Klarus        
Page 1 of 3 123 LastLast
Results 1 to 30 of 88

Thread: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, VIDEO+

  1. #1

    Thumbs up L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, VIDEO+

    Warning: pic heavy, as usual.

    This is a review of the 1xAA L3 Illumination L10 – available with either a XP-G2 R5 (Cool White) or the Nichia 219 4500K 92CRI (Neutral White).





    As you can see above, I have the full set of anodizing colors.

    In addition to the two emitter types, half of the above samples are the 3-level version, and the other half are the 4-level model. This gives me a lot more data than usual to work with, so expect more detailed analyses of sample variation in this review.

    Despite the use of two different emitters, L3 Illumination reports a common set of specifications. As you will see in my detailed testing later in this review, there are a few differences between the emitters.

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

    • LED: Cree XP-G2 R5 Cool White or Nichia 219 4500K 92CRI B10 bin
    • Note: these specs are apparently based on the earlier XP-G R5 model. The dealer's website suggest the XP-G2 outputs will be higher, but doesn't provide details
    • Four modes of output: firefly (0.09lumens, 147hours) -> Low (3 lumens, 30hrs) -> Medium (30 lumens) --> High (120 lumens, 1.5hrs) -> (Tested with Ni-mh battery with actual capacity 2500mAh)
    • Note: 3 mode model removes Firefly mode of 0.09 lumens. Low, Medium, High only
    • Stable current regulated circuit, stable brightness
    • Uses one 1.5V AA battery (ni-mh, alkaline ). 14500 batteries are not recommeded, because they heat up quickly.
    • 20-gram weight (excluding batteries)
    • Made of durable aircraft-grade aluminum
    • Premium Type III hard-anodized anti-abrasive finish
    • Waterproof to IPX-8 Standard
    • Toughened double ultra-clear glass lens
    • Reliable twist switch
    • Candle mode (Capable of standing up securely on a flat surface to serve as a candle)
    • Size: 79mm(length)x17.1mm(dia)
    • Mode switch: Tighten and loosen the head to switch between modes.
    • No mode memory, always starts on Firefly mode
    • MSRP: ~$23-25




    Packaging is pretty basic - a slightly opaque plastic "tube", which contains the light, extra o-rings and a small keychain ring. Printed on the outside of the container are the light specifics (including emitter type, color and mode number). I am not sure if the manual routinely comes with the lights – if not, you can download the PDF instruction manual from the dealer website (although the specs seem out of date in the manual – you are best to rely on the website for those).



    From left to right: Duracell NiMH AA; L3 Illumination L10; Foursevens Mini-AA; Nitecore SENS AA; Fenix E11; Olight S15; Rofis ER12; Zebralight SC52; L3 Illumination L10.

    All dimensions directly measured, and given with no batteries installed:

    L3 Illumination L10: Weight: 20.7g, Length: 79.4mm, Width (bezel): 17.1mm
    ArmyTek A1 XP-G: Weight: 57.3g, Length: 96.8mm , Width (bezel): 23.1mm
    Fenix LD12: Weight: 52.3g, Length: 99.9mm, Width (bezel): 21.6mm
    Foursevens Mini AA: Weight: 17.6g, Length: 78.2mm, Width (bezel): 17.6mm
    Olight S15 1xAA: Weight: 46.4g, Length: 87.0mm, Width (bezel): 23.1mm
    Nitecore EA1: Weight: 50.6g, Length: 84.3mm, Width (bezel): 25.41mm
    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
    Lumintop ED15: Weight: 59.7g, Length: 100.2, Width (bezel): 21.9mm
    Zebralight SC52: Weight 39.5g, Length 79.0mm, Width (bezel): 22.6mm, Width (max) 25.4mm
    Rofis ER12: Wright: 35.5g, Length: 96.2mm, Width (bezel): 18.6mm
    Xeno E03:: Weight: 48.1g, Length 96.7mm, Width (bezel): 21.5mm

    As you can see, the L10 is quite compact for a 1xAA light. This is consistent with its twisty interface.









    The L10 series lights remind me of similar slim-lined offerings over the years from Nitecore, Foursevens, Thrunite, Lumintop and Olight/ITP. Overall construction is pretty basic, but serviceable for a keychain-style pocket light. Basically, you can think of these sorts of lights as the 1xAA equivalent of the common 1xAAA twisty lights from the above makers.

    The L10 does have more knurling than most lights in this slim-lined class. While not overly aggressive (i.e., it won't shred clothing), it does help with grip. The various anodizing colors provide a range of options. The anodizing quality seems quite good on all colors, with no chips on any of my samples. Manufacturer claims hard anodized (aka, type III), which is better than most colored lights (which are only of type II strength). The one exception here is the white version, which is spray-painted over the grey-anodized form. This contributes to a more glossy finish and feel for the white version, and is more likely to chip than the other anodized colors.

    Screw threads are standard triangular cut, fairly fine (like most lights in this class). You can expect some play, but I found the fit to be reasonably good on my samples.

    As you can see, the head of the L10 has a brass pill/base, like a number of lights in this slim-lined class (presumably for heat-sinking purposes). The battery contact surface in the head is flat, and includes a foam cushion surrounding the center disc. A cushion is common on these sorts of lights to reduce battery rattle, and to provide some extra padding. Note there is no spring in the tail region (only a small raised metal post). These types of lights thus all run the risk of crushing your cells if you over-tighten repeatedly.

    There is a lanyard/spilt-ring attachment point at the base of the tailcap, with a cut-out for the keychain split-ring. As such, the lights can all tailstand stably.

    As mentioned earlier, the lights are distinguished by their emitter choice:

    Nichia 219



    XP-G2 R5



    Centering was consistently good on my samples – there was only one in the whole group that was slightly off visibly, but not enough to affect the beam pattern. Both emitter versions seem to come with the same reflector – a relatively shallow one, with an orange peel (OP) coating to smooth out the beam.

    Please see my detailed beamshots later in this review, for each model.

    User Interface

    The L10 uses the same interface as most AA/AAA-twisty lights.

