I've used a blade many time to remove these press fit type switch rings. First a razor blade, then a sharp utility knife. Only time it did some damage was to a D4Ti - marked up the soft copper but wasn't careful about using a clean, un-bent blade initially.
I recall Vinh posted info on using the blade on a NiteCore EDC way back.
I Tested the ROT66 Nichia on my ‘KG-Lightbox’ and managed 5,700 with VTC5A, the XP-L ROT66 measured 8,400 lumens but sadly it arrived with one LED not working.
I wasn’t impressed with the D4 or M43 Nichias but the ROT66 has blown me away. The M43 looked to be emitting a green hue when placed side by side with the ROT66, it’s warmer but not yellowy so.
A word of warning — R9080 Nichia’s, especially the SW45k in the Rot66, make EVERYTHING look greenish/yellowish. Hell they make sunlight look off-color, because they’re pink. I love mine, both in the Rot66 and in some other lights I have, but I wouldn’t hold them up like some yardstick by which to compare others. They’re not good for that.
By the time direct sunlight reaches us down on the surface, its spectral balance is actually above the BBL. It has a peak in the yellow/green area of the spectrum— not perfectly smooth black body radiation.
I asked Hank in the beginning of this year about people reporting green tint in the 219C 5000k versions of his lights and he said it’s true that his current batch (at the time) had a lot of green. But in general, all of my 219C 5000k and 4000k were somkeone greenish and yellowish, which is why I’m not a big fan unless we can get the “E” bin emitters that guarantee below BBL tint.
Your lumen readings appear high. For the 219B 9080 version, Maukka got 4,325 lm at turn on and 3,860 lm at 30s. You should buy his calibration lights to correct your calibration.
Yeah I noticed Maukka’s read a lot lower wih higher head temps too, weird as I’m usually bang on or just below his readings for lights.
I know how you feel. I used to think my extensively “calibrated” ceiling bounce method was spot on. Then with the TA tube, I realize how far off ceiling bounce method is and how ceiling bounce grossly favors throw beams over flood beams. Then I thought my TA tube with the calibration discs applied were reading spot-on until recently after I got the Maukka calibration light kit, I realize my TA tube was reading 16% too high. It’s a shocker and disappointing to know all of my lights perform so much less that I used to believe.
I highly recommend anyone with a non certified lumen sphere to buy the Maukka calibration lights.
I’ve made a light box but tbh no matter how I measure the ROT66 it tests far higher than all my 5 Emisar D4’s and D4S. It looks far brighter. Wonder if Maukka has a duffer.
Maukka tested the D4S 219C version at 2,930 turn on lumens and 2,430 lumens at 30s.
I have the XP-L HI D4S, it hits 4600 lumens in my lightbox. I think Newlumen got a similar figure to give a little indication of where it should approx be hitting. In a straight up ceiling bounce my ROT66 Nichia destroys it and tbh the D4S has a thrower advantage.
Maukka measured his ROT66 in 28c, I measured mine in half that, and tbh his sphere temps may have been soaring if he’d checked calibration with similar powered lights before hand, after 30/60 seconds my ROT66 was reading no where near the surace temps Maukka’s was. He got 50c after 30 secs, mine was 40c!
Has anyone else measured the ROT66 Nichia yet?
I will measure it after receiving it.
I will be gobsmacked if nobody measures over 5,000 lumens.
Newlumen has the 4” TA lumen tube which is the same as mine. Our numbers used to match reasonably well until I received the Maukka calibration light, which is when I realized my TA lumen tube was reading 16% too high. Newlumen has a set of the Maukka lights on order and we will find out soon.
Steel_1024 measured 4,880 lumens with his TA Lumen Tube but not sure if he has a set of the Maukka calibration lights to confirm with. The TA Lumen tubes that weren’t calibrated with the Maukka lights were reading too high.
Charles Lin measured 4,740 turn on lumens and 4,300 at 30s with the TA Lumen Tube. But it might be his TA Tube also does not have correct calibration and reading too high.
I’m not saying your measured 5,700 lumens with the ROT66 219B is certainly incorrect. Just saying there’s a good chance and the only way to confirm the accuracy is to get a set of the calibration lights. It’s very affordable anyways.
My Nichia Rot66 just came in yesterday so I can add some numbers to this:
That’s taken in a TA tube that’s been calibrated using a set of Maukka’s lights, and generally jives with the other numbers given here which seem to range from ~4300-4800. Except for your reading of 5700 lumens, which would mean that yours is ~26% brighter than mine. Almost two entire flux bins higher. I hate to use the word “impossible” but….
That’s pretty good… maukka got 3860 lumen @ 30 seconds… can’t get any better with the TA tube.
4.9% difference, well within the range for random binning variance. In other words identical.
Got my ROT66 with Nichia’s, really nice!
Output at switch-on (it takes 3 seconds before the first read-out): 4300 lm, at 30 seconds 4000 lm. This was on freshly charged 30Q’s.
Wait is that Djozz lumens or calibrated lumens?
Djozz-lumens corrected for a 90+CRI 4000K light from maukka, which is minus 10. But please keep an error of 5, you never know with lumens 
Do you have D4 measurements djozz?
I seem to have a freak ROT66.