Zebralight SC52 (XM-L, 1xAA, 1x14500) Review: RUNTIMES, BEAMSHOTS, VIDEO and more!

selfbuilt

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

SC52004.jpg

SC52005.jpg


The SC52 is the latest refresh of the Zebralight line of 1xAA lights. I recall reviewing the very first Zebralight model ever produced – the 1xAA H50 (XR-E) – some five years ago now. :whistle: That was more of a headlamp model though, and was followed eventually by the traditional forward-facing SC50 (XP-E) and eventual SC51 (XP-G).

The new SC52 is a XM-L-based model, with a revised build and greater output. Let's see how it compares to the earlier models, and the competition.

Manufacturer Reported Specifications:
(note: as always, these are simply what the manufacturer provides – scroll down to see my actual testing results).

  • LED: Cree XM-L Cool White (Nominal CCT 6300K)
  • User Selectable Levels: 3 main levels (High, Medium and Low). Each main level can be programmed to one of its two sub-levels. The second sub-level of the each main levels can be further programmed to different brightness levels or strobes.
  • Light Output:
  • High: H1 280 Lm (0.9 hrs) or H2 172 Lm (1.7 hrs) / 108 Lm (3 hrs) / 4Hz Strobe
  • Medium: M1 50 Lm (7.5 hrs) or M2 25 Lm (12 hrs) / 12 Lm (27 hrs)
  • Low: L1 2.7 Lm (4 days) or L2 0.34 Lm (3 weeks) / 0.06 Lm (2 months) / 0.01 Lm (3 months)
  • Light output are ANSI out the front values. Runtime tests are done using Sanyo 2000mAh Eneloop AA batteries. Light output with 14500 batteries are the same except that the H1 is 500Lm for the first minute and then steps down to 280Lm.
  • Operating Voltage Range: 0.7V - 4.2V
  • Battery: One 1.5V AA (NiMH, lithium or alkaline) or 3.7V 14500 Li-ion battery. Batteries are not included in the package.
  • Parasitic Drain: Negligible (equivalent to 16 years, much less than the self discharging of a battery)
  • Automatic stepping down when battery (AA and 14500) capacity is low
  • Built-in over-discharging protection for 14500 batteries (2.8V cutoff)
  • Electronic soft-touch switch, with a 200,000 cycle operating life
  • Smart user interface provides fast and easy access to all brightness levels
  • Precision machined unibody casing from premium grade Alcoa aluminum bar stocks
  • Proprietary heat sinking design bonds the LED board directly to the unibody aluminum casing
  • Durable natural hard anodized finish (Type III Class I)
  • Tempered optical grade glass
  • Preinstalled bezel down pocket clip
  • Orange peel textured reflector
  • Anti-roll and tail-standing capable
  • Battery power can be locked out by slightly unscrewing the tailcap to prevent unwanted activations or parasitic drain
  • Beam Type: 80 degree spill, 12 degree hot spot
  • Dimensions: Diameter: 0.93 inch (23.6 mm), Length: 3.08 inch (78.2 mm)
  • Weight: 1.4 oz (40 gram)
  • Waterproof to IPX7 (2 meters, 30 minutes)
  • Accessories: two o-rings
  • MSRP: ~$64
Operation:
  • This light has 3 main levels (High, Medium, and Low). Each main level can be programmed to one of its two sub-levels. The second sub-level of each main levels can be further programmed to different brightness levels or strobes.
  • Basic Operation
  • Short click turns on the light to High or turns off the light.
  • Long click (press and hold for about 0.6 seconds) turns on the light to Low.
  • Advanced Operation and Configuration
  • Short click turns on the light to High. Short click again quickly to cycle from High to Medium, and Low.
  • Press and hold to cycle from Low, Medium and High, release to set. When press and hold, the light always cycle from Low to High regardless which level you are currently in.
  • Double click to toggle and select between the two sub-levels for that main level. Sub-level selections (except the strobe) for the 3 main levels are memorized after the light is turned off and through battery changes.
  • The second sub-level of each main levels can be configured after 6 double clicks. Double click (startng with the 7th) to cycle and select different brightness levels or strobes. Short click to turn off the light when finishing configurations. The selections for the second sub-levels are memorized after the light is turned off and through battery changes.
  • This light uses the main LED (flashing 1 to 4 times) to indicate the estimated remaining capacity of the battery. To start the battery indicator, (from Off) short-click 4 times without pause.
SC52003.jpg


The SC52 came in the standard "eco-friendly" packaging that has been standard on Zebralight for some time now. Included in the simple (but firm) cardboard box was the light with removable clip (attached, with screws), two extra o-rings, a one-page instruction sheet, and a notice about battery use.

SC52033.jpg

SC52023.jpg

SC52015.jpg

From left to right: Duracell AA NiMH; Zebralight SC52, SC51; Foursevens Mini AA, Quark AA; Sunwayman V11R+AA extender; Nitecore EA1, SENS AA, MT1A; Rofis JR-12.

All dimensions are given with no batteries installed:

Zebralight SC52: Weight 39.5g, Length 79.0mm, Width (bezel): 22.6mm, Width (max) 25.4mm
Zebralight SC51: Weight: 37.4g, Length 80.5mm, Width (bezel) 21.1mm, Width (max): 22.6mm
Nitecore MT1A: Weight: 54.6g, Length: 104.6mm, Width (bezel): 22.7mm
Nitecore SENS AA: Weight: 26.1g, Length: 82.7mm, Width (bezel): 19.8mm
Nitecore EZAA: Weight 20.9g, Length: 85.0mm, Width (bezel) 16.6mm
Rofis ER12: Wright: 35.5g, Length: 96.2mm, Width (bezel): 18.6mm
Tiablo E2A: Weight: 45.7g, Length: 101.2mm, Width (bezel): 19.9mm
Xeno E03:: Weight: 48.1g, Length 96.7mm, Width (bezel): 21.5mm
Xtar WK25B: Weight 42.9g, Length: 102.3mm, Width (bezel): 22.5mm

The SC52 is slightly shorter overall than the SC51 (which is impressive, given that it also has a deeper reflector and greater height clearance for longer 14500 cells, compared to the SC51).

