Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1xAAA

[milkyspit]

I finally got around to measuring the lux of three contenders for keychain and pocket EDC: Infinity Ultra-G, Arc-AAA Premium, and Dorcy 1aaa. The test was done in a dark room with background reading of 0 lux, at a distance of one meter from the light meter. Note that the Arc-AAA was a factory second. Sorry, but that's all I've got around here!

Now, without further ado...

Infinity Ultra-G: 25 lux
Dorcy 1aaa: 18 lux
Arc-AAA Premium (factory second): 15 lux

Edit: I should also mention that all three lights are using alkaline cells that have no more than perhaps 10 minutes' runtime each.

--Scott

 

[red_robby]

milky
I've tested more then 10 dorcy 1AA, I've got 9 ARC AAA's
2 of those are seconds, 2 are LE firsts , the rest are all
firsts too.
Dorcy is noticeably brighter every time...
and by the way, i don't work for dorcy or anything like that
I just got excited when another company put out a single AAA
led light, and I was not disappointed with it.

 

[Mark_Larson]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

IIRC, the Dorcy 'overdrives' the LED in the beginning so its very bright, but the brightness drops off sharply, in comparison to the Arc AAA, which maintains its "new" level for a period of time.

 

[milkyspit]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

Mark, technically, all three lights overdrive the LED! /ubbthreads/images/graemlins/icon15.gif You're right, though, that the Dorcy 1aaa isn't regulated... or at least isn't as EFFECTIVELY regulated as the other two. It remains to be seen (by someone with a greater mastery of circuit reverse-engineering than I have!) whether Dorcy 1aaa does any sort of voltage regulation, or simply boosts the voltage in a completely unregulated manner... /ubbthreads/images/graemlins/confused.gif

 

[Cougar_Allen]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

See my first post here: Re: Dorcy 1AAA review...

Unless my eyeball gauge is way off, the Dorcy is regulated.

-Cougar :{)

 

[milkyspit]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

Cougar, you know I like these little lights! /ubbthreads/images/graemlins/wink.gif

Several posts above yours, Radellaf presented evidence based on current draw that suggests the Dorcy 1aaa is probably NOT regulated. His post is over here. I do believe what you said about your "eyeball gauge," but the trouble with that gauge /ubbthreads/images/graemlins/grin.gif is it perceives light in nonlinear fashion. In other words, a light source TWICE as bright as another (according to a light meter) would look only a little brighter to the eyes. So in a way, you're both right! Unregulated, but close enough for the eyes to be happy. /ubbthreads/images/graemlins/grin.gif /ubbthreads/images/graemlins/grin.gif

 

[Roy]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

Someone send me the light in question and I'll do a runtime plot on it! /ubbthreads/images/graemlins/smile.gif

 

[Cougar_Allen]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

I'm still a newbie at this and I'm finding it confusing.... I understand that the human eye can't detect a difference in brightness less than a factor of two; I knew that when I did the test, so I tried to gauge as accurately as possible within that limitation. I compared to a stock minimag, and very late in the test I tried changing to a fresh battery and back again and couldn't detect any difference. It wasn't less than half as bright with the used battery as with the fresh one, I'm sure of that much....

I don't know how to reconcile that with Radellaf's observations. He suggested it's wasting power with excessive overdrive through most of the battery's life, which would seem to account for brightness remaining nearly constant, but then how can it be brighter than the ARC AAA with similar runtime if it's wasting that much power? I don't know, maybe the ARC wastes power too?

I found an old light meter in my closet that I might be able to get working. It takes an odd kind of coin or button cell that I'm having trouble finding. But as best I can recall from my photography days, don't light meters use the same geometric scale as the human eye, so the meter won't be able to detect a difference less than 2x either?

I have two of the Dorcys now, so I could try running one and turning on the other occasionally for direct comparison. I have another led light that definitely isn't regulated, too; maybe if I run them both at the same time for a few hours that'll prove something....

I hope Roy can get hold of one to test soon.

