4Sevens Quark Mini AA-2 (XP-G R5) Review: RUNTIMES, BEAMSHOTS, and more!

selfbuilt

Flashaholic
Joined
May 27, 2006
Messages
7,018
Location
Canada
Warning: pic heavy, as usual. :whistle:

Mini2AA008.jpg


Specifications:
  • LED: CREE XP-G R5
  • Material: Type-III hard-anodized aircraft-grade aluminum
  • Lens: Optical-grade glass lens with anti-reflective coating on both sides
  • Waterproof Rating: IPX-8
  • Battery Type: Two AA batteries (up to 3.0 volts combined; batteries are included)
  • Seven Output Modes:
  • Low: 3.0 OTF lumens, 3.4 days
  • Medium: 36 OTF lumens, 8.8 hours
  • High: 180 OTF lumens, 1.7 hours
  • Strobe: 3.3 hours
  • S.O.S.: 10 hours
  • Beacon (Hi): 17 hours
  • Beacon (Lo): 88 hours
  • Dimensions: Length: 5.0 inches, Diameter: 0.7 inches, Weight: 1.0 ounces
  • Ultra compact design
  • Secure, deep-carry pocket clip
  • Superior reflector for the perfect balance between spot and flood lighting
  • Included accessories: Lanyard, spare o-ring, instruction manual, and two AA batteries.
  • Estimated MSRP ~$40
This 2xAA light is a new arrival to the 4Sevens Quark Mini line. It uses a similar interface to the standard 1xAA Mini.

For those not familiar with the Mini family, check out my earlier Mini 123/AA, Mini CR2, and Mini Tint Comparison reviews.

Mini2AA003.jpg

Mini2AA004.jpg


Mini2AA001.jpg


The light comes in fairly typical 4Sevens packaging. Inside the thin cardboard case with plastic insert is the light, wrist lanyard, spare o-rings, two Duracell AA batteries, manual and warranty card.

Mini2AA015.jpg


Mini2AA012.jpg

From left to right: Duracell alkaline AA, Quark Mini AA, Mini AA-2

Mini2AA010.jpg

From left to right: Duracell alkaline AA, 4Sevens Quark Mini AA-2, Lumapower Connexion 2AA, 4Sevens Quark AA-2, Fenix LD20-R4, Eagletac P20A2-II, Nitecore D20.

4Sevens Quark Mini AA-2: Weight: 27.7g , Length 131.8mm x Width 17.6mm (bezel)
4Sevens Quark AA-2: Weight 60.7g, Length 147.0mm, Width (bezel) 22.0mm
Eagletac P20A2-II: Weight 85.5g, Length 161.7mm x Width 25.8mm (bezel max)
LF3XT 2AA: Weight: 114.8g, Length x Width: 150.5mm x 25.0mm (max width)
Nitecore D20: Weight: 109.3g, Length: 154.0mm x Width: 25.0mm (head)

The pictures – and numbers – speak for themselves: The Mini AA-2 is by far the smallest and lightest 2xAA light I've tested. :eek:oo: And no, that weight is not a misprint - the Mini AA-2 actually weighs less than most 1xAA lights!

Mini2AA005.jpg
]
Mini2AA007.jpg


In keeping with its twisty status, screw threads are anodized for lock-out (i.e. tighten to turn on). I had noticed a fair amount of thread play on my early Mini AA and Mini 123 samples, and the AA-2 suffers from this as well (although this particular sample feels "tighter" than most of my other ones).

The Minis have the advantage of generous knurling, which makes it easier to use the twist feature one-handed. :thumbsup: Knurling is fairly aggressive. Lettering is sharp and clear, in bright white against the shiny black finish (type III hard anodizing). No chips or flaws on my sample.

Light can tailstand.

Note the addition of the 4Sevens clip, which is new for the AA-2 model. I like it, as it feels more secure than most clips in this class of light. You would need a fine-tip screwdriver to remove.

Mini2AA006.jpg

Mini2AA013.jpg


4Sevens went with the latest emitter (XP-G R5) for the all members of Mini line, including the AA-2. Reflector is the common shallow light orange peel (LOP). Note that the head of the AA-2 looks exactly the same as the AA head. You should expect the exact same beam pattern.

Which brings us to the requisite white wall hunting ;). All lights are on Hi on Sanyo Eneloop AA, about ~1 meter from a white wall.

Sorry for the quality of the white wall beamshots – I've recently moved, and haven't had a chance to set up a proper beamshot closet yet. What you are looking at below is recently primed drywall.