SC52038.jpg

SC52011.jpg

SC52006.jpg

SC52010.jpg


The overall build design is similar to the earlier SC5x series lights. The most obvious changes are the slightly fatter head and the ridges along the body of the light. I definitely find the SC52 "grippier" than the predecessor SC51/SC50, which were a little too smooth in my view.

Anodizing remains the typical Zebralight "natural" finish (which is a bit more gray-green than most). Note that natural anodizing can be highly variable (e.g., my sample seems lighter colored than most Zebralights I've had).

Zebralight has finally started adding model labels to this series (previously, there were absolutely no identifying labels of any sort on SC50/SC51). The "Zebralight SC52" body label is applied to a a cut-out from the ridge detail, and is very legible.

As before, the SC52 still has a flat-bottomed tailcap (slight indentation now) and an electronic control switch in their head. The switch cover has a revised textured grip, and is more recessed than previously. It also has a firmer switch feel, requiring a more definite click to activate. Taken together, these features should help reduce accidental activations even further. :)

Tailcap threads are anodized as before, allowing for tailcap lockout. :thumbsup: This is important on all lights with electronic switches, since there is always some parasitic stand-by current drain to allow the switch to function (scroll down for a discussion in this case). Threads seem of comparable size and thickness to the earlier SC51.

The light can both tailstand and headstand.

The light comes with a removable metal pocket clip, held in place by two regular Phillips head screws.

SC52009.jpg

SC52018.jpg


As before, the light has a flat stainless steel bezel ring - but the head is wider now, with a wider and deeper reflector on the SC5. Reflector also seems to be less heavily textured than before (at least on my sample, which I would consider a light orange peel – LOP). The cool white XM-L emitter was well centered on my sample.

User Interface

The Zebralight SC-lights have always had a fairly unique user interface. While it may sound a little complex when first described, it is actually quite easy to use. The SC52's interface has been tweaked and expanded slightly from the SC51, but will feel familiar to those used to Zebralight.

On/off and mode switching is controlled by the electronic clicky switch.

Basic Operation

From Off, a quick-click (rapid press-release) of the switch turns the light on in Hi. If you repeatedly quick-click without much of a pause, the light will advance to Med, then Lo with another quick-click.

A slightly-sustained press-hold-release (hold greater than 0.6 secs) turns the light on in Lo. If you hold the switch longer, the light will advance through Med and then Hi, in a repeated Lo > Med > Hi sequence, until you let go of the switch to select the level you want.

At any time when On, press-and-hold to start the Lo > Med > Hi ramp.

A quick-click turns the light off.

Advanced operation

In regular usage, there are 6 possible output modes that are easily accessible, arranged in two sets of Lo/Med/Hi levels – the primary Lo/Med/Hi, and a sub-level set. Note that you can program the second sub-set (as explained in the section below).

To switch to one of the secondary Lo/Med/Hi sub-levels when on, double-quick-click the switch in any given level (e.g., in primary Med, double-click to go to the sub-level Med). The light will remember the selection of which Lo/Med/Hi level was last used, and return to that level the next time it is access (from Off, or as part of the ramp).

For convenience sake, Zebralight refers to the primary levels as L1, M1, and H1, and the secondary sub-levels as L2, M2, H2.

Programming the sub-levels

For the secondary sub-levels, you can actually choose from multiple choices. The primary levels are fixed.

For L2, you have the choice of three different levels. For M2, you have the choice of two different levels. For H2, you have the choice of two different levels or strobe mode (note that the strobe option can't be memorized as part of the standard operation). For the sake of clarity in my tables and graphs, I have referred to the programmable constant-output modes as L2A/L2B/L2C, M2A/M2B, and H2A/H2B. End result is that you actually have 10 defined constant outputs available to you, plus one strobe mode.

To program your desired secondary sub-level, double-click 6 times in any given level. Subsequent double-clicks will now alternate you between your various choices for that sub-level (e.g. M2A > M2B, in repeating sequence). To select the desired secondary sub-level, turn the light off by a single quick-click. When you next access that level, it will be saved to your programmed choice.

I realize the programming above may seem complicated, but you can safely ignore it all and simply use the light in basic mode as a 3-stage light. Or for that matter, as a 6-stage light in two sets of 3. :)

Video Overview:

For information on the light, including the build and user interface, please see my new video overview:



Video was recorded in 720p, but YouTube typically defaults to 360p. Once the video is running, you can click on the configuration settings icon and select the higher 480p to 720p options. You can also run full-screen.

PWM

There is no sign of PWM that I can see, at any output level – the SC52 is fully current-controlled, as claimed.

Note that the early SC51 had PWM on some of the lower levels (although I believe this was eventually replaced with current-control on that model as well). I am glad to see the SC52 is fully current-controlled right out of the gate. :)

Some users have reported seeing a brief flicker or pulse on the lowest possible output level (i.e. L2C, in my terminology). But I have not noticed one on mine – it seems perfectly stable, on all batteries.

Strobe:

SC52-Strobe.gif


The "hidden" strobe is a relatively slow 4 Hz (i.e., more of a signalling strobe than a tactical one).

