Flossing to Eliminate XHP-emitter Multi-die Artifacts

pc_light

Enlightened
Joined
Feb 10, 2004
Messages
600
Location
Old Dominion, USA
Although I've successfully swapped several XHP-50 emitters, despite my best efforts at zeroing in the focus I simple could not eliminate the multi-die artifact/shadow from appearing in the hotspot of my most recent host a Surefire KL6 :banghead:

From what I could gather two popular solutions to eliminating artifacts involve deploying an OP reflector (already in the KL6) and de-doming and/or sanding the surface of the dome itself. (De-focusing the hotspot can also reduce the artifact but that sort of defeats the purpose.)

De-doming has the effect of reducing overall light out the front and/or possibly introducing unwanted green tint in the corona; in exchange for greater emitter intensity and a bright hotspot. Since my light had a relatively deep reflector and the emitter was already well focused I didn't feel this need.

Sanding the dome seemed like a good option but not without its risks. Sanding might also reduce overall output and possibly diffuse the tight hot-spot. Both are bearable trade-offs but an alarming risk that some sanders reported was destruction of their emitters (usually because the emitters were being pushed to/beyond their power limits) because of residue from sanding (or possibly excessive internal reflections?) :eek:

After some fooling around to verify that the artifact/shadow results from the lines between the multi-dies themselves and not some other complex interaction of emitter and reflector, I decided to see if just focusing on the die lines (pun) alone would work. I decided to sand just along/above the separation lines/images themselves.

Here's a photo of the final criss-cross sanded XHP-50 -
gCQczpc.jpg


I cut/used a ~1 mm (<1/32 in) wide strip of dry/wet emery about ~80mm (~3 in.) long. Wet/dry emery is more flexible and follows the contours of the dome better than dry coarse sandpaper. I used 1000-grit which is fine enough to just cloud the surface of the dome. It also slides easier than coarser grades of paper. Sort of like flossing my emitter :D.

The resulting beam still has a slight artifact barely noticeable perhaps when white wall hunting but otherwise it's gone. I do not have any before/after beam comparison shots but from what I've read although visible in person, the artifacts don't show in photos anyway.

YMMV but this should work for just about any multi-die emitter. Not sure if it'd be effective enough on a SMO reflector:thinking:. If you try this and it works for you (or even if it doesn't :() please share your findings. AND, if someone who tries this can take actual before and after measurements to quantify any changes in brightness from using this option to eliminate multi-die emitter artifact that would be really useful.
 
Last edited:

degarb

Flashlight Enthusiast
Joined
Oct 27, 2007
Messages
2,036
Location
Akron, Ohio
Ingenious.

Since the Hotspot is the side emissions, reflected into center, I would suspect it mostly would only partly help the hot spot.

I am not a fan lamp shades. So, the partial sand appeals to me. Probably, should do a ceiling bounce test before and after to see if any drop.
 

pc_light

Enlightened
Joined
Feb 10, 2004
Messages
600
Location
Old Dominion, USA
Thanks degarb, yeah I thought it was worth sharing.

At the time I was so hung-up on getting rid of the artifact that lumen loss was secondary. I figured if de-doming or sanding results in like a 10%-15 loss, surely partial sanding should be less, or so my logic went.

It's hard to compare ceiling bounce without both lights present at the same time, but I should have at least taken photos :ohgeez: but I'm looking forward to someone trying this who can take actual measurements and/or photos of before and after.

:anyone:
 

degarb

Flashlight Enthusiast
Joined
Oct 27, 2007
Messages
2,036
Location
Akron, Ohio
Lux meter, not camera. . Though, I would be interested in the before and after white wall shot.

I cannot see other people trying this approach. Maybe, they can measure the lumen drop.
 

pc_light

Enlightened
Joined
Feb 10, 2004
Messages
600
Location
Old Dominion, USA
FWIW, some "after" photos, these would have been more meaningful if there were some before images for comparison.

It's challenging enough getting accurate beam shots under ordinary circumstances but these images are even less accurate due to the reduced resolution (800px max) required for posting.

With the underexposed shots the center artifact is somewhat evident.
DxO0dp5.jpg


9u9eWC8.jpg


xuKbeE8.jpg
 
Last edited:

pc_light

Enlightened
Joined
Feb 10, 2004
Messages
600
Location
Old Dominion, USA
It looks like the new XHP##.2 generation of emitters pretty much eliminates the need to floss entirely.

See in the image below how the XHP-50.2 has almost no visible die separation.
qL4xgal.jpg


So was Cree's inspiration to fill-in the separation lines with more phosphor the result of a quest for more lumens or to eliminate the die separation artifact? - It's a welcomed improvement either way. :thumbsup:
 

staticx57

Flashlight Enthusiast
Joined
Dec 2, 2011
Messages
1,749
Location
NJ
Richard of Mountain Electronics already built a Convoy C8 with an XHP50.2 and the cross is all but eliminated!
 

sween1911

Flashlight Enthusiast
Joined
Dec 10, 2003
Messages
2,049
Location
Pennsylvania
Hey, funny stumbling on this. I just put an XHP50 in my Blackhawk Gladius Maximus. After shimming the reflector up with a second spacer to get rid of the initial donut hole, I see that tiny crosshair shadow in the hotspot. This is a cool solution.
 

pc_light

Enlightened
Joined
Feb 10, 2004
Messages
600
Location
Old Dominion, USA
Hey, funny stumbling on this. I just put an XHP50 in my Blackhawk Gladius Maximus. After shimming the reflector up with a second spacer to get rid of the initial donut hole, I see that tiny crosshair shadow in the hotspot. This is a cool solution.
Hi Sween1911, thanks for the comment and I hope flossing helps. The Blackhawk Gladius is a classic, built like a tank and worth upgrading. If your result with an emitter upgrade are like mine have been you will be impressed by the jump in output.

Some modders "de-dome" their emitters to try to reduce the shadow, which I considered but I figured flossing could be a first pass option before going that route. The crosshair shadow is very subtle and subtler still will be any improvement from flossing but it's no cost other than a few lumens from added diffusion.

I haven't done any other emitters since the OP now that there's the XHP 50.2 series. Of course now the 50.2's can result in tint-shift, arghhh can't win. I notice mfgs that still use the XHP50, such as Zebralight, put theirs under frosted lens.

Good luck with the rest of your mod and do share any results.
 

sween1911

Flashlight Enthusiast
Joined
Dec 10, 2003
Messages
2,049
Location
Pennsylvania
Checking back on this, I'm experimenting with an XHP50 in an M3 head with a 36mm reflector. Artifacts galore. However, I realize I stuck it behind a smooth reflector instead of the OP one I have. Going to try the OP reflector instead.
 

pc_light

Enlightened
Joined
Feb 10, 2004
Messages
600
Location
Old Dominion, USA
With LOP reflector your results should be similar to the images in Post#5 (reduced but still slightly visible) which coincidentally was from Surefire L6, the LED version of the M3 Combat, which I'd upgraded to XHP50.

I went through my supply of XHP50 and use 50.2's these days with no more donuts!

BTW, Sween1911 how did the Blackhawk Gladius LED upgrade go?
 
Top