I’m not sure we can make that assumption.
If you don’t give him a straight black and white binary answer that is correct 100.000% of the time and that everyone on the planet agrees with;
he will not take your input.
There’s almost no aspect of li-ion cell design or use that can be whittled down that far. It will almost always be “it depends”. 
1 Like
I just removed 2 10+ year old Panasonic NCR18650B individual cells from mostly flashlight service, after I found them getting alarmingly warm during recharge, at a 0.25c charge rate.
The charging source was capable of nearly a 1c charge rate, but those cells in the last few months would not accept more than about 0.3.
Perhaps a ‘warning signs’ of degrading and potentially more dangerous lithium cells would be beneficial.
I know many people just plug and play their charger and wait for a red light to turn green, and one day the ‘battery won’t take a charge!, waaaaaaah.’
As if suddenly something went wrong instead of continuous gradual decline.
But if they had been watching an ammeter and ah/wh counter, or had Infra red vision, they could have seen warning signs.
1 Like
Some round cell chargers measure cell temp fairly well. If the threshold is set well, 45°C perhaps, it can be very handy for finding those about-to-fail cells.
While everyone is looking at this thread, what is the consensus for temperature probe placement from the bms? In my limited experience i have seen the most positive p group “feel” the warmest during discharge and if i get a second probe depending on the bms model i will try to put it in the denseist part if the pack.
Where do yall put your probes? 
currently im probing the most positive as close contact with a middle of the group cell as close the the positive output. Secondary probe at the middle p group in the most buried cell i can.
1 Like
The hottest cells will typically be in the center of the pack but if the main power lead gets hot or there is a lot of current “crowding” into the main lead then a lot of heat can be generated at the bottom or top p-group.
Best to measure in the center of the pack if only a single sensor (use thermal paste and Kapton tape or thermal epoxy) with the sensor firmly touching two cells. Mount halfway up the cells.
4 Likes
My completely arbitrary places are charge port (between the leads on the xt-30 disconnect) and the usually either the positive discharge tab or the tab that’s right in the center of the pack.
I figure if it ever gets hot enough to stop it from charging, it’s probably on fire, since all my bms are run discharge bypassed. It’s mostly just nice fun data to look at when opening the xioaxiang app
2 Likes
Considering a pack failure when charging would likely start with one bad cell it seems having just one temp sensor even a couple cells away from that heating cell would probably not catch it unless the temp sensor is set to a very low temp, no?
It seems best placement would be on the thermally conductive path between cells, on the nickel.
1 Like
I’ll take this as hummies ego friendly way of admitting we were all right about cutoffs and battery preschool is dumb
You’re a bitter guy.
There seems no right or wrong. It’s all just degrees of risk. I recommend you call battery hookup as well as many other cycle reusing businesses and tell them they’re wrong. They told me 1v is a minimum. I’ve often seen 2v as a minimum
If you can’t find places and self-described experts who go with 2v I can list like ten source. I’m not saying it’s right. I’m also not going to tell them they’re wrong.
In light of the growth of dendrites in all cell use… and the ineffectiveness of a single temp sensor… I think there’s dangers all around.
Battery Hookup and other battery scrappers have a very strong financial incentive for their customers to believe that. The more people who believe this, the fewer returns they will see.
Conversely, Molicel/Samsung/LG etc. all have a strong financial incentive for their users to stay within the spec sheet ratings (2.5v absolute minimum, on every manufacturer’s spec sheet I’ve seen) for the safety and longevity of their cells. The more users who stay within these thoroughly tested and known-safe ratings, the more users will be satisfied with the reliability of their cells, and therefore will be repeat customers.
Looking at those two sets of incentives, only one of them has my safety in mind. I think I’ll stick to 2.5v
P.S. - bonus incentive:
If a manufacturer tested and confirmed one of their cells was safe to discharge normally below 2.5v, then they would have an incentive to advertise that, because it would set the apart from the competition and give their cells an extra feature that may be appealing in some applications. The fact that they dont, means the manufacturer has determined that the safety risk of going below 2.5v is greater than whatever benefits they might see from rating their cell below 2.5v
6 Likes
Battery hookup also has an incentive not to blow people up. Maybe liability. Telling me 1v. Maybe they won’t write it down. I’m going to ask them again about why they feel it’s acceptable
The scale of liability between battery hookup and Molicel (for example) is wildly different.
