Review: EagleTac D25A Ti 219 (mid-2014 botch, er, batch)

Official product shot:

(I’ll leave other pictures until later)

The D25A Titanium (with Nichia 219B) is a special-edition light from EagleTac, made once per year. This is the 2014 edition, though it’s from a later batch than the initial run. I ordered mine in mid-January but it seems EagleTac promised more than they could deliver and ran out of stock. I didn’t receive mine until mid-June.

The D25A (especially the titanium special edition) is considered by many to be one of the top EDC lights ever made. It’s basically the same as the regular D25A, but with a titanium body and a white glow-in-the-dark tailcap.

Specs from the Manufacturer:

This is what EagleTac has to say about the D25A-219’s performance:

  • LED: Nichia 219, B11 bin, 92 CRI, neutral white 5000K
  • Output levels / modes: 3 main levels plus turbo and 7 semi-hidden blinky modes.
  • Light output:
    • Main low group (moon mode enabled) : 0.5 lm (150+ hours), 7 lm (20 hours), 55 lm (2.5 hours) OTF
    • Main high group (moon mode disabled) : 4 lm, 20 lm, 55 lm OTF
    • Turbo: 88 lm OTF (1.3 hours)
    • Constant current regulation (non-PWM) for all output modes
  • Other modes: Strobe, Strobe(Var), Flash(Hi), S.O.S (Fast), S.O.S., Becaon, Flash(Lo)
  • Battery: 1xAA, Eneloop, or 14500 cell, 50-51.2mm long. EagleTac also notes, “When using single 4.2V 14500 (protected circuit is highly recommended), middle output mode will become high output mode.” However my protected 14500 cells are an extremely tight fit due to the length.
  • Interface: Reverse clicky plus tighten/loosen head to change mode groups.
  • Dimensions: 3.4 inches (87 mm) long, 0.7 inches (17.5 mm) head diameter, 0.65 inches (17 mm) body diameter.
  • Lens: Impact resistance waterwhite glass lens, anti-reflective (AR) coating on both sides, user replaceable.
  • Warranty: Ten years performance guaranteed.

Visual comparison to other similar lights

Here is the D25A next to some other 1xAA lights, for size reference.
From left to right: Thrunite T10T, EagleTac D25A Ti 2014, Zebralight SC52, L3 Illumination L10, AA Eneloop.

And again, next to a wider variety of lights. It’s worth noting that the T10T and D25A are actually somewhat close to the size of a small 1x18650 light. From left to right: Convoy S7 w/ added SF L2 clip, CNQG brass beauty, Thrunite T10T, EagleTac D25A, Zebralight SC52, JETBeam RRT01, L3 Illumination L10, AA Eneloop.

Measurements

I measured the D25A I received, and here are the results:

Tint: About 4300K or 4400K. The spec said 5000K.

D25A-219 low / “moon-mode enabled” group:

  • Low (spec=0.5lm): 4.67 lm, current-controlled
  • Med (spec=7lm): 18.5 lm, with PWM
  • High (spec=55lm): 93.5 lm, current-controlled

D25A-219 high / “moon-mode disabled” group:

  • Low (spec=4lm): 12.4 lm, with PWM
  • Med (spec=20lm): 24.0 lm, with PWM
  • High (spec=55lm): 93.5 lm, current-controlled

Turbo:

  • Turbo (spec=88lm): 148 lm, current-controlled
    On a 14500 cell (my protected Sanyo 840mAh cell just barely fits if I force it), I got the following (low / moon-mode-enabled group):
  • Low: 4.52 lm
  • Med: 149 lm
  • High: 382 lm
  • Turbo: 402 lm (same as my HD2011-219B on turbo w/ 18650 cell)

The PWM appears to be approximately the same speed as the popular nanjg drivers, probably around 4.5 kHz.

Regular D25A Measurements

I measured another D25A, a cool white XM-L2 model that I think is from 2013. It’s much closer to spec.

Tint: Not sure, blue-ish white with a primary blue halo around the outside of the spill. So, definitely cool white.

Lumen specs are listed as OTF lm / emitter lm, since EagleTac lists both on their site.

