Oh...search for HKJ's guide. His explaination is excellent, and also explains why measuring high currents at low voltage is never going to be accurate :
Vinz, that's a ridiculously awesome high-quality build you did. Great work. What driver did you use?
Vinz, Chicago X and Vinhgyuen54, if you have a lux meter, do try to put some ice on the part where it is between the front bezel and heat fins. You'd be surprised at the numbers you can get. There is actually live for XM-L between low 3 amps and 4.5 amps....as much as 25% with just ice alone. I spent the past 4 hours playing observing this. There has been testing done with blocks of aluminum but then that is room temperature. I am also getting this with my modded Sky Ray SR3800 as it is also direct drive now. (mine is revision 0, same as the DRY)
Ice is not the same as 0 deg C ambient even with airflow over it. (1m per second over a surface makes the surface area twice as large). The ice just cools soaks up the heat via latent heat of fusion. If you have dry ice, that is even better. Even driving the LEDs at such crazy levels with fully charged 4.2V batts, the backing of the LEDs MBPCB never gets warm.
That is to say if you live in a really cold climate like -20 deg C with pretty strong winds, congratulations. But you'd need to wrap up the batteries then to keep them warm.... LOL!
The converse is also true. Drive it at nearly 5 amps without cooling in a warm climate like mine, and you get to the point that you get slightly less lumens than driving at 2.8 amps. More amps means less lumens. I am not surprised, because the LED lumens calculator spreadsheet which someone has done up illustrates this too. But actually truth be told, i am surprised that this phenomenon is true and predicted correctly by the Excel spreadsheet.
Do note that the driver has nothing to do with this, and the batteries also nothing to do with this, i have 6 sets of good triple 18650s when playing with the SR3800.
Ric from cnquality announced that the DRY 4-mode has been back from 'revision' in the factory and should now have a current of 2,8-3A in HI. 2100 as you have a neutral white 4-mode and following the chart (3x850lumen = 2550lumen minus 15% for NW) that's should give around 2100-2200 lumen for this NW 4-mode, which is lot brighter than a skyray am I right? The 3-mode CW at 4,5A then should be a whopping 3000 lumen :o? But the 3-mode is not my thing as I'm too unexperienced, i like the 4-modes too much and I want to be able to use this light as an actual flashlight without having to worry i'll blow myself up.
This light beats the skyray at any point (size, triple batteries, 4-mode)
A last question: as you own both white and NW you agree that NW gives the best colors? CW probably has more of a 'shock' effect to the eye with a reaction: wow this is so bright instead of wow this is bright but i can still see the things I want to see well colored and in detail as CW tends to erase details. Giving a lot of light is one thing but seeing this as they are is another thing so the combination would be great!
Really appreciate your help from the last months with my search for a pocketrocket. I'm glad I waited and didn't buy the Skyray right away. I'm going to order the light (and batteries, charger and kd c8) next week i think when I'm back from a one week trip to the sea :).
The Dry is brighter simply because it has more cells (voltage) with similar current draw, meaning more power to the emitters. With NW, the beam is also more useful since it glares less.
The blowing up part is just a matter using the same batteries and checking that you're not mixing one of different state. Use the Xtar 2600 protected if you're that concerned not blowing yourself up.
You can play around with the spreadsheet here. Note that we are not factoring in driver losses, I am not sure what are the driver efficiencies between these and the SkyRay or Trustfire. Do remember, this is a 3-cell flashlight and fully charged is 12.6V and current drops while voltage drops, while the the other SR/TF are fully regulated and the current rises while the voltage drops so power is maintained overall. (P=I*V) https://budgetlightforum.com/t/-/2525
What I also do is to give additional info to support, I use a light meter to give get a direct “OTF” from the business end, which is probably what matters for everyone. Just take it that the DRY NW at 2 amps is around the same as SkyRay SR3800 CW (150 reflected lux vs 160 reflected lux). If you are talking about 3 amps and the new driver works, it should be a fairly straight line up for your usage in Belgium esp under somewhat breezy conditions as I just checked your weather as of now in August (hey you are always going out to sea, you a sailor/make a living from those delicious crabs in the North Sea or something? :D ), that’d be 240 lux. For info, I get about the same from my TK70, in the 240-250 range.
