Astrolux S43S NW Test Results

This isn’t really a review it’s just some tests and measurements from the copper head S43S with Nichia 219C emitters.

The lumens and amps are at 20 seconds due to the rapid heat build up. Turn on results are a little bit higher, but no big deal.

Top of ramp was 1230 lumens and 8.3A

Turbo was 2000 lumens at 18.6A

That’s some serious amperage right there. I think most folks like Maukka and M4D M4X only measured about 15 amps on turbo.

This is using a button top Samsung 30Q 18650. Lumens are measured on a Texas_Ace Lumen Tube calibrated with one of Maukka’s calibration lights.

Amp draws were made with a thick wire and a UNI-T UT210E clamp meter.

Parasitic drain with switch lights on 106 microamps. With switch lights off I got 24 microamps.

Something no one else has mentioned is that the copper, at least on mine, had kind of a tacky feel. It’s normal when cool, but it feels tacky when hot. Maybe this is some type of coating to prevent the copper from tarnishing?

You can see it in the pics below. Yeah, I can scratch it off with my fingernail. I think the copper had a plastic cover attached during production to keep fingerprints off and the covers were removed before shipping and left a slight tacky film. No big deal.

One thing I noticed is that the older S41 and S42 tubes will only work on the S43 if you use a flat top battery. The S43 tube is a few mm longer. The shorter tubes prevent my button top 30Q from touching. I definitely prefer the knurled S42 tube so I may remove the button top on the 30Q so I can use it.

As mentioned by Tom E, the bezel does seem to block some of the light output. With the bezel and AR coated glass lens removed I saw 1600/2700 lumen (at turn on). With only the bezel on 1260/2100 lumen (at turn on).

This is the fasted heating light I own. Turbo usage is very limited due to the heat. I do like it, though. The NarsilM v1.3 works great and the 219C color looks great. I even like the extra spill light compared to other TIR lens.

Yeah, this is a light that gets really hot, really fast. Definitely limits its use on turbo, and gets painful hot in about 30 seconds. I measured 22uA / 94uA for the standby drains. I didn’t measure current on max (no clamp meter), but your number of 18.6A sounds about right considering how quickly it heats up.

If the XPG3 version runs cooler, it might give it longer run times.

My only real complaint is that it over-corrects when it steps down (gets quite dim), and never steps back up again after cooling off. But, it’s a budget light, so you can’t expect too much in terms of temperature controls.

Yeah, I intentionally updated the Narsil manuals way back to say thermal stepdown, and not thermal control.

I measured a 7% saving on my bezel mods. I used a fixed mode of the max single 7135 to compare/test with. This was my first rev:

After this I sanded the top down further, took a little more off the inner bezel, then gave the inner edge a bevel. It saved a little more to get up to 7%.

This is how it looks now:

What do you think about the weight now that you have it ?

It’s definitely heavy for its size. That’s copper for you. I don’t have scales for accurate weighing, but it weighs about the same as the SP33 (including batteries).

Overall it’s not very efficient. The SP33 with a boost driver can get 2200 lumen at 6A while the S43S needs 18A to get 2000 lumen. That’s about 3 times the amperage for the same output. The SP33 is a much more practical general purpose light.

This flash gets so hot so quick that if it had a liquid cooling system it would need a second liquid cooling system to make sure the first liquid cooling system doesn’t boil off.

:laughing:

I wonder why it has such a high current draw. It seems strange to “only” get a couple thousand lumens from 18A current! :open_mouth:

The Nichia 219C is a pretty small led. You have to push it hard to get decent output. The harder you push it the less efficient it becomes. More and more of the input power gets converted to heat rather than output.

Plus it seems the bezel is robbing output. I get and extra 500 lumen on turbo without the bezel.

Usually the people that prefer the Nichia usually know it’s not going to be as bright as the xpg2 and accept it. Your trading that extra output for the nicer tint.

It’s unfortunate that the bezel blocks off such a significant amount of output. Thanks for the info. 18A is alot of current. My multi-meter only measures up to 2A :person_facepalming: Do you have a link to the best method/equipment to take current measurements for flashlights?

2 things:

1. Use a current shunt, like 10mOhms of 16AWG wire to know the specific voltage drop to calculate the current, on the tailcap. That’s my method, as I don’t need a current meter for now.

2. Use a hall effect clamp meter:

Yeah, but if you get 2000 lumens from four emitters, that’s only 500 lumens per emitter average. Even at 2600 lumens, it would only be 650 lumens per emitter. If it is using 18A to get there, that’s 4.5A per emitter, for an efficiency of less than 50 lumens per watt! That’s atrocious!

According to CREE PCT, the old XM-L in the LOWEST BIN, got an efficiency of 50.7 lumens per watt at 3 amps. So, this 219C is getting less than 50 lumens per watt at 4.5 amps, marginally better than the lowest bin XM-L of decades ago. :person_facepalming:

Jason, our measurements do match up. About 8 amps top of ramp, and 18-20A (on startup).
I don’t have a way of measuring lumens though…

weights without battery:
Astrolux S43

Astrolux S43S copperhead

Well, I mean, the 219C has an efficiency of 69,6lm/W at 4,5A.

If you count the inefficiencies of the circuit, like the dual phosphor bronze springs having higher resistance than a single BeCu spring for some reason, and the less efficient optics, and thinner traces of the driver, no wonder the S43S is less efficient than even the Emisar D4.

Don’t forget the 219Cs in Astrolux’s tests may be lower binned than Intl Outdoors’ LEDs, and what not.

Just got my S43S NW today. My Texas_Ace lumen tube measured the same 1200 top of ramp, 2000 turbo using keeppower 18350 cells. Drains the cell really, really quick :smiley:

I bought the copy cat light with scalloped tube.
Yes it gets quite hot rather quickly as can be seen here…

I’d recommend shaking your new lights and listening for a slight rattle in the head.

Mateminco is not a copy cat. It’s the same light from the same company. Like Honda and Acura, etc…

I recommend everyone with this light to reset the temperature stepdown. In stock form it gets too hot then steps down to about 60 - 90 lumen. That’s no good.

I reset mine to step down a bit sooner. Now it acts much nicer and takes multiple small step downs over time.

I did a video. The 30Q was not fully charged due to a few turbo runs the day before so turbo output is a bit low.

35 seconds at Turbo
23 seconds at 550lm
23 seconds at 285lm
23 seconds at 130lm
Then steady at 87lm

I’m not sure I understand correctly, do you mean if the thermal temperature threshold is set lower, the stepdown will be in smaller steps? Though would step down faster when in Turbo mode?

Yes, that’s correct.

I had to add some music to cover a background conversation. Sorry. There’s no dialog, so you can turn the sound off if you want.

Video on how to change thermal step down.