Standby Drain

A standby current drain is inevitable on the SC52, due to the electronic switch in the head. Here is how the new SC52 compares to the earlier SC50 and SC51:

SC52: 129.1 uA on 14500, 20.2 uA on Eneloop NiMH
SC51: 41.8 uA on 14500, 14.2 uA on Eneloop NiMH
SC50: 7.4 uA on 14500, 2.3 uA on Eneloop NiMH

While there has been an upward trend in standby currents over time on the SC5x series, these values for the SC52 are still quite low in absolute terms. Assuming a standard 900mAh protected 14500, and a 2000mAh Eneloop NiMH, these currents would translate into 9.5 months on 14500, and 11.3 years on Eneloop.

Certainly, the drain is absolutely nothing to worry about on NiMH. The drain is higher 1x14500 (but still not unreasonable). But as always, I recommend you lock-out the light when not in use.

Flicker Issue on 14500

Some users have also reported flickering issues on Max on 14500, but this effect can be inconsistent. Note that the only difference in output between 14500 and NiMH/alkaline is the max mode (H1) runs at a higher level on 1x14500 initially (i.e., 500 lumens). After 1 min, the light steps down to typical max mode level for standard batteries (i.e., 280 lumens). The flickering, when observed in my case, has been on this first minute of 500 lumen output only.

There has been some speculation here as to whether this is due to different 14500 chemistries, voltage or heat. I can confirm that on my one sample, it is fact dependent on BOTH battery voltage and heat.

To explain, let me show you an oscilloscope trace for the first 65 seconds of runtime - on a fully charged AW protected ICR 14500 (~4.19V right off the charger), in my standard runtime testbed with a cooling fan.

SC52-FanCool.gif


Ok, not much to see here – the first spike shows the light On signal at time=1 sec, and the second spike is the step-down to the standard H1 level at t=60 secs.

Not let's try it again without any cooling applied:

SC52-NoCool.gif


Starting ~6 seconds after activation, the light begins flickering – at a rapidly accelerating rate. This rapid flickering continued until the light stepped down at 60 secs – at which point it ceased immediately.

How about a slower spinning fan?

SC52-SlowFanCool.gif


Ok, now you see the flickering doesn't start until much later (~20 secs in), and is much more infrequent (i.e., slowly accelerating). I increased the fan speed to my typical level after ~35 secs, and you can see the flickering stopped immediately.

On the basis of the above, you could reasonably conclude that the flickering is mediated solely by relative heat (i.e., with sufficient cooling, you can prevent it from occurring).

But here's where it gets interesting – if I re-run the test with a partially depleted cell (~4.09V according to my DMM), with no cooling applied, I get this:

SC52-409V-nocool.gif


:thinking: Ok, now all you see are a few flicker spikes starting at the very end the first 60 secs 500 lumen level (which again ceased immediately with step-down). Note again, there is NO cooling applied to run above.

Note also that the actual output above is no different from the earlier runs (i.e., the 4.09V battery is giving the same ~500 lumens as the 4.19V battery did). So this tells you that the effect cannot be due solely to heat (i.e., the heat in the situation above should have been the same as the earlier no cooling trace, given that the outputs were the same).

It therefore seems that there is some sort of circuit issue whereby the light responds to high heat only if the 14500 is near max capacity. On my sample at least, these two variables are intertwined – you can't say the light never flickers <4.1V (it can – but only mildly without cooling), and you can't say it always flickers >4.1V (depends on the level of cooling).

I'm not quite sure what to make of all this, and have queried it with Zebralight. They have responded to me that they are aware of the issue, and are working to resolve it. When I hear more, I will update this thread. In the meantime, if you are finding a similar issue on your SC52, I recommend you either drain some capacity off your freshly-charged 14500 cells, or contact your dealer about a replacement. :shrug:

Beamshots:

All lights are on Max output on Sanyo Eneloop AA NiMH. Lights are about ~0.75 meter from a white wall (with the camera ~1.25 meters back from the wall). Automatic white balance on the camera, to minimize tint differences.

SC52-Ene-Beam001.jpg
SC51-Beam001.jpg

V11R-AA-Beam001.jpg
E03-Beam001.jpg


SC52-Ene-Beam002.jpg
SC51-Beam002.jpg

V11R-AA-Beam002.jpg
E03-2-Beam002.jpg


SC52-Ene-Beam003.jpg
SC51-Beam003.jpg

V11R-AA-Beam003.jpg
E03-2-Beam003.jpg


SC52-Ene-Beam004.jpg
SC51-Beam004.jpg

V11R-AA-Beam004.jpg
E03-2-Beam004.jpg


1x14500 (AW Protected 14500) Li-ion Comparison:

SC52-14500-Beam001.jpg
V11R-14500-Beam001.jpg


SC52-14500-Beam002.jpg
V11R-14500-Beam002.jpg


SC52-14500-Beam003.jpg
V11R-14500-Beam003.jpg


SC52-14500-Beam004.jpg
V11R-14500-Beam004.jpg


Beam pattern is as you would expect for a XM-L light with this size reflector – a fairly big hotspot and decent sized spillbeam. Scroll down for full output details in my tables.

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 any extended run Lo/Min modes (i.e. >12 hours) which are done without cooling.

I have devised a method for converting my lightbox relative output values (ROV) to estimated Lumens. See my How to convert Selfbuilt's Lightbox values to Lumens thread for more info.

Throw/Output Summary Chart:

My summary tables are reported in a manner consistent with the ANSI FL-1 standard for flashlight testing. Please see ANSI/NEMA FL-1 Standard page for an explanation and a description of all the terms used in these tables. Effective July 2012, I have updated all my Peak Intensity/Beam Distance measures with a NIST-certified Extech EA31 lightmeter (orange highlights).

SC52-FL1-Summary1.gif


SC52-FL1-Summary2.gif


As you can see above, the SC52 has class-leading output – for both standard AA and 14500 Li-ion. The max output on 1xAA is particularly impressive. :eek:oo: Throw is reasonable, given the size of the reflector.