    Fully tighten the head and it comes on in its lowest mode (Firefly or Low, depending on the model). Do a rapid twist off-on and the light advances to next level. Mode sequence is Firefly > Lo > Med > Hi (or Lo > Med > Hi), in repeating sequence.

    Turn the light off by loosening the head. There is no mode memory, and the light always defaults to its lowest setting if you wait more than a second or two before turning back on.

    There is no strobe or SOS mode on the L10.

    Video:

    For information on the light, including the build and user interface, please see my 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.

    As with all my videos, I recommend you have annotations turned on. I commonly update the commentary with additional information or clarifications before publicly releasing the video.

    PWM

    There is no sign of PWM at any output level, on any model. The L10 is current-controlled.

    As mentioned above, there is no strobe or SOS mode on the L10.

    Beamshots:

    For white-wall beamshots below, all lights are on Max output on a Sanyo Eneloop NiMH (2000mAh). 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 – except for the Nichia 219, which is using a Daylight white balance on my Canon camera.













    If you neglect beam tint (which is hard to capture on a camera, due to the narrow range of white balancing options), the overall beam pattern is not hugely different. I find the XP-G2 versions to be slightly more "throwy", with a more sharply defined hotspot and slightly greater throw.

    But it is definitely the beam tint differences that you are likely to notice between these emitters. The 4500K on the Nichia is very pleasing (to my eye) neutral white. The higher color rendition index (CRI) should also help display colors more consistently.

    One question that comes up in a lot of my reviews is the consistency of <1 lumen Moonlight/Firefly modes. I have long argued that these are highly variable between samples of the same model (unlike >1 lumen levels, which are very consistent between samples). We have here an opportunity to start to test these differences, as all four of my 4-level models are rated as 0.09 lumens. How do they look? (Daylight white balance below)





    Clearly, there is a HUGE variability in the Firefly output levels among these samples – as I believe there are on all lights with <1 lumen modes. As for the actual measured outputs (and sample variation measures), scroll down to my detailed testing.

    Testing Method:

    All my output numbers are relative for my home-made light box setup, as described on my flashlightreviews.ca website. 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.

    My summary tables are reported in a manner consistent with the ANSI FL-1 standard for flashlight testing. Please see http://www.flashlightreviews.ca/FL1.htm for a discussion, 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).

    --------

    REVIEWER'S NOTE: Normally at this point I would provide a summary table showing how the output and throw of the review light compares to the competition. But we have a wonderful opportunity here to explore inter-sample variability, given that I have 8 samples of the L10 on hand (4 XP-G2s and 4 Nichia 219s, split into 3- and 4-mode versions). Let's work through what this means for us.

    The first thing to note is that L3 Illumination reports identical output and runtime specs for the two emitter versions – although that's because both the XP-G2 R5 and Nichia 219 are still using the originally reported XP-G R5 specs. You might naturally question whether or not that is reasonable.

    The best place to start to answer that question is to look at some output/runtime data. Below is a graph of the 8 lights (colored-grouped by emitter) on Hi, on Sanyo Eneloop:



    The first thing you are likely to notice is that the XP-G2 versions (blue lines) all show higher output on max than the Nichia 219 versions (orange lines). The differences aren't huge, but it is suggestively consistent. The second thing to note is that the runtimes to 10% (ANSI FL-1 standard) don't seem appreciably different (although there is one Nichia that seems to be a bit of an outlier).

    It's always important to actually look at the data in this way, before rushing to do statistical comparisons. It's common in this hobby for people to want to reduce flashlights to just a couple of numbers (e.g., initial output and runtime). But it is often not valid to do that, due to all the differences in regulation and runtime patterns between different lights (see my website's methodology page for a good example of why you need to always look at the runtime data first). But in this case, the consistently tight regulation pattern on NiMH for all my L10 samples suggests we can proceed with some probabilistic statistical analyses.

    In statistics, the standard measure used to describe how much variation there is from the average is known as the standard deviation (SD). You can easily calculate the SD and the average (mean), which gives you a convenient way to directly compare the variation between groups in a consistent manner. Simple sample statistics are commonly reported as the mean +/- the SD.

    For a normally-distributed population (i.e., the class bell-shaped curve for IQ tests, and many naturally occurring phenomenon), you can think of the mean as the center of the curve, and a single standard deviation unit as being the point on either side of the mean where 68% of the samples would fall within. Let's see how the estimated lumens look:



    What is shown above is the manufacturer's common lumen estimates for the four possible modes (first two columns). The next two columns are what I actually measured for each sample, broken down by emitter type (XP-G2 in column 3, Nichia 219 in column 4). These values are expressed as the mean +/- SD.

    A few things jump out here (just looking at third and fourth columns for now). The first is that there is a lot of variability WITHIN the Firefly modes of a given group (although I caution here than the number of samples (n) is lower). For either emitter, the Firefly SD is huge: ~80-95% of the mean value. This is consistent with what you saw in the beamshots earlier, for the four samples. In contrast, the Lo/Med/Hi modes show a far more consistent ~3-10% SD variance around the mean, on either emitter.

    The second major finding is that the absolute mean for each emitter group is not that different (at any level). The possible exception to this is the Hi mode, where there does seem to be a more consistent difference (as we saw in the earlier graph). Is this measurably significant?

    To address that issue, I have performed a very common statistical hypothesis test known as the Student's t test (or simply, the t-test). This test is used to determine if two sets of data are statistically significantly different from each other. It is common to use the probability statistic p < 0.05 as a cut-off for statistical significance. P-values are often misunderstood as the given odds ratio of being right (i.e., a lot of people erroneously think that p = 0.05 means there is a one in twenty chance of being wrong). In fact, when used in this context to compare samples of larger populations, p = 0.05 actually means that only 1 time out of 20 would you expect to see a difference this great between two groups of samples when in fact there wasn't one.

