The SF A2

Not open for further replies.


Flashlight Enthusiast
Aug 2, 2003
Upstate New York

If I had to choose only one light to keep—if I had to sell all of them except one—I would choose the SureFire A2. It’s that good.

And yet, many people regard A2 owners as fools parted from their money for buying such an expensive light that “only has 50 lumens” and “only runs for 50 minutes” and which is mainly an incandescent light (because everyone knows, or else should know, that incandescents are an inferior and dying breed of illumination tool!) Why then do I hold the A2 in such high regard? Allow me to answer that question over the course of this review.

“Only 50 lumens”

Let’s set the record straight: the A2 is putting out something on the order of 75 to 80 lumens. I know this for two reasons. First, a private lab tested several lights in an integrating sphere for Arc Flashlights, LLC, one of them being the E2e. They rated the E2e at 83 lumens on fresh batteries. I own both an E2e and an A2 and I cannot see any difference in total output between the A2 and the E2e running on fresh cells. Second, McGizmo tested the A2 in his integrating sphere as part of the CPF light-meter benchmarking project and arrived at 79 lumens.

That said, even supposing that the A2’s output really were only 50 lumens, the "only 50 lumens" remark would be a stupid comment for someone to make who has never seen the A2’s beam in person. A flashlight beam is more than a lux reading or a lumens total. The beam has a Correlated Color Temperature (CCT), or a “whiteness”; the beam has a profile or shape; it has a quality. A 500 lumen beam can be yellow and unfocused and full of artifacts, and a 50 lumen beam can be beautifully white and long-throwing and smooth. Only someone looking for a powerful spotlight could justifiably make this comment about the A2, sight unseen, but such a person would obviously not even be considering an A2, nor criticizing it for not being what it was never intended to be.

“Only 50 minutes”

The concept and measurement of runtime in a single level light is pretty straightforward: you turn it on and let it go and whenever you think it is too dim you stop the clock and write down a number. This can be when the output falls to 50 percent of the starting intensity, or can be to ultimate run-down-to-nothing dimness, but however exactly it is defined, it is straightforward.

Not so with a dual level light. Out in the field, in use, many tasks require only the low-level, or mostly the low-level with only occasional bursts of the high level. For example, I can walk in the woods and fields at night using mostly the low level from the A2, with only occasional bursts from the incandescent. I could be out walking and exploring all night with the A2, but the E2e, which some would consider to have a longer runtime, would certainly not last me more than 2 hours.

Now it is true that the A2 incandescent lamp can only be run for 50 minutes continuously on a set of fresh 123’s—I don’t deny that—but the point is that you wouldn't normally use the A2 that way, but even if you did, the LED’s would continue to run for many more hours after the incan turned off. In fact, (and this is one of the most important points I have to make in this whole review) at the moment when the batteries no longer support the incandescent lamp, they still have about 25 percent of their energy left. Yes, a full quarter of a tank left.

I determined this by taking the batteries out of my A2 the moment I tried to use the incandescent and it would not come on in regulation. I then installed them in my E2e, turned on, and measured the time until end of cycle dimming. I got a full 18 minutes of runtime, which is about a quarter of the total runtime of the E2e. I also took another set of batteries in the same condition and sent them to SilverFox, CPF’s own master of batteries, and had him run a test on them to determine how much capacity was left. At a 1 amp discharge rate, he found that a little over 80 percent of the capacity had been used when discharging to 1.0 volt, and a little over 70 percent when the test was continued to 0.5 volts. Thus, I believe that the 25 percent figure is an appropriate estimate. In any case, whichever figure you settle on, the point is that you have something left over when your high beam dies on you, by design. There is no need to worry over "what happened to the extra runtime"; no need to try to blame the regulator or the 10 ohm resistor in the tailcap or the holes in the reflector (although, certainly, some light is lost there). The A2 was never designed to be a single level light with a fully regulated runtime that sucks the batteries dry before falling out of regulation. It is, start to finish, a dual-level light.