-Cougar :{)

 

[milkyspit]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

Cougar, we're talking about some subtle stuff here, and definitely some nonintuitive concepts. You're right, one thing that explains some of what you're seeing involves heat. As an LED is overdriven more and more, it tends to generate less light and more heat, and this trend gets worse the farther things are pushed. For instance, the 5-watt X3T emitter in the MR-X flashlight increases its light output by 38% when current is DOUBLED. That's not a great tradeoff! /ubbthreads/images/graemlins/icon15.gif

So if that little Nichia is overdriven then drops back to a more standard amount of current, the light output might only decrease by a little... and your eyes might not be able to see THAT difference at all! You might lose half your current flow and still perceive the light to be almost as bright as when the battery was fresh.

Regarding the Arc-AAA, Doug Owen did some interesting research suggesting that light isn't regulated all that well, either! /ubbthreads/images/graemlins/ooo.gif It probably benefits from the same light vs. heat and human vision phenomena as does the Dorcy 1aaa. On the other hand, Doug found that the EverLED bulb, for example, is a well-regulated light source.

Arc-AAA isn't all that efficient, either. Based on some off-the-cuff calculations, one might expect an LED driven at 30mA to last 15 hours or so on an alkaline AAA cell, but in reality, my understanding is that the Arc-AAA lasts more like 6-8 hours per cell. There's plenty of room for improvement here. If the Dorcy 1aaa lasts at high brightness for nearly 10 hours, it's still not all that close to being perfectly efficient, yet it lasts longer than the Arc-AAA. In both cases, that lost energy generally becomes heat, though you probably wouldn't notice it given the small amount of energy we're talking about. (With something like the Surefire L4, on the other hand, you definitely feel it! /ubbthreads/images/graemlins/mad.gif )

Regarding light meters, the Meterman LM631 meter (popular among many CPFers) can detect the changes in brightness that the human eye misses. I'm not so sure about photography light meters; I've never used them! /ubbthreads/images/graemlins/crazy.gif

Try this when comparing the Dorcy lights: hold both over your head pointed at the ceiling, and turn on first one, then the other, one at a time, like an A-B test. The trick is to look straight ahead and get a sense of the brightness in the room; don't look directly at the lights! When you do this, you're getting a sense of the TOTAL light output (lumens) for each light and eliminating the effect of different beam patterns. This test allows you to compare the total output of even wildly different lights... for example, I discovered that my Surefire L4 and Streamlight Scorpion put out about the same amount of light despite the fact that they have essentially opposite beam patterns! /ubbthreads/images/graemlins/smile.gif

Now that you've got a DMM, you could also remove the tailcap every once in a while, set the DMM to the "10A" current measurement setting, put one probe on the bottom of the battery and the other on the exposed aluminum at the bottom of the battery tube. This will complete the circuit, light the Dorcy, and give you a reading on current flow. Keep records and see how current flow changes as the amount of runtime on the cell increases!

Roy, I've got lots of Dorcy 1aaa lights but am swamped with other things I need to get done in the next few days. I'd also like to use the Dorcy 1aaa to test the runtime apparatus I've been working on, the ROD. /ubbthreads/images/graemlins/grin.gif Eventually I'd be happy to send a Dorcy your way for testing, but if someone else can get you one quicker, that would be better! Thanks for the generous offer, though. Anyone want to get a Dorcy 1aaa to Roy right away?

 

[Cougar_Allen]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

Boy, I have a lot to learn about flashlights.... I tried comparing an alkaline with four hours on it to a nearly fresh rechargeable alkaline in the other flashlight. With the blink on the ceiling test my eyeball gauge was able to detect it was definitely slightly dimmer -- maybe somewhere in the neighborhood of half as bright. I swapped batteries to confirm it was the difference in batteries I was seeing, not a difference in the two flashlights.

Unscrewing the tailcap and applying the leads of my new multimeter (Thanks, Milkyspit!) to the battery and the body of the flashlight, I saw the Dorcy was drawing .849v * .14a = .11886w from the used cell. Pulled the cell and measured no-load volts at 1.290, so that's far from a dead cell. Popped in a fresh cell (no-load volts 1.590) and the Dorcy drew 1.148v * .28a = .32144w from that. So when wattage drops to 1/2.7 brightness drops to maybe somewhere in the neighborhood of 1/2. An unregulated led is expected to be more efficient at lower power.

