I notice the 4 cell 18650 flashlights use 4 18650 cells in parallel.
I’m an old timer and was taught NOT to put cells in parallel. Does this not apply for Li Ion cells?
Obviously it is working OK.
Same cells, same production batch, same charge should be fine.
It was advised to not use batteries in parallel because of the dangers of mixing empty and full alkalines, but for li-ion it’s normally safe to have them in parallel even if the cells are not perfectly matched. But then Of corse it would be best to use the same type with close internal resistance.
Cells in parallel are safer than in series.
Even if the cells have different capacity, they all keep the same voltage during discharging and charging.
With cells in series the voltage of the cells will become different if they have not 100% the same capacity, and that is dangerous (especially with lithium-ion cells).
Welcome to BLF!
I like parallel celled lights, long runtimes and safer then series
But I do use the same cells in them and have written capacity and internal resistance on all my cells and keep a few very nicely matched in the same storage box or lights with parallel configuration.
If the torch you have is not sucking the guts out of the cells then it’ll be fine with mixed cells as indicated above.
However if the light pushes the cells then stick with closely matched cells. Similarly with charging. If a cell is good for 1a charge then you can multiply that by the number of cells in parallel. Mismatched cells I’d probably keep the charge current relatively low as different internal resistance will give the charger a false reading.
Those early warnings with regards to using batteries in parallel were off the early days with primary cells which, supposedly, were not rechargeable and, thence, sh1t could happen when cells would balance-charge its sisters.
With rechargeables, there's absolutely 0 problems even for mismatched capacities and internal resistances as long as:
- Cell voltage ranges properly match.
- Upon setting up the pack, please make sure cell voltages are close enough. According to Ohm's law, when fusing in parallel any two cellpacks (I define a cellpack as any number of cells in parallel here, including a single one):
I = (Vhighside - Vlowside) / (Rhighside + Rlowside)
So, some shitt1ly high current can flow between cells/cellpacks if you set them in parallel without making sure their voltage differential is close enough.
And damm! Use decently thick wires and make sure you make good low-resistance contacts if high current is expected to flow over there!
Cheers 
As for running cells in parallel,
When discharging, “from each according to his ability”
And upon charging, ” to each according to his need”
Cells are happy with this, people, perhaps not…….
Oops, I broke the rules 
When mishandled, LiIon’s generally give problems in 3 areas: over-discharging (damages the cell), charging damaged cells (BIG problems), and un-monitored series usage. Common sense takes care of the rest. As long as the cells are reasonably matched and you don’t allow any cell to over-discharge, parallel use is almost as safe as single-cell use.
If you’re new to LiIon cells, read this safety primer first.
Phil
And usually (Vhighside - Vlowside) approaches ZERO rather quickly.
Of course! The amount of current flow, however, could overheat certain batteries, that's the reason I said it.
Cheers
another failure mode that gets really nasty is if a call develops a hard short.think the contaminated cells that are being recalled.the shorted cell obviously goes into runaway but the unshorted ones are dumping into the same short.yes the ptc on the cells trip but that does little for the one that is venting with flame and the severe heating from running into the short till ptc if one is present trips.they are then primed for failure and the shorted one is now red hot.and its against its neighbors.so in the same timeframe you have all of the cells in that group running away.thats why the tesla packs have fuse links to each cell.