53V cells in the above 5S2P diagram. What gives us a false picture is the p-groups having a variance of 10-15mV, when in fact the cells in the five p-groups have a greater variance (0.29mV). This disparity is due to the interconnected series and parallel bus bars which helps balance all the cells in any pack, but is no indication of the individual cell voltages. If the p-group variance was 30mV the individual cell variance might be 60mV or higher. Apparently the BMS only shuts down the battery when there is a cell failure in one of the p-groups.s
Thus supporting what Amberwolf and others have previously posted that a BMS (BPS) for all practical purpose is for protection and not for balancing the individual cell voltages within 25-30mV of each other ... at best balancing the p-groups within 25-30mV, even if some of the cells in the p-groups suffer from "high" self-discharge with a cell variances greater than 25-30mV.
goatman
10 MW
lets do some math, you stated from above
0.6amps x 12hrs=7200mah
from another of your posts
https://endless-sphere.com/forums/viewtopic.php?f=14&t=110052#p1615369
in order to fully charge 5 cells to 3.99v you would need 12000mah (12ah)
divide 12000 by 0.60amps equals 20 hours of charging required
i went looking at my 30Q test to see how much capacity there is at 3.6v to 3.8v
i didnt have that data for the 30Q but did for the 40T (864mah)
https://endless-sphere.com/forums/viewtopic.php?f=14&t=106550#p1565303
at your 0.6amp charge it wouldve took 7hrs to charge 5-40T from 3.6v to 3.8v
im just saying you probably need to charge those 30Q for another 8 hrs to reach 4.0v
youre in the big capacity bubble right now
0.6amps x 12hrs=7200mah
from another of your posts
https://endless-sphere.com/forums/viewtopic.php?f=14&t=110052#p1615369
in order to fully charge 5 cells to 3.99v you would need 12000mah (12ah)
divide 12000 by 0.60amps equals 20 hours of charging required
i went looking at my 30Q test to see how much capacity there is at 3.6v to 3.8v
i didnt have that data for the 30Q but did for the 40T (864mah)
https://endless-sphere.com/forums/viewtopic.php?f=14&t=106550#p1565303
at your 0.6amp charge it wouldve took 7hrs to charge 5-40T from 3.6v to 3.8v
im just saying you probably need to charge those 30Q for another 8 hrs to reach 4.0v
youre in the big capacity bubble right now
eMark
100 kW
What is apparent is that when balance charging (resistance discharging) is incorporated with this 2S5P pack of cells (suffering with varying degrees of self-discharge) that the capacity shown when balance charging (resistance discharge) is Bogus. For example my balance charger was set at 120 minutes and then automatic "shut-off". After every 2 hrs and then shut-off the capacity was between 1,191 and 1200. So six times (every 2 hrs) over 12 hours the total capacity was 7,174mA.goatman said:
So, although the charger was attempting to add 7,174mA of capacity the balance charging (resistance discharging) was removing around 5,000mA of capacity from the higher voltage cells in attempt to get them in line with the lowest cells' voltage of 3.53V as shown in the following diagram (2S3P=3.53V). Balance chargers are not the most accurate when it comes to capacity readouts even when just used for straight charging. I've got two balance chargers with both showing different capacities when straight charging before shut-off at 2 hrs (my shut-off preference time before starting again until desired pack voltage).
This is the diagram arrangement of the resting cell voltages of the 2S5P pack before bottom balancing and then balance charging from 2.58V to 3.74V. The 2S3P cell was the problem. The 3.44V cell was still at 3.72V after a 2 hr rest; while the 2S3P cell had dropped to 3.66V after a two hour rest. Did i "fix" the infamous 3.44 cell (doubtful) ? Will let you know the on-going self-discharge of the 2S3P (3.66V) cell as well as the other nine cells in the 2S5P pack as shown below ...
1S1P 2S1P
3.56 3.44
1S2P 2S2P
3.70 3.73
1S3P 2S3P
3.53 3.53 (damaged cell when pack was drained for only 2.58V bounce back after 24 hrs.
