Olight M21 (Luminus SST-50) Review: RUNTIMES, BEAMSHOTS and more!

selfbuilt

Flashaholic
Joined
May 27, 2006
Messages
7,005
Location
Canada
Warning: pic heavy, as usual. :whistle:

Manufacturer's specifications, condensed from Battery Junction's website:
  • Luminus SST-50 LED
  • Powered by: 2 x CR123A / 2 x RCR123A / 2 x 16340 / 1x 18650 / 1 x 17670 batteries
  • Dimension: Length: 145.5mm (5.7"), Bezel Diameter: 40.2mm (1.6"), Body Diameter: 24.80mm (0.98"), Weight: 119g(4.2oz), Color: Black
  • Tactical momentary-on forward switch, protruding switch for easy momentary or click to lock on
  • Three digitally controlled constant brightness levels and strobe. 20lm (30hrs)-80lm (8hrs)-500 lm (1.2hrs), Strobe(500lm/2.4h).
  • Auto memorization of specific function for instant access
  • Front removable stainless steel striking bezel and rear striking bezel, removed for daily use, installed for close quarter combat
  • Ergonomic design with solid structure for easy operation, anti-slip body texture that give you a firm tight grip
  • Full orange peel reflector for wide flawless beam
  • Strong / quick to release lanyard to prevent loss and easy access
  • IP68 Water resistant
  • Anti-shattering ultra clear lens, anti-scratching and anti-slip
  • Metal 'cigar hold' ring optimally positioned offer additional anti-roll properties and is pre-installed but removable for smoother body hold
  • Stainless steel pocket clip
  • Operation: For momentary light, gently press (no click) the tailcap push-button switch. For constant light, fully press (click) the tailcap pushbutton switch.
  • To switch functions, click on the light, loosen the head by 1/8 of a circle and then tighten it down, the light will move to the next function. Repeat the same operation for further functions.
  • MSRP $120
The M21 is the "sequel" to the highly popular Olight M20, reviewed over a year ago. How does it compare? Read on to find out … ;)

M21001.jpg

M21002.jpg

M21003.jpg


The M21 comes in a presentation-style plastic carrying case, similar to the M30. The light and all accessories are carefully packaged in cutout foam. Included with the light is a sturdy belt holder, spare O-rings, GITD tailcap button cover, good quality wrist strap, 2xCR123A battery carrier, tactical grip ring replacement cover, and manual.

M21004.jpg

M21005.jpg

M21012.jpg
DminiVX014.jpg

From left to right: Duracell CR123A, AW Protected 18650, D-mini VX Ultra with D65 extender, Olight M21, Olight M20, JetBeam Jet-IIIM, ThruNite Catapult (no extender)

Olight M21: Weight 119.5g, Length 145.2mm x Width 38.5mm (bezel)
Olight M20: Weight: 120.0g, Length 144.0mm x Width: 35.7mm (bezel)

Overall size and weight of the M21 is similar to the older M20, although the head is just a bit wider and longer on the M21. But unless you are comparing the two side-by-side, for all intents and purposes the hand feel is very much the same. This is a high quality light that is comfortable to hold and use.

M21010.jpg


Like the M20, all components seem sturdy and well built. Everything fits together well, with double o-rings and smooth screws threads at all openings (tailcap threads are anodized, allowing tailcap lockout). Tailcap spring is not exposed, but encased within an assembly with a button top. Note that the tailcap is interchangeable with my older M20 – nice to see they have kept the specs consistent for over a year! :thumbsup:.

There is a spring mounted on the positive contact board in the head, so recoil should not be a problem. And newer flat-top higher capacity 18650s should work fine. :)

The bore width is good, so all batteries should fit. A battery magazine is supplied to help prevent rattle with 2xCR123A. Note that the magazine is optional, but if nothing else, it provides a good storage means for backup cells.

Machining and anodizing of the light are top-notch on my sample. The raised checkered portions help with grip, though not as "grippy" as aggressive knurling. The attached anodized clip and grip ring also help enhance grip, and both are removable. Included in the kit is a smooth replacement cover for the grip ring (i.e. the grip ring screw threads would be exposed otherwise). Overall, I would have to say hand feel remains excellent in all configurations.