    Doing multiple t-tests within a larger data set is a statistical no-no, but I feel quite confident in doing the four limited comparisons shown in column 5 above (i.e., are the two emitter groups different for each level, at the p < 0.05 cut-off level?). If you are curious, there are correction methods for much larger multiple comparisons (e.g. Bonferroni), but these tend be overly conservative and not necessary in our limited example. As an aside, most fields of science would use an analysis of variance (ANOVA) F-test for these types of comparisons, followed by a post-hoc test. What many (even in the science biz) don't realize is that the various post-hoc tests all suffer from similar correction factor shenanigans as the simpler t-test methods. Statistical purists decry all post-hoc analyses.

    Back on topic: the t-test results tell us that we can reasonably believe that the Hi mode of the XP-G2 and Nichia 219 sub-groups are in fact different from one another. This is consistent with we observed on the output/runtime graph above – there seems to be a difference between these two groups.

    The t-tests also tell us that there is no apparent difference between the two emitter groups at any of the other levels (i.e., Firefly/Lo/Med). As a result, there is no real justification for treating them separately in this analysis – I have combined all the samples into one joint column (column 6) for these levels.

    Long story short: You can use the bolded numbers in the table above as the best measures of the true mean and variance for each level. Only in the case of the Hi mode do you need to consider these separately, by emitter type.

    All that being said, is the absolute difference between the Hi outputs of the XP-G2 and Nichia 219s all that different? Not really - you are looking at a <10% typical difference. The visual differences in beam profile and tint between these emitters will be far more noticeable than such minor overall output differences.

    So much for output then. How about runtime? For this, I've only done ANSI FL-1 standard testing (i.e., time to 10%) on the Hi mode on Sanyo Eneloop:



    Again, you can see the stats back up what you could see in the output/runtime graph (leaving all the data in). That is, there is no significant different in runtime on Hi between these two emitter classes. I would recommend you stick with the combined column measure above.

    It is nice to see that both the lumen and runtime specs provided by L3 Illumination seem to match very well with my testing. It's also nice to see that there is relatively little variation between samples – these sorts of standard deviations (at these n levels) are quite good.

    Given the much larger difference in beam profile between these emitters, how does the throw compare?



    Ding-ding-ding, we have a winner! There is remarkably little variation in peak intensity throw WITHIN each emitter group – but a noticeable difference BETWEEN emitter groups. In common statistical parlance, the actual p value here would typically be described as "highly statistically significant" (i.e., well below p < 0.01).

    So, while there is a small increase in max output with the XP-G2 version (<10%), there is a much more noticeable increase in peak intensity throw (>35%). But before you get too carried away, keep in mind that a better way to meaningfully compare throw is by beam distance to a common intensity (not raw intensity at a common distance). By the common ANSI FL-1 beam distance measure, the XP-G2s are only a little >15% further throwing than the Nichia 219s (recall: light decays with distance by an inverse square law).

    Now let's roll all that up into my usual simple summary table for NiMH, broken down by emitter group for the L10 (given the statistically significant differences on Hi). Since there is no statistical reason to separate the lights by emitter on Firefly/Lo, I've left those fields common.



    You'll note above that ceiling bounce numbers (which I provide as an additional output measure) suggest a greater output difference between the emitter types on Hi. I suspect this is due to the more throwy beam on the XP-G2 model (i.e., throwy beams tend to get higher ceiling bounce numbers, likely due to direct reflectance off the ceiling back to the sensor). As always, I recommend you go with the much better calibrated lightbox numbers.

    For 1x14500 (3.7V Li-ion), the manufacturer recommends you don't run the light on this source, due to rapid heat buildup. In case you were wondering, here is a comparison table of estimated lumen outputs:



    As you can see, all modes of the lights are much brighter on 14500. I have broken the Med/Hi results up by emitter on 14500, as t-tests suggest these are significantly different (no difference between emitters detected on Firefly or Lo). On the basis of these results, I also recommend you don't run the lights on Med or Hi on 14500, as the output considerably exceeds the Hi level of standard cells. You may thus damage the emitter or battery by the heat buildup (i.e., there is little thermal mass to help cool the light).

    Another other thing to notice above is that the sub-lumen Firefly mode is actually maintained, although it is about four times brighter on 14500 (on average). Interesting, the variability measure stays about the same, which goes to show you that consistency between moonlight/firefly levels improves at the higher levels (i.e., the lower the moonlight level, the more relative variation you are likely to see).

    One last comment about the above - don't get caught up on the implied precision of the lumen estimates. I provide 3 significant figures (sig figs, or sig digs/significant digits as the kids call them today) in that table abpve so that you can get a better feel of variability. I personally limit most of my tables to 2 sig figs, because I don't believe you should be inferring accuracy to the precision of my instumentation (i.e., just because I can measure my lumen estimates consistently doesn't mean it represents the "true" lumen value).

    -----

    Ok, that was a lot of data. I know that was a long preamble to get to this point, but hopefully those of you who are more quantitatively-inclined will have appreciated it.

    Runtimes

    For comparison output/runtimes, I've picked a "representational" sample of each emitter group. In other words, I've picked a sample for each group that was relatively near the mean for that group (for both runtime and output).








    Performance is pretty much what you would expect for a good XP-G/XP-G2 light – excellent regulation, and typical runtime for current-control.

    Since the manufacturer recommends against running the lights on 14500, I have not done additional runtimes on this battery source.

    Potential Issues

    Lights roll easily.

    Although impressively small for 1xAA lights, the L10s may still be a bit large for a keychain light.

    Twisty lights that tighten to activate or change modes always have the potential to be battery crushers. Care needs to be used not to over-tighten, especially when mode switching.

    Like with other lights in this slim-lined class, output levels are much higher on 14500 (3.7V Li-ion). The manufacturer recommends against running the lights this way (due to heat buildup), and I concur. I recommend you do not run these lights on Med or Hi on 14500 – best to limit yourself to Firefly/Lo modes (which are much brighter than on standard cells).

