LG M50LT: Why do EUC Co's keep using this sub-optimal Cell?

[houseofjob]

Gonna leave this Mooch LG M50LT test right here.

 

This cell spot welded into 4P packs has been the common denominator in the majority of recent EUC wheel fires, and now this Mooch test shows the cell's max amp discharge significantly declines the hotter this cell gets (ie. our EUC's do not yet have a proper active cooling solution ala Tesla cars etc.)

 

Which poses the question (smart BMS LiTech / LillyTech / LollipopTech be damned): why does this cell continue to get used in such a high power, high temp application such as a modern-day high power EUC, particularly in power-hungry, heat-building shrink-wrapped 4P configuration pack? (read: well, I actually do know why: $$$$)

 

Obviously, this is not a bad cell outright, but perhaps it is not the right cell manufacturers should be using for our EUC purposes, especially coming from years of the tried-and-true, good old faithful 18650 flavor of cells like the once industry performance standard Panasonic NCR18650GA's.

 

https://www.e-cigarette-forum.com/threads/bench-test-results-lg-m50lt-14-4a-4932mah-21700.974083/?fbclid=IwAR2-DMopoPvoRLKdqpQVMt0drGPXYM1UOP6bqVY8SuSxGt1iRMKrUYxhWOo

 

Bench Test Results: LG M50LT - 14.4A 4932mAh 21700

Test Results 
This is a fantastic performing “energy” cell (high capacity, made for long run times at lower power levels) but be aware of the reduction in its current rating at higher ambient temperatures.

This cell outperforms both the Samsung 50S and BAK N21700CG-50 at 5A. At 10A the Samsung 50S beats this cell by a few percent but the M50LT still performs a touch better than N21700CG-50. This makes the LG M50LT one of the best performing 21700’s we can buy. I have not tested its cycle life yet though.

See my 21700 Ratings and E-Scores table (link below) to see how this cell performed against others.

The preliminary datasheet for this cell lists a 14.4A “Max. Discharge Current” rating but that is only for 10°C-25°C room temperatures! At above 25°C the current rating starts dropping and is only 7.2A by the time the ambient temperature reaches 55°C. These higher temperatures are common in battery packs where each cell heats up the others around it and the cells are in a case or wrapped in insulating materials and heat shrink plastic. Never exceed 70°C cell temperature!

So while you can run this cell at 14.4A if at room temp or cooler I recommend staying at 10A or lower to help improve performance (reduce voltage sag), extend cell life, and reduce the risk of the cell getting too warm.

The top contact of this cell is recessed more than most cells so these cells cannot be directly stacked in a vapingdevice. The top contact metal is also thinner than most cells and the “legs” coming down from it are narrow so I recommend not using this cell where there is a lot of pressure on the top contact.

The two cells I tested delivered 4843mAh and 4850mAh. This is fantastic consistency and is typical of cells from the big manufacturers. The cells I tested didn’t meet LG’s 4923mAh typical capacity rating but they were over the 4770mAh minimum rating.

I strongly recommend reading the datasheet for this cell before using it: https://cdn.shopify.com/s/files/1/0481/9678/0183/files/2020-21700_M50LT_datasheet.pdf

Edited May 20, 2022 by houseofjob

[goatman]

Mooch was lucky enough to get his hands on a couple of the new Molicel P45b for testing, its blowing the the 40T and P42a away. you could check out his testing results if you become a  patron

[mhpr262]

That cell is be perfectly fine for the kind of riding I do... leisurely cruising at 15mph. Manufacturers need to be more straightforward about such technical matters. Nothing wrong with saying "this is long range whel for cruising at medium speeds, it is not meant for extremme power applications". Everybody should be able make more informed buying choices, and how they use the wheel.

Edited May 20, 2022 by mhpr262

[alcatraz]

7A per cell at 55C is just going to create a runaway temperature situation.

The manufacturers are hoping that the rated power output is nowhere near continuous.

They're all peak ratings.

What I'm amazed about is that rated power goes up but the cell configuration stays the same. How long have we had 10-15A 21700 cells in a 4P config? Since the Nikola/MSP. 4 years?

At some point the cells need to follow the rated output. And I mean without an option to choose 10-15A cells.

[alcatraz]

In other configurations like 6P 8P it's not a problem to use 10-15A cells.

With the RS134 volt, were down to 3P even (if it's on the way).

[novazeus]

On 5/19/2022 at 8:59 PM, houseofjob said:

The top contact metal is also thinner

i've been away for 4 years, so is this the problem. the vibration breaking the thin cap and leaking?