So, yes, if you’re only interested in the incandescent part of the A2, the 50 minute runtime is a downside—it means that you would never make use of the full capacity of the batteries. However if you use both levels of the light, as was intended, then you will find that these 50 minutes of incandescent runtime go significantly further than the 75 minutes of the E2e.


OK, enough about those aspects of the A2 that are, in and of themselves, less than outstanding. Let’s move on to those aspects which ARE outstanding.

The A2 incandescent beam

The SureFire A2 is the only production flashlight with a regulated incandescent beam. Think about what that means for a second; think how we all might feel about LED lights if none of them was current regulated and they dimmed down over their runtime and degraded in tint. We might, as a whole, still prefer them to hardware store variety incandescents, but we would certainly not like them as much as we do now. That’s the situation that incandescents are currently in at this very moment. How many of the people who switched to LED’s because they were sick of “**** yellow incans” had only ever experienced alkaline direct-drive drug store cheapo lights, with bad batteries and lamps and a build-up of contact resistance? There is a huge difference between an under-powered and under-driven incandescent and a fully regulated over-driven incandescent. There is all the difference in the world.

When an incandescent is under-driven it is yellowy and relatively inefficient; but when it is driven hard, it is bright and white and relatively efficient. This quality appears in number form as the CCT, which stands for “correlated color temperature”. I had planned on talking about black bodies and color temperature and spectral peaks and so on, but I won’t. Suffice it to say that the higher the CCT, the “whiter” the beam. The best photographic lighting is 3200 to 3400 K. The average light bulb in your house is about 2700 K. Very few production flashlights are above 3000 K even at the start of run. Keep that in mind when I tell you that the CCT of the A2 is 3300 K or slightly higher. Besides being nice and white, a filament driven to this CCT is also very efficient, delivering about 30 lumens per watt.

The SureFire MN20 is also around this CCT at the start of run, and many hotwire mods are at this level or even higher (to wit the 1111 on 6 good NiMH cells, or the 1185 on 9 good NiMH cells) but the difference is that the A2 stays at this CCT for as long as the batteries continue to support the incandescent. The direct-drive lights that hit very high CCT’s at the beginning of the run, fall down to significantly lower CCT’s by the middle and end of the run.

Now, all of that being said, I should point out that a direct-drive incandescent light can still be a wonderful and effective tool. You do not need to regulate an incandescent in order to have a good flashlight. I suspect that is why there are so very few regulated incandescents out there, production or otherwise.

In addition to being precisely voltage regulated, the A2 also has a soft-start feature which limits the current to the filament during the first 50 milliseconds or so after turn-on. This prevents the very high currents which would normally run through the cold filament and weaken, damage, or even blow it. This phenomenon is why almost all incandescent lamps die at turn-on, and this problem has been eliminated with the A2.

The high CCT, voltage regulation and soft-start are all very nice indeed, but they are made all the better by the other characteristics of the A2 beam. The A2 has a beautifully smooth beam that nonetheless throws exceptionally well for such a small light. Some people complain that the hot-spot is not round, but rather oval, but this is an unavoidable characteristic of a transverse filament light which is focused for throw! If the hotspot were made round, the throw would be reduced. It's that simple. Other people complain that the white wall beam profile has a ringed, unappealing corona. And, yes, it is true that the very outside of the beam has several brighter rings of light that detract from the white-wall aesthetic, but, only someone who is a very picky about white-wall beam profile would be overly bothered by these rings. In any kind of actual use, they are not noticeable and do not detract from the illumination effectiveness of the light in any way. Still, if looking at white wall flashlight beams is a favorite activity of yours, the E2e, L4 or L2 would be better choices.

