To melt relatively big optic lens like cute you need some serious power - if we're talking about LEDs light output as source of melting. XP-L HI are not very powerful by today's standards (lm/mm2), so I highly doubt they can melt cute optics. That pretty much can only happen if optics is "dirty" so it absorbs too much energy/light, maybe from remaining flux after soldering, which can evaporate at higher temps and deposit on inner tir surface?
Picture of your soldering work could help to see what's going on.
Other possibility is too hot light case/host , but it must be very very hot to melt PMMA lens (105+C), I doubt EE X6 can become that hot with 3x HI and single battery.
So I measured the current with a full cell and it’s only pulling 12.3A measured with a clamp meter. There is no way a ~4A XPL is creating enough heat for this.
I think you are correct about either vapors from the any flux, IPA, or even the thermal grease must have deposited/reacted with the PMMA and caused a bunch of absorption. I took an old CUTE-3 that had a broken leg and some scratches from my parts bin and installed it later after doing some high-temp “out-gassing” turbo runs and then let it get as hot as I could. It seemed to hold up fine. I’ll have to try again with a new optic… It’s a shame because the one from you had the correct pin height and arrived in pristine condition. Mouser always delivers me imperfect TIRs.
I got my replacement driver and fitted it. All is good now…. except I played around with the lighted tail.
Originally the lighted tail was as dim as it could be. I wanted a little more so I replaced the resistors on the tail. However when connected through the driver it is no brighter at all and in fact had visible pwm.
Do I need to replace the bleeder on the driver as well?
One of the older ICL-1 modules. I cut away the metal casing and swapped out R1 and R2 from 2k to 620ohm. When power is connected directly it seems good but when connected through the driver it’s dim and with pwm.
Does driver has installed bleeder? Are you sure resistors are 620Ohm, they seem to have too low resistance (maybe 62Ohm)? Please check resistor values with multimeter.
That's in any case cheaper repair then, C8 glass thickness can vary from source to source, and probably stainless bezels also can have different heights compared to stock bezel.
Worth to mention here that the new FET switch should be designed without need to additionally modify main flashlight driver…
So far from what I can se in JaredM and Quadrupel version of FET switch it seems that an additional bleeder resistor modd needs to be added somewhere on the driver? If so… I don’t think everyone will want to fiddle with that and modd the driver to get an working FET switch.