Hi, the Fenix is TK35UE v2 as per topic title, but it would have helped to include it in the post. Yes, all light levels.
Hmm, it's been said to me that the torch is designed to work with hi drain batteries, which might not be what I've got, so maybe I'll get some Fenix ones. More expensive though., The Sunwayman worked ok (before) on the cheaper batteries.
I apologize. I don't know how I could have missed that:-(. Clearly I wasn't cookin' on the front burner. I'm glad I wasn't doing something critical at that moment - it could have been ugly!!
Is the Fenix a new light that's had this problem out of the box, or a light that's worked normally in the past and has 'failed' in service?
That Fenix clearly is going to have a rather high current requirement of those batteries, specifically at the higher output levels. At those levels, it will need all they can provide. It is, however significant that the light also shuts off unexpectedly at the lower output levels as well, when maximum current is not being demanded of the batts. That would not typically be the case if the batts were simply not up to the task, as the current requirement of the light should be relatively low at the lower output levels.
Particularly in light of having a very similar sounding issue with 2 different lights, which is itself somewhat odd, is it possible that your 18650s are not in fact being properly charged by your charging equipment? That could be a possible root cause issue in both problems you're having.
If you happen to have a volt meter, it would be interesting to remove them and note the voltage of the cells after the light has shut off as you describe.
Assuming you're running the Fenix on 18650 cells, if you have access to 4 CR123A cells, testing the Fenix using those in place of the 18650s would also be a very worthwhile thing to try, as it would effectively eliminate the specific batts you're using, and any questions regarding their state of charge, as the likely area(s) of the cause.