Forgetting that Molicel is a MUCH bigger fish with significantly more money, and therefore making it a more lucrative target for litigation; just look at the two different markets they both target:
Battery hookup (and other cell scrappers) are selling used packs of unknown quality to a DIY audience with the explicit understanding that their customers are going to be scrapping cells for their DIY projects. Even if battery hookup didnt have some sort of liability release on every sale (which I’m sure they do) it would still be very hard for anyone to bring a case against BH for their own DIY project going wrong. Battery scrapping is an “at your own risk” activity.
On the other hand, Molicel’s customers are massive battery manufacturers building packs for consumer products. These are companies that have the resources to go after Molicel for inaccurate/unsafe tech specs. These companies also have to be extremely sure that their packs are of the utmost safety, because they are going into the most abusive environment known to man: the average household.Therefore, Molicel has lots of incentive to be absolutely sure their tech specs are accurate and safe.
At the end of the day, your endless debate about what is “safe” is pointless. For anyone reading this thread in the future, here’s a quick shorthand:
“Safe” = following the manufacturer’s datasheet exactly. Stay within the rated limits (min and max). Give yourself a decent safety margin within those limits if you want to get max life and reliability out of your cells. Please also keep in mind that there are many, many more factors that go into “safe” liion battery usage than what is on the datasheet. Do lots of research, and always be skeptical of folks who say that the general wisdom about safety guidelines can be ignored.
“Unsafe” = anything outside of the manufacturer’s datasheet. How unsafe? No idea. There isnt good data, so any decisions you make in this regard are going to be entirely up to your own risk tolerance, and unfortunately we dont have great data to judge the risk levels by. Before making any of these decisions, please look up some videos of liion battery fires, and think about how many people could be harmed by a toxic and carcinogenic structure fire originating at your device.
1 Like
You’re wasting your time, he just wants to feel special by disagreeing with folks that make him seem insecure about his abilities and knowledge.
I’m not disagreeing. I wrote from the very beginning asking what low voltage was acceptable given the many different claims given including by those who sold me the cells.
It doesn’t take special knowledge or abilities to do a search. Only through that searching so I hear different opinions. Luckily I didn’t just assume my first answer (from battery hookup) was right.
1 Like
Instances like these, would benefit from a VESC connected BMS, as well as the newer stuff @ENNOID is working on. The shield item looks promising for that kind of thing.
For VESC one wheels, the devs who maintain the Float Package rolled in BMS related triggers for tilt back, rather than using the motor throttling warnings. But for eskates, having the VESC BMS data to throttle acceleration back, is useful. It can be used to signal a battery problem during discharge, either from under voltage conditions or overtemp.
This can be done without the above linked item though, since it’ll work over CAN.
3 Likes
100% this
Cell manufacturers have a huge incentive to produce chemistry’s that can tolerate lower voltages. Also to design and develop manufacturing methods that work at or below 2.5v. None of the LiIon cells can be reliably discharged below this level and also have the energy density available currently or they would advertise this like crazy and then everyone would copy them. The 2.5v minimum i would also bracket with the manufacturer has a huge incentive to place this number as low as possible -
if you look at the discharge curve from the spec sheets there is little to no stored power at this voltage (how i interpret it) and that means a few things to me: obviously there isn’t any useful reason to discharge cells all the way to this point but also it is using an electronic component outside of its spec rating. If it acted in predictable reliable ways after this the spec would reflect that. If it isn’t predictable and there is a significant risk of catastrophic consequences then it’s not worth it to me.
I have a box of dummy bullets, hypothetically, one with 100, one with 10000, one with 1000000. In each box there is one live round. Never will a gun loaded from any of those boxes be pointed at something that isn’t going to die. If the failure mode of cells was always predictable (made of chemistries and methods to prohibit thermal runaway) the find out phase of my fucking around would include some unknowns - no big deal. When my use doesn’t care if a cell will go into thermal runaway it would also be fine (pointing a gun loaded down range with unknown ammunition quality). Just don’t point that unknown gun at your head - the risks prohibit there being a middle ground.
Would be cool for a board to drop down to like 20% power (brakes stay fully operational) when it notices a cell going out of balance or something.
1 Like
Samsung data sheet says can charge a cell if it goes to 1v!
It’s the 35e 18650
The Samsung 30t data sheet shows the same
I asked if it was possible. I got insults and told I was a dumb fuck and too dangerous to be here and that I was starting a toxic discussion. I was told I needed to trust the data sheets. Here it is.