Surefire A2 Runtime Data, for LED-only use, 3 different colors compared

[Atomic_Chicken]

Greetings!

I did an experiment this evening with 3 of my Surefire A2's in order to see if the different colored LED's might lead to differences in run-time when using the LED's alone, without the incandescent lit.

I gathered data for the Red, White, and Yellow-Green A2's (those being the only colors I own). OK... on to the methodology:

I figured that measuring ACTUAL runtime accurately for 3 different A2's running in LED-only mode would be time consuming, and possibly inaccurate. I would have to make certain that all the batteries had the EXACT same amount of charge left in them, a near-impossible task. I began thinking about how I could accomplish this in a scientific and accurate way, and hit upon this idea.

I took a single A2 body, set the tailcap mid-way so that it would allow the LED's to light when pressed, but still have the incandescent light locked out. I then pulled the head and removed the bulb. I ran Alligator clips from the housing and metal ring, and put a digital DC ammeter (Fluke model 77) in series. I then pressed the tailcap button, and took a current reading after about 2 sec. of settling time for my meter. After recording the reading, I removed the heads and lamps from the OTHER two A2's, and attached them to the SAME BODY with the SAME BATTERIES. This will cancel out any battery differences between units, as every set of LED's will be run off the same batteries with the same charge level. Since the LED current limiting resistors are in the head, on the LED ring, I don't need to worry that the different tailcaps would affect anything. I took current measurements for the other two A2 heads, and recorded the results. Finally, because of a nagging thought that maybe even the SHORT time the measurements were taken might have just slightly drained the battery and skewed the results a bit, I repeated the measurements again to verify that the battery voltage (and corresponding current, due to the lack of LED regulation) had not changed. The measurements were identical both times, dispelling this concern.

These are the results:

White A2 - 90.5mA
Red A2 - 80.1mA
Y/G A2 - 87.1mA

The figures were interesting, I expected more variance between colors because of the MUCH dimmer Yellow-Green light, I thought that it might draw quite a bit less current than the other two. WRONG.

The data basically says this:

(1) For the lowest power drain (and theoretically, the longest runtime) when using the LED's alone, the Red A2 appears to be the one to buy.
(2) There isn't much difference in power drain between the White and Yellow-Green versions of the A2, but the Yellow-Green does slightly better.
(3) You can realistically expect 10% longer runtime when using Red LED's over White or Yellow-Green, assuming that the delta-brightness vs. delta-voltage-drop curve for all the different colors of LED's is linear (which might NOT be a good assumption).

Note that the above data is applicable for running the A2 in LED MODE ONLY - and says nothing about full-beam incandescent use. I basically wanted to see if any of the 3 colors did better than the others for long-runtime use in LED only mode... and I think the data pretty much speaks for itself.

Best wishes,
Bawko

 

[leukos]

:goodjob:

 

[SunStar]

Good post AC!!

I would be curious to see beam shots from the 3 with a consistent basis to compare throw, flood and overall brightness.

I think this could really help some folks decide which color is best for them - something frequently asked here on CPF.

 

[Illum]

well done:goodjob:

so the LEDs in A2-RD will last a tad longer than A2-WH

 

[Atomic_Chicken]

Greetings!

Good post AC!!

I would be curious to see beam shots from the 3 with a consistent basis to compare throw, flood and overall brightness.

I think this could really help some folks decide which color is best for them - something frequently asked here on CPF.

greenLED has already done this. See this thread:

http://www.candlepowerforums.com/vb/showthread.php?t=119758&page=1&pp=30

Best wishes,
Bawko

 

[SunStar]

Thanks much! I'm not sure how I missed that one.

Another excellent post by GreenLED.

 

[js]

Atomic Chicken,

Good idea and good effort,

BUT, there's one major flaw in your setup! The Fluke 77 in DC current measuring mode, with the leads set for the 400 mA range, has about 2 ohms of resistance!!!. You can verify this for yourself by having the leads in position for measuring resistance, set the meter to measure ohms, then take the red lead (not common) and put it into the hole for the 400 mA current measuring position. You will get about 1 or 2 or 3 ohms.

I just did this with my Fluke 179 and got 1.8 ohms.

So you are adding more resistance to the circuit.

How much more, though?

The tailcap has almost exactly 10 ohms of resistance, but the LED ring also has resistors in it, and I'm pretty sure that the surface mount resistors in the LED ring vary with LED color. If I remember correctly the yellow green ring has the highest resistance. (So, it would have the least percentage offset.)

So the point is that you are skewing the results differently for the different LED rings! If it were only the tailcap that had the resistance in it, then we could at least say from your results that white>yellow green>red. But we can't say that.

Here's how I suggest you measure current draw through the LED ring:

Remove the tailcap. Now take a precision 10 ohm resistor and put one end on the negative end of the battery stack, and the other end on the butt end of the body. The LED's will light.

Now, measure the VOLTAGE across the resistor.

Then calculate the current using V=IR. Also, measure the resistance of the 10 ohm nominal resistor with you Fluke and use that number in the formula (although it shouldn't be much different than 10 ohms if its a metalic oxide resistor)

You should get just about 110 mA for the white LED ring with fresh batteries.

 

[Atomic_Chicken]

Greetings, js!