Since I don't have a lot of comparator data on other 1xAA XM-L lights, here are some summary tables for recent 2xAA and 1xCR123A/1xRCR lights I've recently tested (mainly XM-L-based). You can directly compare the output and throw values for the SC52 in the tables above to the tables below.

SC52-FL1-Summary3.gif


The SC52's ~290 estimated ANSI FL-1 lumens on max puts it a bit under most of the max outputs of the newer XP-G and XM-L 2xAA lights I've tested. That's to be expected, given that it is easier to have a sustained power draw from 2x AA cells in series than it is a single AA (i.e., you can't drive a 1xAA light as hard).

SC52-FL1-Summary4.gif


A similar pattern can be observed for most the 1xCR123A lights – typically, these have fairly similar max outputs to 2xAA lights.

SC52-FL1-Summary5.gif


On 1xRCR, the better comparable would be the ~540 estimated ANSI FL-1 lumens I get on the SC52 on 14500 Li-ion. That puts the SC52 in much the same category as many of the recent 1xRCR lights I've tested. This is as you would expect, as there is effectively no real difference between RCR and 14500, except perhaps greater runtime/capacity on some 14500s.

Lumen Estimate Summary

As with a number of my recent reviews, here is a summary table showing what my ANSI FL-1 lumen estimates are for the SC52:

SC52-Lumens.gif


Zebralight's reported ANSI FL-1 lumen specs seem pretty consistent with my output estimates (although my estimates seem to be a bit higher at the high end, and a bit lower at the low end). Part of this may simply be variability from one light sample to the next (e.g., it is hard for manufacturers to reliably reproduce the same sub-lumen levels in every light).

I know there's been some discussion here as to the accuracy of various lumen estimates. To expand on this in the case of the SC52, I have done some additional detailed comparisons to a Fenix light in post #2. :wave:

Output/Runtime Comparison:

Let's start with standard AA cells (NiMH Eneloop, alkaline, and Energizer L91 lithium):

SC52-HiEne.gif

SC52-MedEne-1.gif

SC52-LoEne-1.gif


SC52-HiL91-1.gif


SC52-HiAlka.gif

SC52-MedAlka.gif


Ok, there's a lot of data up there. :sweat:

The general take-home message is that the SC52 is a real output AND efficiency leader on standard cells. :eek:oo: A second point is that the SC52 is typically quite well regulated (i.e. flat-regulated) on standard cells, except on max output (especially on alkaline, as expected).

It is particularly impressive to see how long the SC52 runs on the Med-Hi levels, compared to equivalent output levels on other models. To help you compare to the Zebralight specs, here is a table comparing runtimes to the ANSI FL-1 standard of time to 10%:

SC52-Runtimes.gif


Like with the Zebralight lumen output specs, the ANSI FL-1 runtimes seem fairly consistent with my testing as well. :)

As always though, I recommend you actually look at the full runtime graphs for more info on output level changes over time. Given the range of possible regulation patterns, it is important not to get fooled by looking at just two (rather arbitrary) points from the ANSI FL-1 standard – the full graphs tell you infinitely more (as I explain on my Testing Methods page).

1x14500:

SC52-Hi14500.gif

SC52-Med14500-1.gif


On 14500 Li-ion, overall output/runtime efficiency of the SC52 is good, and seems pretty much in keeping with other lights in this class. However, the SC52 again shows very flat-regulation – which not all of the other lights were able to maintain.

As shown in my output summary table in the preceding section, Zebralight has also done an excellent job keeping output levels consistent between standard cells and Li-ions. The only real difference is on max (H1), where the SC52 shows a higher initial output for 1 min, before stepping down to the standard cell H1 level.

Normally that would be the end of my runtime comparisons, but I have been asked to show how the SC52 compares to other classes of lights. I provided summary tables for 2xAA, 1xCR123A and 1xRCR in the preceding section, and thought I would include 2xAA below (given the comparable battery chemistry).

2xAA Lights:
(NOTE that the SC52 is a 1xAA light, and all the others below are 2xAA!)


SC52-2AA-HiEne.gif

SC52-2AA-MedEne.gif


SC52-2AA-L91-1.gif


SC52-2AA-HiAlka.gif

SC52-2AA-MedAlka.gif


A couple of points strike me in this comparison of the 1xAA SC52 to the 2xAA lights. The first is that max output of the SC52 is not that much lower than the 2xAA lights – but standard cells get exhausted quickly at this level. You can really see this on the alkaline graph – there is simply no way that a 1xAA light is going to be anywhere near competitive to a 2xAA light on max drive levels on alkaline.

The second point is that the very high efficiency of the SC52 on its Med levels gives it the opportunity to compete fairly well with a lot of the 2xAA lights. No, it doesn't run quite as long for comparable output, but it is surprisingly close for a 1xAA light. Again though, you are best to run the light on NiMH or L91 lithiums to see the best performance here.

UPDATE FEBRUARY 1, 2013: Upon request, I've also plotted the max output of the SC52 against common XM-L-based 1xCR123A lights in my collection:

SC52-MaxCR123A_zps758cec05.gif


Please note that I have plotted *BOTH* L91 lithium and NiMH Eneloop for the SC52 above. In my testing, many 1xAA lights show slightly lower max output on L91 than they do on alkaline or NiMH. I'm not sure why (maybe to do with voltage?), but it is a common observation I've noticed over the years. :shrug: In the case of the SC52, I estimate ~290 ANSI FL-1 lumens on NiMH, and ~255 estimated lumens on L91.

I suggest you compare the full curves for how the lights perform, but in general terms, the SC52 is toward the middle of the 1xCR123A pack for initial activation output levels on max. Of course, regulation on lithium-based CR123A cells is typically flatter than NIMH.