Mini2AA-Beam001.jpg

Mini2AA-Beam008.jpg


Mini2AA-Beam002.jpg

Mini2AA-Beam007.jpg


Mini2AA-Beam003.jpg

Mini2AA-Beam006.jpg


Mini2AA-Beam004.jpg

Mini2AA-Beam005.jpg


As expected for a XP-G emitter in a shallow reflector, the light has a wide spillbeam with reduced throw (compared to most other lights).

Again, the Mini AA-2 beam pattern is exactly the same as the earlier Mini AA (same head dimensions). However, the AA-2 max output is a lot higher (beam comparison not shown – scroll down to the runtimes for some numbers).

User Interface

The Mini AA-2 uses the same interface as the earlier Mini AA and Mini 123 – which is in turn the same as the original 1xAAA Preons (the new 1xAAA Preon ReVo is slightly revised).

Basic operation is controlled by twisting the head tight against the body to activate the light, loosen to turn off. Do a twist off-on in under 1 sec and the light advances to its next mode in this order: Lo > Med > Hi mode. Wait more than two seconds before re-activating the light after turning off, and it returns to default Lo (i.e. no memory mode).

There is no strobe or SOS mode to worry about in the regular interface, they are "hidden" away if you want them. To access, complete the Lo-Med-Hi twist sequence twice in under two seconds (i.e. Lo > Med > Hi > Lo > Med > Hi). The additional modes then present themselves in sequence as Strobe > SOS > Hi Beacon > Lo Beacon (twist off-on to activate each in sequence).

Note that there is no memory for "special outputs" either – if you turn off the light for more than 2 seconds, you will be back at constant output Lo when next you re-illuminate. So no worries about getting stuck accidentally strobing yourself here. :thumbsup:

PWM and Strobe

Med
2AA-MedPWM.gif


Lo
2AA-LoPWM.gif


Like all the earlier Mini and Preon lights, pulse-width-modulation (PWM) is used for the Lo/Med modes. PWM frequency is unchanged at a reasonably high 2.42 kHz across the line (i.e. you won't notice it in use).

2AA-Strobe.gif


Again, strobe remains ~10Hz across the Mini/Preon lines.

Beacon modes are 5 rapid flashes followed a single flash at 10 sec intervals, at either full power (Hi) or lower power (Lo).

Testing Method: All my output numbers are relative for my home-made light box setup, a la Quickbeam's flashlightreviews.com method. You can directly compare all my relative output values from different reviews - i.e. an output value of "10" in one graph is the same as "10" in another. All runtimes are done under a cooling fan, except for the extended run Lo/Min modes (i.e. >12 hours) which are done without cooling.

Throw values are the square-root of lux measurements taken at 1 meter from the lens, using a light meter.

Throw/Output Summary Chart:

2AA-Summary-1.gif


Output/Runtime Comparison:

2AA-HiEne.gif

2AA-LoEne.gif


2AA-HiL91.gif


2AA-HiAlka.gif

2AA-MedAlka.gif


The Mini AA-2 has incredibly good runtime for the output on its Hi setting. :eek:oo:

Unfortunately, overall efficiency on the Med level is somewhat below average for this class of light. This is likely due to the less efficient PWM method for reducing output – it doesn't really compare to the current-controlled lights (e.g. regular 4Sevens Quark, Fenix, Olight, Eagletac). However, the higher output bin R5 at least puts overall runtime performance within an acceptable range.

But I imagine a lot of people would like to know how it compares directly to the 1xAA Mini:

MiniAA-AA-2-Runtimes.gif


Potential Issues

Body wall construction is fairly thin, in keeping with the ultra-light build. As with the other member of this Mini family, I suggest you be mindful of this when handling.

There can be some play in the threads for all members of the Mini line (which can lead to accidental mode switching by pressing the head). Although my AA-2 sample seems better than most of my earlier single-cell Mini lights, it is hard to tell from one sample.

Preliminary Observations

The 1xCR2/CR123A/AA Minis have carved out a nice little niche for themselves in the ultra-light market of lights (sorry for the pun ;)). Certainly, the user interface and build are remarkably similar to this 2xAA version (e.g. the heads look the same, act the same). All the positives I pointed out in my Mini AA/123 review apply here as well. There's also the addition of a new, good quality pocket clip on the 2xAA build. :thumbsup:

But do people really need want/need an ultra-light 2xAA light? For most non-flashaholics, I typically recommend a standard length/weight 2xAA light, with multiple modes and a clicky switch. But users in the ultra-light world (e.g. travelers, backpackers, etc) look at things a little differently. For them, the emphasis is on performance for the size, with reliability and ease of use as key criteria (e.g. twisties typically preferred, as clickies can break on the road, etc).