On the other hand, if you value actual performance in the field, the A2 shines, out-throwing the E2e and L2 by significant margins, while still having a practically artifact-free beam. I have some beam shots of these three lights at two different distances which clearly illustrate the A2's superior throw, and they will appear below shortly, but first I need to say a few words about the method I used to capture these shots, because it is non-conventional. I first posted about this in my thread A new flashlight beam metrology and you can check out the link for more detail, but the method is simply to separate the flashlight and the camera by moving the camera closer to the object being illuminated. This allows a more even light to fall on the cameras field of view, making it easier to capture a good image, and it also shows the differences in throwing ability much more clearly and dramatically than the conventional method which places the camera and flashlight at very nearly the same place and pointing in the same direction. Two sets of pictures follow. The first set was taken with the flashlights on a tripod placed at 33 feet away from the tree being illuminated. The second set was taken with the tripod at 66 feet away. In both sets, the camera tripod was 11 feet away from the tree, at an angle to the beam-line so as to avoid shadows falling upon the tree. The camera was also angled slightly upwards from horizontal. The lights were held onto the flashlight tripod top by a rubber band, so the beam centers are not exactly the same from shot to shot, unfortunately. The variation is fairly small, though, and I think the shots are plenty consistent enough to be useable. So without further ado, here they are:

A2 at 33 feet:

L2 at 33 feet:

E2e at 33 feet:

Now for the shots at 66 feet, taken in the same order. Note that the A2 at 66 feet is pretty much as good as the E2e at 33!

A2 at 66 feet:

L2 at 66 feet:

E2e at 66 feet:

Also note that the color rendering is quite different, even between the E2e and the A2, although the greatest difference is between the E2e and L2. I took a daylight shot of the tree in order to show the "true" color of the object, but unfortunately I did not place the tripod in the same place. I wish I could have just left it there for the time required to have the beam shots and the day light shot be in EXACTLY the same location, but I didn't want to risk it. I cropped the photo below so that the tree trunk is more or less the same distance from the right edge of the photo:


In my opinion, the camera has exaggerated the differences somewhat. The L2 did not lack the reds and yellows quite as much as shown, and the E2e and A2 did not accentuate the reds and yellows as much as shown. Even so, this does partially illustrate the differences in color-rendering. I will let you decide for yourself which is better.

In addition, I took daylight shots from the point of view of the lights at both the 33 foot location and the 66 foot location in order to give you a feel for the throw distances involved. Here is the view from 33 feet:


And here is the view from 66 feet:

While I was out taking these shots, the coyotes were also out and about, sometimes sounding disturbingly close. Now, I was not afraid of being attacked by coyotes, but I can tell you that it sure was nice to have the A2 with me in order to tell if any were coming close to investigate. The L2 was completely useless for this task, and the E2e felt inadequate. Only the A2 had the throw that made me feel in command of the situation. The distances involved in the coyote scouting were probably three or four times the 66 foot distance I used in the beam shots above, but if I had tried to take shots with the lights so far away the pictures would have turned out too dim to be usable. So just take my word for it: at those distances the A2 not only ruled the E2e and L2 roost, but did so uncontested. The other two simply could not reach out that far.

Of course, there's more to a flashlight than throw! And the L2 and E2e are great lights and have their strengths, but just note that the A2 can throw much better than either the E2e or the L2. This concludes my discussion of why the A2 has such a superlative beam. I now move on to discuss the next superlative characteristic of the A2:

The A2 switch and ergonomics

Can I just wax poetical for a moment? Can I just say how simply AWESOME the A2 switch and hand-grip ergonomics are? From the first moment I had the A2 in my hand I knew that it was a perfect fit, a dream come true. And I mean that somewhat literally. I started out my SF experience with a D2, which is just a 6P with a pocket clip, and I could simply never get used to the grip. The feel was just wrong, and my hand kept slipping back towards the rear of the light with the pressure required to keep momentary mode activated on the LOTC. I took off the pocket clip to try to make things better. No luck. So I traded up to a C2-HA to take advantage of the grip ring. That, I thought, would solve things. The grip ring was key. But no, the C2 also simply never felt right, and was never comfortable to operate. So I sold it, and next tried the E2e. This was certainly better, but although it is acceptable, I was never fully satisfied with it: the light is too short for a good grip, and the LOTC spring tension is a bit too high. Nor did the clickie tail cap improve matters. I found the clickie uncomfortable to operate, mostly due, I think, to the insufficient distance between the thumb and first finger when gripping the light. So, I actually prefer the LOTC on the E2e.