Potential Issues

Zebralights all use an electronic switch, and therefore require a small stand-by current when fully connected. The standby drain on SC52 is low enough on standard cells as to be completely negligible (i.e., would take over a decade to drain an Eneloop). However, on 14500, the drain is bit higher – enough to fully drain a cell in under one year. Note that you can always physically lock out the light at the tailcap.

My SC52 sample had flickering issues at the "500 lumen" Max mode on 1x14500, with freshly-charged cells. As explained earlier in the review, this flickering disappeared <4.1V, and was highly dependent on the cooling state for Li-ion cells between ~4.1V and ~4.2V. I suspect there is some sample variability here, as only some users have reported similar issues near the 14500 max capacity level. Zebralight is aware of the problem, and working on a resolution - I will update this thread when I hear more.

The SC51 was known to have accidental activation issues (i.e., when I carried it as an EDC, I found that the SC51 could turn on if something pressed against the switch in my holster/pocket). I have been carrying the SC52 for better part of a week now, and have had absolutely no accidental activations yet. It thus seems like the revised switch design of the SC52 (i.e., smaller button, more recessed, firmer press need to activate) is effective at ameliorating this previous issue.

Preliminary Observations
The SC52 is a very impressive update to the SC5x-series of 1xAA lights. :party:

I was impressed with the XP-G-based SC51 when it was first launched – especially with its class-leading max output and its extremely efficient runtime performance and good regulation pattern. It did have a few limitations though – no 14500 support, a rather smooth finish, PWM on lower modes (at least originally), and an annoying tendency toward accidental activation when carried as an EDC (typically on max!). I'm happy to report the SC52 has improved in all the above areas. :)

Let's start with the output – the XM-L-based SC52 is again a new class leader on standard AA cells. Regulation pattern is excellent (i.e., perfectly flat on all levels except max), and output/runtime efficiency has reached an even new high for the class, at all levels tested. :eek:oo: I don't know how Zebralight has managed to squeeze so much runtime out of their current-control circuit, at all levels now.

The support for 14500 is appreciated, although there are a few quirks here. My sample showed a flickering issue initially on fully-charged 14500 during the ultra-Hi max output stage. This flickering was both voltage and temperature dependent, and was easily resolved by draining off a small amount of capacity from the freshly-charged cell first. Regulation on 14500 was excellent, with perfectly flat output at all levels (something a lot of other multi-power lights can't match). Overall efficiency was very good, in keeping with other good current-controlled lights (but just very good - not at the outstanding efficiency levels seen on standard AA batteries).

I'm also glad to see that Zebralight appears to be quite accurate in their specs – including ANSI FL-1 output and runtime measures, according to my testing (see post #2 for an additional discussion of lumen estimation).

Build-wise, I like the new "ribbed" body design, which improves grip. I also personally like the larger head – in fact, I'd like an even longer body (I have fairly longer hands and fingers, so can find overly small lights problematic for handling). The more recessed switch (with firmer feel) is also greatly appreciated – no accidental activations so far, when carried on my belt. This is what caused me to give up on EDCing the SC51, so I'm glad to see they have addressed it here on the SC52.

So far, I'm not coming up with any negatives for the SC52, compared to its predecessor SC51. Beam pattern is bit more "floody" on the XM-L-based SC52 compared to the XP-G-equipped SC51, but peak throw remains pretty similar on max (thanks to the extra output).

It is frankly amazing to see a 1xAA light than can – on standard batteries – nearly match the output of many 2xAA or 1xCR123A lights. Coupled with its outstanding efficiency and excellent regulation patterns the SC52 is a real competitor for these other classes, not to metion its own 1xAA class. :wave:

----

SC52 was supplied by Zebralight for review.
 
Last edited:

selfbuilt

Flashaholic
Joined
May 27, 2006
Messages
7,005
Location
Canada
I've noticed there has been a bit of discussion around here as to how accurate Zebralight's ANSI FL-1 output ratings are. As this is a separate issue from the actual review, I thought I would do some analysis in a second post.

As you can see from the summary table in my review above, my SC52 lumen estimates actually match the Zebralight specs fairly well. That said, I've also seen comments that my lumen estimate scale is inflated, compared to some sources. That's certainly possible, since my estimation method is based on a calibration of my lightbox to certain specific makers and testers (see my Lumen Estimation page for more information). If the lumen values for the group of lights used in that calibration were biased one way or other, than my estimation method would be similarly biased.

But my key point has always been that it doesn't matter for relative output measures, as those remain consistent across all my reviews (i.e., regardless of the absolute value of the calibration, the relative relationships always hold consistently). I also recommend you don't focus on one arbitrary output time point (i.e., ANSI FL-1 30 secs), but rather look at the full runtime graphs for how relative output changes over time.

In any case, I think that it is unlikely that there is a large systematic bias in the absolute values of my lumen estimates, given that I used what were generally agreed upon as reputable lumen sources for the calibration standard at the time (again, see the link above for details). In particular, Fenix was one of the sources used in my calibration, and their lumen measures were very consistent with the best-fit line of my own estimates (although this was based on their pre-ANSI FL-1 lumens). Note that I matched the time of lumen measurement as best I could, and gave priority to later time points where available (i.e., I only use initial activation maximums if that's all that was available for a given light). Looking at the recent results (including this review), there may be a slight inflation of my lumen conversion scale - but it's very hard to say without testing the lights in a proper integrating sphere. :shrug:

In this case, I have been asked to comment on how the SC52's lumen estimates match the Fenix LD20. I happen to have a LD20 XP-G R4, which was rated by Fenix as 205 lumens at the time. Let's start with my calibrated lightbox estimated lumens:

SC52 Hi1 = 280 Zebralight lumens = 290 Selfbuilt Lightbox estimated lumens
SC52 Hi2A = 172 Zebralight lumens = 190 Selfbuilt Lightbox estimated lumens
Fenix LD20-R4 Turbo = 205 Fenix lumens = 205 Selfbuilt Lightbox estimated lumens

EDIT: As jirik_cz points out below, this "205" from fenix was pre-ANSI FL-1. Unfortunately, I don't have any of their later LD20 lights, so don't how they would compare to this one. I would assume the LD20-R4 should have a slightly lower ANSI FL-1 lumen rating, but it's hard to know what that would have been. If we had true apples-to-apples ANSI FL-1 output lumen values, I suspect the LD20-R4 would probably be rated somewhere close to the SC52 Hi2A.