    Preliminary Observations

    The L10 by L3 Illumination is a nice slim-lined 1xAA light, very similar in functionality to a number of competing lights by Olight/ITP, Foursevens, Fenix, Thrunite and others. What really distinguishes the L10 is the stock option of the latest Cree emitter (XP-G2 cool white), or the popular Hi CRI Nichia 219 neutral white emitter. Decisions, decisions.

    One nice feature is the use of current-control for the low modes on the L10. Although this is becoming increasing common, some of the earlier slim-lined designs from other makers used PWM (with potentially visible flicker). My L10 samples were all fully flat-regulated and visually stable on all levels, including Firefly.

    Another difference here is that you have the option of a 3-level (Lo/Med/Hi) or 4-level (Firefly/Lo/Med/Hi) model. The only difference is the addition of the extra Firefly level at the start of the mode sequence. Note there is no mode memory, and the lights all start at their lowest level. Mode spacing seems good to me (and I personally like the dim Firefly mode).

    Finally, you have the option of a range of anodized colors. Many lights remind me of the famous quote from Henry Ford - "any customer can have a car painted any colour that he wants so long as it is black." The colors here are all reported as hard anodized (aka type III), and I certainly saw no signs of chips or damage on any of the ones I've handled. Body design is good for this class, with better grip than some (i.e., can easily use one-handed).

    The manufacturer doesn't recommend the use of 14500 Li-ion batteries (because of the high heat, due to the much higher output). I concur, and recommend you stick with standard AA cells (NiMH, alkaline or L91 lithium). Performance-wise, the current-controlled circuit worked consistently well in my testing. Runtimes are excellent for the class, and regulation was relatively flat.

    Beam pattern is quite good on both emitter types, a fairly general purpose beam. I personally prefer the slightly smoother hotspot-to-spill transitions of the Nichia 219 – not to mention its neutral white tint and higher CRI. But that choice is yours.

    As an aside, it was nice to be able to do some detailed sample variation testing on the L10. I don't get that opportunity very often, and the results here match my previous experience. To wit, Firefly/Moonlight mode output levels can be extremely variable between samples of the same model, but most regular modes are very consistent. In my experience, this significant variability is true of all lights with <1 lumen modes (particularly <0.1 lumen modes), which is why you see so many disagreements here about the actual lumen level of various people's n=1 samples. Note that I didn't directly test runtimes at the low levels on the L10, but I have previously had the same experience on that front as well (i.e., runtimes are generally very consistent at higher outputs, and very variable at low ones).

    A nice addition to the simple, twisty-style class of 1xAA lights. Not much to criticize here, it seems like L3 Illumination has gotten this model right.

    ----

    L3 Illumination L10 lights were supplied by sbflashlights.com for review.
    Last edited by selfbuilt; 12-20-2013 at 08:10 AM.
    Full list of all my reviews: flashlightreviews.ca. Latest hobby: whiskyanalysis.com. New: Selfbuilt's Summer Sale!
    Gratefully accepting donations to my battery fund.

  2. #2
    Flashaholic doctordun's Avatar
    Join Date
    Jan 2006
    Location
    Red Oak, Texas
    Posts
    306

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, VIDEO

    This is an awesome review. Thank you.

  3. #3

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, VIDEO

    Thanks for the great review! I have one of these and I really like it, so much so that it's in my EDC rotation. However, I am concerned about the durability of the foam cushion around the head / battery contact area - especially since I carry it and use it a lot. What are you thoughts on this?

  4. #4

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, VIDEO

    Thx for reviewing this CPF favorite! I've had one of the 4 mode 219 versions for some time and have always felt it was a great value for a "pocketable" Nichia 219 Hi-CRI EDC light, being rare...and was always curious as to the true max output of this light. My moonlight mode has been, IMO at a perfect 1-2 (est) lumen level...and I was surprised to see the variability you observed among several test samples!

  5. #5

    Default L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, V...

    Great review Selfbuilt. FWIW, I got 48 hrs on a low mode test with a 2000 mah Eneloops, far exceeding spec, on an N219 version....one of the most efficient 1xAAs in my collection.

    ...As an aside, it was nice to be able to do some detailed sample variation testing on the L10. I don't get that opportunity very often, and the results here match my previous experience. To wit, Firefly/Moonlight mode output levels can be extremely variable between samples of the same model, but most regular modes are very consistent. In my experience, this extreme variability is true of all lights with <1 lumen modes, which is why you see so many disagreements here about the actual lumen level of various people's n=1 samples. Note that I didn't directly test runtimes at the low levels on the L10, but I have previously had the same experience on that front as well (i.e., runtimes are generally very consistent at higher outputs, and very variable at low ones)...
    Appreciate you taking the time to look into this, and especially the side-by-side photo which perfectly illustrates the variability. As a sub-lumen/AA/14500 collector that tests his own lights, I know I have been one to address this repeatedly.

    I'm wondering if output consistency between individual samples increases significantly as lumens increase - that certainly seems true based on your SD statistics here. In other words, I think lights with "brighter" sub-lumen levels might show much greater sample consistency than "dimmer" ones. For example, here's a side-by-side of a sample of my AA sub-lumen collection with Manuf. spec/my measurement (which I've posted elsewhere):



    I think I've had 8+ copies of the Quark 0.3 XML; 4 of the Quark 0.2 XPG; 3 of the D25A 0.25 XML ('12); and 5 of the T10 0.1, and although I admit that I have not metered each sample, to my eye, they appeared stone consistent when I tested them side-by-side for proper operation upon receipt (I stock up for gifts, esp during the great rebranding sales).

    I do not have multiple copies of any light that fell below the Thrunite Firefly 0.09 level, however, as I personally just find it to be less useful, and was frankly disappointed by how far off spec the lights seemed to be (also had the S15, which was similar to your 2nd example, if you recall). Lastly, perhaps sample consistency is more prevalent among certain manufacturers than others?
    Last edited by reppans; 11-29-2013 at 10:05 AM.

  6. #6

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, VIDEO

    Quote Originally Posted by rje58 View Post
    However, I am concerned about the durability of the foam cushion around the head / battery contact area - especially since I carry it and use it a lot. What are you thoughts on this?
    This is a common concern on lights that use a foam cushion. But there is an easy fix if one falls off or gets too squished over time: Dr. Sholl's corn pads.

