Emisar D4V2 Review Supplement (the pickier details)

I have had my D4V2 for a while now (since April 25, 2020) and I want to contribute some points about this flashlight and the options I got that are not well discussed in the reviews I have seen. I tend to be a detail-oriented perfectionist, so I am perhaps more concerned about certain details than most. Hopefully others like me will find this helpful. I will pass over the main points of the flashlight that have been well discussed. Overall, I am well pleased with the look, feel, and performance of this, my first enthusiast light. I am very much enjoying the flashlight itself, as well as the Andúril interface, which is a lot of fun to play with and has many fun modes (I particularly enjoy using Candle mode to light my bedroom in the evening while I listen to an audiobook). I have found Andúril to be intuitive and easy to learn considering its complexity. I also love the RGB auxiliary LEDs, and the tiny, comfortably pocketable size of the flashlight.

If you would like any further information about the flashlight, feel free to ask. Here are the main reviews I have read/watched:

The flashlight and flash kit arrived in Indiana, USA from China 16 days after they were shipped (I ordered them directly from Intl-Outdoor). I selected the Sand color, the raised switch retaining ring, and the SST-20 4000k 95CRI emitter. I also got the following accessories: 10623 floody optic, SS bezel, pocket clip, and magnet in the tailcap. Except for the optic, I don't have anything new to say about the accessories (and I haven't used the clip), but feel free to ask about them. When you swap the bezels, don't forget to move the glass lens (and transparent O-ring) to the other bezel. It was so clean and transparent, I almost didn't notice it still mounted inside the bezel.

The flashlight works perfectly, and so does the flash kit, which I used to update Andúril. I checked the thermometer and it was already calibrated correctly. There was no documentation at all, but that is not a problem as I know how to find it online, and I have no use for paper copies anyway.

Beam Quality, 10623 Floody (Frosted) Optic, and Aux LEDs
My one (minor) complaint about this flashlight is that I do not much care for the beam quality with the default optic. (It's a shame because the crystal clear optic makes the aux lights sparkle beautifully.) The hot spot has subtle but, for a perfectionist, distracting and unappealing rays protruding from it. This is caused, I suspect, by the fact that the small size of the flashlight's head forces the optic's lenses to be too crowded against the mounting pins and requires them to have cutouts in their sides to accommodate the mounting pins (see image above). The injection molding channels also intrude into the sides of the lenses slightly. The D4SV2 is larger in diameter and does not have these cutouts. (Some sacrifices must be accepted to have four LEDs in such a tiny flashlight.) Moreover, the flood has subtle rings in it and is a bit dimmer toward the outer edge than I like. (Note that if you are not a perfectionist, and don't examine and critique your flashlights' beam characteristics, you will almost certainly not notice or care about these beam artifacts. They are not at all bad.)

However, I bought the "10623 floody optic" with the flashlight, since I knew I'd want a floody beam, and it removes all beam flaws and gives me exactly what I want. Contrary to what the name suggests, the "floody" optic does not focus differently or produce a wider flood (the flood is already nearly 180°), it just has a frosted texture on the outside that smooths out the beam profile, greatly improving it in my opinion. It doesn't hurt the throw much. The rays around the hot spot are just blended together into the corona, the hot spot is blended outward slightly into the corona, and the flood artifacts are smoothed out. The result is a beam with no artifacts or significant contrasts anywhere, with a hot spot/corona that blends very smoothly into the flood. The flood is made just a little brighter and more useful out near the edges, and so even on dim settings it gives me good peripheral vision. Yet it is not wastefully bright. There is still a good concentration in the center. I highly recommend getting that optic.

To remove the optic, unscrew the bezel, then tap the head of the flashlight against something hard to jostle out the optic (it fits fully and somewhat snugly inside the head and does not fall out easily, and there is no way to get a grip on it unless you use a suction cup).