D25A-XML2 low / “moon-mode enabled” group:

  • Low (spec=0.5lm/1lm): 0.43 lm, current-controlled
  • Med (spec=9lm/11lm): 12.1 lm, with PWM
  • High (spec=85lm/121lm): 121.9 lm, current-controlled

D25A-XML2 high / “moon-mode disabled” group:

  • Low (spec=4lm): 5.86 lm, with PWM
  • Med (spec=20lm): 23.62 lm, with PWM
  • High (spec=85lm/121lm): 121.9 lm, current-controlled

Turbo:

  • Turbo (spec=141lm/200lm): 201.4 lm, current-controlled

This was, mostly, pretty close to the “emitter lumens” values specified on EagleTac’s site. Their “OTF lumens” specs seem low though, compared to the ZL scale my light box is calibrated to.

Discrepancies

So, I noticed some pretty large differences between the official D25A-219 specs and my own measurements.

Lumen output
There is no moon mode. Not even close. That mode is so far off spec that it’s actually brighter than the “low” mode on my other lights.

My light box is calibrated to a Zebralight lumen scale. This is known to be significantly different than the EagleTac lumen scale, by a factor of 40% or 50. In other words, 100 EagleTac lumens = 140 to 150 Zebralight lumens. So, I expected my measurements to be about 40 brighter than the spec.

But instead, my measurements showed values as high as 834% higher than spec (in “moon” mode), and as low as 68% higher than spec (for turbo).

The 14500 performance is not as EagleTac described: “When using single 4.2V 14500 (protected circuit is highly recommended), middle output mode will become high output mode.” In reality, the middle output mode becomes turbo, high becomes super-turbo, and turbo becomes a marginally brighter super-turbo. Oh, and my protected cells don’t fit. It’s longer than the 51.2mm maximum length spec and I was almost sure I’d break something if I twisted it far enough to activate turbo.

Tint
Compared to my other lights, the tint appears to be about 4300K or 4400K. The nearest tint neighbors I can compare against are:

  • A ZL H51w, spec’d at 4200K (seems accurate).
  • A ZL H52FW, spec’d at 4400K (and I believe it).
  • Two L3 L10-219 units which both get a pretty consistent 4500K.
  • Two custom-made wide-spectrum lights centered at about 4600K and 4900K.
  • Two custom-made Nichia 219B-based lights, at around 4800K to 5100K.
  • Two XP-G2 3B lights at about 5200K.

The tint seems very close to the ZL H52Fw, but yet looks pretty yellow compared to a L3 L10-219.

As far as I can tell, the color temperature is ~4400K, but (and this is a big but) the CRI is nowhere near 92 so it looks a lot less white and a lot more yellow. The emitter inside is definitely a a Nichia 219B and I believe the emitter itself probably puts out a 5000K beam with 92 CRI. However, the D25A ends up much warmer and less white.

Why? I suspect the culprit is that double-AR-coated lens.

This D25A has a very unusual quirk: It has a bright blue secondary corona around the outside of its normal beam. I think the lens may actually be splitting the different wavelengths of light apart, bending blue to the far outside and letting yellow through un-hindered. This gives it a yellow, lower-CRI beam and an annoying blue halo… which kind of defeats the entire point of using a nice high-CRI emitter.

Other discrepancies:
Despite claims that the lens is user-replaceable, it appears to be solidly fixed in place between the bezel and the reflector, meaning I’d probably have to remove the pill to change the lens. However, maybe it’s possible to unscrew the pill if I use pliers padded with rubber or leather. And if this is the case, I’m pretty sure I want to replace the lens with a regular piece of glass instead of this funky CRI-killing blue thing.

EagleTac says “Constant current regulation (non-PWM) for all output modes”, but I found rather obvious PWM on 3 of its 6 main output levels.

EagleTac describes the emitter as being from the “B11” output bin. However, it’s a Nichia 219B, not Nichia 219A. The 219A went up to a B10 output bin, and the 219B uses a different naming scheme entirely. I believe the name of the bin is actually called “D200”.