M21006.jpg


The light can't tailstand - but that's not uncommon for "tactical" lights, since tailstanding rings tend to obstruct easy access to the clicky switch. Light comes with a forward tactical clicky with good tactile feel (momentary on, click for lock-on).

M21013.jpg


The light features a scalloped stainless steel bezel ring, which could serve as a strike bezel. The Luminus SST-50 is clearly visible well centered in the reflector. :thumbsup: The reflector is fairly heavily textured, which should provide a very smooth beam. Given the larger die of the Luminus LED, I do not expect it to throw as well as the original M20 (i.e. harder to focus to a fine point).

And now for the requisite white wall hunting ;) … all lights are on Max with OP reflectors on AW 18650, about 0.5 meters from a white wall.

Note: the tint of the M21/M20 comparison pics is thrown off by the very green M20 - my M21 more closely matches the tint shown in the other light comparisons.

M21-Beam25-1.jpg

M21-Beam25-2.jpg

VX-Beam25-1.jpg


M21-Beam100-1.jpg

M21-Beam100-2.jpg

VX-Beam100-1.jpg


M21-Beam1600-1.jpg

M21-Beam1600-2.jpg

VX-Beam1600-1.jpg


The M20 has what I consider to be a very standard beam profile for a tactical light - moderate spillbeam, reasonably bright spill, strong hotspot with good throw. The M21 certainly carries on this tradition, with a slightly wider spillbeam and less focused hotspot, consistent with its emitter/reflector combination.

Although hard to see above, the M21 has a fairly premium cool white tint. Early M20s tended to have R2s with the greenish WH cool white tint (which is throwing off my camera in the first set of comparisons). My ThruNite Catapult and Lumapower Dmini VX Ultra (both SST-50-equipped) tend to be just slightly on the purplish-side of premium white. For those of you not familiar with tint bins, please see my Colour tint comparison and the summary LED tint charts found here.

Here are some close-ups at 5m to better show you the hotspots. Both lights on Max on 18650. The height of each image is ~1.25m.

M21-Beam5m-1.jpg

M21-Beam5m-2.jpg

M21-Beam5m-3.jpg


Pardon the spackle marks (this is from an unfinished part of my basement :laughing:). But as you can see, the M21's hotspot illuminates an area about twice as wide as the M20 at 5m. The width of the hotspot is about 1m wide for the M21, half-a-meter for the M20, at this distance.

User Interface

UI is identical to the M20, and very straight-forward - press the forward clicky for momentary on, click for lock-on.

Light moves between modes by a simple switch of the head - loose/tighten the bezel to move to the next state. Sequence is: Lo - Med - Hi - Strobe, repeated in an endless loop. Light features mode memory, and retains the last setting used (so you can always have it come on where you want).

There is no sign of pulse width modulation (PWM) on any output mode - I believe that the light is current controlled. :)

M21-Strobe.gif


Strobe is a fairly tactical (and certainly annoying) 9.5 Hz. :green:

Testing Method: All my output numbers are relative for my home-made light box setup, a la Quickbeam's flashlightreviews.com method. You can directly compare all my relative output values from different reviews - i.e. an output value of "10" in one graph is the same as "10" in another. All runtimes are done under a cooling fan, except for the extended run Lo/Min modes (i.e. >12 hours) which are done without cooling.

Throw values are the square-root of lux measurements taken at 1 meter from the lens, using a light meter.

Throw/Output Summary Chart:

18650-Summary1.gif


18650-Summary2.gif


18650-Summary3.gif


As expected, the Max overall output of the M21 is higher than the original M20. The low-level is also slightly higher, but not significantly so. As predicted from the beam shots, center beam throw is indeed reduced on the M21 - but still quite acceptable for a general purpose/tactical light.

Note that my lightbox seems to be underestimating the relative output of the M21 compared to the other lights (the ceiling bounce results suggest the M21 is actually ~40% brighter than the M20 on max).

Output/Runtime Comparison:

Note: Effective January 2010, all CR123A runtimes are now performed solely on Titanium Innovations batteries. You can compare the generally excellent performance of these CR123A cells relative to the Duracell/Surefire cells used in all my earlier reviews here. I have marked all the new runtimes of lights with Titanium Innovations CR123As on the graphs with an "*".