Note that I also do ceiling bounce tests for all my lights, to give you an independent measure from my light box. I have just re-measured ceiling bounce results for both lights at 30 secs post-activation, using my NIST-certified Lux lightmeter:

SC52 Hi1 = 290 Selfbuilt lumens = 55.5 lux ceiling bounce
SC52 Hi2A = 190 Selfbuilt lumens = 36.5 lux ceiling bounce
Fenix LD20-R4 Turbo = 205 Selfbuilt lumens = 38.5 lux ceiling bounce

Now, the ceiling bounce numbers don't help you with figuring out absolute lumen estimates – but they again correlate fairly well for the relative relationships. In the example above, the LD20-R4 is 5.5% brighter than the SC52 Hi2A by ceiling bounce (consistent with my lumen estimates which say it is ~8% brighter). Similarly, the SC52 Hi1 is 44% brighter than the LD20-R4 on ceiling bounce (consistent with my lumen estimates suggest it is ~41% brighter).

It's hard to demonstrate this visually, since our relative perceptions of output are not linear (see my discussion here of the power relationship that defines human visual output perception). Also, the differing beam profiles of the two lights introduce a huge confound to visual comparisons, as the LD20 has a narrow (and correspondingly brighter) spill, and a more "throwy" hotpsot (thanks to smaller XP-G emitter).

Beamshots are an extremely limited aid in this regard, but here's a simple comparison of output of these two lights under various camera exposure settings. To facilitate visual comparison, I have moved the lights closer to the wall (to diffuse out the beams somewhat), and adjusted depth so overall spillbeam diameter is roughly similar (so you don't get fooled by a wider or narrower spill).

SC52-Hi2A-LD20-Beam001.jpg

SC52-Hi2A-LD20-Beam002.jpg

SC52-Hi2A-LD20-Beam003.jpg

SC52-Hi2A-LD20-Beam004.jpg


SC52-Hi1-LD20-Beam001.jpg

SC52-Hi1-LD20-Beam002.jpg

SC52-Hi1-LD20-Beam003.jpg

SC52-Hi1-LD20-Beam004.jpg


Again, the lights are at different distances to the wall, to try and equalize the overall spillbeams to allow you to more easily compare overall output visually. This is distinct from all my other beamshots, which are a standard distance from the wall to allow you to compare different spillbeam widths. In any case, I think the above direct comparisons are consistent with what my lightbox and ceiling measures both report. :)

So, by all measures at my disposal, it would thus seem that Zebralight's lumen values (for the SC52 at these levels) are at least pretty comparable to Fenix's lumen values, as well as my own lumen estimates. At least it gives you a way to calibrate your expectations.

P.S.: Despite the above, I continue to recommend that people do NOT focus on max lumens estimates at the rather arbitrary time point of 30 secs post-activation (i.e. the ANSI FL-1 standard). It is far more important to pay attention to the actual output/runtime graphs, as these show you how the relative output changes over time. On max, a lot of Iights tend to drop off rapidly in output, or step-down after a couple of minutes. As such, you risk being misled if you simply look at ANSI FL-1 output and runtime values for a given light. I discuss the importance of comparing runtime graphs visually on my Testing Methods - Runtimes page on flashlightreviews.ca. :wave:
 
Last edited:

jirik_cz

Flashlight Enthusiast
Joined
Jul 29, 2007
Messages
1,605
Location
europe
Thanks for the SC52 review :) It is probably one of the most efficient AA lights on the current market.

In this case, I have been asked to comment on how the SC52's lumen estimates match the Fenix LD20. I happen to have a LD20 XP-G R4, which was rated by Fenix as 205 lumens at the time.

LD20 R4 was rated at 205 lumens, but that was before Fenix adopted the ANSI standard. The next R5 model which used 7% more efficient LED was rated only 190 ANSI lumens, so the LD20 R4 was probably only around 177 ANSI lumens.
 

reppans

Flashlight Enthusiast
Joined
Mar 25, 2007
Messages
4,873
Stellar Selfbuilt.. thank you so much for including 2xAA and 1xCR123 light data given that the ZL specs really compete in that realm. Also, thank you for addressing the lumen scale thing, I know I've been one of the voices behind it. Whatever the scale used, as you have mentioned, the important thing is that your data is consistent and relative, and so can be used as an accurate common base from which to compare between lights.

Very, very much appreciated.
 

GordoJones88

Flashlight Enthusiast
Joined
Nov 26, 2011
Messages
1,157
Location
Tennessee
Thanks SB for yet another excellent review, best ever.
You really stepped outside the box and gave us some meat to chew on.
Your runtime graphs truly give insight into how these lights work.
 

Badbeams3

Flashlight Enthusiast
Joined
Sep 28, 2000
Messages
4,389
Superb review. We are so lucky to have Selfbuilt and our other excellent reviewers.

Regarding the light, solve the flickering issue and a total win.

Spotted only one error: "Again though, you are best to run the light on NiMH of L91 lithiums to see the best performance here." ...Did you mean "or"...or is this a new hybrid batt? Lol...

Well, been waiting for this review, guess it`s time to place my order. Dis is mine nixt lite. Soon as my dealer gets them back in stock...
 