Other quirks

This D25A has an X-shaped beam pattern. Most lights use a reflector to collimate their light, and the resulting rays move forward (ideally) in parallel, shaped like ||. Some lights go for a floodier look though, and use a V-shaped beam. But the D25A’s photon rays actually cross each other like an X about 70mm to 90mm in front of the lens. This makes for a strange beam pattern, especially when used up close. Any closer than that and it has a donut hole, or at 70-90mm the hotspot has a secondary hotter spot in the middle, and beyond 100mm or so, the hotspot finally starts to smooth out.

Fresh out of the box, the light was clearly untouched since the lens didn’t even have a speck of dust on it, nor any sign it had ever been touched by skin. But the knurling was … dirty.

Pictures

The light, from a side view.

This is the same shot, cropped to better show the knurling. It’s brand new, yet somehow looks… dirty.

Beam pattern
Here’s the beam pattern, showing off the bizarre blue corona around the outside of the spill:

And a couple from behind:

I wanted to see where the blue-ness was coming from, so I took a few pictures from inside that blue corona. These four pictures gradually move the camera through that part of the beam:



I also took a couple pictures directly into the light from up-close, to see if any weird colors could be seen. Something around the outer edge is also causing some blue-ness:

Moon mode comparison
The lights in these shots are, from left to right:
D25A Ti 219, XinTD C8 V4, ZL SC52, L3 L10-219, Thrunite T10T NW w/ DC-Fix

But the usual three-exposure approach really doesn’t show the difference in brightness very well. So, I turned on the overhead lights and took another picture so you can see just how much difference there is between the D25A “moon” mode and … well, pretty much anything else on the market:

Beam shots compared to other lights:
The D25A will be on the left side in each of these pictures. Also, I used the “fluorescent” white balance on my camera, which shows things pretty closely to how my eyes see them. If it looks yellow in the picture, it will generally look yellow to me in person too.

First, let’s see how the D25A Ti 219 looks next to a L3 L10-219. Three shots, with different exposure times.

The D25A’s blue halo is pretty visible in the first couple shots. Also note, on the right, the L3 L10’s subtle halo is visible too. It has a very slight secondary corona caused by off-angle reflections off the very bottom edge of the reflector, at the narrow end. It’s nothing compared to the D25A’s massive bright halo though.


And here the warmer tint of the D25A is obvious:

That last shot was a bit close to the wall, so let’s see it again with both lights pulled back a bit:

Okay, now compare the D25A to a Thrunite T10T NW. Two exposures, and then the third is the same but pulled back from the wall a bit more. The T10T seems throwier with a cleaner and brighter hotspot.

And farther away from the wall:

Now, somewhat by accident, I ended up with some DC-Fix stuck over the bezel of my T10T. I was using it as a cutting template for another light, it didn’t fit, and I stuck it there for temporary storage and forgot about it. When I tried the light later, it produced a beam that I think I like a lot better than its usual beam. This shows the diffused T10T NW on the right, but is otherwise the same setup:

Glow in the dark button
One neat thing about the D25A Ti is that it has a nice glow-in-the-dark white rubber switch cover. It’s pretty nice, as far as GitD items go, and glows much better than the cheap tailcap buttons on most lights.

So, in running-gag BLF style, I bring you… The End.

After a fresh charge:

Final thoughts

Okay maybe that wasn’t really the end.

Pro:

  • Nice titanium body.
  • Good screw-on clip.
  • Easily bright enough, and has a decent “low” and “medium” mode.
  • Has a Nichia 219B inside.
  • The D25A Ti has an unusually nice white glow-in-the-dark tailcap. Not quite as cool as if it were also titanium, but better than any other GitD button I’ve seen. It glows almost as well as the Zebralight GitD rubber headband.
  • Small-ish size, though still larger than I expected.

Con:

  • No moon mode, despite promising one.
  • Most levels way out of spec. I really wanted moon, ~3lm, 20-30lm, and 100+ lumens. I don’t care about any other modes.
  • Not actually high-CRI; looks yellow instead of white.
  • Not actually 5000K; more like 4400K.
  • Bright blue ring around the outer edge of the spill.
  • Doesn’t actually fit protected 14500.
  • Awkward, inconvenient method to switch between mode groups. Actually, the fact that it has mode groups at all is kind of a downside, since that means my two most-used modes are in different groups (or they would be, if it actually had a moon mode).
  • The blinky modes are pretty awful. I was hoping it would at least have a decent beacon mode (low-brightness blink for ~2ms with 2 or 3 seconds of blackness in-between, run time measured in months, good for finding the light in the dark)… but no, every last blinky mode on it seems useless to me.
  • Took 5 months to arrive, due to EagleTac promising more than they could deliver and then apparently having trouble making a new batch. I kind of wonder if they eventually gave up and shipped anyway, despite problems in this batch.

