I now have it apart.
The reflector and glass etc retained as previously described. The electronics module on the right is retained by epoxy against an FT 101 type compound. The torch is not designed to support disassembly.
🔗121030[linkphoto]121030[/linkphoto][/link]
Removal of the battery was more difficult and required drilling of the casing where the LED backplate fits, to accommodate a drift.
🔗121031[linkphoto]121031[/linkphoto][/link]
The battery as expected is a CR123 which has a management PCB mounted on the base and external connections to the LED in the form of metal strips, the whole covered in a second shrink wrap. The battery gave 2.51V and inducing some charge into it from the Lab PSU enabled it to light a test LED, however I could not get it to accept charge in the normal way.
Continuing to charge it I got it up to 3.7V and then, lo and behold, it charged off the supplied magnetic USB connection.
I think that before reassembling, which will require 'permanent' sealing, I will discharge it and see if it accepts normal charging. I may have to replace the management chip at the bottom of the battery, this is a standard item I have in stock, so no big deal.
On the face of it, it looks like I have repaired the unrepairable.
My avatar is a poor likeness.