Looking at it through that lens, the question really becomes: is there a performance advantage for the extra size/weight of a second cell? To answer that, I suggest you take a look at the Mini AA and AA-2 output/runtime comparison graph.

As you'll see, the Mini AA-2 has exactly the same runtime on Hi as the Mini AA – but with at least twice the output. I suspect this is what a lot of users like to see on Hi – twice the output for the extra weight of carrying two cells. :thumbsup:

But what about Med and Lo? As my Med results show, both output and runtime are increased on the Med mode the AA-2 (i.e. output is only partially increased, giving some runtime advantage as well). I haven't test Lo, but as the output levels are nearly identical, I would expect quite a bit longer runtime in 2xAA format. :twothumbs

These results are significant, as I don't think there would be as much of a market for the AA-2 if it didn't have an absolute runtime advantage at Med-Lo (i.e. I doubt most ultra-light users could justify the extra size and weight without greater runtime at some levels). As such, I recommend 4Sevens clarify the relative runtime numbers on their website – the currently listed specs don't seem to accurately capture the differences (i.e. the Mini AA runtime specs seem a bit high :rolleyes:).

The other possible ultra-light comparator here would the AAA class of lights. Although not available in the same physical build as the Mini, you can get the same circuit and relative performance in the 1xAAA and 2xAAA format as the 4Sevens Preon. But while a lot thinner and lighter, I would caution that AAAs only hold ~1/3 the capacity of AAs.

For its intended use, my main suggestion for the Mini AA-2 would be to improve the relative efficiency of the circuit on Lo/Med modes (only Med tested here). I may have been spoiled by all the current-controlled lights, but I would like to see longer runtime for the output. ;)

But definitely no complaint about the Mini AA-2 performance on Hi – the runtime results are outstanding, on all batteries. :eek:oo: I am particularly impressed by the regulation and runtime on standard alkalines (again, while we may eschew them, a lot of the mainstream world still uses 'em).

From my perspective, the AA-2 is a successful extension of the Mini line into the dual-cell world. At the various least, it provide another lightweight option for extended runtime and output on standard AA cells.

----

Quark Mini AA-2 was supplied by 4Sevens.com for review.
 
Last edited:
Nice work SB, as usual. I don't suppose you could do a comparison shot with a Fenix L2D Q5? For me this looks like my bike commuter light of choice if the hot spot is dispersed at distance. Mind you, I take your warning about handling the light with care. Might not take a crash too well.
 
I don't suppose you could do a comparison shot with a Fenix L2D Q5? For me this looks like my bike commuter light of choice if the hot spot is dispersed at distance.
I'll see if I get a chance when it gets dark tonight.

In the meantime, the LD20-R4 is a comparable to the L2D in the sense of overall spillbeam width and max center beam throw. The actual beam profile looks quite different (i.e. more sharply defined hotspot on the L2D, lots of beam rings, lower overall output), but the general dimensions of the beam are not that different.
 
The light comes in fairly typical 4Sevens packaging. Inside the thin cardboard case with plastic insert is the light, wrist lanyard, spare o-rings, two Duracell AAA batteries, manual and warranty card.

I'm assuming it actually comes with two AA batteries? ;)

selfbuilt said:
For its intended use, my main suggestion for the Mini AA-2 would be to improve the relative efficiency of the circuit on Lo/Med modes (only Med tested here). I may have been spoiled by all the current-controlled lights, but I would like to see longer runtime for the output. ;)
Yes, I was thinking the same thing, and was hoping to see better medium performance.

The size and weight of this light sure is nice though.

Anyway, another great review selfbuilt. :thumbsup:
 
Thanks selfbuilt! :wave:
Your reviews are always reference class - the combination of runtime graphs, beam shots, size comparisons and discussion tell me everything i need to decide if i open my wallet yet again :grin2:

And the timing was perfect - in a shopping spree i just snatched the Preon ReVO SS NW, and oggled the Quark Mini AA-2 - it would be my first 2x AA light (the others are single-cell AA/AAA/18650, except the iTP A6).

However after seeing the runtime in medium mode i will wait until 4sevens upgrades the minis with the circuit of the ReVO - which seems consequential to me after the "upgrade" from Preon to Preon ReVO.

In all other aspects i like this light - very small and light, nice brightness and runtime in high, and a very fair price :)
 
Last edited:
Your reviews are always reference class - the combination of runtime graphs, beam shots, size comparisons and discussion tell me everything i need to decide if i open my wallet yet again :grin2:
Thanks, glad you find them useful.

One additional point - I understand some users have had issues with the spring resistance being too low. This can apparently result in the light changing mode by shaking it (i.e. momentary interruption).