M21-Runtimes.gif


Taken in isolation, the runtimes on all batteries seem reasonable and consistent with what I would expect. Note that like the M20 (and many other multi-power lights), the M21 bypasses the protection circuit cutoff on 18650 batteries. You should stop any run on 18650 once the light begins to dim noticeably, to insure you don't over-discharge your batteries.

M21-Hi18650.gif
M21-Med18650.gif


M21-HiRCR.gif

M21-MedRCR.gif

M21-LoRCR.gif


M21-HiPrim.gif

M21-MedPrim.gif


The SST-50-equipped M21 is definitely brighter on Max than any of the single-die Cree lights, but the difference is not as high as you might expect (~25% according to my lightbox, ~40% according to ceiling bounce). The Lumapower Dmini VX Ultra and ThruNite Catapult (not shown) are both SST-50-based lights driven harder on Max – but at the expense of runtime and with greater heat generation. The M21 likely strikes a good balance for long-term stability.

Overall, runtimes seem pretty reasonable for the output levels - the M21 doesn't seem quite as efficient on Med as the original M20.

Potential Issues

I can't really think of anything significant. :) The M20 has proven itself a reliable and robust light, and given the balanced approach to managing output with the SST-50 version, I am sure the M21 will prove the same.

Preliminary Observations

I'll get right to the point – I think Olight has another winner on its hands. :twothumbs

The M21 is really a direct port of the Luminus SST-50 to the M20 design, with a slightly larger head and revised reflector design to accommodate the new emitter. Otherwise, the body parts are interchangeable, and even the relative output levels are not so different (more on that in a moment).

The M20 has a lot of "legs" in the flashlight world. I first reviewed it 15 months ago, yet it remains one of the better choices in the general purpose/tactical 2xCR123A market. Moreover, Olight has managed to keep the quality of construction and availability of extra options roughly constant over that time – which is something of a feat for this market.

The main advantages of the M21 over the M20 are due to the Luminus SST-50 emitter, and include greater output on Hi, smoother beam pattern, and (in my experience) lesser likelihood of extreme tints (especially green ones). Potential drawbacks to the M21 are the reduced throw (due to the larger emitter die), lower Med/Hi runtimes, and potentially greater heat.

Olight seems to have resolved the latter issue by not driving the M21 as hard as other Luminus SST-50-equipped lights (e.g. ThruNite Catapult, Lumapower Dmini VX Ultra). While you still get a significant output boost compared to the M20's Cree XR-E R2, in my testing the M21 is no more than ~40% brighter overall (and perhaps less). Those expecting something closer to twice as bright will have to look at other lights.

But this is not a bad thing – personally, I am a little concerned about how heavily driven some of the other SST-50 lights are. Only time will tell, but I suspect Olight has struck a good balance in maximizing the life of the LED (and consequently the light) with a reasonable set of output levels (and thus heat). :thumbsup:

At the end of the day, the choice between the M20 and the M21 comes down to the differences in beam pattern and output. I don't know if there are enough compelling reasons to "upgrade" your M20 to a M21 – that's your call - but I think new users will be happy with the build of either. :)
 
Last edited by a moderator:

jhc37013

Flashlight Enthusiast
Joined
Feb 17, 2009
Messages
3,268
Location
Tennessee
Very good review as always thankyou. I have seen and tested the Olight M20 Ti version and was wondering if you know of anything different about the Ti M20 and the M21. Everything points to them being the same inside. Thanks again your stuff is priceless.
 

DM51

Flashaholic
Joined
Oct 31, 2006
Messages
13,338
Location
Borg cube #51
Great work as always!

Olight have stuck with a good design and used the new emitter to improve the performance. I like your point that they haven't gone crazy and tried to wring the maximum possible output from it - that is a sensible approach. 2-3 years ago, no-one would have believed 500 lm from a light this size was possible.

Moving to the Reviews section...
 

HKJ

Flashaholic
Joined
Mar 26, 2008
Messages
9,715
Location
Copenhagen, Denmark
As always a nicely done review, but I have a few comments:

What are you showing on the oscilloscope trace, I would expect a flat line when you have no pwm (Not the strobe)?

The output matches my measurements on the M20 Ti (that was the predecessor to the M21). In this, rather small, light it is probably a good idea to run the SST50 at lower power levels, but in bigger lights I do not see any problems with running it at higher power.
 

sfca

Enlightened
Joined
Aug 10, 2009
Messages
572
Location
Westcoast
Quick question regarding perceived brightness:

If one was standing say 10ft out and within the hotspot but on either side - not direct-center, which flashlight would be perceived as brighter the M20 or M21?
 

