Astrolux S1 (BLF A6) measurements

UPDATE 31.1.2016: added color rendering graphs and new lumen measurements using a diy 30 cm integrating sphere

UPDATE 28.1.2016: updated the beamshot and spill cie graphs with more accurate measurements and differentiated between the hotspot corona and the outermost spill corona, faster sampling rate (2 sec) on the Turbo stepdown graph

UPDATE 27.1.2016: added info about runtimes and temperatures to the end

Astrolux S1 is better known as the BLF Manker A6 (Special Edition). The rebranded version is supposed to be identical to the A6 that was put together by BLF forum members. It has an advanced driver with extensive functions. Read more about the firmware and operation at http://toykeeper.net/torches/blf-a6/final/blf-a6.txt

The Astrolux S1 differs from the original in that it comes with a 18350 battery tube to shorten the light. The aluminum finish is excellect for the price (currently $30). A lanyard and a non-reversable pocket clip are included as well as two extra o-rings for the 18350 tube. There’s also a spare tailcap switch.

S1 weighs 76 grams with the 18650 tube and 55 grams in the 18350 form. Dimensions are 121/88,2 mm x 24 mm.


The light has a claimed output of 1400-1600 lumens in direct drive turbo. This can be achieved with a spring bypass and using high current 18650 batteries. I only managed about 1000 in the stock form with the batteries I had on hand (Eagtac 3500 mAh, Sanyo NCR18650BL, LG MH1). Turbo mode steps down to High 2 after 45 seconds.

The rubber switch is located in the tailcap. Deep press activates and shuts down the light and half presses (taps) switch modes. Longer half press cycles the modes backwards. In total there are 4 or 7 modes depending on how the user has programmed the light.

In group 1 the modes are Moonlight - Low - Medium 1 - Medium 2 - High 1 - High 2 - Turbo. In group 2: Low - Med - High - Turbo. Special modes (10 Hz strobe, battery check, biking flasher) can be accessed from the moonlight mode with a 0,5-1,0 second tap (cycle backwards). I accidentally named the lowest mode firefly and not moonlight in the images, I’ll fix it later.

The emitter is Cree XP-L with a choice of three tints: 1A (6500 K), 3D (5000 K), 5A (4000 K). The light I tested is the 3D variant.


The beam angle is 78° with a hotspot of 22°. Color temperature is noticeably cooler in the spill area and greenish around the hotspot. I wouldn’t call the beam beautiful tint wise, but the hotspot isn’t too bad.

There is PWM in all modes but turbo, but it is not visible even to the most sensitive people. The switching frequency is high and the output doesn’t drop to zero as is the case with traditional PWM. So no worries there.



Because of high frequency and narrow modulation PWM is invisible to the eye even in moonlight mode. Top: absolute (DC coupling, horizontal center line 0 lumens), bottom: zoomed in (AC coupling)

See all the scope screenshots here:


18650 battery, Astrolux S1 (18650 tube), Eagletac D25LC2, Zebralight H600Fd III


18650 battery, Astrolux S1 (18350 tube), Eagletac D25LC2, Zebralight H600Fd III

Measurements

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 valostore.fi. 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
x,y = coordinates on a CIE 1931 chart

Spectral distribution

Color rendering (Turbo)
See explanation here: CRI vs. alternatives with measurement examples

Tint

Different brightness modes

Tint towards the edge of the spill and corona.

Runtime (Turbo & High 1)


The light steps down to High 2 after 50 seconds, so I didn’t measure High 2 separately.

Zoomed in

First two minutes on turbo.

Temperature
The light gets quite hot in Turbo/High 2. When running without a fan I measured 54°C on the head at 5 minutes and 62°C at 10 minutes. Grip was 53°C at 10 minutes. I consider anything over 50°C uncomfortable to hold with bare hands for more than a couple of seconds.

After turning on a fan at 12 minutes, the head dropped under 50°C in 3 minutes while the grip temp had decreased down to 44°C.

Thermal imaging of the first 5 minutes

I am liking your tests. Thank you!

FYI....get a Samsung 30Q and bypass the spring and that light will be a different light

Really impressive review. Thanks for posting.

Good Day maukka,

Great review with Very useful measurements.

Thank you Very Much,

George

Thanks for another great set of numbers, on a well-known light :-)

maukka - I love these CRI and color temp tint charts you’ve been doing. I would like to see measurements on more neutral white lights and also cool white lights, just to see how they compare. Maybe you can compile the charts into some kind of database that people can look up.

Thanks for the informative review! Just got mine a few days ago too, and I’m very impressed. Looking forward to trying a spring bypass (it’ll be my first - might practice on a cheaper light first…)

Thanks for a very excellent review

I will compile all the measurements in to tables and graphs at some point.

Here’s the tint graphs for Eagletac D25LC2 Nichia and Zebralight H600Fd III that are neutral:

And a couple more lights in the same graph:

I understand most of it, but just what does D50 - D55 mean? - Duh… :slight_smile:

Love your posts! - Thank you!

-Chuck

They are simulated daylight illuminants. More info: Standard illuminant - Wikipedia

Thank you. - Looks like I have a lot to learn :wink:

-Chuck

Impressive testing!

Can you tell which instrument you used for color measurements?

Sorry, forgot to add information about the measurement devices used.

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.

Thanks, i1Display Pro seems to be affordable, found it at 200$.

I guess you just shine the light into its diffuser, from a distance?

Exactly. The i1Display Pro is a colorimeter, ie it has three light meters with filters inside. It is not as accurate as a true spectrophotometer but for output testing it is fine. Although if you don’t need a colorimeter for monitor calibration but just for runtime or lux measurements, I would probably just get a loggin lux meter like Extech HD450 that is comparable in price but easier to use (no 3rd party software required). But then again, with the right software the i1Dispay Pro can roughly estimate color temperature too.

Oh we just found BLF-selfbuilt :stuck_out_tongue:

Thanks for this review man.

Something for you, Djozz ? are you reading this?

More info on runtimes and temperature at the end of the first post.

Excellent testing info. Thanks.