1S4P 2S4P
3.65 3.62
1S5P 2S5P
3.70 3.70
Will post the resting voltages of these ten 2S5P cell voltages at the end of the month. Then get a better idea if this latest experimental testing of these 30Q 141 self-discharge cells may have lessened (temporarily?) that 3.44V cell from its previous "high" self-discharge; while that one 3.53V cell is junk; while the 3.44V cell survived. Will be interesting how much more its on-going self-discharge is come 2/28.
goatman
10 MW
so the 3.44 cell is the only one thats been discharged all the way to 2.5v and recharged to 4.2v while the other 9 were discharged to 2.5v and then partially charged
the cell that is still self discharging, be interesting to try and do the full discharge/charge just on that cell
i noticed that it would take 2 capacity tests to straighten out the curve
the cell that is still self discharging, be interesting to try and do the full discharge/charge just on that cell
i noticed that it would take 2 capacity tests to straighten out the curve
eMark
100 kW
Yes, it was the only cell of the ten discharged to 2.5v and even below 2.5v. using my manual discharging device (1.25A) and then "recharged to 4.2v" FULL using the MiBOXER. The ten cell 2S5P pack (including that 3.44v cell) was also discharged well below 2.5v, with bounce back of 2.58v only after 24 hours, but charging was stopped with p-group voltage of 3.74v ... after twelve hours of balance charging at 0.2C (0.60 amps).goatman said:
As previously posted one of the 3.44v cells' test was discharging it well below 2.5v until it's bounce back resting voltage was only 2.51v after one hour rest. The result was that there wasn't the loss of capacity as much as one would think. My tests seemed to show that a 30Q cell with leaky capacity does not limit it from restoring capacity ... only to leak some its available capacity again.
The following is from my February 11, 2:39pm post that you didn't get a chance to read ...
I do remember one post last year where the builder observed a p-group (or cells) with higher self-discharge than normal. He said that the self-discharge went away, but no clue as to why. Not sure if he was referring to the self-discharge of a p-group or an individual cell. That is the only post i've seen at ES about higher 30Q self-discharge than normal disappearing (normalizing itself).
Have you come across any threads at ES or on other forums where a 30Q pack with higher than normal self-discharge somehow corrected itself to whatever one would consider was a more acceptable rate of self-discharge?
eMark
100 kW
You're not the only one confused :?999zip999 said:
I was probably the most favorable ES advocate of the newer Samsung 30Q "141" cell that was first shipped to suppliers in 2019. It was my optimistic belief that Samsung is just as adamant about zero tolerance for defects as is Japan (Panasonic/Sanyo) as evident by Elon Musk's use of Panasonic cells in Tesla. Now it's lookin' like i was over-confident in believing that Samsung had solved the "high" self-discharge being reported with the "136" runs over the past years that have plagued the 30Q "136" production runs varying from production run to production run.
My reason for building a VRUZEND 10S3P 30Q "141" pack was for experimental testing having high hopes that the newer 30Q 141 cell was the best high energy cell for the money and i wasn't alone. Here is a thread in which spinningmagnets expressed a similar viewpoint ... https://endless-sphere.com/forums/viewtopic.php?f=14&t=108430 ... with the assumption that the newer 30Q 141 cell was the best buy for the money ... spinningmagnets, goatman, myself and others have generally been favorable toward 30Q as the energy cell being the best buy for the money.
If my mindset wasn't so inclined to problem solving i wouldn't have purchased a VRUZEND V2.1 kit solely with the idea of using the newer 30Q "141" cells to prove that these cells were a better buy for the money than any of the other high energy cells. However, now 999zip999 and others wonder if 25R is actually a better buy for the money than 30Q 141 cells. Has anyone ever posted on ES that they've experienced their 25R DIY build suffering from "higher" than normal self-discharge. What's your opinion on 25R cells ever suffering from "high" self-discharge to the extent of 30Q 136 cells.
For my conservative etriking LG MJ1 would have been the choice for my VRUZEND DIY build. Now i still have 20 "good" 30Q 141 cells showing no signs of self-discharge for more than a month. AND 10 30Q "6 KH1T" cells of questionable quality that at this point i still plan to integrate with my reassembled 10S3P 30Q pack and report back in July after 100 more charge/discharge cycles.
Will report the current resting voltages of the 20 "good" 30Q 141 cells on Monday. I tested them yesterday with all 20 cell voltages the same as on 1/5/21.
goatman
10 MW
eMark said:
no to 30Q fixing itself
if 1 cell is bad in a p-group it will take out the entire p-group, even the good cells
when you built the 2s5p i thought it wouldve been a good test to see if doing the 0.6amp discharge/charge 4.2v/2.5v would be able to save a pack without disassembling it and replacing cells but you stopped early on the charge without completing the process because of length of time it was taking
by having the 3.44v cell paralleled in the group it will be brought down by the self discharging cells
why did you switch from 5s2p to 2s5p
your hobby charger would be able to do the test in less than half the time at 5s2p
