[selfbuilt]

Fenix TK45 Review (XP-G R4 Testing Sample): RUNTIMES, BEAMSHOTS and more!


Reviewer's Note: The pre-release Fenix TK45 was provided for review by Fenix and the Canadian Fenix distributor Notosora International Imports Ltd. Please see their websites for more info. Note that my review is of a pre-release engineering version sent for beta-testing.

Warning: pic heavy, as usual

Note: Fenix has sent a number of these out to CPF members to review, so I urge you to check out other reviews posted here (especially for the additional beamshots).

Manufacturer Specifications: (taken from the Fenix manual and website).

• 3 x Cree XP-G LED (R5) (note: review sample is R4 output bin)
• Four types of lighting mode: Turbo (760 Lumens, 2Hrs) -> Low (8 Lumens, 232Hrs) -> Mid (95 Lumens, 22Hrs) -> High (312 Lumens, 6.5 Hrs) (Tested with 2500mAh Ni-MH battery)
• Three types of flashing mode: Strobe, SOS, Warning flash
• Max beam length: 200 meters
• Digitally regulated output - maintains constant brightness
• Intelligent memory function
• Outstanding heat sinking performance offers long working time in Turbo mode
• Uses eight 1.5V AA (Ni-MH, Alkaline) batteries
• 202mm (Length) x 39.2mm (Diameter) x 56mm (Head)
• 324-gram weight (excluding batteries)
• Made of durable aircraft-grade aluminum
• Premium Type III hard-anodized anti-abrasive finish
• Waterproof to IPX-8 Standard
• Toughened ultra-clear glass lenses with anti-reflective coating
• Sidewinder dual switch system
• Capable of standing up securely on a flat surface to serve as a candle
• Included accessories: Lanyard
• MSRP: tbd

The TK45 is the latest Fenix addition to the high-output/multi-emitter class of lights. Following up on their earlier 8xAA MC-E-based TK40, the 8xAA TK45 uses three of the current highest output single-die XP-G emitters (R5 in the shipping versions, R4 in the pre-release beta sample tested here). The light is thus suited for those of you looking for high output on common AA batteries.





I have no idea what final packaging will look like – my light (clearly labeled “testing sample”) came in a fairly sturdy plastic case with a “beta testing instructions” sheet. I would expect the regular set of Fenix extras on the shipping versions.

Clearly, the styling of this light is distinctive.




From left to right: Duracell AA alkaline, Fenix TK45, ITP Polestar A6, Lumapower Mentor (2xC), Olight M30 (no extender), EagleTac M2C4

Fenix TK45 Weight: 307.3g (no batteries), 514.7g (with batteries)
Fenix TK45 Dimensions: 202mm, Width (bezel base) 50.6mm, Width (tailcap) 44.0

Overall weight (sans batteries) is not as high as most of my 2x18650-sized high-output lights, but seems hefty enough with batteries installed. Note that the TK45 definitely weighs more than the 6xAA ITP Polestar A6 (~210g without batteries).




The light has anodized screw threads at both ends of the battery tube, but that doesn't really matter since the current is carried through the battery carrier (lock-out is still possible, scroll down for details). Screw threads are thick and square-cut, so they should be durable and stable.

There are no flaws in the black type III anodizing on my sample, and lettering is clear and sharp.

Consistent with the relatively low overall weight, body tube walls are not as thick as most of the 2x18650 high-output lights I have reviewed. But they do seem thicker than the budget 6xAA ITP A6.



Grip is very good on the light. In addition the overall checkered pattern of the handle, each segment has a large number of tiny concentric ring ridges running along. Rather distinctive, this pattern certainly adds to the enhanced grip.





As you can see, the light uses a battery carrier to hold the 8xAA cells. Quality seems ok – but note the wall construction is all-plastic, aside from the contact plates and springs. The batteries also have a tendency to pop out easily.

As an aside, the battery carrier introduces a lot of rattle if you shake the light laterally when its fully loaded. Sounds rather like a maraca!



The TK45 uses electronic switches in the head to control on/off and mode switching. These are conveniently located right below the head. The right switch controls on/off and mode selection, and the left controls output levels within each mode.



As you can see, the light can tailstand thanks to the flat base.




Ok, no doubt about it – this is one unusual looking light.

Fenix refers to this model as the “MINI-GUN” - presumably in reference to the early Gattling Gun design (with its rotating gun barrels). From on top or behind, it doesn’t look so strange – but when looking head-on, the three independent bezels on a common base is … ummm … unique.

My first thought when I saw the design was to wonder if they simply stuck three LD10/LD20 heads on there. But I can assure you that’s not the case:



Each bezel on the TK45 is wider than a standard LD10/20 head, and seems to have a thicker wall diameter. The reflector is also noticeably different – wider and shallower (i.e. not as deep), and with a larger opening around the reflector.