ARA

Enlightened
Joined
Jun 24, 2009
Messages
210
:bow::bow::bow: Thanks mate for such a great review. Its always informative to read your reviews. appreciate your efforts. :)

cheers
 

selfbuilt

Flashaholic
Joined
May 27, 2006
Messages
7,005
Location
Canada
Thanks for the support everyone! :grouphug:

Olight have stuck with a good design and used the new emitter to improve the performance. I like your point that they haven't gone crazy and tried to wring the maximum possible output from it - that is a sensible approach. 2-3 years ago, no-one would have believed 500 lm from a light this size was possible.
In this, rather small, light it is probably a good idea to run the SST50 at lower power levels, but in bigger lights I do not see any problems with running it at higher power.
Yup, I think it is very sensible in this case. And it's true, the much larger ThruNite Catapult should be able to handle the higher power - it was the smaller D-mini VX Ultra I was thinking of in my comments (and other potential lights of this 2xCR123A-size class).

FYI, my D-mini VX Ultra review should be up by end of day tomorrow ... :whistle:

What are you showing on the oscilloscope trace, I would expect a flat line when you have no pwm (Not the strobe)?
Ah, that's likely just noise in my setup. :)

My oscilloscope is just a freeware piece of software that monitors my sound card's mic-in port ... and my high-tech probe is dollar-store microphone where I've stripped off the head and wired in a red LED instead. :laughing: I've also recently upgraded computers, and am just using the built-in sound features instead of dedicated card (used to run an SB Audigy), hence why things have likely gotten a bit "noisier".

And I enjoy your reviews too - keep 'em coming! :)

Quick question regarding perceived brightness:
If one was standing say 10ft out and within the hotspot but on either side - not direct-center, which flashlight would be perceived as brighter the M20 or M21?
If you mean around the edge of the hotspot (i.e. the corona), then no question - the M21 is brighter. Its hotspot is much broader, and illuminates an area about twice as wide by 10ft out.

Just tested it with a lightmeter (since tint differences make it hard to judge by eye), and the difference is clear. Even if I move more toward the end of the respective hotspots, the M21 maintains an output advantage at all areas just outside the immediate centre-beam hotspot (where the M20 is of course brighter). I'll see if I can take some pics once it gets dark ... :whistle:

UPDATE: Here are some close-ups at 5m to better show you the hotspots. Both lights on Max on 18650. The height of each image is ~1.25m.

M21-Beam5m-1.jpg

M21-Beam5m-2.jpg

M21-Beam5m-3.jpg


Pardon the spackle marks (this is from an unfinished part of my basement :laughing:). But as you can see, the M21's hotspot illuminates an area about twice as wide as the M20 at 5m. The width of the hotspot is about 1m wide for the M21, half-a-meter for the M20, at this distance.
 
Last edited:

Geban

Newly Enlightened
Joined
Sep 17, 2009
Messages
39
Location
Stockholm Sweden
Thank´s for a great review!:clap:
Just got my M21, and I haven´t had the opportunity to try it out really yet.
But what I have tryed indoors, it seems to be a winner. :D
 

Olef

Newly Enlightened
Joined
Dec 12, 2009
Messages
86
Location
England
Nice one Selfbuilt, excellent review as always. I've been waiting for this one!

One thing puzzles me. I know manufacturers routinely overstate runtimes etc, but Olight state -

"Functions: Three brightness levels plus strobe, 20lm (30hrs)-80lm (8hrs)-500 lm (1.2hrs), Strobe(500lm/2.4h)"

Comparing Olight's stated medium 80lm for 8 hrs to your runtime graphs there is a huge difference. You got 2h34m on med with AW RCR's and 5h34m with 18650 neither of which comes close to Olight's claims. Furthermore, can Olight really be that cynical to quote maximum brightness figures on one hand with the optimum battery config of 2 x RCR and yet in the same sentence claim runtimes that have to be with 18650, giving much less brightness, and even then nearly double what you have measured?