Then came the A2. It was the grip and feel I had been wanting all along. The hand-grip to thumb-spacing is just right, and with the hand in position, the little finger rests just so near the bezel. It just feels right. It feels good. And the extra length over and above the E2e size is actually part of the secret, as well as the lower spring tension in the tail cap switch. To my taste, the tail cap spring has just the right amount of tension.

More importantly, the switch not only feels good, it IS good. This two stage switch gives you instant access to two levels of light from just one location. There is no need to click through some annoying LED ON--OFF--INCAN ON--OFF, etc. sequence. Nor is there a need to access another switch or selector ring in order to get the mode you desire. In addition, you can have both modes be momentary, or go to constant on LED, incan momentary, or constant on LED and incan, or, lockout incan with momentary LED, or lockout both incan and LED. I don't think many people realize how amazing an achievement this is. This level of engineering is rarely found in something so prosaic as a flashlight switch.


Which brings me to my next topic: this light really is an engineering marvel. Now, obviously, it's not an engineering marvel in the way that the space shuttle was an engineering marvel, or the Brooklyn Bridge, when they were unveiled. It is not pushing the state of the art. I say the A2 is an engineering marvel because it is one of the most, if not THE most technically sophisticated flashlights ever made. I say this for three main reasons: Willie Hunt's LVR regulator, the 2-stage LOTC, the MA02 A2 LA, and the quality of components, build, and machining. Let's take them one at a time:

Willie Hunt's LVR

Who the heck is Willie Hunt, anyway? you may be wondering. He is an engineer at SureFire, LLC, and he is the designer and creator of the LVR pulse width modulated regulators for use with incandescent flashlights. The name "LVR" actually stands for "lightbulb voltage regulator." You can check out his webpage at Willie Hunt's Lightbulb Voltage Regulators for more details and for some great information. The short of it is that the LVR3 series of regulators are micro-processor controlled PWM voltage regulators that are highly efficient, small, precise, and sophisticated.

So what’s going on with the LVR, anyway? Well, if you have a DMM, it is very easy, fun, and educational to try the following experiment. Remove the head of the A2 from the body of the light. Looking down at the top of the body, you will see a center contact, a ring outside of that, and the threaded top outside of that:


The ring is positive, directly from the top of the battery stack. The center is connected to the FET drain, and the FET source is connected directly to the body tube of the light, which is electrically and mechanically the same as the threaded top of the body of the light. The FET gate is controlled by the LVR circuitry. If any of that went over your head, "FET" stands for Field Effect Transistor. It is essentially a switch. You put something on one side of it (the drain), something on the other (the source), and then by applying voltages to the gate, you control whether or not there is a connection between the source and drain (a short-circuit=switch closed or "on"), or not (an open circuit=switch open of "off") or in the case of a linear circuit, something in between—a resistance, as opposed to a short or open circuit. The LVR is a switch-mode circuit, and so the FET is always either on or off or transitioning very quickly between those two states. (More on this in a moment).

So anyway, getting back to the experiment, the center is the regulated negative terminal for the incandescent lamp, the threaded top of the body is the negative from the bottom of the battery stack through the switch, and the ring in between is the positive from the top of the battery stack and is a direct connection. Thus you should note right away that there is no regulation circuitry between the batteries and the LED ring. When you activate the switch to the first position, there is a 10 ohm resistor from the battery stack negative to the LED ring negative, and a direct connection to the battery stack positive. Close the switch all the way, and the 10 ohm resistor is replaced by a short circuit, or in other words, a direct drive to the LED with no extra resistance. You may be able to prove this to yourself, by removing the lamp from the head, reinstalling the head without the lamp, and activating the switch. In some A2's, the LED ring contacts will mate with the middle ring and body of the light even without the lamp in place, and you can see that taking the 10 ohm resistor out of the circuit by depressing the switch all the way makes the LED's brighter than they are with just the first position activation. But I digress.

