I was trying to smooth out artifacts and the corona but at the same time turn it into something similar to the Acebeam L17. Trying to get more of the hotspot and less corona to minimize artifacts in the beam.
Trying to get more throw if possible with a 10 or 15 degree TIR.
What is the difference between frosted or pebbled? Off topic but I might get a TIR for LH351D and 219c. Not sure if I want frosted or pebbled.
Hum, the type of TIR to make it similar to the L17 would need to be different than those on the store.
Eventually, something like this:
https://www.aliexpress.com/item/1005002196259882.html
This kind of shape lens is - eventually - more what want (notice there seems to be a hole in the front, that will make the beam narrower as the L17). No 100% sure, though.
Frosted TIR optics are those which seem to have been put on the freezer and the front seems fosted (opaque). They give you floody beams without artifacts.
EX: 3.73€ |15/30/38/45/85 grad 20MM TIR objektiv, matte oberfläche, für 3535 LED (xp,sst20,LH351D,etc)|Tragbare Beleuchtung Zubehör| - AliExpress
Pebbled/beaded TIR optics are those that seem to have small bubbles on the top, that will also give you floody beams, depending on the type of degree you choose, more widespread or slight more narrow.
EX: 3.73€ |10/15/20/25/30/38/45/60/85 grad 20MM TIR objektiv, perle oberfläche, für 3535 LED (xp,sst20,LH351D,etc)|Tragbare Beleuchtung Zubehör| - AliExpress
Both of these will not have “hotspots” and will eventually reduce beam artifacts and shifts. Depending on the LED too, of course.
How would the beamshots differ between the two? Output? I can’t seem to find any information where someone has tried frosted TIR optics or pebbled/beaded TIR optics. I know the most popular option is a 45 or 65 degree TIR.
My EDC is currently (and has been for year ) is a custom FW1A with a sliced sst20 4000k. I was first using an OP S2+ reflector, then a modified S21A OP reflector, and now a pebbled narrow TIR from the Convoy store. I’ve bought so many 20-22mm TIRs in search of the perfect beam (for me). What I’m targeting is an artifact free medium/narrow spot with only the faintest spill. 9-11° is the upper range and down to 7-8 is okay. I’ve bought each variant of the LEDiL Leila XP optics, and have been very unhappy with their results with this emitter, especially considering the $ spent. Thinking of swapping in a 351d sliced to give a better beam with the larger LES, but don’t want to give up CD/LM. With the aforementioned Convoy optic, I’m seeing the best results according to my goals, but there are a few stray assymetric spill artifacts that annoy me, so the hunt continues. I think they are from the mold injection points. BTW, beam is about 9 or 11 degrees, depending if I count the ‘corona’, which is really more of a fuzzy edge
So @supreme, I think I understand what you might want from such a setup. CD/LM without artifacts and spill that causes foreground glare in close quarters. I personally can’t get behind the CW osrams in anything but a dedicated max throw setup, but if you don’t care about CCT or CRI go for it. Not much lost if it doesn’t turn out the way you like it. Its your flashlight! 
I don’t have numbers for you, but I can say that both reduce the output if compared to a light with a reflector. I am sue there are numbers out there comparing lights in the same situation or theorizing about that.
I did some close up tests a while ago using different kinds of TIR optics. Essays with TIR lenses on Convoy S2+: XML2 + XPL-HI [Update] - #15 by MascaratumB
I will try to find my frosted optic an take some photos to show you how different they can be in terms of beam.
I just did some output measurements to compare losses last night and rough ceiling bounce tests showed the difference between the reflector and narrow optics was minimal. About 3%, if you would believe that the noise of my measurement isn’t that much or more…
BTW, I’m not saying that all TIRs will perform this way, but my samples didn’t suggest a huge loss.
Thank you in advance!
Well, and there I was, in name of the flashlight science, experimenting some non scientific stuff 
So, I picked my Skilhunt M200 with Samsung LH351D led and did the test with different TIRs, and stock reflector. Also made some “ceiling bounce” - if you wanna call it that way, with the different configurations.
Luxmeter configured to Lux measurement.