Despite first appearances, Fenix has clearly engineered the head for this light. This give me hope that the beam profile will be reasonably smooth.

Beamshots

To start, here are some “white-wall” beamshots, comparing the TK45 to some of my other high-output lights up close. The distance is taken ~0.5m from a while wall, and is designed to show you the spillbeams:


















First thing to notice is that the new 3-head design of the TK45 does NOT produce any hotspot or spillbeam artifacts up close. This is very impressive.

In my experience, multi-emitter lights that use partially-shared reflectors (e.g. Eagletac M2XC4) produce a “daisy-flower” pattern from the overlapping beams. Even lights that use separate reflector wells in one large unit (e.g. Lumapower MVP 3xCree) tend to have some artifacts. Good job Fenix!

UPDATE: for some additional up-close white wall beamshots, scroll down to post #25.

To better compare the hotspot and spill of the lights, here are some outdoor shots focused on a point ~ 10 meters from the lights. Note that these were taken at different times for different reviews, so they may look a little different (e.g. I planted a tree at the end of last summer ).









And here are some lower exposures to better show you the hotspots:









Clearly, the TK45 is not a great thrower – you wouldn’t expect it to be with this design. But it does have one of the smoothest and most even beam profiles I’ve seen in a high-output light. If you want a "wall of light," this would be a good choice. Frankly, it’s hard to believe this is achieved with three separate heads.

UPDATE: Some additional long-distance beamshots, to show you how the light compares to others in its class.

Please see my recent 100-yard Outdoor Beamshot review for more details (and additional lights).






User Interface

As previously mentioned, the Fenix uses two electronic switches to control on/off and mode selection. These are located just under head, in the traditional location of mainstream consumer flashlights.

The right button is the main switch. Click it to turn the light on/off. The left button is output selection switch. Click it to advance through the output states.

Note that the TK45 has 2 sets of output modes. The constant output mode is the default setting when you turn the light on. Press the left switch to cycle through Lo – Med – Hi – Turbo output, in a repeating cycle. The light has output level memory, and will come back on at whatever level you turned it off in.

The light also has strobe/SOS modes, but these have thoughtfully been relegated to their own mode that you never have to see if you don’t want it. To access these flashing modes, double-click the right button while the light is on. The left button will not cycle you through Strobe – SOS – Warning Flash.




Strobe was measure at high (and annoying) 15.6Hz, and the “Warning Flash” at a nice and comfortable 2Hz.

Note that the left-button-selected output level memory is independent for the two modes (i.e. it remembers both the constant-output level and the flashing level setting separately). And no matter which mode you are in when you turn off, it always comes on at the last memorized constant output mode. So there is no risk of you accidentally strobing yourself if you last used that mode- you won’t see strobe again unless you do a double right-click from on to re-enter the flashing modes.

I am unable to detect any sign of PWM at the constant output levels, suggesting the light is indeed current-controlled like other Fenix offerings.

UPDATE: I don't normally report these, but here are my measured current draws for all output modes, including the stand-by current when off but with the carrier still connected:

Turbo: 1.012A
Hi: 290mA
Med: 77mA
Lo: 5mA
Standby drain: 0.055mA

Testing Method: All my output numbers are relative for my home-made light box setup, a la Quickbeam's flashlight reviews method. You can directly compare all my relative output values from different reviews - i.e. an output value of "10" in one graph is the same as "10" in another. All runtimes are done under a cooling fan.

Throw values are the square-root of lux measurements taken at 5 meters from the lens, using a light meter, and then extrapolated back to estimate values for 1 meter. This will be my standard way to present throw on these types of high output lights from now on. The beams don't really have a chance to fully converge until typically several meters out.

Throw/Output Summary Chart:



As expected, the throw of the TK45 is less than most of the competition. Although the lux meter says it is roughly comparable to the MC-E-based Olight M30, I think you will find that throw is less at greater distances.

What was a bit of surprise to me is the overall output – both my ceiling bounce and lightbox report that the TK45 is brighter than any of my MC-E/P7/SST-50 lights.

In fact, both methods suggest that the TK45 is at least 15-20% brighter than either the ThruNite Catapult V2 or Eagletac M2C4 on max, my previous two high-output champs (well, the SST-90-equipped Olight SR90 doesn’t really count ). Note that part of the issue may be the wider spillbeam pattern of the TK45 (i.e. thrower lights tend to get lower scores in my lightbox and ceiling bounce tests). But these results do match what I see by eye in the ceiling bounce tests - the TK45 does a better job lighting up the room than other lights.

A note about the Lo output – this is surprisingly low for a high-output class light. It also has a funny quirk – only one of the three heads lights up on low. But which one lights up rotates every time you enter the mode (i.e. it remembers the last head to light up, and moves along in sequence).

