【convoy】S21E XHP50.3 HI available now

I got the same L21B (CSLNM1.TG, 12-group driver) and the driver looks similar to the one in your picture.

Using UT210E clamp meter, and with a fully-charged 21700 battery, I measured up to 5.4Amps tailcap current (not 6 Amps).
I'm not sure if this is the 5Amp or 6Amp driver. The box label indicates "5A 12-group" though.

Beamshot of the Convoy L21B (Osram CSLNM1) (5 Amp driver? - although I measured around 5.4 Amps max tailcap current when using a full charge battery)

for reference, this is the how the scene looks during daytime

distance as measured by Google Maps


I measured this with my UNI-T 383S:
1488 x10 lux L21B CSLNM1 at 6m = 535680 cd
Throw: sqrt(535680 cd x 4) = 1463m

And Manker for reference;
4831 lux MC13 White at 6m = 173916 cd
Throw: sqrt(173916 cd x 4) = 834m

Distance and height could be off by +/- 10cm when placing the flashlight at one end of the room and taking the measurement. I used Max mode and moved the beam across the sensor.

The distance is critical and does vary the numbers quite a bit. I get 155,800cd from my MC13. So yours is either performing really well, or your numbers a little high (or mine low)

I’m using a Vapcell M11 18350 charged to 4.20V

I was holding the flashlight on top of two boxes. :person_facepalming:

Simon, can you tell us when the 12 mode driver will land for the T3… will be waiting for that one.

As far as I know, the Convoy L21, be it B or A, uses a ∅22mm driver. Then, the ∅22mm “SST40” drivers (12-groups, 2-modes, 4-modes) claim up to 6A of output current, i.e. they should limit current exactly to a maximum of 6A.

We already know there have been some shenanigans involving the aforementioned driver family. However, d_t_a claims that his L21B only goes up to 5.4A, which is abnormal to say the least… ¯\_(ツ)_/¯

Just received the Convoy M21E in the mail today. One little glitch, when turned off 1 of the 4 quadrants of the XHP70.2 stays on. Unscrew the tail cap and it fades out in about 2-3 seconds. Tighten the tail cap and that one quadrant comes back on.

I measured 6.42A with a 40T and 6.5A with a Vapcell T50:

Purple ring might be AR coating? I’m a little worried that 6.5A might be pushing the CSLNM1 too hard.

That purple is AR coating. And i think your LED will be fine. heres why i say that -

man, Tim that reflector focusing is Tight! And althought the osram data sheet says 5 amp max, we know all mfg spec sheets under rate their max pulse amperage in order to achieve the most life out of the leds..

I think youll be fine at 6.4 amps, although it is high and 6.0 amps would be better, i found this test on blf from 2018 from member Luminarium Iaculator https://budgetlightforum.com/t/-/53735

They say that the cslnm1 on an FET driver with samsung 30q lasted remained at 7 amps with no losses for 3 minutes multiple times. and that also the emitter performed best between 5.8 - 6amp.

I have also run an M21c with 6amp sst40 driver which i installed a CSLNM1 white into, and it was a BEAST, the LED never got close to AngryBlue (TM), and was one of my most impressive throwers. (i shouldn't of sold it)

This is unfortunate.. Mine should be in any day now.. i will be pretty dissappointed in this. I dont even want the xhp70 version, i just had to see the host in person and it was worth $35 for me to do that lol.

I ordered a bare xhp50.2 4000k at the same time to throw into it to see which one i like better

Honestly, once Simon is done with the new 3 volt driver, i will buy it and be swapping out that xhp70 for ...(fill in the blank) CULNM1, CSLNM1, SFT40, etc etc etc.

Ordered 8 holsters from Convoy Ali store today. Thanks, Simon for offering these. I can’t understand why it’s so hard to find decent holsters.

Ditto. His holsters are excellent. I’ve got ten of them on my flashlight battle belt. I really appreciate the velcro and snap button “closure” on the belt loop as well as the extra little elastic pockets on the sides for extra batteries.

6.4-6.5A is quite high.

I previously mentioned I got 5.4Amps (using a similar setup as what you did, and also using the same UT210E setup) - I previously just used an ordinary Liitokala Lii-40A 21700 battery (not so high-drain).

I tested again using a Samsung 40T (4.20v full-charge), and got 5.8 Amps max with UT210E. It would appear it is a 6 Amp driver used on the L21B (CSLNM1).

inhave a dozen or so from Simon, his are fantastic quality! And a great price

yea, and 6.4-6.5 makes no sense. Like Barkuti mentioned above, it should cut off at 6.0a ... unless it's an underperforming 8amp driver (I've heard that happen before too, but not this low it was more like 7.2 iirc).. but that would seem like a huge blunder- installing the 8amp driver instead of the 5amp driver??


Here are the KW CSLNM1.TG (1mm², 3030 footprint) and KW CULNM1.TG (1mm², 4040 footprint) datasheets. Maximum specified currents are 3000mA for the 3030 and 3300mA for the 4040, at 25°C.

Then, here is the Reflow conditions tested of Osram KW CSLNM1-KW (White Flat 1mm2) thread by djozz. In such a detailed test, all of the 9 emitters' peak output hovered between a little bit above 5, and a hair above 5.5A. With this information in mind, it can be easily understood that driving it above 5A is absolutely unnecessary.

Best performance between 5.8 and 6 amps? That's what luminarium iaculator said in his own review, where he also “found” peak performance at 10A :???: for the KW CSLPM1.TG. In my experience, though, these claims are incorrect. You can see the djozz test. Also, I found the KW CSLPM1.TG (2mm², 3030, first units from led4power) to be better driven at just ≈6.5A, where it performed better than at 8A. However, this is my particular experience, and since there are no formal tests of the 2mm² versions (as far as I know), you can take this as you like.

I did tests of 4 different Osrams: HHL532, CSLNM1, CSLPM1, and CULPM1.

1mm² are peaking around 4.75A and 2mm² are best at about 6.75A. 3030 vs 4040 doesn’t honestly seem to make a real world difference.

I’m sure instant turn-on numbers max out at up to an amp or so higher, but this is not sustainable beyond 1 or 2 seconds. I take my readings at ~5sec, or basically right after they settle in. Continuing from there, they’ll gradually fall as the heatsink warms up. Bone cold, instant startup values are highly misleading and not something you should spec a flashlight to, but they will look impressive on paper.

Simon, why have all versions of 3X21A and the SBT90 version of 4X18A been removed from your store? 4X18A SBT is one of my favorite lights.

We talked about the possibility to made 3535 and 5050 size 3V Leds from Getian at ~1800k Cri 90 R9 >60,for me is enough.
Important would be negative Duv at very low current and blue peak as low as possible.

My M21e just arrived!

So far I like this a lot. My first ramping Convoy and I now understand the comments here about the odd speed, but I don't hate it.

The fit and finish is typical Convoy perfect. The style is nice and it's a little smaller than I thought it'd be, but not really small. Almost exact same size as Astrolux ft03 mini.

Can't wait for night. And I grabbed a 4000k xhp50 to throw in it this weekend to see the difference in throw!

Will report back once I've tried it out.

nice work Simon!

oh - Texas Shooter- my xhp70 does turn off all the way. But it does it Very Slowly. I thought it wasn't going to but it does.