The point is that the center contact is controlled by the regulator. So, what happens when you turn down the switch to the position where the LED's would be on (if the head were installed), but the incan would be off? You might expect to find about 6 volts DC across the middle ring to body tube (LED's on), and 0 volts from middle ring to center contact (incan off). But you will find a suprise in store for you. There is also about 6 volts from the middle ring to the center contact. Why? Because the gate of the FET is pulled down, and the FET is "on." So why wouldn't the incan be ON in this position?

The LVR is on so long as it sees enough voltage across its inputs. The inputs are at the positive top of the battery stack, and at the negative ground return path of the body tube, which is on the other side of the switch from the battery stack negative. So if no current is flowing through the 10 ohm resistor, the LVR sees enough voltage to turn on. However, the instant current starts to flow to the LED’s, the 10 ohm resistor will drop enough voltage so that the LVR will turn off, because it does not have enough input voltage to stay on. It’s that simple. Or rather, I should really say, that it is that elegant.

And for those of you who have often wondered what is going on when your A2 appears to have three levels (LED, low incan, high incan) depending on how the switch is pushed or on the luck of the draw, here is what is going on: there is some extra resistance in the circuit. (Assuming you have fresh full batteries installed, of course--low batteries will exhibit this low incan mode even if all contacts are clean and everything is in working order) The extra resistance could be due to contact resistance in the LOTC floating contacts or in the battery stack or somewhere else. Or due to a broken spring contact. But somewhere there is extra resistance—just small enough so that the LVR sees enough voltage to allow it to turn on but too much to keep it on. So it turns on and starts to draw high current through the lamp, which causes a larger voltage drop across the switch and contact resistance, which lowers the LVR input voltage, which causes it to turn off; and the same thing happens over and over again. Sometimes you push hard enough, or the position of the floating LOTC contact on the end of the body tube is such that there is a small enough resistance to run the incan in regulation. But other times, there is too much resistance, and you put the LVR into an equivocating state of confusion about whether it should be on or off, resulting in a sort of in-between state, and thus the so-called "low-incan" level. If you remove the switch and simply short circuit the battery stack negative to the body tube, I bet you will find (in these situations) that it works fine. This indicates the need to clean the contacts inside the switch with a soft cotton pipe cleaner, and the butt end of the body of the light with a cloth or paper towel.

Let me talk about how the LVR actually regulates while it IS on and functioning properly with no extra contact resistance in the circuit. If PWM voltage regulation has you mystified, read on. It is really very simple in principle. First off, note that filaments will radiate light as long as they are hot enough, whether or not electric current is flowing through them at the time. The electric current only serves to make and keep them glowing hot. The incandescent lights in your home, for example, run directly from the 60 cycle AC line voltage. This means that 120 times a second, there is zero current flowing through the filament, and yet we don't notice this fickle drive source at all. This is simply because the time it takes the filament to cool down to, say, half its starting temperature, is quite long compared to 1/120th of a second. So as long as whatever crazy stuff is happening with the supply voltage and current happens fast enough, the eye and the filament won't notice, and will only see a sort of average of the applied power. I will simply be using the term "DC equivalent" to avoiding having to discuss RMS power and integrals and the like. The LVR switches the FET on and off rapidly in order to maintain a precise average DC equivalent voltage at the filament. In other words, it messes with the supply voltage in such a way that the filament sees the equivalent of whatever constant DC voltage set-point has been programmed into the uC of the LVR, and it does so rapidly enough that the eye sees no difference between that, and an actual DC voltage of the specified value. How does it do this?

