To check paper money on the go, I would choose a Jetbeaum UV AAA.
It works great, is true 365nm and can even be modded with a ZWB2 glass to get rid of the visible light.
the Nitecore has a 3W UV LED
or better 3.4-3.8V and 700mA, maybe a little overdriven
so expect 600-800mW 365nm radiant flux from it
to get 3000mW radiant flux you need a 10W LED which costs like 50-60$ a piece
just do a simple tailcap current measurement or compare the produced heat with a 10W white flashlight
There was some thread I (thought I) was following a while back about a BLF build of a UV light with 2 or 3 different wavelengths. Does that ring a bell for anyone?
X3, where did you get that 10mm ZWB2 filter?
I just found 20mm ones at affordable price.
Anybody seen this Jaxman:
JAXMAN E3 365nm UV
This would be ideal form factor, even filter is already included. Host looks exactly like Jaxman E3/Thorfire TG06.
I agree UV Convoy S2+ lights up everything pretty good but for money checking it’s an overkill imho - but OTOH I do not have the possibility to check $100 bill 
Some weeks ago I got this light from ebay:
Led UV Flashlight 365nm Purple Ultra Violet Light Torch Blacklight Lamp AA
Can’t say if this is really 365nm - maybe or maybe in the 385nm range. Light looks very different compared to convoy - it gives off strange greenish-white visible light. First time I turned it on I thought it was a scam and they put some very weak and off-color chinese white LED. But was pleasantly surprised that it actually produces a lot of UV light and is able to fluoresce banknotes and other things easily (again - have no access to $100 bill to check). Alas, mine came without any orings, but the housing is surprisingly sturdy regarding the price.
true 365nm is what the US 100 redline is all about
fwiw, here are the US paper notes colors, from a 375nm UV that is having difficulty with the red line on the $100
Sounds like the AAA Jetbeam has a lot going for it. Thanks!
I have the MecArmy SG3 which has, I believe, a genuine 356nm UV mode. My newish Euro/ UK passport (soon, god help us, to be UK only) lights up like a Christmas tree - some very weird and wonderful markings.
The thing is, I’d like the Convoy S2+ UV and filter, but can’t work out how best to get this from Simon using Paypal. I’ve never used Alibaba before, but do remember seeing something about ordering from him direct. Any suggestions welcome.
Ah, found it. It was Discussion: Group Buy for Custom BLF Triple Emitter Flashlight. LG 365nm, LG 385nm and Nichia 90CRI NVSW219BT-V1 R9050 D220, but looks like it never got off the ground.
If anyone can help figure out the UV output level for the SG3 and Jet UV, we could begin to rank them on output
true 365nm UV
Convoy S2 UV 18650, 3000mw input, 1000mw output
Nitecore Cu6 18650, 3000mw input, ?mw output
Jetbeam Jet UV AAA ??mw input, ??mw output
Jetbeam UV AA ??mw input, ??mw output
MecArmy SGN3 ??mw input, ??mw output
not true 365nm UV
Nitecore Tube
Nitecore MH27 UV
As Convoy’s official unofficial spokesperson I’d like to hear any input on the S2+UV that needs adjusting.
I know very little on the UV topic but Simon (owner of Convoy) is a friend of mine so if something needs changing, please talk to me.
Is the Nichia 365nm in the S2+ being over driven?
I don’t have one personally but obviously many posting in this thread do so please speak up and I’ll do what I can to make things better.
@Woody
Send me a PM and I’ll help you out with the ordering details. Ordering from Ali freaked me out at first too but it’s not as scary as it seems and you don’t really even have to use their system at all if you want to go the PayPal route.