Conclusions:

1) The Dorcy is not regulated at all, not even partially.

2) The eyeball runtime test I did earlier was not worth as much as I thought it was. (But I can do better with the blink on the ceiling test.)

3) Voltage regulation doesn't make as much difference as I thought it did, not in practical terms.

I can think of a number of experiments I can do just with the equipment I already have, and once I get that light meter working or get a photocell to hook up to the multimeter ... a variable power supply will be useful, too; I can improvise that easily ... I can see hours of fun ahead; I can waste as much time testing flashlights as I do testing knives! I've been playing around with overdriving various PR base krypton and xenon bulbs; now I want to quantify how much more light I'm getting....

Should I post my tests of PR bulbs in this forum or the mods forum or the electronics forum or where? I've already figured out if I cross-post anything on this website I shall be cast into the outermost darkness....

I notice when I test amperage with the tailcap off the flashlight lights up; when I test voltage it doesn't. I guess that means multiplying them together doesn't really tell me how much wattage the flashlight is using, does it.... The voltmeter adds so much resistance the flashlight doesn't light so the voltage it draws without the voltmeter in series must be higher. If I connect the cell to the flashlight body with a wire and put the voltmeter in parallel will that give a more accurate reading?

-Cougar :{)

 

[milkyspit]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

Cougar, way to go! That's some good testing you're doing. /ubbthreads/images/graemlins/thumbsup.gif

There are a couple benefits to regulation that we don't talk about enough. First, it can protect the LED from overdriving! For example, the regulation circuit in MR-X guarantees that the X3T emitter will get a maximum of 1.5A current draw, which is already quite high! /ubbthreads/images/graemlins/mad.gif The emitter is extremely well heatsinked and probably handles the overdriving in this case pretty well. But without the regulator in place, fresh cells might pump much MORE current through the X3T and cause bad things to happen to it! /ubbthreads/images/graemlins/ooo.gif So a regulator in that sense acts like a sort of specialized surge protector.

Another benefit to the regulator is that it keeps the cells from spewing out all their energy in the beginning of runtime, when it would only be wasted as heat (or worse!), and essentially saves the excess energy for use later, when it's needed! Think of this in terms of horse racing. If the horse starts the race at top speed, he'll usually run out of energy and be soundly beaten by the end. Instead, the jockey holds the horse back a little at first, then entices him to empty the tank, so to speak, in the final turn. This way the horse uses his limited energy much more effectively. And so it is with a regulated flashlight! /ubbthreads/images/graemlins/grin.gif

Regarding your measurements of current and voltage by removing the tailcap: strictly speaking, it's inaccurate to measure voltage this way. The rule of thumb is voltage ACROSS, current THROUGH. You hookup the way I described for current because the DMM needs to be "in the loop" of the circuit to measure the current running THROUGH it. However, voltage needs to be measured not in series with the circuit, but rather in parallel with it. That would mean somehow touching the DMM probes to the positive and negative terminals of the battery WHILE the light was running normally! Such as with the tailcap on. This, of course, is pretty difficult to do... you'd probably need to make some temporary connections with wires to get access to everything for that test. If you have the ambition, go for it! If not, the "loaded" current (current while running) and "unloaded" voltage (battery's voltage when pulled out of the light) will still give at least a partial picture of what's happening.

Regarding photocells, the CdS ones (cadmium sulfide) are available cheaply from Radio Shack. ($1-2.) CdS photocells are like the ones used in the "electronic eyes" used on nightlights to turn themselves on at dusk, for example. They're the little round things with the zigzag wire pattern visible across the surface. If memory serves, you'd measure the resistance of the photocell, which would change depending on how much light hit it! Easy to test, though note that response isn't necessarily linear. (In other words, half the resistance doesn't necessarily mean double the light!) They'd be good for doing some sort of brightness over time runtime plot.