    Believe it or not, the classic round ones are a good size to prevent AA rattle. They are bit wider than the L10 stock pad diameter, but could be easily trimmed slightly if needed. A quick internet search tells me these seem to be more oval-shapped now, so some trimming will definitely be required for those.

    Quote Originally Posted by cyclesport View Post
    My moonlight mode has been, IMO at a perfect 1-2 (est) lumen level...and I was surprised to see the variability you observed among several test samples!
    I wasn't. Although lights near the 1 lumen level are usually pretty consistent, all bets are off once you go <0.01 lumens. A slight difference in circuit or emitter characteristics could have a signficant impact on output (and runtime).

    As an aside, this is why you will see so many disagreements here about the relative output of <0.01 lumen lights. I suspect the individual samples are varying widely, so you don't have any consistent reference standard to work against (i.e., there is no one light everyone can agree on, since they all suffer from some degree of sample-to-sample variability). I would point out as well that most commercial integrating spheres are designed for higher lumen levels (i.e. in the ~10-10000 lumen range). Manufacturers would need to use an entirely different sphere to accurately measure their sub-lumen levels, to get reliable numbers.

    Quote Originally Posted by reppans View Post
    I'm wondering if output consistency between individual samples increases significantly as lumens increase - that certainly seems true based on your SD statistics here. In other words, I think lights with "brighter" sub-lumen levels might show much greater sample consistency than "dimmer" ones.
    Yes, I would think so.

    As a side issue, although I haven't reported on pair-wise comparisons, I did notice a possible trend for a given light to consistently deviate across levels (i.e., if it is relatively bright on one level, it tends to be so on the others, compared to the group mean). But this certainly broke down by the time I got to Firefly - for example, the brightest Firefly sample was also the dimmest Hi.

    I'll double-check the data and see if I can find time to do an ANOVA to confirm ...

    EDIT: I just checked, and there is no statistically significant pattern here after all. So there is no evidence of consistency of variance across levels for the repeated samples.

    Lastly, perhaps sample consistency is more prevalent among certain manufacturers than others?
    A good point - I have no idea how representational the L10 results are to other manufacturers. We would need to see multiple samples of a given model to know for sure.
    Last edited by selfbuilt; 11-29-2013 at 01:14 PM.
    Full list of all my reviews: flashlightreviews.ca. Latest hobby: whiskyanalysis.com. New: Selfbuilt's Summer Sale!
    Gratefully accepting donations to my battery fund.

  7. #7
    Flashaholic* psychbeat's Avatar
    Join Date
    Nov 2009
    Location
    SF norcal
    Posts
    2,759

    Default L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, V...

    Awesome review!

    I think my 219s moonlight is on the lower end of the spectrum - still useable tho.

    The 3 mode is better for non flashaholics for sure.

    The white paint does chip a bit but with the gray ano underneath I think it will have a nice "patina" after a couple of years hard use

    I wish they made an 18650 version!

  8. #8
    Flashaholic* tobrien's Avatar
    Join Date
    Dec 2005
    Location
    Georgia Highway 441
    Posts
    4,861

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, V...

    thanks selfbuilt! excellent work as always!

    I think I'll have to pick up a 219 model. I really, genuinely believe there is no beating the beam on a 219 in low-output scenarios.

    The XP-G2 is great, don't get me wrong, but the hotspot is too strong when you're in the dark and don't need a ton of light (IMO). Similarly, when you don't need much light, you can't go wrong with the "slightly smoother hotspot-to-spill transitions of the Nichia 219" (quote from selfbuilt's review).

    The 219 works great in these smaller lights in my experience.

    edit: is there no blue or green model for the 219 variant?
    Last edited by tobrien; 11-29-2013 at 11:26 AM.
    aka Edgar Allan Bro, Brosama Bin Liftin, Walter Crunkite, Bro Namath, Teddy Brosevelt, and the Tomahawk Crunkmissile.
    my lights - review of PrecisionWorks - that's Gucci Mane in my avatar

  9. #9

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, V...

    Quote Originally Posted by tobrien View Post
    edit: is there no blue or green model for the 219 variant?
    I believe all anodizing colors are available for either emitter, and for both 3/4-level versions. I suppose the only issue is how much stock they can possibly carry of each, given so many permutations.

    BTW, I've justed edit my earlier post where I speculated about a possible consistency of deviation for individual samples (across each level). ANOVA confirms there's there is no clear pattern (p = 0.25).
    Last edited by selfbuilt; 11-29-2013 at 01:16 PM.
    Full list of all my reviews: flashlightreviews.ca. Latest hobby: whiskyanalysis.com. New: Selfbuilt's Summer Sale!
    Gratefully accepting donations to my battery fund.

  10. #10

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, V...

    I always appreciate your reviews but this one was over the top! I was thinking "I am such a geek" as I took in your compiled data with fascination!

    I'm sure this review was a lot of work; thanks for putting it together!

  11. #11
    Flashaholic*
    Join Date
    Dec 2007
    Location
    USA
    Posts
    1,785

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, V...

    In light of the crAA 14505's short positive terminal, does it work with the foam protection of the head?
    That shouldn't overheat this 20 gram light, right?

  12. #12
    Flashaholic
    Join Date
    Dec 2012
    Location
    Maryland
    Posts
    123

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, V...

    Selfbuilt, nice review as always!!
    Last edited by stevieo; 11-30-2013 at 11:37 AM.

  13. #13

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, V...

    How did I not realize selfbuilt never reviewed on of my favorites until now? I have two black L10s with the Nichia 219. As irony would have it, my most often used light with the best tint and beam is the cheapest
    GOOD TINT!

  14. #14

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, V...

    Quote Originally Posted by rickypanecatyl View Post
    I'm sure this review was a lot of work; thanks for putting it together!
    Yes it was, but its fun to have more data to examine.