Imagine that you are the operator of a mechanical flashlight switch, and imagine that you can turn that switch on and off more or less instantly, and that you are fast and alert enough to do this many, many times a second. So, you turn on and current starts to flow, then you turn off, and current stops. And you wait for a moment, then turn back on, and wait for a couple moments, then turn off, then turn on, and so on, and on, many, many times a second. And the amount of time you hold the on and off states depends on how hot and glowy and bright you see the filament being. When it just barely starts to be a bit dimmer than you like, you make your ON times a touch longer, and when it starts to be just a bit brighter than you like, you make your OFF times a touch longer. At some point, as the battery voltage drops, you reach a state where your ON times are very, very long, and your OFF times are very, very short, and you are simply passing the battery voltage directly to the lamp. After that, the lamp will of course start to dim despite your best efforts at regulation. Or perhaps you will have had instructions to turn off once that state has been reached.

THAT, in essence, is pulse width modulated regulation. And because the switch drops no power when on (because it is a short circuit and thus has almost no resistance), and drops no power when off (because it is an open circuit and no current is flowing), and switches so very fast between these states, the regulator is very efficient. There is pretty much nowhere to drop any power! In point of fact, the LVR3's are better than 98 percent efficient, and will handle up to 10 amps of current.

The 2 stage LOTC

I have mentioned this a number of times before, but I want to talk more about how much machining and engineering design must have gone into this switch. For those of you who are familiar with all the trouble Arc Flashlight had with the Arc4 switch, or with the trouble that the first FireFly’s had with the o-ring engaging before threads, despite the simplicity of the switch (a simple twisty), you may begin to have an understanding of what the SureFire engineering team must have faced when trying to implement this two stage switching concept, and bring it into reality as a perfected, functional, reliable switch. I'm no machinist, and I’m no expert on thread types and CNC machines and lathes, but I can tell you that the threads on the A2 tailcap are a special type of thread. According to a post of McGizmo’s, they are “Acme lead screw type thread” and because of this “Longitudinal displacement per revolution (pitch) is increased in a very robust and dependable fashion with this choice of thread.” In other words, there is very little play in the mating of the threads.

But don’t take my word for it! Remove the LOTC from your A2 and look at the threads on the end of the body from the side. Check out their profile. Now compare them to a regular old bolt, or even the threads on your Arc LS or mini Mag or Q3. See any difference? You should notice that the A2 threads are just about rectangular in profile. Most threads are triangular, or are saw-toothed shape. And this is only the tip of the iceberg. The simple truth of the matter is that this 2 stage switch all by itself is mechanically more sophisticated and more expensive to produce than most flashlights in their entirety.

Let’s take a look at the internals of an A2 switch. There are seven major components: The rubber cover, the threaded plastic retaining ring which holds it down, the metal body, top and bottom injection molded pieces, between which are sandwiched a circuit board with three upwards protruding tabs or fingers, and the spring, which is riveted onto the bottom injection molded piece. The top injection molded piece is captive inside the top cylindrical housing of the metal body. It travels in one direction as far as the underside of the cap glued on top of its shaft, and as far in the other direction as the shoulder of the piece itself. In order to remove it from the metal switch body, the cap must be broken off from the shaft, and then the two injection molded pieces and circuit board and spring will all come out the front of the metal switch body with a little persuasion. Here is a picture of the top of the switch with the retaining ring and rubber cover removed:


Note that while it may look as if the center cylinder could be unscrewed from the main switch body, it is in fact all one solid piece, and what you are seeing is the glue/sealant which was underneath the edge of the rubber cover.

And here are the internals of the switch after the cap was removed and the internals were pushed out of the switch body:


Those three metal fingers attached to the circuit board between the two black plastic pieces are electrically connected to the center spring via a 10 ohm resistance. When the switch internals move forward enough for these to touch the end of the flashlight body, then the LED circuit is completed through the 10 ohm resistance, but the LVR does not see enough voltage to turn on, and so the incandescent stays off.