The SST-20 4000k 95CRI emitter gives what was at first a surprisingly warm light, but only because I was accustomed to cooler lights. I quickly became acclimated to it and I like it. Now my Fenix LD12's 5500k, slightly yellow-tinted emitter, which used to appear pleasantly warm to me, looks cold (though not unpleasantly so) and I perceive the SST-20's light to be neutral (unless I have been using the Fenix for a while and then switch back to the Emisar, which makes the Emisar's light look very warm for a minute until my eyes acclimate to it again). I notice no color tint at all, even at the lowest brightness. It is just a pleasing, slightly golden white.

Comparing with the Fenix, colors do look in general more vivid, and of course, much warmer in the SST-20's light - more like they do in sunlight. As reported by others, reds and browns in particular look much richer and more accurate in the 4000k, high-CRI light. I at first thought green plants looked prettier in the Fenix's light, but while they do pop more, and look by contrast duller and yellower in the SST-20's light, I realized after paying attention to what I was seeing, and to how I perceived the same colors during the day, that the Fenix's cold light emphasizes the blue in plants and mutes the yellow and brown, which in a way looks prettier, but also unnatural. The SST-20 makes greens look approximately like they do in sunlight, and they appear natural and beautiful when I haven't just been comparing to the Fenix. (It's interesting how the eyes auto-acclimate to whatever light is available, making it seem normal until compared to a different type of light. It's an example of how important frame of reference is to experience. This is analogous to many, perhaps all, modes of perception and experience, it occurs to me. There are great philosophical implications here beyond the sense of sight.)

The aux LEDs, except for the blue, are bright enough at their high setting that I can easily read small print by them with the light four to six inches from the text (the frosted optic makes this easier as it smooths out the aux LED's artifacts). I was hoping they would also be bright enough to project onto the ceiling for a fun party effect, but while they do dimly reach the ceiling and nearby walls in a dark room (even from the floor), the glow is not bright enough to be much fun, and the red LED barely casts any ambient light at all. For the first week or so I left the aux LEDs on in rainbow mode, but later I decided that I preferred the voltage mode. Either mode makes the flashlight fun to look at on its shelf as I walk by it, but the voltage mode also tells me the battery's status, which is helpful to know.

The bottom of the ramping brightness (moon/firefly mode) is just bright enough that in a very dark environment, with my eyes adjusted to the dark, I can easily see my surroundings and walk confidentially. The aux LEDs on high are about equally bright (except for the red) and I can also use them as a moonlight mode to walk around in the dark. The difference is that they have no hot spot but are pure flood, and so I can not target them on anything.

Using Standard, Low Discharge Rate Cells
I am using standard 18650 cells (Samsung ICR18650-22F 2200mAh) from a laptop battery, which, according to specs, deliver a maximum of 4.4A, way below the recommended 15A. I was concerned that the flashlight might cause these low-current cells to overheat, but even after five minutes on turbo (resetting to turbo twice near the end after the flashlight throttled down), the cell was only slightly warm, which was almost certainly due entirely to the heat of the flashlight body. Stepped Level 7, which draws 6.06A according to 1Lumin's review, also did not make the cell more than slightly warm after a 20 minute test. So while I do not know what, if any, long-term effect there might be on these cells from drawing that much current, they appear so far to be performing very well in the flashlight, and do not appear to be over-stressed. I didn't get this flashlight for its turbo brightness anyway, and I don't anticipate much need of it. I use turbo only occasionally and briefly, just for fun.

Turbo is much brighter than the top of the ramp (I expected the top of the ramp to max out the cells' current limit), and is very bright. I do not have the instruments to measure either the current or the lumens, but I can say that on turbo at room temperature with the flashlight in my hand and using a fresh battery (4.2V), I can see the light ramping down slightly within 20 seconds, and a lot at about 30 seconds (temp limit set to the default 45°C). So it appears that I am somehow getting close to the maximum 3,000 lumens, since the reports say that the heat throttling begins during that time period. That is only the case with a freshly charged cell, though. By the time the cell reaches around 3.6V, it runs so much cooler on turbo that it takes 90 seconds or longer to throttle down. Maybe this would not be the case with a high-drain cell. I have not had the opportunity to try one.