    Quote Originally Posted by MichaelW View Post
    In light of the crAA 14505's short positive terminal, does it work with the foam protection of the head?
    That shouldn't overheat this 20 gram light, right?
    The positive button of the 14505 CRAA is not sufficient to make contact on my sample (although I suppose you could remove the foam). Would certainly be better than 14500 Li-ion, but I'm sure Hi would still be contra-indicated.

    Quote Originally Posted by stevieo View Post
    you have typo in MANUFACTURERS SPECS section.
    Thanks, fixed.
    Full list of all my reviews: flashlightreviews.ca. Latest hobby: whiskyanalysis.com. New: Selfbuilt's Summer Sale!
    Gratefully accepting donations to my battery fund.

  15. #15
    Flashaholic StudFreeman's Avatar
    Join Date
    Sep 2011
    Location
    Maine
    Posts
    185

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, V...

    This is probably my favorite selfbuilt review yet. Entertaining and informative. The L10 219 sure is a great value!

  16. #16

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, VIDEO

    Quote Originally Posted by selfbuilt View Post
    A good point - I have no idea how representational the L10 results are to other manufacturers. We would need to see multiple samples of a given model to know for sure.

    Just wanted to get back to the question of sub-lumen variability. Here's a quick shot of four Quark LV XMLs on 0.3 lms (2 CW, 2 NW) and three 2012 D25A NW XMLs on 0.5 lms (I meter them to be ~0.25 lms). The furthest right is dimmest, but it has an unusual "PWM" issue (or some sort of high frequency strobe/pulse/flickering) on moonlight mode - gave this one to the Mrs. with ML mode programmed off.



  17. #17
    Flashaholic doctordun's Avatar
    Join Date
    Jan 2006
    Location
    Red Oak, Texas
    Posts
    306

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, VIDEO

    Selfbuilt, would you have an estimate of runtime of the Nichia 219 medium 30 lumen level with and eneloop 2000?
    Short of running a test, is there a formula for something like this.
    Thanks.
    Last edited by doctordun; 12-02-2013 at 11:02 AM.

  18. #18
    Flashaholic
    Join Date
    Nov 2012
    Location
    Munich/Germany
    Posts
    164

    Default

    The white ones are painted, not anodized!

    Gesendet von meinem GT-N8000 mit Tapatalk 4

  19. #19

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, VIDEO

    Quote Originally Posted by reppans View Post
    Just wanted to get back to the question of sub-lumen variability. Here's a quick shot of four Quark LV XMLs on 0.3 lms (2 CW, 2 NW) and three 2012 D25A NW XMLs on 0.5 lms (I meter them to be ~0.25 lms). The furthest right is dimmest, but it has an unusual "PWM" issue (or some sort of high frequency strobe/pulse/flickering) on moonlight mode - gave this one to the Mrs. with ML mode programmed off.
    Thanks, good comparison. Of course, the L10 Firefly level is lower than my various Quark or Eagletac D25x-series lights, so the effect could still be due to the ultra-lo level here (as opposed to being solely manufacturer-specific).

    Quote Originally Posted by doctordun View Post
    Selfbuilt, would you have an estimate of runtime of the Nichia 219 medium 30 lumen level with and eneloop 2000? Short of running a test, is there a formula for something like this.
    Not really - you actually need to do the runtime test to find out for sure. I have not done the Med mode on eneloop - and may not have time, given how busy I am with other lights. That said, most AA lights around this output level are not too different in term of runtime and regulation on alkaline and eneloop, so you could use that data as a rough indicator (with eneloop likely lasting slightly longer).

    Quote Originally Posted by wertzius View Post
    The white ones are painted, not anodized!
    Yes, that was pointed out in the review and the video.
    Full list of all my reviews: flashlightreviews.ca. Latest hobby: whiskyanalysis.com. New: Selfbuilt's Summer Sale!
    Gratefully accepting donations to my battery fund.

  20. #20
    Flashaholic
    Join Date
    Nov 2012
    Location
    Munich/Germany
    Posts
    164

    Default

    Sry, i don't know how i overlooked that!

    Gesendet von meinem GT-N8000 mit Tapatalk 4

  21. #21

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, VIDEO

    Quote Originally Posted by wertzius View Post
    Sry, i don't know how i overlooked that!
    It's ok, there's a lot of text up there.
    Full list of all my reviews: flashlightreviews.ca. Latest hobby: whiskyanalysis.com. New: Selfbuilt's Summer Sale!
    Gratefully accepting donations to my battery fund.

  22. #22

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, VIDEO

    Quote Originally Posted by markr6 View Post
    How did I not realize selfbuilt never reviewed on of my favorites until now?
    Me too, only just noticed it in his sig as the latest review, and what a review! Much of the stats stuff just went out my ear though. Although I like warm and Nichia 219 has no apparent tint that I can see, I really like this light for much more than just the emitter, flat regulation for a start.

    I did some lumen calculations for my 219 based off your LD10 R4 review, wasn't too far off but not close. I think this is the new set of figures I'll use to calibrate my new set up should I even bother. It seems that most new lights eventually make it to your hands, and shows the recognition from the industry as to high quality and value of your reviews.

    The run time on high is interesting. I calculated high based on the current draw and I should only get around an hour. Of course, what I realise now is that it is coulud be the resistance in my probes as the multi meter is in series and I take the reading with batteries fresh off the charger. In any case, I get 1.9A on eneloops but around 1.5A on anything else. To get 1 hour 23 minutes, it pretty much has draw to 1.5A, yet I can't see any sort of initial peak on your runtime graph? Could it be some sort of ramp up on eneloops or just my eneloops (gen 1)? It seems to hover around 1.8/1.9 for the first 30 seconds roughly (holding probe and compressing the foam is hard). I notice that with a depleted alkaline that it some times takes a couple of seconds for high to switch on.

    In any case, to get the output out of the 219 on high, it'll need to put 500mA to 600mA to the LED (going by memory as I can't find the spec sheet) so safe to say the efficiency is good?