However, when the internals are pushed forward even farther, the ends of the fingers bend down and touch the coppery colored pads directly underneath their ends. These contacts are short circuited to the spring, and thus cause the 10 ohm resistor to be taken OUT of the circuit path. Thus the LVR will see enough voltage to turn on and power the incandescent (assuming the batteries are fresh enough). Here is a close-up of one of the metal fingers and short-circuiting pad contact underneath it:


The internals can be pushed forward by simply moving along with the switch body as it is twisted in, or they can be pushed forward by a force applied to the cap on the end of the shaft on the top injection molded plastic piece. And alternately, if the switch body is turned back far enough, then the internals can be too far away for the tabs to be bent until they touch the contacts underneath them no matter how hard the rubber cover is pushed (incan only locked out, LED momentary), or even too far away for the tabs to make contact with the body at all (both LED and incan locked out).

I hope that it is obvious why I say that this switch all by itself contains more design, engineering, and machining than most flashlights in their entirety. If it’s not obvious, here’s why: two injection molded plastic pieces, one circuit board, one molded rubber cover, and a fairly THIN complex metal body that needed to be machined from both ends. The thinner walls and smaller dimensions of this piece mean tighter tolerances which means more precise and expensive machining. More than that, though, is all the design that had to go into making this switch work reliably and repeatably. As a finished, fait accompli it is easy to dismiss it as easily done, but I am quite sure that it was not at all easy to design, build, and test, and that it probably involved many hours of work from high-paid engineers and technicians.

In short, the A2 switch is a marvel.

The MA02 A2 lamp assembly

We are all a bit jaded about SureFire lamp assemblies; we take them for granted, or even consider them to be over-priced, over-designed units that are just another way for SureFire to extract more money from its poor, ill-used customers.

Not so.

These lamp assemblies are, in fact, ASSEMBLIES! As one of the few people on CPF who can pot his own lamps, let me assure you that the SF LA’s are something special. Take a look at the MA02, for example:


First, notice that the bare lamp itself is potted into the plastic housing. Next, notice that there is an outer metal ring and a center rounded contact. The lamp filament is perfectly centered relative to the axis of the plastic housing, and is at exactly the right height above the shoulder of the housing so that when the LA is installed in the head of the A2, the filament is properly focused. There is simply no easy way to accomplish these two, simultaneous alignments, and if you think there is, then I challenge you to do it. And tell me how so I can steal your method for myself. :devil: Seriously, though, a lot has gone into making the MA02, above and beyond the lamp itself, which is a very nice high pressure xenon lamp with enough halogen added to it to slow the rate of tungsten deposition on the glass, but not enough that the pressure would need to be lowered in order to prevent corrosion of the filament and support wires.

So before you take the MA02 for granted, or even dismiss it, take a careful look at it and start thinking about how you could do it cheaper or better!

Quality of build, components, and machining

The A2 is as well made as flashlights come these days. The aluminum is of the highest grade, with hard anodizing and chem coating. The lens is pyrex, the reflector is machined aluminum with vacuum aluminizing and undercoating. Everything is o-ring sealed. And so on. I could go on, but let me just show a couple pictures. Here is a cutaway of an A2 which I stole from one of Size15s posts:


And here is a nice picture of the machining that went into the head and reflector, which I stole from one of McGizmo’s posts:


Go to a machine shop and try to get quotes for something like this.

And anyone can see all the detailing that went into the outside of the body, especially at the top where the head threads on, and where the pocket clip is attached.

But there is so much more that is unseen or unknown. Some of it I know about and have tried to convey, but almost certainly there is a lot of stuff that I don’t know and thus couldn’t mention. One thing I can tell you by way of example is that not only is the regulator in the A2 a highly efficient and sophisticated piece of electronics, it is also covered in a conformal coating which is water proof. According to a post I read here on CPF, PK went to some trouble to find the right coating for the LVR. And the same care and attention went into the rubber for the switch covering and into the choice of spring and into the plating on the contact points. And then there is the LED ring with surface mount resistors.

And on and on.

Quite simply, to my mind, it is amazing that the A2 was ever envisioned, designed, developed, and marketed. I am even tempted to say that the A2 really shouldn’t exist in this cut throat commercial and cynical world. But it does, and I for one give thanks for this small miracle of a flashlight.