EDIT: I corrected the description of the Turbo thermal ramping and subjective temperature. I previously said it only started ramping down at about 30 seconds, and that it became too hot to hold comfortably at around 50 seconds, and so I adjusted the temp limit to 40° for normal use. It turns out that this was because the internal thermometer was not calibrated, but was reading 8° too low. It was calibrated when I had last checked it. Maybe it lost calibration when I updated the firmware (I can't remember if I first checked the calibration before or after the update). The factory reset corrected the calibration to within a couple degrees. I then calibrated it a little more accurately. Now the throttling happens sooner (as I would expect) and the flashlight does not become uncomfortably hot with the limit set to 45°. With the thermometer wrong, the limit was effectively set to 53°!

Some Runtimes
I was curious about the runtime for the 18650 cells I am using, so I ran a few tests. I ran the tests until the voltage reached about 2.9V, at which point the flashlight could no longer maintain the brightness I was testing, but stepped down to a low level. If I get my hands on a multimeter, I can estimate the remaining runtimes. There is a trend in the four stepped levels I tested, though. Each level down runs about twice as long as the higher level.

Cells: Samsung ICR18650-22F 2200mAh (with 88.4% capacity remaining, or 1945mAh, according to BatteryInfoView while these cells were still in the laptop battery). Emitter Type: SST-20 4000k 95CRI. Andúril version: 2020-03-18. Lumen measurements are from 1Lumen's review. Note that if 1Lumin's amperage measurements are correct (he was using the same emitters as I am using), I should have gotten much lower runtimes on Levels 5 - 7 than I actually got. I don't know what the cause of this large discrepancy is. Level 4 is right on the mark. Maybe ToyKeeper adjusted the highest levels in a more recent version of Andúril. In that case, the lumen measurements are also incorrect.

  • Stepped Level 7 (1212 lm) - 47 mins (I set the thermal limit to 50° and cooled the flashlight with my hands with the help of ice to prevent it from stepping down, and I checked occasionally to make sure it was still at Level 7 (I'm that bored). At 3.1v the battery could no longer deliver the current needed to maintain the brightness for longer than around 30 seconds at a time. When it stepped down, the voltage read 2.9v for a few seconds and then bounced up to 3.1v. This would probably not be the case with a high-amp cell.)
  • Stepped Level 6 (682 lm) - 1hr 25mins (required a higher temp limit than 45° and my hands bleeding off heat to prevent stepping down)
  • Stepped Level 5 (298 lm) - 2hrs 55mins
  • Stepped Level 4 (96 lm) - 6hrs 6mins (this is the lowest brightness at which I can point the flashlight at the bedroom ceiling beside me and still read a book minimally comfortably by the ambient light)

Voltages at Level 4 at each hour of the test, as reported by the flashlight's Batt Check mode (began at 4.2V): 4.0, 3.8, 3.6, 3.6, 3.5, 3.1. A few minutes after the last check the flashlight stepped down automatically (I was reading by its light at the time), and the voltage read 2.9V.

Raised Retaining Ring
I was slightly concerned that the raised button ring would make the button a bit difficult to press with my thumb, but it does not at all. The ring is 15mm (19/32") across the rim, and the end of my average man-sized thumb fits it very well. I have had no difficulty in clicking out the various commands. The "ball" of my thumb (where my thumb-print is) does not fit well, but that causes no inconvenience. I can still click the button; I just need to press harder and so I would have trouble trying to click out long sequences rapidly enough. For those, I merely have to adjust the angle of my thumb a bit, which I probably would do anyway as it is a more comfortable and dexterous position. The button sits well below the rim of the ring (at least a millimeter, maybe two), and I don't think there is any significant chance of it turning on unintentionally in a real-life situation.

