At 10 Amps you would lose 0.15 volts here, with some unavoidable contact resistance even more.
@sharpie
post a pic of your DMM and leads please
FWIW, one can measure lumens with a cheap lux meter and a white box and a bit of time to assemble and calibrate. It’s usually easier than reliably measuring amps.
Yep, just checked mine (Nichia S41 from the second batch) and am getting 8.5A-9A with an Efest 18350 and 10.5A with a 30Q 18650 (batteries fresh of the charger).
Measurements taken with a quality DMM and short, heavy-gauge, solid-copper leads.
Maybe not the best test equipment, but it has proven to be spot-on with other factory lights/drivers (measures right @ 3.04A with an 8x7135 Qlite, etc.).
while you wrote i was in the mancave...
I don’t have an accurate way to measure amps over about 2 or 3, but if I do a math-based estimation, here’s what I get:
- 1720 OTF lm measured on an Efest 18350 cell
- 1720 / 4 = 430 OTF lm per emitter
- Add ~10% to cover optics losses: 473 emitter lumens
- Check djozz’s 219B measurements
- There’s a data point at about ~473 lm and 2.4 amps
- 2.4A * 4 = 9.6 amps
Or, without the 10% optics boost, it looks like it’d come out at about 8.8 amps.
So, I’d expect a measurement to be somewhere near that. It’s reassuring to hear that others have measured 8.5A to 9.0A with the same light and battery.
It’s not a direct measurement so don’t put much stock into it. However, I’ve found in the past that similar estimations tend to come out very close to reality, thanks to the high-quality data djozz provides.
Edit: I haven’t attempted a measurement at 30 seconds. With this much power in such a small light with a small battery, I doubt a 30-second turbo measurement would mean much. The output curve almost certainly has a steady decline all the way from zero to when the cell runs dry, and it shouldn’t really stabilize anywhere. That’s the nature of direct-drive hot rods, especially with the heat and power limitations of something so small.
Does anyone know of a magnetic ring that fits on these?
For what purpose?
You could put a ring-shaped magnet around it, or maybe try to fit one into the tailcap somewhere, but neither one seems like it’d be very useful for attaching it to metal surfaces.
Magnetic rings are sometimes used for brightness selection, like on the V11R and RRT01 and Darth, but the S41 has no electronics inside to use that as an input mechanism.
One could also add a grip ring, but those typically aren’t magnetic.
I think perhaps I just don’t understand the question.
Something like what is available for the convoys. Just a simple magnet that goes on the tail cap so you can stick the light to things and work with both hands.
I suspect he is looking for a magnetic ring to go on the end so the flashlight can be held by a magnetic surface.
Bob
Ah, that makes sense. Those look pretty nice. 
I’ve never seen one wedged around the button like that before.
It looks like that area is about 18mm outer and 14mm inner on the S41. I haven’t found any of that size online though.
Yeah, I miss having a magnetic tail after having one on my olight s2 baton. I find that I still carry it a lot just because of the magnet even though I have lights that outperform it in every other way.
Followed your debate and out of pure curiosity topped off a purple Efest 18350 and a 30Q.
On my Uni-T UT210E the results were:
1/ BLF SE A6 XP-L dedomed on 30Q = 5.64A
2/ Astrolux S41 4*XP-G2 on 30Q = 11.02A (hot!!!)
3/ Astrolux S41 4*XP-G2 on Efest = 7.60A
The S2 is a pretty nice light, even if it may not be the brightest. I use an Olight S1 quite a bit.
For the times I need hands-free light, I generally carry a Zebralight headlamp. Or I have a DQG Fairy on a necklace chain that I can hold in my mouth for quick tasks. But it’s uncommon that I have a metal surface to attach a light to, so I haven’t used the Olight magnet much.
It’s unfortunate that magnets are rare on tail-clicky lights, and hard to find in the right sizes.
I have a few lights with magnets in the tail. First thing I did when I got them was remove the magnet.
I carry my lights loose in my pocket. I don’t want to risk accidentally sticking the light in the pocket with my wallet and having it demagnetize my credit cards.
Also, a strong magnet makes it harder to get the light out of my pocket as it will stick to my keys or pen.
If you’d be OK with a magnet on the outside of the tailcap, maybe shrinkwrapped to it, ’oogle
segment magnet
arc magnet
— there are lots of partial-cylinder shapes made to go around things.
You’ll find them if you take apart some (DC) electric motors.
Try an image search under those results
Please use correct calculations for your discussions: 9.4A divided by 4 gives 2.35A perled. In the graph you linked that corresponds to just under 520 lumen (times 4 is 2080), count 10% loss for the optic, say 5% for the glass lens and 2% for the edge of the optic that falls under the rim and you’re left with 1700 lumen OTF in ideal conditions. This is right after switch-on, expect some loss by heating up as well.
I’m sure you can find something here. Some can even withstand heat. I found this place in one of Hoops posts.
k&J Magnetics
Consider that the difference at turn-on between the Nichia S41 and the A6 is maybe 400 lumens (1700 v. 1300). Due to the way the eyes work you need approximately a 20% increase to notice any difference in brightness at all. The 400 lumen difference barely above that threshold. Result is that even with a ceiling bounce test the slight increase in lumens will be barely noticeable.
Now compare the hotspot intensity between the 2 lights. The A6 has a single reflector. It produces a comparatively narrow, but intense hotspot with considerable throw. In contrast, the S41’s hotspot is much wider… but also much dimmer. If you compare just the intensity of the hotspots, the A6 will look much, MUCH brighter than the S41.
Why this difference? Keep in mind that the A6 is using a fairly modern emitter heavily driven with a single reflector. In contrast, the Nichia S41 is using a fairly old and low output emitter. That’s not to stay the Nichia 219B is a bad emitter. It’s actually quite good… but it’s claim to fame is high CRI, not sheer output.
A direct drive pocket rocket like this is always going to have a short battery life. FET drivers tend to be about maximum output, not long runtimes.
Regarding heating up: I see you have also noticed that the copper pill gets hot really really fast. This doesn’t actually having anything to do with the light being inefficient or not. Copper conducts heat really well and the S41’s pill has an uninterrupted direct thermal path to the back of the star.
In many other lights, the LED is mounted in a brass or aluminum pill which then touches the sides of the aluminum body. There isn’t a direct unbroken thermal path from the star to the outside of the light, or if there is, it is made of aluminum rather than copper. In the S41 it feels like the outside of the pill is VERY hot… and it is. However, it’s hot because it’s nearly the same temperature as the back of the star and the driver. Too hot to touch? yes. But the LEDs are actually probably quite a bit cooler than in other lights that don’t have such a large chunk of copper.
In my modded E14 with H17F driver, I have the thermal stepdown’s temperature sensor set to step the output down at the absolute lowest possible temperature (45 degrees C). The head still gets too hot to touch. It’s almost like they need a plastic cage around the copper head to prevent fingers from accidentally touching that part of the light until it cools to a safe level.
So far everyone else has reported around 8-9 amps with the Nichia S41. You appear to be the only one reporting 6 amps. It may well be you got a dud, or there is an issue with your measuring setup.
Yes, I admit I much prefer the 4500K high CRI 219Bs to these 5000K ones. And even at 5000K, I actually prefer the tint of some other emitters (XML2 3D). I prefer a slightly rosy or brown tint with no hint of green.
I’m pretty sure “direct drive quad turbo” and “sustained” are mutually exclusive, especially on a small host with a small battery.
The turbo mode may not be very practical, with a runtime of only 4 minutes at 9 amps, but it makes a great stun burst mode.