I've tried to reproduce this on mine, only managed to get it to happen once with rapid and vigorous shaking. But this was far beyond normal handling - I was about to give up when it finally happened.

As an aside, Mrs selfbuilt walked in just before this point ... needless to say, I got a strange look until I explained. :laughing:
 
I received my mini aa-2 a few days ago. This little light is great! Selfbuilt's reviews are always top notch. However, they are known to induce a buying spree in some of the CPF members. :paypal:

This light has just the right amount of features and power to satisfy a wide range of users. Strobe modes are cool, but much nicer for the general users when they are hidden until needed as in the mini series.

lovecpf
 
Eric,

I enjoyed this review, as I always enjoy your fine work.

Thanks again for all you do for us around these halls of light! :thumbsup:

Best regards,

Andrew
 
Great review - I've been on the fence about getting one of these but looking at those runtime graphs, I think I may just have to go for one now!
 
Thanks, glad you find them useful.

After searching the net about two years ago i came across your review of the Nitecore D10, and bought it. At that time the "flashaholic virus" didn't bite me yet.
But a year after i thought: lights are kind of fun - maybe i need a bigger one, and after that a smaller one, and a headlight, and so on...:laughing:

Most of my buying decisions i base on reviews, and in my opinion only series of tests with always the same methodology and test setup are really useful to understand a product in context. There are only a few reviewers for each type of product who do this, and in combination with your attention to detail your are "it" for flashlights :wave:

But exactly because you have such a big database of lights, the runtime graphs get a little bit unclear - maybe a diagram with selectable overlays for each light would help - i don't know if this is possible in this forum, with javascript and a checkbox for each flashlight. Then the reader could compare only the lights he is interested in.

As an aside, Mrs selfbuilt walked in just before this point ... needless to say, I got a strange look until I explained. :laughing:

I can imagine - "honey - why are you shaking the light so violently???" :nana:
 
I replaced my stylus pro with this light as my EDC and never looked back. Great review.
 
Thanks for the support everyone. :grouphug:

But exactly because you have such a big database of lights, the runtime graphs get a little bit unclear - maybe a diagram with selectable overlays for each light would help - i don't know if this is possible in this forum, with javascript and a checkbox for each flashlight. Then the reader could compare only the lights he is interested in.
I've had much the same thought myself, but the limited functionality of vBulletin wouldn't allow it. Frankly, just too busy right now to even consider building a dedicated site with javascripting, etc. Will have to stay on the back burner ...
 
I've had much the same thought myself, but the limited functionality of vBulletin wouldn't allow it. Frankly, just too busy right now to even consider building a dedicated site with javascripting, etc. Will have to stay on the back burner ...

What program do you use for generating the runtime graphs? Would you be willing to offer the data for download so that inquiring readers could generate the graphs with only the data important for them?

Also integration of lumens over time could be done for comparison purposes - currently i'm estimating the area by multiplying the mean value (estimated) with the runtime to 50%.

In the mean time I'll look for online resource for automatically building interactive plots of measured data. Thanks again for your continued dedicated efforts :wave:
 
What program do you use for generating the runtime graphs? Would you be willing to offer the data for download so that inquiring readers could generate the graphs with only the data important for them?
It's a little off-topic, but ... ;)

The problem with any testing methodology is planning ahead of time all the possible permutations in presentation. If I had known I was going to do this many reviews where I started out, I would have organized my data collection and presentation systems differently! Now, it would be a fair amount of work to go go back and organize it all in a way to allow on-the-fly direct comparisons between the entire dataset over time (i.e. my systems are somewhat fragmented).

This doesn't apply just to runtimes - I frequently "waste" a lot time re-doing beamshot comparisons, which I wouldn't have to do if I had set a proper beamshot studio. But I've recently moved, so that would have screwed things up anyway. Ditto "celling bounce" ... I've found a way to approximate the values I obtained in my previous house, but frankly need to redo as many as I can in new controlled environment. On the plus side, almost finished building my basement, so should have facilities set up soon for both beamshots and ceiling bounce.

It's still my plan to re-organize all this eventually - I just need to find the time when I'm not actually testing new lights (or busy working for a living :whistle:).

Also integration of lumens over time could be done for comparison purposes - currently i'm estimating the area by multiplying the mean value (estimated) with the runtime to 50%.
Easier for me to do of course - I actually did area-under-the-curve measures for my quick CR123A battery comparison. The problem you will find is that unless you have perfectly flat regulation, you will have difficulty accurately integrating by just output and time to 50%. In most cases, the actual runtime curves are really the best way to compare - reducing curvilinear data to a couple of numbers typically reduces information too severely.