Updating Firmware
I screwed up the first firmware update attempt, fumbling the flashlight head and losing connection before the update was finished. The software glitched into an infinite loop of error messages, but I simply closed and reopened the command window, reconnected the flashing connector (holding it more carefully this time), and re-ran the command. It correctly flashed the new version, which I confirmed by making the flashlight report the firmware version (15 clicks, giving the version in "YYYY MM DD" format).

By the way, I encountered some confusion when installing AVRDUDE, so I will describe it in case anyone else has this problem. While first installing it, I installed it into the AVRDUDE directory I created, carelessly deleting the auto-created MHV AVR Tools subdirectory in the Setup dialog's installation path (I changed the installation path to "c:\AVRDUDE" instead of "c:\AVRDUDE\MHV AVR Tools"). For some reason, this seemed to cause avrdude.exe not to be found. And I looked in the directory and it was indeed not there. It is, in fact, in the bin subdirectory, not in the main install directory, and the command window will find and run it if you install the application into the MHV AVR Tools directory that is auto-created. I finally figured that out during one of my reinstallation attempts and then it all ran correctly.

Also, note that while the Emisar D4V2 Flash Kit Instructions here on BLF by Terry Oregon (https://budgetlightforum.com/t/-/57807), provide links to ToyKeeper's Andúril hex files, as TK releases new versions those links will (unless Terry updates them) still point to old versions. I used the base, directory URL of the files (http://toykeeper.net/torches/fsm/) and found a later version to download. The file name you want to look for is in the format "anduril.[YYYY-MM-DD].emisar-d4v2.hex", and if you want the Nichia 219 version, look for "-219" after the flashlight model. UPDATE: Terry added a note about this and a link to the directory at my suggestion.

Don't forget to perform a factory reset after updating the firmware! My flashlight’s thermometer was calibrated correctly when I first checked it, and while I can’t remember, that was probably before the update. I did not perform a factory reset after the update, and I just found that it was reading 8° low! After the factory reset I just performed, the thermometer was correct to within a couple degrees. I then calibrated it a little more accurately.

Factory Reset Procedure (from ToyKeeper):

"Let the light settle to room temperature, then hold the button while tightening the tailcap to connect power. Then keep holding for ~3 seconds until it does a bright flash. After that, the temperature sensor should be calibrated and the light is ready to use."

Wow . Would had taken me an eternity to write . I’m waiting for my 4th d4v2 and still are procrastinating on using my brand new flashing kit . I’m still in the awe phase . Currently testing with the 18350 tube and it’s fantastic . I ’m using the 10623 too but with the xpl-hi .
Have fun .

I have just one D4V2 with the raised ring. A couple observations:

  • The ring was not removable. I can remove the flat rings fairly easily with the tip of the nail file from a Victorinox mini used as a lever between the button boot and the ring. But this technique did not work at all with the raised ring.
  • I found the upper edge of the raised ring to feel a bit sharp. It’s not enough that it would cause any injury, but it felt uncomfortable to me.
  • I used sandpaper to grind down the upper edge of the boot to make it flatter. It’s still high enough to completely shroud the rubber. Entire grinding job was easy. Took about 15 minutes and no disassembly was needed.

Welcome to BLF!

Thank you for your valuable detailed and well organized review.

I got mine modded with the "throw optic" and W2 emitters, and I am amazed at both the power and throw from this instant ultra classic light.

Thank you! This is my first forum thread (I’ve never been much of a participant in any forum), so I appreciate the welcome and encouraging words!

There is a throw optic? Which one is that? And how does it perform? That might be a useful/fun option to have occasionally. I almost bought the D1 in addition to the D4V2, but mostly just as a novelty to play with like a lightsaber. I decided instead to wait and see if there will be a version 2.

Heh, I did spend a few days drafting it, and I made quite a few edits since posting it. I don’t have much else to do lately, but also I enjoy writing in general, and writing product reviews in particular, when I have the right motivation. I always appreciate a detailed product review that covers the finer points of a product that most people probably aren’t picky enough to care about, so I like to do that for others in the hope that someone like me will find it and be glad I wrote it.