[houseofjob]

58 minutes ago, novazeus said:

i've been away for 4 years, so is this the problem. the vibration breaking the thin cap and leaking?

not sure, I was just re-posting, but the overall theme seems to be this cell was never meant to be put under such amp load & vibrational stress as a PEV application, even though the cell is highly regarded by most reputable battery authorities, but more for it's more applicable usages, say, in a stationary hundreds-of-cell low amp powerwall, etc.

[alcatraz]

It's good you guys bring up vibrations and long term effects. Can you imagine showing a video of an euc being ridden HARD to an LG engineer?

"Hey LG, there's a problem with your cells!"

What about control board components? That can't be good there either. Does anyone remember that episode of Top Gear when they drive european cars on those African washboard roads? The BMW and Volvo just literally fall apart in kilometers.

Edited December 6, 2022 by alcatraz

[mhpr262]

When I do longer, relaxed tours on my V11 I drain the cells over a timespan of two and a half to three hours, so nowhere near its rated maximum discharge current. For me and my use case that cell is a perfect fit.

[0000]

Its a point worth making, cell  suitability can be highly dependent on rider behavior just like practicing good/safe charging habits is for ensuring long term battery pack health. This especially the case if the batteries selected by the design engineers didn't respect all use cases such as aggressive riding or racing in addition to lower/higher bound operating temperatures.

That said, as a general rule a fast wind motor (high speed) is going to stress the batteries more requiring higher amounts of current to equal the torque output of a slower wind motor (high torque) at the same speed.

Knowing this, its generally safer to put these high energy 21700 cells in a wheel with a HT motor vs a HS motor which will tend to stress the cells more given the same rider behavior.

High power cells like 40Ts and P42As are a much better and safer fit for low parallel group packs (4P and less) as a rule of thumb, but particularly for fast wind motors.

I don't know if anyone has reported combustion issues (like that's a small thing) with the M50LT vs the old M50Ts I think it was.

As for vibration damage, I haven't read anything about the "jellyroll" cased inside of the aluminum can ever getting damaged from vibration. Is that even a thing? I'm sure the battery manufacturers perform these kinds of tests to ensure a robust design, but I'm just assuming. Don't care to look up on my phone.

Generally speaking, its usually the cell interconnects or spot welds potentially failing from repeated mechanical actuation or vibration where potential shorts against the cell cans could most likely occur.

[novazeus]

if batteries get extremely hot during discharge, and they like naked batteries for heat dissipation, so they wouldn't double shrink wrap them,

could the heat make the shrink flexible enough to get enough movement to break the thin cap where the spot weld has weakened the metal.

just a guess.

[Cerbera]

12 hours ago, Vanturion said:

As for vibration damage, I haven't read anything about the "jellyroll" cased inside of the aluminum can ever getting damaged from vibration. Is that even a thing? I'm sure the battery manufacturers perform these kinds of tests to ensure a robust design, but I'm just assuming.

That sounds correct to me. Cells are so densely packed they are  individually very solid, and a group of them bound together in their plastic blocks and shrink-wrap are even more so. I would very surprised if they were moving at all. pretty much regardless of what the ride is doing...

[Pingouin]

Hello guys,

Correct me if I'm wrong, but except from the NYC Kingsong S20 fire, no EUC with the 21700 LGM50LT where involved in a fire.
It seems there is a confusion with the LGM50T used in the older V11, old RS, S18, Nikola 100V, MSP, first batch EX, first batch Monster PRO.

To my knowledge, it is only the older LGM50T that were involved in most EUC fires reported. Of course it doesn't meen they are fireproof (no Li-ion cells are) but they seem to be safer than the older LGM50T cells, the datasheet seems to say that the LGM50LT are safer than the Samsung 50E for example.

Regarding the datasheets, the LGM50T are rated at 7.3A max discharge at optimal temparature, while the LGM50LT are rated at 14.4A which is basically double.

https://www.dnkpower.com/wp-content/uploads/2019/02/LG-INR21700-M50-Datasheet.pdf

Maybe @Chriull @Marty Backe @mrelwood could confirm ?

Edited December 8, 2022 by Pingouin

[dycus]

On 5/19/2022 at 5:59 PM, houseofjob said:

 (read: well, I actually do know why: $$$$)

This is, really, the only answer to this thread.

Big numbers sell wheels! Few people would look at two otherwise-identical wheels, then buy the one with the lower battery capacity.
Well-informed buyers may know that the high-discharge cells are the safer option, but I would wager most buyers either think "I don't need high-discharge for the riding I do" or "the company wouldn't make an unsafe design". Neither of those are true, unfortunately, as even moderate acceleration or speeds can overstress these high-capacity cells, and every current EUC company has had fatal flaws in their wheels. They are not engineered or tested to rigorous standards.

The risk (and history) of fire doesn't seem to deter customers much either! Begode wheels sell like hotcakes.

