That’s such a small difference tho. That could be normal self discharge from just be a few months of sitting on a shelf.
You could charge the few low cells to 3.52v and then let them sit for 2 weeks to confirm if any actually have higher rates of self discharge. It’s more work but you wouldn’t end up recycling a few good cells.
When 99 are one voltage and that one outlier is different, i see it as a runt that should not be used.
For the cost of a single cell, why would you risk it? Its crazy.
I’m not saying you’re wrong, because I do see your point.
It also seems like this is easy to test and confirm if the cell is fine. Because self discharge is an easily measurable thing. So if you weren’t in any particular rush, why not just charge the runt up to whatever the others were at and see if it maintains the same voltage/ self discharge?
It just seems unlikely to me that you’re going to see a very high bad cell rate for new cells from molicel, but I’m not an expert on batteries.
Maybe @Battery_Mooch wouldnfind this convo interesting
Because I value my time 
Ain’t nobody got time for that!
Youre not wrong and its probably fine, but as Al said, nobody has time for that when youre trying to get down and build a pack
I just use the low one(s) for weld checks then go into vape or flashlight rotation if they dont get too mangled from weld tests.
Moli’s come so consistent that when an outlier of .05v differential comes along its hard not to look at it sideways.
Was already tracking but just not much to say. A lot depends on personal preferences as we don’t have a ton of data to base decisions on.
The most oft quoted failure rate for big-name round cells is one in a million to one in ten million. But a higher self-discharge rate cell isn’t a failed cell. It’s not even a problem unless the difference from the other cells is large.
If one cell is 50mV below all the others then manufacturing dates and batch numbers are important. If same batch then check dates. If they’re all four months old then 50mV difference is a lot worse than if they’re all two years old.
I should mention that the only thing that unbalances a pack is the difference in the self-discharge rates of the cells…that’s it. So if crappy BMS (no balancing, lousy accuracy, very low balancing current, etc.) is being used then a tighter grouping of discharge rates can help to extend the life of the pack. If a good BMS is used then any differences in the rates doesn’t really matter except that less balancing is needed if the pack isn’t being charged often.
What to do is up to each of us but 50mV in a year old, or more, cell isn’t a worry for me. If a lot more recent than that then I might consider using that cell for something else if all the others were very close to each other.
But my OCD side would completely understand someone not using a cell even just 10mV off if all the other cells were really tightly grouped together. Certainly no harm doing that. Well, no harm to anything other than our wallets. 
I guess I’m getting 170 cells then. LOL.
2 motors. Each motor has its own 60A controller. I’m probably going to upgrade to a stronger controller and shunt that controller. Those 2 controllers are going to the BMS then the BMS to the battery. I think I’m using the BMS discharge. I want to be able to control how much amps is going to the controller.
I just bought the parts. Amazon says that it’ll be here on Friday.
This is generally regarded as not the safest route, and the experienced esk8 builders here all run the bms bypassed for discharge. From what I understand it increases the chances of a cutout and sending you into a crash.
This is easy to do through existing telemetry devices - davega, metr, robogotchi, megan, even just a bluetooth module and “yourstruly” app.
I’ll be doing post throughout my build. I’ll keep in touch and you can let me know if there’s a better way on doing stuff. This is my first battery build so me doing post throughout my build is a way for the esk8 community to look over my shoulder just in case I do something wrong or if I’m missing something.
Thank you for the idea of bypassing BMS. I’ve experienced cutouts before. It’s not fun.
Hey man. Do you mind telling me which website you bought your JBD BMS? Like a link to the website? Thanks. Also, how long did you have to wait? LOL.
I bought mine from @Skyart
I got mine direct from JBD off AliExpress. Arrived in about 7-10 days.