Distance to the ceiling (white) = ~1.68m
The flashlight wasn’t moved from its place during the test. The output in every test was turbo.
1 - No lenses nor reflector nor bezel = 23 lux
2 - Reflector + glass lens + bezel = 60 lux
3 - Pebbled TIR + bezel = 45 lux (less intense and more widespread beam than the frosted, as you’ll see below)
4 - Frosted TIR + bezel = 59 lux
5 - Angle TIR (25 degrees) = 64 lux (more intense in terms of light concentration, and less widespread spill than the reflector)
Some beamshots comparing:
Reflector > Angle TIR (25 Degrees)
Pebbled TIR > Frosted TIR
Please note that the TIR optics were sitting on the gasket, not directly on the PCB, so that may change the beam a little bit. Still, you are able to see what they do.
This is also a beamshot of a Angle TIR (45degrees) sitting on a XP-L HD CW emitter (with gasket too).
It seems pebbled produces more spill while frosted tends to produce a smaller and tighter overall beam vs pebbled.
Basically pebbled for more flood, frosted for a narrower overall beam which is not as floody.
Yup! I might give a 10 degree TIR from Simon a try in the S2+ with CW osram. I’m trying to have two cheap 18350 flashlights for short night walks. One thrower and one flood. Should be achievable with two S2+’s for less than $50.
Yup, the frosted - at least in my case - has a less wide/floody beam! It is probably that larger domed leds, or dedomed leds, will produce differences, though.
But I’d say that a 60º Pebbled lens will produce a wider beam (you can chose one less floody)
May I suggest you one thing?
For the shorter throwy light, why don’t you go for an S2 (not S2+) with Smooth reflector?
Check this post where I compared an S2 with smooth reflector and Osram W1 and an S2+ with smooth reflector and XP-L HI Leds.
You can see that the S2 has incredible throw! And with an 18350 tube you’ll be able to “shrink” it!
Just a suggestion, though!
Simon said he was unable to put the Osram with the S2 when I asked. I don’t want to build one myself
Somebody here would help I’m sire if you said where you are located
When I did it the reflector wouldn’t focus properly and my beam had a lot of artifacts. My flat white did focus with the S2+’s reflector though.
How would a clear 45 degree or 60 degree TIR affect the beam?
I believe this is what Simon was referring to when I asked him about it and suggested different options to possibly make it work.
At this point, I’m ok with the S2+ and just trying a TIR. Just sure if this would fit.
CREE XML XML2 LED TIR Since it has the 10 degree clear TIR
or if I should order the 10 degree from here 3535 LED
For flood, what is the difference between 45 degrees and 60 degrees?
I am not sure if I understood the question well, but I’ll try to reply.
Looking to the examples I gave above, namely the clear TIRs of 25º and 45º (last one), you can see that these clear TIRs produce an hotspot but there is some spill around it which is more or less defined depending on:
a) how the TIR sits on the gasket or PCB
b) the type of (dome, dedomed, footprint)
c) the angle of the TIR itself (10, 25, 45, 60).
Besides this, there is the outer spill which is also more or less floody depending on the above criteria.
As you can see, the 25º lens has a more centered beam (although not completely defined), while the 45º has a defined hotspot and a defined “spill” around it.
Check this thinh. maybe it can help as it is slightly standardized!
https://www.reddit.com/r/flashlight/comments/5pc9ct/convoy_s2_tir_lens_1015254560strip/
BTW, it doesn’t have clear 45º and 60º TIRs, though.
I think I get it, a clear TIR changes the hotspot, corona, and spill depending on the degree.
Some manufacturers choose to use a TIR instead of a lens. Manker MC13 and Acebeam L17 for example?
Using a clear TIR on the S2+ would replace the AR coated lens. You would modify the hotspot, corona, and spill to your liking by replacing the lens with X degree TIR you wish?
Honestly it sounds like you should just buy a sample pack from simon and play around with them. See what you like.
The sample pack is this one which I’m planning on ordering.
CREE XML XML2 LED TIR
I’m not sure if it will fit so I’m also looking at other TIR Simon sells with 3535 in the listing title such as 3535 LED