So why would anybody stop using high-capacity cells in low-P configurations when it might result in lost sales?

[0000]

2 hours ago, Cerbera said:

Cells are so densely packed they are  individually very solid, and a group of them bound together in their plastic blocks and shrink-wrap are even more so

Yeah from what I've seen on my RS, the cell interconnects are recessed beneath the surfaces of the plastic holders, although the parallel connections leads on each pack to the BMS are a little more exposed at the top and bottom of the pack.

So basically, while the top, bottom, and front and rear surfaces are supported with foam, you still see some rubbing and wear marks against the large surfaces of the packs themselves against the plastic shell. This isn't a big deal unless you ride in very wet conditions as the wearing of the heatshrink can eventually create an opening for water. It would be another matter if the wear was occuring near the positive electrode of any of the cells as bridging to the cell can would be the catastrophic short scenario. Again, just speaking about the RS pack design here as that's all I've personally seen or looked into.

Unless you get a puncture from catastrophic damage, its unlikely you'll have a problem with the interconnects assuming spot welds from factory assembly were good and QA performed.

Anyway, what I was trying to say earlier is if engineering performed their job in respecting performance limits of the cells chosen vs the amount of power they're permitting the motor controller to draw, rider behavior shouldn't matter. But for those unlucky to own and use devices where this wasn't the case such as the with some M50T applications, whether you have an slow wind vs fast wind motor and rider behavior are likely to be a factor in your safety.

In any case, since I'm sure many of these PEVs are still floating around out there, there's a non zero chance this topic will keep popping up and creating unnecessary confusion.

Edited December 6, 2022 by Vanturion

[novazeus]

2 hours ago, Pingouin said:

To my knowledge, it is only the older LGM50T that were involved in most EUC fires reported. Of course it doesn't meen they are fireproof (no Li-ion cells are) but they seem to be safer than the older LGM50T cells, the datasheet seems to say that the LGM50LT are safer than the Samsung 50E for example

thanks for that! 

[houseofjob]

On 12/6/2022 at 1:09 PM, Pingouin said:

Regarding the datasheets, the LGM50T are rated at 7.3A max discharge at optimal temparature, while the LGM50LT are rated at 14.4A which is basically double.

https://www.dnkpower.com/wp-content/uploads/2019/02/LG-INR21700-M50-Datasheet.pdf

argh, this is the much earlier/older 2016 M50 Datasheet, not M50T (see the date year is 2016 on the PDF header), EUC wasn't even remotely wide scale using 21700 back in 2016, and the battery manufacturers usually release revisions every 1-2 years for the same model cell.

 

Ugh, so confusing with all these yearly revisions but same/similar cell model names, so let's put this debate to rest with actual facts here (not directed specifically at you here @Pingouin, just this argument gets frustrating without hard fact comparison):

 

2016 LG INR21700 M50 (https://www.dnkpower.com/wp-content/uploads/2019/02/LG-INR21700-M50-Datasheet.pdf) 

• Max Discharge Current:
  • 5-45ºC: 1.5C(7,275mA) (page 4)
  • 45-60ºC: 1.5C(7,275mA) (page 4)

2018 LG INR21700 M50T (http://queenbattery.com.cn/index.php?controller=attachment&id_attachment=111)

• Max Discharge Current:
  • 10-25ºC: 3.0C(14,550mA) (page 7)
  • 25-55ºC: 1.5C(7,275mA) (page 7)

2020 LG INR21700 M50LT (https://cdn.shopify.com/s/files/1/0481/9678/0183/files/2020-21700_M50LT_datasheet.pdf?v=1650029106)

• Max Discharge Current:
  • 10-25ºC: 3.0C (14,400mA) (page 5)
  • 25-55ºC: 1.5C (7,200mA) (page 5)

2021 LG INR21700 M50LT (https://f.hubspotusercontent10.net/hubfs/6584819/Cell Spec Sheets/LG/Lithium Ion LG INR 21700 M50LT 2021-LSD-MBD-b00002_21700_M50LT_PS_Promotion_V1.pdf)

• Max Discharge Current:
  • 10-25ºC: 3.0C (14,400mA) (page 5)
  • 25-55ºC: 1.5C (7,200mA) (page 5)

 

Also, to add to the confusion, EUC distributors, from what I remember last research, are all over the place referencing which exact cell is being used, the circa 2016 LG M50, circa 2018 M50T, or more recent M50LT. I'm not even confident they themselves indeed know which exact cell was used in these fire incident wheels now, and what with the demand for batteries, I highly doubt an 2018-2019 produced Gotway wheel is utilizing some kind of ancient storage of the 2 year old 2016 LG M50 cell, as that would have been all used up on 2016 PEVs (I also remember stuff like the Z10 couldn't even source proper 3.5Ah cells, having to resort to like ~3.2Ah max, due to battery shortage somewhere around this time).

