Light (sans battery) rode in my pocket on a dive to the U-853 this weekend. Max Depth was 117 Ft. I did 5 min of hang between 10 and 20 Ft. (I think this is one of the most likely times for a leak). It came out dry, so I got it most of the rest of the way open and will post some measurements and pics soon. Hopefully.
My assessment is that it wouldn't be a bad backup light as delivered, but if my life depended on my light, I would choose something else. I'd also advise that before counting much on it that it be torn down, cleaned, threads inspected and cleaned up as necessary, and have all the O-rings replaced. You might keep the one behind the lens and the one that provides friction on the magnet ring.
The light was shipped with double 23mm OD x 2mm thick silicone O-rings on the tailcap. After measuring the gland size, these do not appear to be the appropriate size rings. In fact, a standard (in the US at least) -024 o-ring which is nominally 1-1/4 OD and 1/16 thick (but really 0.070" thick) appears to be an ideal fit. I might have this size on hand from previous projects, but if not I will get nitrile -024s before diving again with this light.
I've got the light apart now...
The junction between the body and head also has two O-rings, and the threads had locker on them, so I had to use a strap wrench to open it. Fortunately the flats on the body provided a nice place to put it in the vise, and a little padding prevented any damage to the finish. The O-rings are 26mm OD x 2mm, despite the gland being smaller diameter than those on the tailcap. Perhaps the rings were installed in the wrong places. It looks like a standard -021 ring would be a very suitable replacement. Yeah! But I do not have this on hand. Boo!
The magnet ring is captured between the body and head and holds a single stack of 2 neodymium magnets, each 5mm dia x 1mm thick. I was quite surprised to see that the ring is actually two pieces, inner and outer. I couldn't tell until I pulled the magnets out and looked in the hole. Fortunately the dimensions would prevent the outer ring from falling off if the two become separated. Not that it looks like that is likely to happen. I mention the magnets because the magnets in my first full-custom dive light (DL1) have degraded quite a lot over several years. Despite being fully nickel plated, they have rusted, partly disintegrated, and lost a lot of strength. I tried to replace them before the U-boat trip, but they expand when they corrode, and so won't come out of the ring, and they haven't yet disintegrated enough to easily dig them out.
Friction on the ring is provided by a 30mm OD x 1.5mm o-ring, the same one as behind the lens. There isn't an ASA standard size that's comparable, so it's probably easiest to use the metric sizes, if they ever need to be replaced. I wouldn't replace either of mine unless it became necessary. However, the friction ring will be exposed to debris from the water, especially iron from wrecks, which will be attracted to the nearby magnet, so it may need more frequent replacement.
I worry that the magnet ring may get jammed by sand, bits of iron attracted by the magnet (yes, this DOES happen on wrecks), etc. Sand (I think) quickly jammed the focusing head on my second full-custom dive light (DL2), by jamming between the smooth ID and OD of the head and body on the water side of the O-rings. DL1 has a magnet ring (DL2 uses a piezo switch), and I've had to clean it out during a dive. Fortunately I had the accidental foresight to cut the ring into two half-circles and hold the two halves onto the body using an o-ring like a rubber band. I did this so the body wouldn't have to separate at that point to install the magnet ring. It has the advantage that the ring can now be removed, cleaned, and re-installed during a dive, and I have had to do this several times. Getting sand out of it only takes a few seconds. A single-piece ring can't do this, so if it gets jammed, it's jammed for the duration of the dive. I'll keep you posted over the summer as to what happens.
The lens is 1.160 x 0.075" glass, with an o-ring behind it, in a nice groove to keep it in place. The retaining ring which screws into the bezel in front of the lens could stand to be bored out a little. It's not much of an issue with the stock reflector, as the ID is the same on both parts. But if using an aspheric, either standard or Fresnel, the added room could let out a little more light. I'd think about boring it from the current 0.904" to about 1.025"
The emitter is on a slightly undersized 16mm round aluminum 'star'. I can't tell if a full 16mm round would fit in the space or not. The star is labeled XP-E ZL 16. I doubt it's really an XP-E, though it looks like one. There is significant debris on the dome of the LED. The solder paste used to attach the LED to the star was applied in excess, and has squeezed out and formed a large number of solder balls which are loosely attached to the board by the solder flux residue. Cleaning after assembly would remove this, but they didn't bother. Also, the very thin fiberglass laminate on top of the LED's ceramic substrate has been damaged and fibers are frayed. It looks like this happened after assembly, as there's a big scratch on the star along the same line. Sounds like some pretty horrific assembly practices.
Now, off to take some pictures.