Last edited:

cyclesport

Enlightened
Joined
Feb 14, 2012
Messages
676
I'll add my voice to the other grateful readers and thank you for yet another detailed review of a very popular light. I too have been interested in the posts of some with flickering on H1 using fully charged IMRs/ICRs and had not noticed it with my SC52 until today when I popped a charged Sanyo based Intl' Outdoor 14500, 840mAh ICR and noticed that it dropped from H1 to L1 within seconds. No flicker just a drop-down. Oddly the battery level indicator only showed 3 flashes with this battery as well? Took it out and measured 4.07v. When the battery finally reached around 4v it functioned normally. Another exact same IO Sanyo battery as well as 7 other fully charged 14500s of diff brands function fine as well.
 

Mr Floppy

Flashlight Enthusiast
Joined
Feb 19, 2007
Messages
2,065
Another great review. I know you don't normally do this but how hard does it work an AA battery? On the H51w, it is drawing around 2.2A-2.3A at the tailcap on an Eneloop and 2.5A on a Sanyo 2700. I found that the H51w was very fussy (i.e hard to get the highest mode) with a lot of NiMH AA batteries other than Eneloops.
 

StandardBattery

Flashlight Enthusiast
Joined
Sep 2, 2007
Messages
2,959
Location
MA
Nice Review, I like the runtime tests. I'm a little sad to see that this light is not that well suited to the Lithium cells when used on Hi. The SC51 is much better for EDC with Lithium primary in my opinion even if it gives up some brightness.

From your analysis it looks to me that the flickering is totally from heat buildup probably on the regulator chip. I suspect your cooling is not effective enough to cool the circuit enough. The excessive heat appears to have two sources; the overall excess heat generated by the LED, and the excess heat created by the regulator itself when the input voltage is high. These two significant heat sources are contributing different amounts of heat in your various tests that exhibit the flicker issue, such that the temperature of the regulator circuit is too high.

Amazing output for this light, I quite like mine but I don't have any 14500, I thought this light might be the first to make me buy a 14500, but I think I'll pass on that for now. Maybe by the time the 5K version ships, I am quite disappointed in the performance on the Lithium primary because if I was to EDC this light, that is the cell I'd want to use... I guess this one is Eneloop power for me. I was quite impressed with how well it did on Alkaline. It looks like they paid less attention to the Lithium and more to the Alkaline, NiMH, Li-Ion chemistries.

Looks like there are some typos from cut and past in the legend for the Medium level graphs as the modes are still listed as HiA and HiB. I think I saw some other typos in the legeneds, but I forget now.

Thanks for the extensive testing.
 

selfbuilt

Flashaholic
Joined
May 27, 2006
Messages
7,005
Location
Canada
Thanks for the support everyone, glad you are enjoying the review. :grouphug:

LD20 R4 was rated at 205 lumens, but that was before Fenix adopted the ANSI standard. The next R5 model which used 7% more efficient LED was rated only 190 ANSI lumens, so the LD20 R4 was probably only around 177 ANSI lumens.
Thanks, a good point - ANSI FL-1 spec would certainly have been lower than max OTF lumens. Unfortunately, I don't have the later lights to directly compare in my lightbox, so don't know how they perform. But in my experience, a higher output bin does not always translate into a directly proportional increase in output (and conversely, a lower output bin may not be that much lower in output spec). Not sure why, but it likely involves a combination of effects of the circuit design and other LED characteristics (Vf maybe?). So while it would have been lower, it's hard to know what the ANSI FL-1 spec would have been for the R4 version (unless Fenix released that info at some point, for the older light?)

As an aside, I should mention that my lumen estimation conversation method was based on a comparison of specific time points post-activation. I matched the time of lumen measurement as best I could, and gave priority to later time points where available (e.g. if I had specified 30 sec or 1 min post activation, I choose those over initial activation for the comparison). I only used initial activation maximums if that's all that was available for a given light. In any case, they were always matched to the same time point in my lightbox testing.

Stellar Selfbuilt.. thank you so much for including 2xAA and 1xCR123 light data given that the ZL specs really compete in that realm. Also, thank you for addressing the lumen scale thing, I know I've been one of the voices behind it. Whatever the scale used, as you have mentioned, the important thing is that your data is consistent and relative, and so can be used as an accurate common base from which to compare between lights.Very, very much appreciated.
Thanks. Yes, at the end of the day, I'll I can say is my relative scale is consistent over time. Even if it does seem slightly inflated, not sure there's much point in trying to come up with a new calibration - it would likely just create confusion between the "new" numbers and the "old". The consistency and backward-compatibility is more valuable in my mind than the absolute correlation.

Regarding the light, solve the flickering issue and a total win.
Agreed. Zebralight tells me they are working on it ... I'm sure they will sort it out (hopefully soon). I have certainly seen many other cases of new light launches (from various makers) where there have been flickering issues on Li-ion. Typically tends to be due to some specific circuit component not being quite up to spec - can take a bit of time to isolate, but once they do, it is usually resolved for good by changing the component. I'll let you know more when I hear more. :)

Spotted only one error: "Again though, you are best to run the light on NiMH of L91 lithiums to see the best performance here." ...Did you mean "or"...or is this a new hybrid batt? Lol...
Thanks, fixed that one. I'm sure there are a lot more ... it's a long review to proof-read. :sleepy:

I found that the H51w was very fussy (i.e hard to get the highest mode) with a lot of NiMH AA batteries other than Eneloops.
Interesting. Unfortunately, I've stuck with Eneloops-only NiMH for years now. I found the old high-capacity/high-self-discharge NiMH didn't hold up well to repeated use, and Eneloops have been rock solid performers.