 

All that aside....

Unless there is some magical vastly improved 2022 LG M50LT cell whose data sheet I couldn't find, I don't see any vast difference between M50T vs M50LT, unless someone can dig that magic 2022 cell revision data sheet out, or can produce someone's battery review testing that contradicts the Mooch 2022 M50LT testing report which clearly aligns with the various M50T & M50LT spec sheets stating max discharge dramatically reduces in half the farther you get above 25ºC.

And I am not inclined to, again, believe that, what with all the supply shock demand of lithium batteries that still exists even today, that these circa 2018+ manufacture fire incident wheels were using 2-year old circa 2016 LG M50 cells, unless someone can prove that as well, backed up with actual proof.

 

Edited December 8, 2022 by houseofjob

[novazeus]

https://www.kingsong.com/products/ks-s22-18-inch-series

https://www.kingsong.com/products/ks-s18-18-inch-series

according to kingsong's website, the s18 got m50t and the s22 got m50lt.

one, do u think that is true, and two, do u think the "t" makes a difference?

Edited February 5, 2023 by novazeus
addition

[novazeus]

what is elon using?

that's why i'm not wasting brainpower on this. some humans waaaaaay smarter than me sb doing forensics and figuring this shit out. 

going to mars but i might burn ur apt complex down.

https://www.tesla-fire.com/

[novazeus]

guy on the flashlight forum said this:

Many subvariants were available in the LG “M50“production line. The first “M50” was rated for 500 cycles (LG product spec info) with relatively high IR, then comes M50T with only 300 cycles (LG product spec info) and the same IR. The latest subvariant I know is “M50L” and “M50LT” which should deliver 1000 cycles (rumors) and ca 30% lower IR. M50LT is very good cell in its category and now it is available at an interesting price. You can check the results of ongoing cycle life tests on another discussion forum.

and i found this at queen battery,

LG INR21700M50LT(life cycle 1000 times)

LG INR21700M50T(life cycle 500 times)

 

[alcatraz]

Does high IR directly translate to poor current tolerance at hot temperatures?

Maybe that's it. The resistance for current to travel through the cell can only convert into heat. 

Also, what happens to IR after a few hundred cycles. While most cells out there degrade gently, perhaps M50T completely tanks, with much higher IR over its (shorter) lifespan.

[jamy76]

Hello everybody,

This is a link youtube about battery cells for comparaison between range/ long range/ long life EUC.

 

[supercurio]

14 hours ago, jamy76 said:

Hello everybody,

This is a link youtube about battery cells for comparaison between range/ long range/ long life EUC.

 

Note that this table is full of errors and the source can't be trusted - doesn't understand and doesn't try either.
When observing this video being posted I wrote a detailed comment indicating the errors, how and why but as usual my comment got deleted.

It's best to ignore this channel as source of information entirely, most of what is published here is based on misunderstandings. Same for the work being done here: avoid.
As source of inspiration for creative mods, it's great.

In short
The cycles are at wildly different charging and discharging which is why you see the very robust Samsung 40T with a miserable 250 cycles at 60%: only because in the datasheet it is discharged and.. charged at freaking 6A (which is a lot). In this datasheet, Samsung gives a data point for a high drain / fast charge use case which makes sense given the type of cell. But then you can't compare it with other cells charged at ~1/3 the speed.

I highly recommend @BatteryMooch as source of information tho, and its excellent report on most of the cells out here. That, you can consider correct and all tested.
https://www.e-cigarette-forum.com/blog/blog/mooch.256958/

Edited March 4, 2023 by supercurio

[meepmeepmayer]

The overlap between competent people and people who wear a lab coat (wow, real scientist!) for making an online video of writing on a blackboard (wow, real lecture!) tends to be very small

[0000]

I'll echo what @supercurio said (+1 to the Mooch) and add the following (if I haven't already linked it earlier in this thread):

The best publicly available information I've ever come across on 18650 and 21700 battery cells directly comparing cycle life using defined/equal charge and discharge currents in controlled labratory conditions with the same equipment is from poster Pajda on endless-sphere.com.

Here's the links to his threads:

https://endless-sphere.com/sphere/threads/cycle-life-tests-of-high-energy-density-cylindrical-cells.114486/

https://endless-sphere.com/sphere/threads/cycle-life-tests-of-high-power-density-cylindrical-cells.114473/

No youtuber is likely to come close to sharing anything near this level of detail and rigor regarding consumer available lithium ion battery performance due to the time investment required to obtain this data and it's a wonder Pajda posts it for free IMHO.

Edited March 5, 2023 by Vanturion