Yah thats me. In case you’re wondering about the throw, at turn on i got about 192Kcd or under 900 meters. At ANSI its down to 834 meters . The T9R throws much better, not surprisingly, but these aren’t ‘bad’ numbers for a warm CCT led with a large (almost 13mm²) LES vs the sbt90.2 9mm². Again, matter of preference, but the Luminus is srill a better choice I think for a thrower until Jack can get better bins of this ffl909mx or just get the 6500k version of the ffl909mx.
2 Thanks
Thanks for the informative numbers and for saving me 100 bucks. 100 is the high end of our budgetlight forum 
and it has to be something more special than what we are seeing. A pretty face (that Firefly red color) only goes so far.
2 Thanks
In @Sirstinky 's review, note how Duv is always mentioned in context of output. That’s because with SBT90.2 the Duv changes greatly with output, becoming more negative as brightness increases. I hope the picture below shows this well.
SBT90.2 in Wurkkos TS30S Original (sadly no longer made, but clearly lottery winner) becomes remarkably more magenta/rosy at Turbo output. This change is less so with TS30S Pro, and least with my green monster FFL909MX 6500k.
I think someone has made a distinction of Duv in the forward spill (more magenta) vs collimated beam (still slightly green), maybe @QReciprocity42? My next step will be to measure the Duv in hotspot vs spill.
2 Thanks
Still measuring and checking - this is more tedious and time consuming than I expected. Some preliminary observations:
1. SBT90.2 in my Wurkkos TS30S non-Pro first version is the tint-lottery winner if less green tint is what you’re looking for. This light is no longer made so if you are a lucky owner, hold on to it like dear life (just kidding).
2. At high output/Turbo level, even though spill of SBT90.2 in Wurkkos is rosy with negative Duv, the hotspot remains Duv positive. Personally this doesn’t bother me much in a thrower, in actual use. The target is so far and much dimmer such that any green coloration is not easily noticeable and bothers me a lot less than say a pukey green close-up flooder.
3. TS30S Pro also improves as output increases. T9R has the greenest tint among these 3 (its saving grace is the fantastic 1.3 km ANSI throw, the very nice T9R host, and Lume1 buck driver).
I don’t understand why we can’t see the highest Flux Bin LEDs within the best Chromaticity Bins.
Are they simply not available? Not orderable? Too expensive?
From SBT90.2 spec sheet:
For example, I searched extensively for a 5000K rosy SFT-25R G2 Flux Bin LED. While it technically exists according to the spec sheet, it seems unattainable in practice. Does this mean flashlight manufacturers genuinely prefer cooler, greenish-tinted CCTs? Or is there another factor at play that I’m unaware of?
One possible explanation is that when a new LED emitter enters the market, the initial batches available tend to have higher CCTs and a greenish tint. Since it’s impractical to order fewer than 500 or 1000 LEDs at a time, flashlight manufacturers are likely to use these early batches for an extended period before they can source higher-performance, better-balanced versions. If the emitter remains on the market long enough (before being replaced by a newer model) manufacturers may eventually release “v2” versions of their lights, incorporating more efficient LEDs from the highest Flux Bins and CCT options at the same time.
Just a possibility.
Binning is largely up to random chance. If you tried to make 100 identical LEDs, they will all have some variation in efficacy and tint. If you draw an LED out of 100 randomly, it’s improbable to get (A) a sample that is significantly higher-binned than average (i.e., attains maximum binning). It is also improbable to get (B) a sample with unusually good tint. These events are improbable because the distribution of efficacy/tint will not be uniform, but very light-tailed: the probability of getting higher than x efficacy (or better than x tint) drops superexponentially fast in x, if you assume a Gaussian distribution.
To get both (1) highest bin and (2) usually good tint is even more improbable than the product of the probabilities for getting (A) and (B), since the events are negatively correlated: efficacy tends to vary inversely with pinkness of tint. So if you get 5% chance of scoring a high bin and 20% chance of scoring good tint, there is less than 1% chance of scoring both. That’s why you don’t see emitters with top bin and good tint floating around often.
3 Thanks
I see. Thanks for sharing.
Nice discussion. Data of 519a showing the 9080 bin consistently having lowest output.