Nice Review, I like the runtime tests. I'm a little sad to see that this light is not that well suited to the Lithium cells when used on Hi. The SC51 is much better for EDC with Lithium primary in my opinion even if it gives up some brightness.
Well, to be fair, I've only done the highest two modes so far on L91. Let me try the Hi2B level and see how it does - that's when alkaline performance really started to pull away from the crowd. I'lll do it tomorrow and update the graph. Also, keep in mind as well that my relative output scale for the lightbox is not linear (i.e., the Hi1 and Hi2A are quite a bit brighter than the lower levels - my relative scale is not in estimated lumens).

Looks like there are some typos from cut and past in the legend for the Medium level graphs as the modes are still listed as HiA and HiB. I think I saw some other typos in the legeneds, but I forget now.
Thanks, I see what you mean - some of the "Hi2B" should actually be labelled M1. I'll fix the legends tomorrow, when I'm back in front of my main computer.
 
Last edited:

Swede74

Enlightened
Joined
Sep 30, 2011
Messages
577
Read, savoured and reread.

Roger Ebert was the first film critic to win the Pulitzer Prize for Criticism. If there were one for flashlight reviews, I can think of no worthier candidate for it than Selfbuilt.
 

ruf997tt

Newly Enlightened
Joined
Jan 11, 2013
Messages
30
Thanks for a timely and excellent review. I can't wait to get mine and maybe I should order another as I think I am really going to like this light and as good as it seems it might be in short supply for some time to come.

:thumbsup:
 

TweakMDS

Enlightened
Joined
Jun 18, 2012
Messages
504
Location
The Netherlands
Great review! Certainly one of the most anticipated reviews - of a light I already own :D

I was one of the people doing my own tests on lithium primaries - comparing it with a few Eagletac lights. With that, the Zebralight came in relatively lower, but it's very plausible that the D25A Ti CLickies are simply brighter on lithiums than their NiMH rating, so I'm glad you were able to address these concerns.

Thanks again for your time and effort in these reviews. I expressed my gratitude by hitting the donate button on your website for a small amount. I hope more people join in with that ^^
 

LEDburn

Enlightened
Joined
Jul 26, 2010
Messages
265
Man! Awesome review, as usual. Thanks for spending all the time on it.

There's been a lot of people waiting for this review - I am glad the lumen arguments can finally be laid to rest.

Zebralight, and obviously selfbuilt too, have outdone themselves yet again!!!
 

dts71

Newly Enlightened
Joined
Nov 16, 2006
Messages
191
Thanks for a great review - much appreciated!
I think many of us have been waiting to see this one.
 

hazna

Enlightened
Joined
Sep 24, 2007
Messages
610
Location
Australia
I notice the sc52 doesn't come with a headband... does the sc52 fit inthe sc50/sc51 headband?
 

selfbuilt

Flashaholic
Joined
May 27, 2006
Messages
7,005
Location
Canada
Read, savoured and reread. Roger Ebert was the first film critic to win the Pulitzer Prize for Criticism. If there were one for flashlight reviews, I can think of no worthier candidate for it than Selfbuilt.
Thanks! :laughing:

I was one of the people doing my own tests on lithium primaries - comparing it with a few Eagletac lights. With that, the Zebralight came in relatively lower, but it's very plausible that the D25A Ti CLickies are simply brighter on lithiums than their NiMH rating, so I'm glad you were able to address these concerns.
Interesting. It's a shame Eagletac never sent me the D25A to review (I gathered at the time that they had difficulty keeping it in stock, which is why it wasn't sent out with the other models). The D25Cs that I have are certainly very highly driven, so it wouldn't surprise me that the D25A was as well. But unfortunately I don't have any data to compare. :shrug:

Thanks again for your time and effort in these reviews. I expressed my gratitude by hitting the donate button on your website for a small amount. I hope more people join in with that ^^
Thanks, always appreciated as well! :)

There's been a lot of people waiting for this review - I am glad the lumen arguments can finally be laid to rest.
Well, I wouldn't say that exactly. ;) Despite how it may sound above, I'm not trying to shut down any discussion of relative or absolute lumen level estimates. I think it's important to critically evaluate ALL claims of purported "lumen" values, to see if they hold up to scrutiny. My commentary above is just to add to the existing literature on my own method, so people can better understand my estimates (and the effort I put into making them the best I could).

Truthfully, I resisted coming up with an estimated lumen scale for the longest time. I preferred (and still prefer) that people look at the full relative output/runtime data to get a better feel for how a light performs over time. But once a lot of people started posting their own "lumen" values here (some carefully considered, others not so much so), I realized I had to make an effort to provide a comparable scale. The point is that I did the best I could at the time to come up with an appropriate conversion (just like I've done more recently for high-output throwers that don't fit in my lightbox).

Of course, even if you had the most accurate lumen estimates, I find the arbitrariness of two simple ANSI FL-1 measures frustrating. It can easily be misleading as to a light's true performance, unless you look at all the variables in-between. That's why the output/runtime graphs remain the staple of my reviews. :)

I notice the sc52 doesn't come with a headband... does the sc52 fit inthe sc50/sc51 headband?
Good question. There was no headband included on my SC50 or SC51, so I can't say for certain. There was of course the one that came with my H31 (i.e., the contemporary 1xCR123A headlamp version of that series), but I don't know if they made different size attachments for the CR123A vs AA lights. I would guess that they used the same attachment, as the head and tailcap pieces weren't that different in diameter. I just tried that one on my SC52, and it would be really hard to get over the head - but you could do it easily enough from the tailcap, if you removed the pocket clip first.
 

Pvt. Pile

Newly Enlightened
Joined
Oct 1, 2012
Messages
23
Location
London
Nice one Selfbuilt! Ordered one before I even finished reading the review. I think we could call this THE most anticipated light review so far this year.
 
Top