Have you any idea what is going on here? Have Olight really overstated this light so badly or could the unit you have be flawed in some way? :thinking:

Olef
 

selfbuilt

Flashaholic
Joined
May 27, 2006
Messages
7,005
Location
Canada
Comparing Olight's stated medium 80lm for 8 hrs to your runtime graphs there is a huge difference. You got 2h34m on med with AW RCR's and 5h34m with 18650 neither of which comes close to Olight's claims. Furthermore, can Olight really be that cynical to quote maximum brightness figures on one hand with the optimum battery config of 2 x RCR and yet in the same sentence claim runtimes that have to be with 18650, giving much less brightness, and even then nearly double what you have measured?
I'd say it's a pretty sure bet they are using 18650 for their runtime estimates, and likely newer high-capacity 2600-30000mAh ones. I still use the older 2200mAh AWs, to allow backward comparison to my earlier data. With higher capacity 18650s, and taking into account variation in emitters, circuits, etc., those runtime estimates seem quite believable.

Not so sure about the actual lumen estimates ... although I know my lightbox is hardly linear above Med settings, I don't think Hi is really more than 6 times brighter than Med. Just a guess, but I'm thinking that 500 lumen estimate is a bit of an overestimate - something in 350-400 range (tops) is more likely. Similarly, the 80 lumen Med also seems like an underestimate (probably more like something in the 100-120 lumen range). The 20 lumen low sounds about right, but might actually be even a bit lower (which is why I haven't done runtimes - should last a long while). But someone with a calibrated light sphere would have to make those determinations - I'm just guestimating here. ;)
 
Last edited:

Olef

Newly Enlightened
Joined
Dec 12, 2009
Messages
86
Location
England
I'm just guestimating here. ;)

And doing a superb job too, thank you. I look forward to every one of your reviews :goodjob:

I was just surprised by the disparity in the runtimes you achieved and those that Olight publish, and I guess a little disappointed in both Olight and the M21. That aside, I like my M21 a lot. It's well engineered, has a no-nonsense UI and chucks out a amazing amount of light for its size - plus its floody nature makes it so very useful for close / mid-range.

It's still going to be my bedside light, but now I might invest in extra AW cells so I can keep it running :)

Olef
 

MattK

Flashlight Enthusiast
Joined
Dec 30, 2004
Messages
3,027
Location
Connecticut Shoreline
Great job as always!

Olef - most factories use the highest capacity 18650's for runtime tests - 2600-3000mah is common. As selfbuilt points out he uses 2200mah in his tests so I'm not sure why you are disasspointed in Olight. Also, the factories often use unprotected batteries when doing these tests (Japanese 2800+mah 18650;s aren't tey available in protected form).

Wouldn't you be using some of the highest capacity products to test your products to state the best possible runtime?
 
Last edited by a moderator:

Geban

Newly Enlightened
Joined
Sep 17, 2009
Messages
39
Location
Stockholm Sweden
Yesterday I tryed my new M21 at work for the first time, and it´s quite impressive.

Not the best thrower, but a whole lot of light. :D
 
Last edited:

Olef

Newly Enlightened
Joined
Dec 12, 2009
Messages
86
Location
England
Olef - most factories use the highest capacity 18650's for runtime tests - 2600-3000mah is common. As selfbuilt points out he uses 2200mah in his tests so I'm not sure why you are disasspointed in Olight. Also, the factories often use unprotected batteries when doing these tests (Japanese 2800+mah 18650;s aren't tey available in protected form).

Wouldn't you be using some of the highest capacity products to test your products to state the best possible runtime?

I certainly would try to show my product in the best light (lol) possible. I just find the stated medium of 8 hrs a bit too overstated based on Selfbuilt's excellent review. 5h20 on 18650 and 2h34 on RCR's are both a long way off 8hrs. That, to me, is a little like paying out for a 150mph sports car and finding under real world conditions it cannot even make the ton!

Especially as AW 18650's cost the equivalent of over $19 a pop in the UK :sigh:

Olef
 

MattK

Flashlight Enthusiast
Joined
Dec 30, 2004
Messages
3,027
Location
Connecticut Shoreline
Well the key is that selfbuilt is using rather old 18650's - most everyone is using 2600mah + cells these days. Selfbuilt, as noted above has only stayed with the 2200's to lend his review runtimes consistency so it's not fair to fault the factory for that.
 
Last edited by a moderator:
Top