Source: page 7, https://led-ld.nichia.co.jp/api/data/spec/led/NVSW519AT-E(6102D)R70%20R8000%20R9080.pdf
This is a similar but different discussion: the R9080 bin results not from variance in production, but from deliberate engineering, so the probability of finding a R9080 sample of the 519A is not low. The phosphor used is different. But the reason for low output is similar to the reason that pinker tints have lower output: less green/yellow and more blue/red.
1 Thank
Is there an explanation for this inverse relationship between efficacy and pinkness of tint? Does it have to do with the higher amount of phosphor needed to achieve a desireable CRI or tint, and this thicker layer blocks/reduces output?
Efficacy is lumens per watt. Per watt of radiant power, green/yellow will attain higher lumens than blue/red because human eye is more sensitive to green. So a closer match to pure 555nm green light, the higher the efficacy.
Magenta tint and high CRI both divert power away from green and toward blue/red, hence the drop in efficacy.
3 Thanks
Final result. Although my LM3 does follow pro reviewers’ numbers well, pls keep in mind it’s still amateur-grade equipment. Take it FWIW pls.
The FFL909MX is very green, but its throw at 1.3 km is noticeably longer than both TS30S and TS30S Pro . Whether this is due to the FFL909MX LED’s intensity or the T9R host’s reflector (largest and prob. deepest), I don’t know.
4 Thanks
Hold your light up in the air. It really helps to give a lightsaber effect. No warranty if very dry and clean air of course !
2 Thanks
Just ran across some beamshots of a couple Firefly E90’s with 909MX on reddit HERE, and I’m surprised by the nasty ring artifacts. I am no expert but my 4 lights with single large TIR (X1Lx2, Convoy M21H, IF22a) have the nicest and smoothest hotspot in comparison to my reflector lights.
And smooth hotspot is what I thought to be expected of single large TIR optic. Explanation anyone?
Not all TIRs are created equal–some are just much better or much worse. And some TIR’s interact better with some emitters than others.
In this beamshot, the weird hotspot boundary is suggestive of improper focus. But there is no excuse for the ring artifacts–the TIR was designed in a way that leads to unwanted secondary reflections.
In general, smooth TIRs are pickier than reflectors because there are two air-solid interfaces (LED to interior of TIR, and exterior of TIR to air), which leads to many unwanted reflections.
2 Thanks
It’s the design of the optic and how the led is situated. All the bigger optics I’ve tested have some degree of artifacting-some worse than others. The Wurkkos TD01 and E90 in particular are bad. Acebeam has the best TIR (large size) TIRs around for beam cleanliness.
3 Thanks
The TD01 optic is also insanely lossy–in particular, the TD01C has only 60% the output of the L21B with the same emitter.
TIRs do seem to get pickier with size. The Carclo triple TIRs are tiny and none of them are known for being ringy. The Convoy 20mm single-emitters TIRs exhibit a lot of variance in quality: the 5050 one is very good (no rings, no tint shift), but all the 3535 offerings are mediocre at best, producing a combination of rings and tint shift.
1 Thank
It depends on the led also. Domed vs no dome, CSP (like 519a, lh351D, ffl505 or ffl351, cree xhp50/70) vs Lateral design (sst and sft series, sbt90.2, ffl909mx). Plus the optic design (narrow or flood). It’s tricky to get right.
3 Thanks
Plus, you always lose output with TIRs. The bigger, the worse it gets. The more the light has to reflect or refract, the more you lose. Bigger TIR have a longer distance (inside) for light to go, so it’s less photons out the front. It’s been my experience that TIR vs reflector lights, the reflectors are always brighter (in single and multi configurations).
4 Thanks
Yeah with larger TIR lights, the throw is always worse compared to a reflector light of the same diameter. Though for smaller lights (S2+ size), a well-designed TIR can out-throw the stock reflector and make a nicer beam.
It’s unclear to me which one is more lossy, a small reflector or small TIR. I would guess that the TIR is still more lossy since a small TIR relative to the LED implies that a lot of light from the boundary of the LED will be going in directions that eventually end up as stray light. With reflectors there is no such dependence, and the loss is constant.