Output/Runtime Comparison

Here’s how the TK45 output levels compare:



Good spacing of levels, and excellent regulation on all levels on Eneloop NiMH.

I’ve only done the one run on alkalines, on Hi, for comparison purposes. But you can see the excellent regulation and runtimes on all tests. I just wouldn’t recommend Turbo on alkalines, since I doubt the cells could handle that kind of load for long.

Here's how the TK45 compares to other members of the high-output class (most on 2x18650 2200mAh).





Well, this is impressive. Although many enthusiasts would dismiss multi-AA lights when it comes to high output, the 8xAA TK45 is certainly holding its own. More than that – output and runtime of the TK45 is typically greater than any of my 2x18650 lights (when run on the older 2200mAh cells, at any rate). Although again, the floodier nature of the TK45 may be skewing my lightbox results slightly (i.e. thrower lights tend to score a bit lower than floody lights in my lightbox).

Potential Issues

The three separate heads mean there is more area for dirt and debris to accumulate. Not likely an issue for most people, but could be a problem if you plan to use it in rugged environments.

The light uses a battery carrier for the 8 AA cells. The carrier is mainly plastic. The cells also have tendency to pop out easily when you are loading it.

The battery carrier has quite a bit of rattle when loaded with cells in the light. I don’t have the TK40 to compare, but this is more than I’m used to.

Due to the electronic control switches, there is a continuous parasitic drain when the batteries are installed. I am not setup to measure the specifics, but there is an easy work-around - unscrew the tailcap or the head a full turn or more when not in use. Thanks to the overall tolerances in the head of the battery carrier, this is sufficient to cut the current flow.

UPDATE: With HKJ's help, I've managed to get current draw and stand-by drain readings (see post #15-1. The standby drain was measured at 0.055mA, which means your 2000mAh Eneloop cells would be drained in 4.15 years.

Final packaging and extras are unknown at this time.

Preliminary Observations

I will be honest – I wasn’t expecting to be impressed with this light, given the unusual head design and 8xAA configuration. But despite these aspects, and a few other quirks, there’s definitely a lot to like here.

Let’s start with the output – it is brighter overall than any of other 3xCree/MC-E/P7/SST-50 lights in my collection at the moment. How much more is harder to say, given the floodier beam pattern on the TK45, but my ceiling bounce measures do match what I see by eye. The TK45 also lasts longer on 8xAA Eneloops than any of those other lights on 2x18650 (albeit the older 2200mAh cells in my testing).

Of course, max output isn’t everything – so what are the lower levels like? Here again, Fenix efficiency pulls through. I was impressed with the runtime and regulation at all levels (which are well spaced apart). I was even pleasantly surprised to see Hi mode could run so well on standard alkalines (I don’t recommend Turbo on alkalines, since I doubt they could last long).

I also wasn’t expecting much from the beam – in my experience, multi-emitter setups generally have some degree of artifacts in the beam (especially in the corona and the periphery of the spillbeam). But none of that is present here – the beam is beautifully smooth. It really is a true "wall of light."

Of course the flip-side is that throw is relatively low – but typically not much worse than the MC-E/P7 lights that use shallow reflectors to compensate for the dark center void (e.g. the ITP A6 or Olight M30, which use the same reflector).

How does this light compare to the Fenix TK40? Honestly, I can’t say, not having tested the TK40. But from the specs, the output has clearly increased on the TK45. I would expect the TK40 to throw better, though. I would welcome others with direct experience of both to comment further.

How about the interface? Again, I like the placement of the buttons near the head, and the fact that all the strobe/SOS modes are relegated to a hidden mode (thank you! ). I’ve never been thrilled with having to contend with a flashing mode when cycling through constant output levels on other lights. One small point – the buttons aren’t labeled on my engineering sample, so there’s no intuitive way to know which one does on/off and which one does level selection until you give it a try.

So what don’t I like? Well, battery carriers are never fun at the best of times, and this one has enough rattle when fully loaded to make you feel like playing a rumba or cha-cha-cha. The body of the light is also less substantial feeling than most of the members of this high-output class – but it’s still a few notches up from the budget ITP A6. And of course, “beefiness” is largely a matter of preference – as long as you aren’t planning to use the TK45 as an impact weapon, you should be fine.

I also find it hard to embrace the aesthetic of the triple-mount head, but that’s really a cosmetic issue. I can assure you that I certainly like what’s coming out of that head, and that really is the true measure of a flashlight.

Bottom line, the TK45 is a nice addition to Fenix’s high-output line of lights. Not sure if there’s enough here that would cause existing TK40 owners to upgrade (i.e. more output - but likely less throw - and a new and improved interface), but it does give new consumers a choice of beam patterns. And I think the performance on 8xAA cells is enough to give dedicated Li-ion fans something to think about – output, regulation and runtime performance certainly exceeded my expectations in this case.