    Quote Originally Posted by doctordun View Post
    Selfbuilt, would you have an estimate of runtime of the Nichia 219 medium 30 lumen level with and eneloop 2000?
    Short of running a test, is there a formula for something like this.
    Thanks.
    I'm getting 250mA from the battery tail cap measure, so it is pretty much as the runtime in the alkaline graph. Around 8 hours. Of course, nothing beats an actual runtime graph :wink:

    Quote Originally Posted by selfbuilt View Post
    Finally, you have the option of a range of anodized colors. Many lights remind me of the famous quote from Henry Ford - "any customer can have a car painted any colour that he wants so long as it is black."
    It's a quote that has been perhaps misquoted. The original attributed quote is "you can paint it any color so long as it is black". The customer bit was somehow added. Then there is the issue as to whether he ever said it, as the model T came out with different colors before they were all apparently painted black to reduce costs. Old page from Henry Ford museum. http://www.hfmgv.org/exhibits/showro...8/model.t.html
    Last edited by Mr Floppy; 12-03-2013 at 06:26 AM.

  23. #23

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, VIDEO

    Quote Originally Posted by Mr Floppy View Post
    Of course, what I realise now is that it is coulud be the resistance in my probes as the multi meter is in series and I take the reading with batteries fresh off the charger. In any case, I get 1.9A on eneloops but around 1.5A on anything else. To get 1 hour 23 minutes, it pretty much has draw to 1.5A, yet I can't see any sort of initial peak on your runtime graph? Could it be some sort of ramp up on eneloops or just my eneloops (gen 1)? It seems to hover around 1.8/1.9 for the first 30 seconds roughly (holding probe and compressing the foam is hard).
    Hard to say. I don't bother with tailcap current draws any more, as I know they are not accurate with my meters/cables. The fresh-off-the-charger could also be a factor. But as you say, nothing beats an actual runtime ... it's hard to troubleshoot why an estimation method is not accurate.

    It's a quote that has been perhaps misquoted. The original attributed quote is "you can paint it any color so long as it is black". The customer bit was somehow added. Then there is the issue as to whether he ever said it, as the model T came out with different colors before they were all apparently painted black to reduce costs. Old page from Henry Ford museum. http://www.hfmgv.org/exhibits/showro...8/model.t.html
    He added it himself. I took the quote wording from wikiquote.org, which links to the full text of his autobiography (which is a verifiable source). He wis referring to his own recollection of what he had said at the time, so it is of course possible he misquoted himself. ...
    Last edited by selfbuilt; 12-03-2013 at 07:54 AM.
    Full list of all my reviews: flashlightreviews.ca. Latest hobby: whiskyanalysis.com. New: Selfbuilt's Summer Sale!
    Gratefully accepting donations to my battery fund.

  24. #24

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, VIDEO

    Quote Originally Posted by selfbuilt View Post
    Thanks, good comparison. Of course, the L10 Firefly level is lower than my various Quark or Eagletac D25x-series lights, so the effect could still be due to the ultra-lo level here (as opposed to being solely manufacturer-specific).
    Yes, agreed but I thought it was important to establish some consistency among the "brighter" sub-lumen mode manufacturers since the following general statements indicates high variability among most sub-lumen lights.

    ....One question that comes up in a lot of my reviews is the consistency of <1 lumen Moonlight/Firefly modes. I have long argued that these are highly variable between samples of the same model (unlike >1 lumen levels, which are very consistent between samples)....

    ....Clearly, there is a HUGE variability in the Firefly output levels among these samples – as I believe there are on all lights with <1 lumen modes...
    But to the point on ultra-lows and manufacturer consistency, here are the Thrunite T10s (probably the L10s closest competitor) which I have on hand and they're rated the same 0.09lm/147hr as the L10, although I meter the T10s as accurate. (BTW, I promise I am not cherry picking - the QAAX, D25A, and T10 are the only models I have 3 or more copies of and I am displaying all of them.)



    I know I have repeatedly raised the sub-lumen variability question with you, but I just don't see it, and I've had over 2 dozen (across 4 models) pass through my hands, but perhaps it's because I do not repeat purchase lights with <0.09 lm lows. What really confounds me and started these questions are the late model Zebralight lights (ie 52 series). They're supposed to have among the brightest 0.34 lumen moonlight modes, but mine meters (see pix above) at ~ 1/5th of the equivalent QAAX spec (although it probably explains how ZL claims 3x the runtime of the QAAX and D25AX). I believe you metered both the QAA2X and SC52 0.3 mode as having the same output?

    Sorry to belabor the point, but I use "bright" sub-lumen modes more than any other and just find that manufacturer spec exaggeration, or huge sample variability, in the 300%+ "off" range to be..... disappointing. No offense to the L10, I love my 3-mode N219 - gorgeous beam, super efficient, superb value, and with SB's "corn pad" idea, good for long-term use.

  25. #25
    Flashaholic* TweakMDS's Avatar
    Join Date
    Jun 2012
    Location
    The Netherlands
    Posts
    505

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, VIDEO

    Love the review again. Thank your for that and for keeping up the scientific approach! I hope you will at least be rewarded by getting to keep this nice set

    I have a soft spot for single AA/AAA/CR123 lights, so this one has immediately been an interesting one, especially the Nichia 219 option (and - to be frank - only that one).
    However, to me it falls a bit in the middle between twisty AAA keychain lights, and full fledged EDC AA's with a clicky and pocket clip, such as the Eagletac D25A. What I mean to say is that it's too big to be a keychain light (for most) but it also lacks a pocket clip to chain it to the inside of my jacket.

    With that, a light like this is immediately battling a few other very interesting lights for pocket space, like the Eagletac D25A Clicky Ti, Zebralight SC52, Spark SG3 and the Olight S15-Ti. Apart from price and weight, it's not winning many of those battles.