Ready for me to spend some more of your money? /ubbthreads/images/graemlins/wink.gif

Take a look at this thread I did a while ago, which reviews a neat power supply available from MPJA.com (a surplus electronics warehouse) for $60. This particular PSU is nice in that you can adjust both voltage AND current! Perfect for flashlight experimentation. I would also suggest a pair of banana jack test leads available from Radio Shack for a few dollars, as they plug into the terminals on this PSU easily. I think I mention the RS part number for those toward the bottom of that PSU thread. Highly recommended! /ubbthreads/images/graemlins/grin.gif

As far as where to post, sometimes you're just not sure. It's more important to post in exactly ONE place than to get the place exactly right. Personally, if what I'm doing involves electronic components, I usually put it in the electronics section, and if it involves putting together some custom light or experimenting with a light-generating circuit, I put it in the mods section.
/ubbthreads/images/graemlins/icon3.gif

Sorry I keep dropping these long posts on you! /ubbthreads/images/graemlins/icon15.gif

Yes, there's lots to learn about flashlights, especially the ones that are no longer just a bulb and a battery. Lots of subtleties. As far as knives... well, I'm afraid I don't really understand how you could spend all kinds of time reviewing one of THOSE. /ubbthreads/images/graemlins/icon15.gif I'm sure it's just a reflection of my own ignorance on those matters! Maybe you could teach me... sort of a knowledge trade, you learning flashlights, me learning knives? /ubbthreads/images/graemlins/thinking.gif

Not sure I've even got a single genuinely good knife in the house. What I'd love to find would be a pocket knife that's both "good" by your standards and affordable! The affordability test is that I shouldn't be losing sleep at night worrying about the economic impact of losing the thing...
/ubbthreads/images/graemlins/sleepy.gif

 

[TCG]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

Cougar. with reguard to measuring voltage and current in the Dorcy aaa, have you considered taking it apart? The body of
the Dorcey is reversable.

To take mine apart, I removed the tail cap, and used a butter knife across the slots. This combined with tight grip on the bezel broke the threads loose. If this does not work, try wrapping the head in a dishtowel and using pliers.

Once the circuit is out its trivial to measure either voltage across the circuit or current flowing through it. On mine, I get 1.253 volts and 118 ma with the original battery. I get 1.492 v and 198.7 ma with a new battery. Note that both the current and voltage measurements seem to be dropping quickly with the new battery. These measurements were taken within about 3 seconds of turning the light on.

Keep in mind that your DMM actually measures current by measuring the voltage drop across a internal shunt resistor and calculating current from the value of the shunt resistor and Ohms law. The current draw without the meter attached will be slightly higher.

With two DMM's and an extra pair of hands, one could measure bothe voltage and current at the same time. For some truly usefull data, unsolder the led and measure current and voltage there as well. Then one could come up with efficiency curves for the boost circuit as a function of voltage.

ps. I just used my LM631 and see 15 and 20 lux at one meter for the above two batteries.

pps. What exactly is "partial regulation"? Regulation to me suggests a feedback loop and something compensating for voltage drop of the battery by pulling more current. Is "partial regulation" just marketing hype? Its like being partially pregnant.

 

[Cougar_Allen]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

Thanks for the tips, guys. Re knives, go to www.bladeforums.com -- the Candlepowerforums for knife knuts! There's a considerable overlap of members.... You'll find there are some knives that are both pretty good and very affordable, and you can marvel at the lunatics who spend all our time making knives, testing knives, reviewing knives, arguing about knives ... /ubbthreads/images/graemlins/cool.gif

I thought I was the only one who didn't understand what partial regulation is. /ubbthreads/images/graemlins/faint.gif

-Cougar :{)

 

[milkyspit]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

TCG, glad to see some consistency in our lux readings. Remember, I used a battery with about 10 minutes' elapsed runtime on it, and measured 18 lux, right in line with your 20 lux for new battery, 15 lux for the original one. Sounds like the makings of an accurate reading! /ubbthreads/images/graemlins/grin.gif

That butter knife trick is clever, too. I used two pairs of pliers to get my Dorcy apart, and in the process ripped the large rubber grip plus scratched the side of the head. /ubbthreads/images/graemlins/frown.gif But I've got 8 more of these little guys, so it's not that great a loss! /ubbthreads/images/graemlins/smile.gif Next time I do it your way.