The Trustfire 3T6 is at 130 lux.
But one thing you need to note is that this is for newly charged batteries. For eg I have written how I Direct Drive my SkyRay SR3800. On my meter I am getting 4.6amps from 4.2V Solarforce v2 2400mAh cells unloaded, and 3.6amps from my XTAR 18700 2600mAh at 4.2V unloaded. If you look at the discharge curve of batteries, it drops from 4.2 to 4.0V quite fast. On the SR3800, it drops from 4.6A down to 3.7A when the voltage drops from 4.2 to 4.01V. (of course when loaded the voltage sags). After that fast drop, the discharge follows a much gentler curve from 3.9V down.
https://budgetlightforum.com/t/-/2325#comment-44727
The SkyRay 3800 when DD (ie no driver at all, pulls amps direct from 18650s) starts off at 310 lux but drops quite fast to 270 lux in less than 30s, and then slower from 26x to 250. If i pack the head with 2 ice cubes, i can maintain 290-300 for quite long, and any drop is due to batteries, not heat. That performance is brighter than a TK70.
The DD 3-mode driver should be very similar in performance. Basically the heads are the same.
So Direct Drive could be a good choice if you are looking for performance in a cold breezy environment where additional cooling would let you take advantage of that additional lumens available from 3 amps to 4 amps range, for that 1 -2 min or so. This is what I found out last night. It needs to be cold and windy enough so that it is pulling away enough heat. Neutral White is really preferably for outdoor use. But if you want the max brightness and you are shining on concrete buildings and stuff, then CW is a good option too.
The solution is very simple. Buy an additional driver from Ric, and then you can decide if you want 3 amps or direct drive.
I was thinking the exact same thing: the Dry might start out brighter, but with the other lights being regulated, the Dry might be less bright after a minute. Someone would have to plot the output curve to know for sure.
I also asked Ric how much a separate 3-mode driver is, in case I want more output than the 4-mode driver gives me. He told me $4.50 US.
So my bright idea is to get a Dry 4-mode (maybe...), and if I don't like it, I can swap the driver for a 3-mode, or wait and see if they improve the 4-mode driver or even introduce a 5-mode driver (4 steady outputs plus strobe--that's what I'd like best) sometime in the future. This host, and the NW LEDs, and the 3-cell side-by-side configuration, are good, so just changing the driver could give you almost a whole new light.
I could, based on maybe a resolution of 2 minutes usage and then take current and lux. But based on my experience with my SKy Ray SR3800 modded to DD, i don't think so. Probably way over the 40th minutes mark then the regulated lights would pull a lot of current to pull ahead of the DD lights. Remember this light has a energy density of 3 x 18650 vs the other's 2 x 18650, that's a 33.3% advantage already.
Seriously it is all about packing the most "punch" in the shortest length which also by coincidence results in a form factor that makes it feel more solid to hold.
Sorry, I was thinking of the NW 4-mode Dry only, since that's the one I'm interested in. :)
The NW 4-mode starts out less bright than a CW triple XM-L running DD, so it might not take that long for the NW 4-mode to drop below those regulated lights. In fact, 2100, if your NW 4-mode is a representative sample of all the NW 4-mode Dry lights, it starts out below the Skyray already, and is not much brighter than the Trustfire 3T6.
Ric is also sending me the new 4-mode 3A driver in addition to the 3-mode DD, this new 3-mode 3A driver should push out 2.8A-3A, better than the 2.0A I am getting, so i'd need to test it out. (Scheven_architect also mentioned that a few posts above this one).
As long as all the contents are packaged, it's coming fast via DHL and as long as it gets collected before 9pm, it should be able to catch the daily 2AM flight to SG via HK Air (DHL merger or something) and i will get it the next afternoon 3-6pm. :)
You need to note that if it is like 2.8A like my WW, it takes a heck lot of time to drop to say 2.2A (I estimate this to be the level where the SR3800 is just as bright), as i have mentioned this has a energy density of 3 x 18650. You definitely need to look at the battery discharge graphs in order to grasp it.