User Lexel had problems with damage to the led (possibly from overheating), perhaps you should PM him to know more. The uv led on the convoy is rated for 500mA while it’s running at 700mA. Usually that’s not a problem but that led appears to be quite heat sensitive.
the Convoy S2+ UV is 40% overdriven
this LED has no thermal center pad
it is soldered on a normal very thin aluminum star
my Nichia LED is permanently damaged and dying slowly now
even soldering it on Djozz copper board thats directly soldered to the pill was not enough cooling for long time operation
just do a 19 Ohms 1W resistor parallel to the LED to eat 200mA
The Nichia 276A that is used in the Convoy is specified at 700mA continuous, and that is what it is driven at in the Convoy. But the quality of the MCPCB used in the Convoy is very poor; the two solder pads are very wide apart so that straight under the led is an air-gap. Some people use the custom two-pad board that I have had made, that has the solder pads much closer to each other and should be quite optimal for the 276A, but even then people find that the led is not happy. The 276A apparently is a very sensitive emitter.
yeah I think I did damage mine on the stock LED board, its degrading more and more
I used that light often >15 minutes continous cleaning bathroom and kitchen to find the ugly stuff
the Nichia LED has 19.1K/W thermal resistance
max temperature is 85°C
so driven at 3.84V and 700mA = 2.7W electrical power
lets say this stock LED board and pill has 5K/W
running the light in my hand I can feel the heat, so about 40°C
specced 780mW UV light are only at Ts=25°C
but Ts=Body temperature 40°C+2.1W*5K/W=50.5°C
above that the UV radiance decreases a bit so say 600mW radiant flux
2.1W*19.1K/W=40.1K
Tj is then 50.5°C+40.1K=90.6°C
so we are already above its maximum rating of 85°C, probably the stock LED board is worse than 5K/W
lets compare to an overdriven overdriven LiteOn
I got a LiteOn with ZWB2 driven on 1050mA on a Noctigon star in a S6 with shelf
the LiteOn has 5.1K/W temperature resistance with its thermal pad in the middle
so 3.54V and 1050mA = 3.72W electrical power
about 900-1000mW radiant flux
head temperature 45°C after running some time
2.82W heat generated
the Noctigon DTP star with arctic silver 5 and a lot pressure on it as I twisted the bezel very hard because the ZWB2 is thicker than the stock glass
lets say the shelf and copper star adds 1K/W
so 45°C+2.82W*6.1K/W=62.2°C
The LED has a Tj of 90°C 5 more than the Nichia
Thanks for the info guys. I’ll talk to Simon. Do you think if it were on a better mcpcb it would be okay at 700mA continuous or should I recommend he both increase the quality star it’s sitting on and lower the current a bit? We’re using a good DTP copper star in some of the S2+ line-up now but that’s using XM-L2 and XPL-HI emitters. I assume the little Nichia UV emitter won’t work on that same star?
No, the Nichia UV leds have a two-pad lay-out without central thermal pad, same as the Nichia 119 leds, so DTP is not happening. I assume that 700mA is fine if the two pads are placed closed together, and it helps if the trace thickness is 2 oz (=0.07mm, instead of the usual 1 oz).
Perhaps 500mA is safer but then a 7135 chip based lineair driver can not be used. 350mA (1 chip) is a bit low I guess.
an easy solution is to have a resistor eat like 100mA and use a star with better layout and copper core
my diode is behind a ZWB2 filter but does not emit light on almost a third of its surface
in German TLF multiple people complain about damaged Diodes
Thank you djozz and Lexel. I’ll bring this up to Simon and pass on the details you’ve shared. Hopefully we’ll see an improved version in the near future. I’ve wondered about these UV lights for a while but don’t really have a need for one…not that lack of “need” has slowed down my flashlight collecting. My wife keeps asking “Do you really NEED another flashlight?”! 
Hand prints on the ceiling does sound fun though! :laughing:
If Simon is upgrading the UV S2+ anyway, a nice addition would be a ZWB2 filter as a lens (he already stocks those filters). The Nichia has only little visible light but still the UV-fluoresence is washed out a bit by it.
If it was only hand prints on the ceiling!, it is stains litterally everywhere, in all sorts of colours, and without the UV you see nothing, and usually the stains do not come off easily.