T6 vs U2 vs U3

Okay, I posed this question at the end of another thread (where it’s lost), but the feeling seemed to be no.

Keeping the existing driver, will changing the emitter from a T6 to a U2 or U3 make any appreciable difference in output?

I changed a T6 to a U2 in a light with a toasted emitter. Going by my sometimes-flawed memory, the light seems now to maybe 10-20% brighter. Or maybe the soldering/wiring was better than the old one (with increased efficiency — sending more juice, resulting in more light). OR perhaps my memory is just plain bad, and I’m dreaming.

As there’s always the possibility of not having enough wire to change the emitter or other mistakes that could render the existing emitter/driver nonoperable, I don’t want to change some emitters willy-nilly — if it won’t make any difference. Plus, U2 emitters are $6.x a whack, and “Cheap” is my middle name. :wink:

Any help would be greatly appreciated.

In terms of brightness; T6>U2>U3

EDIT: Woops, Pok is right; T6<U2<U3, got my symbols mixed up.

In terms of brightness; T6<U2<U3 :wink:

T6 Versus U3 in same host, same driver, same battery voltage:

Sorry, I don't have a U2 comparison, but I'd only assume it would be somewhere in between.


Beamshots were taken at just over 1 meter distance with Pentax K100D f/11 ISO-200 1/30 sec. and white balance forced to "daylight"

I also verified there were NO overexposed pixels in either image. What you see is what it did.

The host was literally the same one, it was a before and after mod beamshot, in the same exact light. The only possible variable is how deep the led ended up sitting in the reflector hole, and maybe centering, although they "looked" centered to me in both instances. For a few more photos and another comparison including the SSC P7, my original post is here.

Hey thanks for the quick answers. Then I wasn’t losing my mind. (Well, I still may be, but that’s different story.) There is a difference, even if the driver and host stays the same!

Great beam shot …. But realistically how much gain are we looking at here with an emitter upgrade (from t6 -> u2): 10% - 20% - 50% ??? (50% - I wish!).

For example, my X8 is a hair dimmer than my hd2010 (yes I know the beam patterns are different, but generally speaking), would the emitter upgrade be enough to push the x8 past the 2010 or make it equal to the 2010? In other words, is it worth doing?

I can hear you asking. Well, why doesn’t he just do it himself and find out? Because by pulling the existing t6, I lose my main point of comparison, so it will be hard to judge the upgrade result, except by memory.

Thanks again.

7% (i.e. a factor of 1.07) from one bin to the next (in the order T4, T5, T6, U2, U3, U4).

About 7% difference between bins. Of course that could also be .1% difference or 13.9% difference between say a T6 and U2. Always assume the very top bin is clustered closer to the bottom of it’s brightness bin …especially when they are in short supply. It’s just the nature of the normal distribution of bins that come out of the manufacturing process. Oh and Cree’s flux measurement error according to the datasheets…+–7%. Yep the bin sizes are basically the same as the measurement error. :wink:

Also don't forget how human eyes perceive brightness levels so a 10% difference might look subjectively like closer to a 5% difference.


Hey thanks to everyone for their help.

So has anyone switched out an existing T6 emitter for a U2 or U3? Is it worth it? Were you happy with the results? Were you blown away by the results? Was it waste of time and money?

Would you switch out a T6 in an X8 for a U3? Ke-5? Keygos m10, small sun zy-t08? Those types of lights. I’m thinking biggest bang for the bucks. Changing out the emitter in top performers.

Maybe the old computer processor upgrade percentage applies here. I always told people that it’s not worth upgrading your processor unless it ends up being more than a 20% speed difference. Anything less, and you won’t notice. You might not even notice 20%.

Thanks again.

While the higher bin should always be brighter, I’m not sure you can pin a figure to expect.

If you look at the tables here: http://flashlightwiki.com/Brightness_Bins

A top of the line U2 is 1040 lumens (3000mA) while the lowest U3 is also 1040 lumens, so you’d see little or no difference.

However chances are you’d end up with a middle’ish U2 and a middle’ish U3 so there should be some difference, but it could vary quite a bit. Using the example above you could see a 130 lumen difference if you went from 975 to 1105 lumens which is the extreme from worst U2 to best U3.

If the camerasettings were the same, and assuming the die size and dome of the T6 and U3 are identical, these pictures suggest that the U3 led has a bit more throw, suggesting that it is positioned better in the focus of the reflector than the T6??

I put a U3 1C in my HD 2010, T6 based from TMART. I think it made a difference but didn't blow me away. Crelant's new 7G5CS has a U3 they rate at 920 lumens, while the U2 was rated at 895 - I suspect the U2 was a U2 1A.

Added more information to my beamshots above to help clarify.

If its the same tint, upgrading the emitter is always worth it, you never know if your T6 is average, under average or over average, so just upgrade it.
No point comparing two different lights though as they have different drivers.

>>>> they have different drivers

And different reflectors. :wink:

Think of it this way: it’s basically impossible to tell if a given light has a T6/U2/U3. If you can’t even tell, why bother?

Sorry - was trying to point out with the Crelant light, going from a U2 1A to a U3 1C only meant 25 lumens increase, it appears. From reading/postings on the new U3 model, it looks like they did nothing to increase lumens, so the 25 lumens is probably purely from the LED change.