9x vs 12x testing results for which ones were better for lux and lumens.

I'm a big fan of multi-emitter lights and have built dozens of them. I get asked this question every once in a while so I thought I would answer it once and for all. Or at least to my satisfaction.

What is the effect of adding more and more emitters to the mix? To answer the question I went ahead and built two very similar multi-emitter lights. I used identical emitters, drivers and wires on both of these builds. De-Domed XM-L2's on noctigons with a BLF Zenor mod DD ultra-low resistance FET drivers with ramping mode. I used the same set of freshly charged Sony 30amp 26650's. After testing one light I recharged them for the second test. So the exact same batteries were used and in both cases the batteries were fresh off of the same charger. I even measured and cut the wires so they would be the same.

So what did I find? Surprise, surprise surprise. Not a dimes worth of difference in lux or Lumens at 30 seconds!!

Lux was measured from the exact same spot and so was lumens. In both cases I made sure to keep the meter in the same spot and also set the lights in as close to the same spot as possible. Lux was 146 for the first light and 148 for the second. Well within the possibility of me having placed on a fraction of an inch differently. Lumens were both essentially the same at 8,750 and 8,780. I make no claims that my actual lumens or lux are perfectly correct. So actual lumens/lux might be higher or lower but the comparison is accurate. Lux and Lumens do not seem to be effected by the addition of the extra 3 emitters.

Now on to opinion; It's my belief that in this setup the total lumens and lux are being limited by battery power rather than number of emitters. Of course this would change if you went from 3 emitters to 12 but when it comes to deciding to mod either a 9x or a 12x I would suggest sticking with the 9x because it's cheaper to do so and the results will be roughly equivalent.

I also tested a J-20 up against the 12x that I have pictured. I did manage to get heavier wires in the J-20 and that made a difference in lumens and lux. But the major difference was the J-20's ability to handle the heat. Both Lux and Lumens were maintained for a longer period with the J-20 vs the standard 12x trustfire. Is it worth it to go with the much much much larger J-20? Well that is up to you, it's a lot to carry around and either light will give you the ability to do about the same for a short time. Of course the J-20 also has the advantage of using 32650 batteries. The top performing multi-emitter light once modded is the J-20 but it loses out on comfort when carrying. It really needs a sling, whereas the standard 26650 multi-emitters don't. They are easy to carry and if you mod them to include a side switch like I do they are just about the perfect security light.

12x

I have a question you may be able to answer.

18sixfifty,

I have a no name 15 emitter XML (not XML2) light. it is DOA. my question is: how are the factory emitters wired?

15s1p (un likely)
1s15p (possible)
or some combination of sets.

I ask because in looking for how to test it or how to spec a replacement driver this will matter.

my problem is that I have been able to get the driver out (left wires attached) but I can not get the head open so I am limited to what I can find out about the emitters themselves.

Do you have any insight?

thank you

Ronin, Is this a 4x26650 or 3x26650? Either way though it’s not likely that it’s wired completely in series or parallel. Probably something like 3s5p or 5s3p. It’s possible it’s got a boost driver and in that case it would be wired in Series but not likely, boost drivers for 15 emitters are few and far between.

Chances are there are phillips head screws holding the reflector in place. If you look into the head of the light they should be there someplace.

Hey 18sixfifty!

What I’d like to know is, how you did the side-switch install/conversion. I assume a drill press was involved. My J20 really should have had one from the factory.

Interesting write-up. I would have expected a few % difference if each LED for the 12x were driven less hard than the 9x just due to being supposedly more efficient.

Nice experiment. Thanks for sharing the info.

Thanks, I was surprised too. The one thing I haven’t gotten around to testing yet is battery run time on low. I wonder if the 12x will have better run time?

Just used a regular hand drill. I put one on that J-20 I built too. Drill press would have been a lot easier. The switches are just Button Switches, they make all kinds of them.

Thanks for testing, that is a neat test! I'm not surprised that 12 or 9 emitter do not matter much, but I am surprised that you got virtually identical results

And how about 18 and 20 emitters?

Its a 4x26650 long (series) tube.

I found a 1.5mm Allen screw in the center of the head (have not been able to loosen it yet)

yes with a nominal voltage of 14.4 and a fresh off the charger of 16.8 volts I agree it is not a 15 emitter boost driver.

so like you said the best bets are 3s5p or 5s3p.

Sooo

3s5p (9.9vf) with a buck driver OR

5s3p (16.5vf) with a boost driver

with a range from-to 11.2-14.4-16.8 volts in the “battery”

Do I have that right?

As the tube comes as 2x 2(cell) tubes and the tail cap fits on each I think maybe it is also designed to run on 2x26650 which would mean its a 5s3p boost driver (right?)

Sounds right to me. Did you ever run it off of two batteries? If you did and it worked correctly then yep it’s got to be right.

Thanks for the testing and report 18650. Good info. I've always prefered the beam pattern of the 9xT6 as it is more throwy. That may change with the Nichia 219C though.

Ronin42,

My 15xT6 is wired 5S3P for the emitters. It has a very insubstantial boost driver. I will be converting mine over to 3S5P, with DD and 3 cells. Plan on using small emitters (probably 219C's).