I think what I really wanted was a bare-metal L3 L10-219B with a screw-on clip. That would be just about perfect. I thought the D25A Ti 219 could give me something very close to that, but it has been pretty disappointing. The Thrunite T10T NW I received last week is actually a much closer fit for what I wanted, despite its significantly larger size, mode memory, and lack of a “low” mode between moon and medium.

If anyone else has a D25A Ti 219, or even a non-titanium D25A-219, I would really like to know if your light has the same issues as mine. Specifically, the lack of a moon mode, the blue ring around the beam, and the lower-CRI yellow beam.

Update:
I got an aluminum D25A which has a removable lens. If I hold that lens in front of another Nichia 219B light, it makes the tint warmer and yellow instead of white. From this, I conclude there is basically no point getting a high-CRI EagleTac light until they fix their lenses or provide one without AR coating, since the lens destroys the CRI.

Thanks for the review! A pity that this light does not perform as you hoped, I like it very much.

I have this same light, but the 2013 xp-g version, and that one can toggle between two low modes, one pwm-controlled, not really moonlight, and one much lower one without pwm. To toggle, with the head loose you cycle three times through the modes. I wonder if that is build into your light also.

Yes, to switch between the “moon mode enabled” mode group and “moon mode disabled” mode group, the process is: loosen the head a bit. Turn the light on. Quickly tighten and loosen the head three times, making sure to go all the way to turbo and back each time.

This is how I obtained the two different groups of measurements in the review.

The light can also toggle mode memory, by doing sort of the opposite. Tighten the head, turn the light on, then loosen-tighten three times. I haven’t tried this though, since I don’t want mode memory. (or so the manual says, anyway… but I just tried it and it doesn’t seem to enable mode memory on either the regular or turbo modes)

Edit: Though the printed manual said it had an option for mode memory, the web site says this is D25C only.

I think this would be a much, much better light if I were to replace the double-AR-coated lens with a plain sheet of glass (assuming I can get the lens out somehow). It’d also be very nice to have a moon mode, but I don’t think that’s something I could fix on my own.

Just guessing away (I'm a mere onlooker on what the electronics guys are doing at blf), the moonmode is current regulated, so perhaps there is a physical path on the circuit board taking care of that, and a resistor might be replaced by a higher-ohmed one to lower the moonlight?

Thanks for the review, ToyKeeper! I had considered one of these until I heard of it having poor anodizing.

The blue halo, is the lens AR coated? I get the same halo on my Rustu 108S after switching the lens to AR coated (it is coated on both sides). It only appears when the light is pretty close to the wall:

Chloe, yes, the lens is AR-coated on both sides. I think that is probably the cause of the blue ring and lower-CRI beam. The AR coating may be acting like a prism, splitting different wavelengths of light to emit at different angles. On a high-CRI light, this is a Bad Thing. It seems to lower the CRI by like twenty points, to the level of a typical Cree LED.

I have other lights with AR-coated lenses though, and none of them make a bright blue halo. The worst is my JB RRT01, which has an extremely faint tertiary corona with a blue-ish tint, but I hadn’t even noticed it was there until I went around checking all my lights for halos.

djozz, you’re probably right about the resistor but I don’t think I can get at the guts of the driver. I don’t even know if I can get the pill out. I’m hopeful that I can at least swap the lens for something less blue though, since the specs state it has a user-replaceable lens. I think that means the pill is supposed to unscrew, though with just my hands it won’t budge. It’s probably glued in place.

I’ll probably try to contact EagleTac before attempting to break the glue. I’ve also asked the vendor if they might be able to check other units for the bright moon mode and blue ring, but I’ll understand if they don’t want to turn mint-condition stock into tested-once stock.