This concludes the first part of my discussion of the SF A2. In later posts I will provide a lot of links to important past CPF threads on the A2, and more technical details, such as the DC-equivalent voltage set-point of the LVR and power being applied to the lamp, and the current going to the LED ring, and various non-destructive tests that can be done on the A2 for fun and enlightenment. And also some discussions of other technical aspects of the A2. But for now, I really just wanted to get this much posted, as I could have kept adding to this post and re-writing and revising it and so on until I found that it was late 2010 before I ever finished. It’s not perfect and it’s not all that polished, but some may find this post worthwhile. So there it is, and I hope you all enjoyed it.
Last edited:


Newly Enlightened
Jan 6, 2005
Took me all of 20 mins to read this post....
And almost made me go out there and buy a A2 for EDC!!

if I got the $$$!


Mar 29, 2004
That was awesome. I just want to say thank you. Surefire should say it to you because I'm probably going to buy one now.


Dec 19, 2005
Fantastic, thank you Jim!

This really does show the A2 in a whole new light (sorry), and indeed the sort of work that goes into high end lights such as those made by SF.

Part of what appeals to me, and I imagine most flashaholics, is the level of technology and engineering required to make a really good light. Never before have I seen these elements described so well.



Sep 27, 2005
Long Island, NY
Thank you so much for your post js. I am sure that now, alot more people will see the "light", and seriously consider the A2. I dare say that of all my SFs, the A2 is my favorite. As a matter of fact, I own two A2s. One for me and one for my wife. Its that good, and its the only light I truly trust and personally chose for my wife to EDC :rock:


Flashlight Enthusiast
Dec 14, 2005
Tucson, AZ
Except for a big honker like a Magcharger, Ultrastinger, or Tigerlight (to keep in the car I don't have), the A2 is the only incandescent I'd ever consider owning, mostly because it's regulated. I'd probably own one already, if I weren't spending all my money on LEDs.

Great write-up!


Flashlight Enthusiast
Apr 8, 2004
This thread should be entitled, "Ode to the A2" or "The A2 Panegyric" :)
Nice work, Jim! :goodjob:
Last edited:


Mar 29, 2002
Gloucester, New Jersey
Awesome post Jim. As you may or may not remember, when the A2 first came out, I couldn't stand the light, the cost, the hype. I had gone back and forth.......... I don't like it, I do like it, It's just OK, etc. That was a couple years ago and throughout all my selling and trading, I've always made sure to have at least (1) A2 in my possession. It truely is a one-of-a-kind light and it my main light when the wife and I do night-time Geocaching, hikes or walks along the trail.


Aug 29, 2000
Kettering, England
Wow! :bow:

I don't have much time online at the moment and I am so glad the one post on CPF I did read was this one.

I've always had a thing for the A2 as some of you may know.

Thank you Jim!

Al :D


Mar 26, 2004
La Tiquicia
That is an excellent review, js! :clap: (now I want one :p) Thanks for sharing all the technical info on how the light works. I'll have to read this again a couple of times.

Incidentally, the new BD hybrid (incan/LED) headlamps are also regulated (by Willie's circuits too, IIRC). Not your "tube flashlight", but the regulation is there.


Newly Enlightened
Aug 30, 2004
I ALMOST feel bad that I just spent a big chunk of my light budget on a Ti CR2 Ion and a Glaius. Note that I said 'ALMOST'. I'm not unhappy with my decision at all but I think I now have another "need".

Great post!


Flashlight Enthusiast
Jan 18, 2003
Saved to my hard drive for posterity's sake. Excellent review.

I've had my SF A2 for several years, it is one of the earlier versions with a S/N below 800. I've bought and sold many SF lights over the years, but the A2 is one I always plan to hang onto. The A2 and my early L1 (S/N below 200) are two of my favorite lights and complement each other very well. In fact, the SF A2, SF L1, ARC AAA and my Tigerlight FBOP handle 99% of any situations I could ever need a flashlight for, with the A2 being able to cover about 90% of those situations by itself. It has to be the most versatile light I own, by far.

Not open for further replies.