I am very interested to know the results of your test of the replacement 4-mode driver. Me being skeptical, I'm not yet convinced that the new 4-mode drivers are different from the one you got. Hopeful, but not convinced. :) As you mentioned, sometimes just differences in measuring equipment or technique make a big difference in the measurement. You measured 2 A, but Ric or the factory might have measured 2.8 A with the exact same light, using a different meter and leads.
But since you've done so many measurements (using the same equipment every time, I assume) on different lights, we can trust that if you read 2.8 - 3 A with the new driver, then the 4-mode driver has indeed been improved, and the NW 4-mode Dry should stay brighter than the (stock) SR3800 for a long time as you said.
When I see proof of that, I'll finally be able to click that "Check Out Now" button!
It occurred to me this evening that I used to spend $60-70 on a little 1xAA Fenix. The Dry is only $75 or $85, not including batteries. Seems like a good value. It's hard for me to spend $85 now though, because I've become accustomed to spending $20-30 since I discovered budget lights. Oh, and I also have a restriction: I'm not allowed to buy another light for a long time after this one, says my spouse, so I have to make the right choice!
Clearly wrong priorities. BLF > spouse.
Well, my spouse also told me not to buy already, but i still bought. Seriously of all the hobbies I have been into, this is really the cheapest.
I have a colleague who spends approx 2-3k a month of liquor and women....that's worse. Another Hifi acquaintance lost 0.8 million to a China woman as alimony compensation, worstest!
The local casino(s) at Resort World Sentosa/Marina Bay Sands by LVS just recently opened, lots of sad stories. Whenever i go there to take events/weddings @ the ballrooms, sometimes I see at the main entrance uncles/aunties applying medicated oil to their foreheads and noses, and consoled by friends I believe. Very sad....
Thing is, just buy something for the missus....it should be ok.
I also get a little nervous using multiple cells even if I'm being careful. However, the funny thing is that Tesla owners should be laughing at all of us then, because they're sitting on over 6,000 18650 cells 
And we worry about our measly 2 or 3 cells 
But seriously.. I still worry a little. haha
When I was chatting with Ric he listed this light first when asked what his favorite flashlights of recent are. He also listed a few Yezl lights along with the DQG II. Of course, he's pretty invested in the Yezl brand.
This man speaks from experience.
Can anyone tell me even approximate lumen outputs for the 4 modes high-med-low please?
It's real difficult to say for now, when i get the new 4-mode 3A driver circuit i will do some measurements.
I would like to do a preliminary report that at -15 deg C with a bit of breeze like 4-5kph (aka refrigerator), it is safe to use the light. Quite a few of you get this kind of conditions I guess.
I used the Sky Ray 3800 in DD 3-cell which is extremely similar to this DRY, with 3 x Solarforce 2400s at about 4.03V starting, and the ending current was 2.8A at about 3.92V cell voltage. 4 mins of gunning. The head was probably about 40 deg C at most, extremely cool running. Excellent heatsinking from LED MBPCB -> backplate to host, this time no ice. Since it is DD, no driver inefficiency, just a 6-lanes expressway to fun. :D
I think I need a better DMM. Maybe Uni-T UT71D. And I still can't find my $@#%@%$ Fluke 62 IR thermometer lost during the moving to my new house!!
I can tell you, even with HID background, this one is shi* bright with DD, provided you can cool the head. I am getting just 30% lower lumens than my 65W Ebay/Aliexpress HID. I use an gel-pack wrapped with rubber bands and blanket due to tropical night time temperatures and it was able to maintain the lux (though the place which i use it is always windy, usually 10kph and sometimes 20kph windspeeds).
All you really need is a wire with known (low) resistance. Get clip probe and measure voltage across it when connected through tail. The fundamental problem here is the resistance through the meter and resolving it by buying a meter of unknown resistance is a poor solution.