Thanks for the review ToyKeeper. If you manage to get the head off, I’d be curious to hear what the tint is like with the lens removed. Do you think the bezel depth could have anything to do with the blue corona you’re seeing?

Also, I’m sorry to hear the moonlight mode is as bad as it is. I don’t expect something as good as a Zebralight in terms of output and runtime, but I was hoping it would at least have a sub-lumen moonlight mode, especially considering it’s a limited edition light.

About the blue ring, it is on my ET too. I don't think it affects cri in a negative way. It is a bit of double reflection happening to a minor portion of the light that strikes the lens at an angle (light coming direct from the emitter going to the edge of the lens), almost all of the light (more than with a non-AR coated light) , and especially light coming from the reflector forming the hotspot, just goes through the lens unaffected: all wavelengths equally well.

Thanks for the review. I got the Ti XM-L2 version and I’m not sure what to make of it. Mine won’t work on regular Eneloops and only works on 14500, which is OK I guess.

The moonlight mode on mine is only about 5 lumens, so it’s still acceptable as a low-low to me. It’s also got an ugly, ringy beam with a blue-ish tint. The output is impressive, though, and it’s super-lightweight. And of course they make the best deep carry clips of any manufacturer. So I have definite mixed feelings about it.

Still trying to get a good stacked focus shot of mine:

You could try cleaning the knurling with a mild abrasive like toothpaste or jeweler’s rouge and a small brush (if you will keep it, with all the flaws).

Thanks, it’s interesting to hear that other units also have the no-moon and blue-ring issues. Sigh. I’m glad mine works with Eneloops, at least. No Eneloops would be an immediate deal-breaker.

Also, how many pictures did you stack? Looks like maybe about 8 images?

My early 2012? version ones on moonlight mode w/ a AA do what a zebralight h31 does on it's lowest low . On a 14500 I like how it becomes a 2 mode light . it runs about 25 lumens low and 600+ on high . Even though i know it's coming I'm always impressed when I change modes .. If I had to change one thing I'd change tint and cri .. overall I like them.

Thanks for the review.

My early 2014 Ti XML-2 Nichia clicky is warmer than my HDS 170 Nichia, which seems more whitish by comparison.

No blue tint that I can see.

No moon mode. But low suits me fine.

EagleTac 14500 750 mAh fits like a glove.

It’s a great little back-up light for pocket-carrying as long as you don’t need to run it on high for more than a few minutes — they do warm up fast!

I was expecting something more like the 2012 version; reppans posted a pic of its moon mode compared to other lights. It’s a bit on the bright side (maybe 0.6 lm), but it was still under a lumen and pretty close to spec.

(from Zebralight H52w / H52Fw / H52 / H52F review )

that’s a great review.
i don’t think the weight diff justifies
the price jump from alum to titan.

i EDC an alum d25a xp-g2 and i can’t find anything about it
that i would change. i love the ease of twisting the head to get to the hidden modes. i love the beacon mode way at the end of all the hidden modes. i think the beacon mode can flash for months. it is so small that sometimes i can’t find it in my front pocket. why carry an AA light that’s big as a 18650 light ?

i have an ultrafire AA light that has the same head twisting -
mode changing feature. it would be a good host

My 2013 D25A (NOT Ti) has the moon mode.

It also has the purple/blue halo, but it is very minor. Much less than what you show in your photos.

Anyone know how to unscrew the head on a D25A? EagleTac support told me to try that to see how the beam looks without the lens, but it’s glued in place and the usual padded-pliers trick didn’t work.

I think I’ve heard about melting the glue first with heat, but I’m not sure how to heat it up without causing any damage. Just hold a clean soldering iron on the head and slowly rotate to different spots to heat somewhat evenly all around? Maybe roll it around on a clean pan on the stove/range? And then, I’m guessing, pick it up with gloves on and try to unscrew it by hand? How likely am I to damage parts inside, like melting the lens O-ring or damaging the emitter or unsoldering things?

Pour boiling water over the head from a kettle, maybe put it in a ziplock first if you’re not confident about waterproofing.
Some people boil the whole thing in a pan after freezing it.

I applied a blow torch (moderately) on the head of my MaratacAA copper to loosen the pill, it did not damage the led, o-ring or electronics.