Astrolux S41 measurements (Quad Nichia 219B, 18350/16340/18650)

EDIT 2016/3/6: verified the pwm frequency, which is 17.2 kHz on all modes except moonlight (8.6 kHz) and turbo (no pwm).

Astrolux S41 is a quad emitter hotrod light manufactured by Manker. It is equipped with either high CRI Nichia 219b or cool white Cree XP-G2 LEDs. My light is the Nichia variant purchased from The light is also available as a Manker E14.

Manufacturer specifications

  • Battery type: 1x 18350, 16340 or 18650 with an optional tube
  • LED: 4x Nichia 219B 5000 K (also available with 4x CREE XP-G2 6500 K)
  • Body material: Aluminum/Copper
  • Switch type: Reverse tailcap clicky
  • Waterproof: Yes, IPX8
  • Lens: TIR
  • Protection: Low voltage warning (2.9-3.1 V), reverse polarity
  • Modes: Selectable 4 or 7
  • Special modes: 10 Hz strobe, battery check, biking flasher
  • Mode memory: Selectable
  • Length: 80 mm
  • Width: 24 mm
  • Weight: 90 g

The S41 driver is the same as in the popular BLF A6/Astrolux S1.

Astrolux S41 with a 18650 and a 18350 battery.

The light accepts protected and unprotected cells. In addition to 18350s 18650s can be used with an option 18650 tube. There’s a double spring in the tailcap.

Please note: lumen measurements are only rough estimates
My diy 30 cm integrating styrofoam sphere has been calibrated using a Fenix E05 on high with manufacturer’s claim of 85 lumens. Verified with an Olight S10 that has been measured with a Labsphere FS2 integrating sphere by Results may be more inaccurate with especially throwy or floody lights.

For spectral information and CRI calculations I have an X-rite i1Pro spectrophotometer with HCFR for the plot and ArgyllCMS spotread.exe for the data. For runtime tests I use spotread.exe with a custom script and a i1Display Pro because it doesn’t require calibration every 30 minutes like the i1Pro.

Explanation of abbreviations
CCT = correlated color temperature, higher temperature means cooler (bluish)
CRI (Ra) = color rendering index consisting of 8 different colors (R1-R8), max value 100
CRI (R9) = color rendering index with deep red, usually difficult for led based light sources, max value 100
TLCI = television lighting consistency index, max value 100
CQS (Qa) = Proposed replacement for CRI, RMS average of 15 color samples
CRI2012 (Ra,2012) = Another proposed replacement for CRI, consists of 17 color samples
MCRI = Color rendering index based on the memory of colors or 9 familiar objects
x,y = coordinates on a CIE 1931 chart

After 35 seconds on turbo with a 18650 battery, my lights drops down to moonlight and doesn’t recover before switching it back off and on again. The backward mode cycling with a long press also stops working after overheating.

*Spectral distribution

Color rendering

Beam and tint

Tint in different brightness modes is consistent. Mode 1 = moonlight, Mode 7 = Turbo.


High 2 and Turbo

Medium 2 and High 1

Is this the nichia numbers? I would like to know the difference between xpg-2.

Very nice info, thanks for the review! :slight_smile:

A fan blowing on the light.

This is very interesting. A lot of work, I guess. Thank you for this useful information!
How did you measure PWM?

great job maukka!

I measure PWM with an oscilloscope and a photosensor. Since I didn’t have time to do it with the S41, I just inserted the Astrolux S1 numbers I did earlier in to the table. I’ll verify this soon.

The numbers seem rather low though, frequency should be 18 - 19kHz with this driver. The med2 value looks like a copy/paste error, according to Toykeepers code all modes except lowest one should run with fast pwm with the frequency mentioned above .

Excellent review


Excellent review with new/unique style.!

Ok, thanks for the heads up. I’ll double check the pwm.

Out of curiosity I checked the pwm frequency of a ATtiny13a sitting on my breadboard, and my old DMM says 15 kHz. Can’t use my oscilloscope for double check, it is even older, analog and totally de-calibrated. Attiny datasheet says the internal clock oscillator calibration has been done in factory - but not very accurate I guess …

maukka, two thumbs up for your detailed review!! :THUMBS-UP: :GRAD:
Thanks for that!

To get the specified clock tolerance you have to use the default internal oscillator (9.6 MHz for the tiny13). The calibration factor for the other oscillator is there, but you have to apply it manually in firmware. Look for the OSCCAL register.

Very nice review!

Great! I read through with interest.

edit: just wondering, did you see readings from initial flash lumens when power ON?
Over 1K lumens @30sec aint bad for this size of light anyway 8^)

Ah, that’s interesting. I just fused the ATtiny mentioned above to 9.6 Mhz and the pwm frequency reading is now very accurate. I guess I can’t read the calibration value from Attiny program code?

Thanks for the excellence review @maukka

Great review! Thanks.
I love this tiny high cri quad copper head!

Yes, the turbo lumens at power on are in one of the runtime graphs:

For the 18650, the numbers are 1364 lumens at 0 sec and 1175 lm at 30 sec.