With measuring current with a DMM, I learned about the potential for interference the hard way, after thinking my original DMM was defective and buying another! /ubbthreads/images/graemlins/icon15.gif At any rate, now standard practice is to use the HIGHEST acceptable current measurement range on the DMM to minimize this effect. On many DMMs, the highest range is "10A" or even "20A," plus you need to plug the red test lead into a special "10A" or "20A" socket.

Partial regulation? Could be marketing hype or a great way of describing what you SEE (not what happens electrically), depending on whether you're in the "half-full" or "half-empty" crowd. There is no technical definition for such a thing that I know of, and in fact such regulation technically isn't "regulated" at all. You're right... regulation means maintaining a desired output characteristic despite a varying input.

It may simply be that a circuit using voltage regulation has diminished current flow as the battery drains, then at some point voltage begins sagging, too, and when it falls near Vf the light appears to "suddenly" dim into a much lower "moon mode." Sounds like what the user sees happen, but electrically it's a little different ballgame!

Cougar, thanks for the link!

 

[Phaserburn]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

Milky, have you ever worked as a teacher? You do a good job! /ubbthreads/images/graemlins/blush.gif Have you tried a lux reading on the Ultra-G with a li-AA? I have an Ultra-G en route from that nice little Countycomm sale, and I wondering what brightness might be picked up from that kind of cell.

 

[milkyspit]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

Phaser, aw, shucks, thanks for the compliment! /ubbthreads/images/graemlins/blush.gif

Wonder how an old guy like me breaks into the teaching world at this point? /ubbthreads/images/graemlins/thinking.gif

Just for you, I pulled the test equipment out again and tested using a lithium AA cell. The results?

Infinity Ultra-G (lithium AA): 27 lux
Infinity Ultra-G (alkaline AA): 25 lux

That's probably not what you were expecting. /ubbthreads/images/graemlins/icon15.gif

What does it all mean? Well, it's certainly true that the lithium AA makes the Ultra-G brighter, and I would even go so far as to say that I think I can see a difference. But my light meter tells me a different story, one in which the lithium AA only makes Ultra-G marginally brighter. Is the increased brightness worth $2 for the lithium AA cell vs. $0.25 for the alkaline AA?

There are other ways to look at this of course. For instance, you may decide that the long runtime of the Ultra-G makes the difference in price for lithium vs. alkaline a non-issue. And the lithium AA is certainly a more rugged cell. It performs MUCH better in cold weather, has dramatically longer shelf life, and ought to provide higher capacity, running brighter for longer than an alkaline ever could. In fact, the lithium will likely still be BRIGHT at a point when the alkaline would be DEAD! Plus the lithium cell is lighter weight, though the fraction of an ounce difference is probably not the key selling point here.

So the decision's up to you, but them's the facts. /ubbthreads/images/graemlins/grin.gif

 

[this_is_nascar]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

My biggest plus for the lithium AA cell has always been the difference in weight. Put an alklaline in your Ultra or ARC-AA, than loosely bounce it around in your palm. Now do the same test with a lithium installed. It's like night and day.

 

[Phaserburn]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

Nascar, I will be trying that shortly. Having the lithium cell "assist" in regulation is also a plus.

Milky, kudos as always. I've seen runtime plots of the Ultra-G on alks, but not on a lithium AA?

 

[milkyspit]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

Nascar, neat! I didn't think anyone would care about the weight difference in lights this small.

Phaser, maybe Roy has some runtime graphs using AA lithiums? At any rate, you could expect a much flatter discharge curve. The typical lithium discharge involves a drop in output in the initial minutes followed by a very long, very flat stretch, and a steep drop to nothing at the end. I sometimes refer to lithium AA cells as a "poor man's regulator." /ubbthreads/images/graemlins/grin.gif

 

[this_is_nascar]

Re: Lux measured: Infinity Ultra, Arc-AAA, Dorcy 1

Yep, weight is important, especially if it's being carried in a pocket. For some reason, even the slightest difference in weight makes a big differences carrying this light in the front pocket of a pair of Dockers.

Just for grins and tickles, I took one of my newly acquired AAA to AA cell adapters to power my ARC-AA with an AAA cell. That wieghs even less than the AA-lithium. I know it's counter-productive as to why the ARC-AA even exists, however it was a cool excersize.