    But I do wish more manufacturers picked their modes as well as this light has them, and more of them used the Nichia 219. Maybe I can still get myself to pick up a few L10's, just because of the shiny colors and nice emitter

  26. #26
    Flashaholic MojaveMoon07's Avatar
    Join Date
    Sep 2009
    Location
    U.S.
    Posts
    463

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, VIDEO

    Quote Originally Posted by TweakMDS View Post
    Maybe I can still get myself to pick up a few L10's, just because of the shiny colors and nice emitter
    I have the orange L10 219, and the orange is very handsome. Except for one speck, the color on the body looks brand new even though I have used this daily for the last two or three months. The color on the head, however, is about 50% gone.

    Quote Originally Posted by TweakMDS View Post
    With that, a light like this is immediately battling a few other very interesting lights for pocket space, like the Eagletac D25A Clicky Ti, Zebralight SC52, Spark SG3 and the Olight S15-Ti. Apart from price and weight, it's not winning many of those battles.
    The S15 and the Spark SG3 and SG5 have mode memory. One of the reasons why the L10 219 is so useful to me is that it always starts on low. I would prefer it if no manufacturer ever used mode memory.

    For my needs, there is an important difference between the L10 219 and the D25A and S15. For an every day carry, I don't want to own a AA flashlight that does not have (a) if not a setting of approx 50 lumens, then a setting of approx 30 lumens and (b) a low of between 1-3 lumens. For me the lumens at (a) and (b) are so useful in my daily routines that they are my most used levels of brightness.

    Using selfbuilt's measurements in his (review) of the D25A2, the first group starts at 8.9 lumens and then jumps to 90 lumens while the second group starts at 0.2 lumens and then jumps to 65 lumens. In his (review) of the S15, the S15 starts at 0.12 lumens, jumps to 8.4 lumens, and then jumps again to 110 lumens.

    Regarding the SC52, I'm discouraged by how many people here have had a green tinted $ 64 SC52w. In complete honesty, I don't think I've seen even one testimony about something off about the tint of the L10 219; this $ 25 flashlight seems so far to have had 100% consistency in the tint among every customer that has posted on this forum and another. My family owns flashlights in the $ 60-something range; between now and when I joined this forum, I still regard $ 60-something as a premium-priced flashlight. By that I mean I have high expectations which is why I regard the SC52 tint lottery as not just disappointing but very disappointing.


    Quote Originally Posted by TweakMDS View Post
    What I mean to say is that it's too big to be a keychain light (for most) but it also lacks a pocket clip to chain it to the inside of my jacket.
    There is a flashlight holster that I have that I think is perfect for this size AA flashlight and for AAA flashlights. At batteryjunction.com , search for "Ultrafire holster" and you'll find a result for a "UltraFire Belt Clip Flashlight Holster - fits Olight T10 & T15 and similar small lights". It seems well made, it's light, has a wide sturdy belt loop on the back of the holster, has a loop at the top as a second way to carry the holster, and is barely bigger than the L10 so there's no bulk created by carrying the L10 in it. What I'm offering is an imperfect solution to your situation, but because of the aspects of it that I listed it might make your situation manageable.

    What I also like about the size of the L10 is that I can carry it in the side pouch of my Case Logic TBC-403 (I keep my cell phone in the main pouch) along with a Thrunite Ti2 and still have room for a third similar sized light.

    On high the L10 219 projects light to the same distance that my 4sevens MiNi AAx2 XPG on high does when I have to take trash out to the dumpster in the alley at night behind our residence.

    After daily use of the Ti2 and the L10 219, I see the AAA, a flashlight the size of the L10, and a full fledged EDC AA as three separate categories. These two fit my needs so well that a full fledged EDC AA is something that I turn to only for what for me are out of the ordinary situations that those two aren't right for.

    On a misc. note, I can comfortably hold the L10 pinned under my chin against my chest if I need my hands free.
    Last edited by MojaveMoon07; 12-04-2013 at 01:35 AM.

  27. #27
    Flashaholic doctordun's Avatar
    Join Date
    Jan 2006
    Location
    Red Oak, Texas
    Posts
    306

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, VIDEO

    On a side note, you can order a clip with the light for $1.50. It is not very robust, but works fine for me in my shirt pocket. The L10 is so light, that I hard know it's there. Oh and I love the Nichia 219.

  28. #28
    Flashaholic* EngrPaul's Avatar
    Join Date
    Sep 2006
    Location
    PA
    Posts
    3,679

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, VIDEO

    I have 5 Cree XP-G2 and 3 Nichia 219. All four-mode.

    I have noticed a definite trend:
    *The LED's on all Crees are off-center, but the beam is acceptable.
    *The LED's on all Nichias are centered quite nicely.

    I could see dirt on the emitter lenses of some of them with the naked eye. Under a microscope I discovered it's unmelted solder paste and is also smeared elsewhere around the base of the emitters. There are also small "curly fry" shavings of metal and white plastic adhered to the lens.

    One of my samples had the front glass o-ring sticking up on the air on arrival. I could not tuck it in, so I had to pay to ship it back and wait for a replacement.

    A variety of output flux on the lower two levels within each group. I expect this may be partly due to the Vf of the emitter, partly due to variations in the driver.

    Gorgeous anodizing and an overall good value, especially the Nichia version which takes the advantage in beam spread and tint. I consider the pocket clips a must-have.

    Thanks for the runtimes and the review.
    Last edited by EngrPaul; 12-07-2013 at 03:00 AM.

  29. #29

    Default L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, V...

    Pocket clip is a must have if you are ready to glue it in place with some epoxy... otherwise is too easy to come off to rely on it.

  30. #30
    Flashaholic doctordun's Avatar
    Join Date
    Jan 2006
    Location
    Red Oak, Texas
    Posts
    306

    Default Re: L3 Illumination L10 (XP-G2 R5 or Nichia 219, 1xAA) Review: RUNTIMES, BEAMS, V...

    Quote Originally Posted by riccardo View Post
    Pocket clip is a must have if you are ready to glue it in place with some epoxy... otherwise is too easy to come off to rely on it.
    Actually I used what is called a removable glue dot under the clip and it isn't going anywhere. What is nice about the glue dot is that it is easily removable.

Page 1 of 3 123 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •