I’m afraid I don’t have any resistors or anything I can use. The center contact on the charging port is recessed and it’s extremely unlikely that the contacts would be shorted.
Shorting them is not a problem, only low resistance like from dirty water, mud etc.
And you get or what? 3Ah batt running 1.2A led , minus 15% driver efficiency( T2 mode). Cri index do not effect runtime at all.
Very nice review! I never pulled the trigger to buy one (yet)
At the end your are comparing it to Sofirn SP40 (that was a flashlight I liked a lot, even with the missing modes… because of the ease of swapping emitters), you may try the newer Sofirn HS40 that is quite nice (but only available with SST40 emitter for now)
I’m glad you enjoyed the review! HS40 does look pretty interesting I hope Sofirn offers it with some different emitter options (presumably LH351D) soon.
I think I already measured this somewhere, if I recall correctly the short circuit protection triggers around 0.5Ω, above that the current passes. I just tested with 100Ω and the current is 40.4mA with a 4.15V cell.
I measured the efficiency of the driver here :
(Not counting springs)
Not sure what T2 mode is, I think I used the default modes.
Aside from that, good review, those H04 and H04 mini are pretty good, on the light side for 18650, most of the competition is heavier, some significantly so, modability is pretty good, altgough the glued tube (and so non interchangeable tube) is disappointing. The interface is okay in B mode, but the lockout when holding from off is annoying as I always expect the light to ramp up.
Regarding the clip on holder that people rave about, I find it uncomfortable after a not so long use duration, I much prefer the classic silicone style holder and I use a ZL one with my H04s.
Personally I think the H600 IV is better, I have no use for magnet nor integrated charging, it’s a bit lighter, better build, better efficiency, actual thermal regulation and especially higher sustained output.
I’m not sure if Grizzly’s sample has the same driver, high CRI LH351D should produce 790-900lm at 2,9A and 460-480 at 1,2A. Unless we’re taling about the least efficient optics on earth (25-30% light loss).
CRI affects runtime by decreasing LED efficiency.
With fixed power output ?
If it’s the same driver it should push the same 2,9A and 1,2A in two highest modes.
The runtime should actually be longer with LH351D 90CRI because its Vf is a bit lower than the XM-L2.
I have a recent H04, I could open it to check but I’m pretty sure they use the same driver as the older ones from which I made the measurements.
We are talking about H04 flashlight with fixed power output modes, right? So how cri will effect runtime? Output and quality of lumens - yes. But runtime…
1 Thank
CRI affects general efficiency and therefore - braightness/runtime ratio. That’s it, I didn’t say anything else, you’re too fixated on that part.
Thing is - based on brightness I’m not sure it’s fixed at the levels mentioned in other posts. What we have in the review is estimated brightness, runtime and battery capacity. In T2 mode output stabilizes at about 350 lumens, with 90% optics efficiency we’re talking about 390 LED lumens. 90CRI LH351D needs 950-1000mA to produce those 390lm. With 3000mAh battery we should see close to 3 hours of runtime (minus driver efficiency) meanwhile low voltage stepdown kicks in after just 2 hours. Something is very inefficient there and that’s my observation.
Because H04 have async. buck driver , not the boost or buck -boost. It cant output 1.2A from start to the end. T2 mode probably will stepdown at about 3.2V .
You’re right Quad, and maybe a little higher, I measured a dropout voltage of 0.26V at Iout= 1.2A but with the wires and springs it should be a bit more, according to Maukka’s test his sample of 5000K CRI90 had a Vf=2.97 at 1.2A, I think we can assume that it would start to drop out of regulation around 3.3V at the cell’s contacts, which according to HKJ’s test happens around 2400mAh for a new 30Q at 1A.
In grizzly’s test it starts to step down at ~2.1 hours. In my test the average input current on T2 is about 1.1A, so 1.1x2.1 = 2.31Ah. Slightly lower than the 2.4Ah we should get.
But I tested with 4×SST20, which have a lower Vf, in reality the average input current would be a bit higher with a single LH351D, and so we would get probably very close to the 2400mAh the 30Q is able to deliver until 3.3V.
Regarding the output though, not sure what’s up, against what light you calibrated your sphere tactical grizzly ?
Hmmmmm excuse me BUT Armytek wizard pro E21A > Skilhunt H04 RC 
Modded with E21A out of the box and for 35$ shipped?
1 Thank
I’m using a Texas Ace calibrated lumen tube. I take brightness measurements on each mode with a full battery. Then, I do runtime tests with the ceilingbounce app, and I add a multiplier in Google Sheets to make sure the turn-on lumen number in the runtime graph is the same as the turn-on lumen number I measured on the tube.
We really need a right angle light just like this but with Anduril (and while I’m making wishes, NO on-board charging)
Man, I hear that. I desperately want Hank to make a dual-channel headlamp. I’d be happy with right angle or a regular headlamp, but I want Anduril 2 on a headlamp, especially with dual-channel functionality like some of Hank’s latest lights.
Anduril have bad mode spacing for headlamps, because of linear stepped mode. Its possible to set up, but its not ideal . It works with 7 steps only and ramp ceiling about 1Amp.