[Hmm, these replies aren’t making clear to whom I am responding. This was in response to Lightenzaza.]

Ah, I see how this works. I need to hit the Quote button instead of the Reply button so that the original message appears in my reply. I’m new at this.

Did you get the ss bezel ?

Vinh at SkyLumen offers a slightly throwier optic for the D4 (and maybe other quads?). I don’t think it is produced by Carclo as the one used in the D4 is listed as their option with the tightest beam. Or perhaps he raises/lowers the height the optics sit at to change focus and increase throw?

Out of curiosity, why did you buy four of them? Are you trying out different emitters, or are they for use in different locations, or are you just addicted :wink: ? I am new to the enthusiast flashlight world and I learned about “flashaholics” who collect large numbers of flashlights. I’ll have to be careful myself not to go too far with this. I’ve always loved flashlights, but only got serious about them this year (well, the Fenix LD12 from 2013 was semi-serious; it was my first non-WalMart-grade flashlight). Fortunately, my interest in collecting things mostly phased out a couple decades ago, so I should be ok.

Yes I did. I figured I might as well get the bling, since I’m treating myself! I like the elegant look of it. I also got the magnet in the tailcap (which I like, but I didn’t have anything new to say about it for the review), and I got the belt clip, but I haven’t used it.

Why do you ask?

Took me only a few hours after receiving my alu d4v2 that i wanted the brass version . The ti version was a natural next and then the colored ones were a must . This must look crazy . Craftmanship is very nice . The design and smallness is calling me . The ui is my preferred one to date. Using the right batteries you can have an impressive turbo . The 18350 tube create an interesting substitute . The aux lights are nice . The switch is good . And the price is ok .

Because the look is much better with it, with the raised ring as a recall .

I see. The Ti does look nice. And I really like Anduril, too. It’s a lot of fun, and easy to learn. It looked complicated when I read about it, but I had a feeling I would get the hang of it quickly, and I did. I’m a programmer and computer techie, so I’m pretty good with this sort of thing, and ToyKeeper did a great job designing it.

I should have paid more attention to your signature. That answers my question! #1, #3, and #4 in your collection were ones I considered before the D4V2, and Matt Smith’s FW3A review on YouTube was my introduction to BLF and the world of flashlight enthusiasts!

Since you’re a collector, and own the FW3A, can you tell me if the somewhat rough beam profile with the rays protruding from the hot spot is common to compact flashlights using an optic rather than a reflector? Is that something more experienced buyers would have expected? I actually read one user review that mentioned that, but nobody else has, so I’m guessing it’s either to be expected, or most people don’t care. How is the FW3A’s beam?

The beam of my fw3a is more round then the d4v2 with regular optics . In fact it’s almost perfect circle . But this affects only your white walling activities ; corners won’t be obvious in regular usages .

It is common in triples/quads with clear, throwy optics; and to a lesser extent in lights with triple/quads using reflectors too. The FW-series has a much smoother beam because the stock optics are frosted. With clear, narrow-spot optics they will have artifacts in the beam - at close range - as well.

The D4V2 comes with a Carclo 10622 optic. It features a beautiful beam pattern, but not quite as much throw.

Carclo also has the 10621 optic which is throwier, but has an uglier beam pattern with more rings. The 10621 quad is basically the same optic shape as the popular Carclo 10507 for triples and produces a similar beam pattern. My guess is Vinh is swapping in the 10621.

I’ll be damned, always though the D4 came with the narrow-spot clear.

My (minor) gripe about the Emisar D4v2 (and D4S, etc) is the way the lanyard hole is open to the back. Tailstanding the light requires the lanyard to be pulled away from the tail in a two-handed operation, otherwise the light will invariably sit on the cord and be unstable. A lanyard hole accessible from the side, like on this Skilhunt, is a better solution.