As an aside, since you seem to be interested, my lightbox is not linear for output (i.e. doesn't correlate linearly with lumens). However, I once compared my measured values to Novatac 120P manufacturer lumen specs, and computed a remarkable strong correlation to a specific power relationship.

Since then, I have compared that formula to lumen values reported here for those who have calibrated integrated spheres. The dataset of lights we have in common is quite large. Although there is some variation, on the whole the formula holds very well between ~1-400 lumens (i.e. some people's spheres seem consistently slightly above my curve, some slightly below). But on the whole it averages out.

So if you want to "guestimate" the lumen output of lights from my lightbox's relative overall light output (ROL), simply apply this power relationship: Lumens = 0.6 * ROL^1.3

It works pretty well. :wave:
 
Offtopic warning... :sssh:

The problem with any testing methodology is planning ahead of time all the possible permutations in presentation.

This doesn't apply just to runtimes - I frequently "waste" a lot time re-doing beamshot comparisons, which I wouldn't have to do if I had set a proper beamshot studio.

Even with planning ahead it's not always possible over a longer time - i tried to do comparisons of CPU coolers over a longer time to see the progress made, but after a year or so a CPU socket is "deprecated" and you have to switch your platform. Using a socket-independent heater and measurement would work, but wouldn't be representative of real-world results (surface curvature, with or without heat spreader, etc.).
Same applies to technical equipment and locations (beamshots, ceiling bounce room, etc.).

It's still my plan to re-organize all this eventually - I just need to find the time when I'm not actually testing new lights (or busy working for a living :whistle:).
Oh, my - good luck with that! I had to give up the moderation of a forum because of "working for a living"...

In most cases, the actual runtime curves are really the best way to compare - reducing curvilinear data to a couple of numbers typically reduces information too severely.
I know, and I usually do that. But for sorting by efficiency one number is very useful, especially if the curves differ much - e.g. longer runtime but lower output.

As an aside, since you seem to be interested, my lightbox is not linear for output (i.e. doesn't correlate linearly with lumens). [...] So if you want to "guestimate" the lumen output of lights from my lightbox's relative overall light output (ROL), simply apply this power relationship: Lumens = 0.6 * ROL^1.3
Thanks, that's very valuable! :wave: I'll update my Excel sheet - all higher output lights were treated unfairly by my linear approximation. I even added a column "fun-factor" where i squared your output multiplied by runtime to represent more closely what tickles my interest... :ohgeez:
 
Written by Mostly on 11-15-2010 10:07 AM GMT

In the Duracell Med Alkaline graph, should that be 9hr 41 min on the Mini 2-AA instead of 1hr 41 min?

With the price drop to $39 on these, it's getting very hard to resist. Yes, there'll probably be a version with the updated Revo circuit one of these days... but likely (judging from the Revo) it won't be in this price range.
Written by selfbuilt on 11-15-2010 11:05 AM GMT

Mostly said:
In the Duracell Med Alkaline graph, should that be 9hr 41 min on the Mini 2-AA instead of 1hr 41 min?
Whoops, so it should. Just fixed. Whoops, so it should. Just fixed. :)
Written by Redheddedstranger on 11-17-2010 09:09 AM GMT

i had just ordered this the other day. just now noticed the write up.. great detail. props on this
Written by MonteroGraves on 12-07-2010 12:37 PM GMT

thank you for this review. I am a light noob and the hard numbers in your charts really helped my understanding of the different terms and technical aspects about flash lights that i've been learning about. I love my new quark mini AA so be sure ask 4sevens for some commission because I can tell you that you are responsible for selling atleast 4 of them. :santa:
Written by Helmut.G on 12-18-2010 04:04 AM GMT

Since this thread has got relatively few responses I wanted to say that this great review has helped me a lot!

Good work selfbuilt!

About the light I think the extremely small size and especially the MAX mode runtime an the duracell alkies is very impressive!

No current regulation is a drawback of course but you can't have everything. At the very affordable pricing this is an exceptionally great light judging from the review!

I have ordered one and also received it very fast to germany from 4sevens.com but won't open and try it yet, it's a christmas gift for my dad.

When he went on a biking trip this year he asked me for a good flashlight to take with him.

I gave him a Fenix L2D and MC10 (the one-AA anglelight) to choose (AA because it was a long trip so he could buy new batteries).

He took both lights in his hands and, without considering any other aspect, he said, holding up the MC-10, "I'll take that one with me, it has less weight."

I guess he's gonna like this one.
 
can anybody tell me if the end cap can be switched to a clicking kind,to work like the Tactical model?
 

Latest posts

Top