Atomic Chicken,
Good idea and good effort...

Thank you!

Now, on to your concern about the test method.

You say that the Fluke will add resistance to the circuit when measuring current. This is true - all methods of current measurement other than inductive coupling (used by clamp-style meters) will introduce a small amount of additional resistance to the circuit being measured. However, I do not think that this will introduce errors to the results because:

remove the tailcap. Now take a precision 10 ohm resistor and put one end on the negative end of the battery stack, and the other end on the butt end of the body. The LED's will light.

Now, measure the VOLTAGE across the resistor.

Then calculate the current using V=IR.

Your method also adds resistance to the circuit, EXACTLY like the Fluke does when it measures. Basically, the Fluke (and other multimeters) will add a small (typically under 3 or 4 ohms) resistor in series with the circuit being measured, then measure the VOLTAGE across the resistor in order to obtain a reading. The multimeter circuitry will do the necessary I=E/R calculation internally to display the correct current measurement from the voltage it measures.

So... basically you are recommending a method for me to use that is ALREADY being used by the multimeter.

Also, because it is always adding the SAME resistance regardless of which LED head is being used, the various different power-drains of the different heads should still accurately reflect in the resulting current measurements. Think of it this way: If the tailcap resistance is 10 ohms or 12 ohms (meter added), it is irrelevant when comparing the runtime of the differing LED heads... because if it's 10 ohms, the runtime will be shorter on all of them, and if it's 12 ohms the runtime will be longer on all of them. However... the runtimes should all be comparitively the SAME percentage different from each other regardless of tailcap resistance.

You should get just about 110 mA for the white LED ring with fresh batteries.

This does not surprise me - I had quite a bit of use on the A2 I used as the battery holder, so I suspected that the current measurements I took wouldn't be close to what they would be if the batteries had a "full tank". However, I was taking the readings as a comparison of power-draw between the different colored heads (which will cancel the battery voltage out), so this didn't concern me nor do I think it would be all that much different percentage wise from full-tank readings.

Thanks for the suggestions - please tell me if I am in error with regard to any of my assertions.

Best wishes,
Bawko

 

[js]

Bawko,

No.

The 10 ohm resistor is not an additional resistance to the circuit.

The 10 ohm resistor replaces the 10 ohm resistance of the tailcap. To use my method, you REMOVE THE TAILCAP and put a 10 ohm 1/4 watt resistor in its place.

Both tailcap and a 10 ohm resistor have 10 ohms of resistance. So there is no additonal resistance added to the circuit with this method. The only difference is that with a 10 ohm resistor you can easily measure the voltage across it, whereas it is impossible to do that with the actual tailcap.

Clear?

 

[Burgess]

Thank you for your efforts, Atomic.


Subjectively, if you could have only ONE Surefire Aviator A2,

which LED color would you choose ?


And . . . .

Of the 3 colors which you currently own,

which one is your LEAST favorite color ?


Just tryin' to decide here.

 

[Atomic_Chicken]

Greetings!

Thank you for your efforts, Atomic.

Subjectively, if you could have only ONE Surefire Aviator A2,
which LED color would you choose ?

And . . . .

Of the 3 colors which you currently own,
which one is your LEAST favorite color ?

Just tryin' to decide here.

Wow... you ask the easy questions, don't you!

Seriously - this is an amazingly hard thing to answer. When I initially went in to a local store to replace my A2 last Monday, I had decided that I would probably buy the White LED version. However... when I got there, they had both White and Red versions available, so I just HAD to compare the two in the store. You know what? I just couldn't decide - after a while I just gave up and bought both.

If I could only have one, though, I think it would have to be the White. The reasoning is that you can always take a White and add a red or yellow-green filter to it, but you can't take a red or yellow-green and add a white filter!

White seems to be the most useful for all-around utility. It had excellent color rendition and is plenty bright for 3/4 of my uses as a flashlight - without the need to activate the incandescent beam. A lot of people object to the "angry blue" tint of the White LED A2 - but I don't personally find it very objectionable at all.

Red is unquestionably the best for preservation of dark-adapted night vision. No other color will allow this - only Red will not bleach out the protein that builds up in the rod cells of your eyes that allows for dark-adapted vision. However... I don't particularly care for the "world view" under red light - it's somewhat hazy and indistinct (at least to my eyes) and I find red difficult to read with. Depth perception and color rendition are not very good under red light either, at least for me.

Yellow-green is a fantastic colored A2. It is probably the most useful color other than white - it gives good color rendition and is dim enough that it won't "shock" your night vision immediately. If you only hit the light sporadically and for really short durations, you could probably maintain your night-adapted vision fairly well with the yellow-green light. I could be wrong, but my experiments so far have indicated that the yellow-green A2 is very pleasant to use in pitch-black and I can still see pretty well in the dark after using it for short bursts. Where the yellow-green A2 really shines (pun intended...) is for detail recognition. It's VERY easy to read under yellow-green light, and depth perception and shading are extremely well preserved. If you can't tell, the Yellow-Green A2 has rapidly grown on me - I really like it a LOT.

My personal ranking, from 1-10 of the A2 versions would go like this:

White LED: 10
Red LED: 7
Y/G LED: 6.5

I rank the Red above the Yellow/Green version ONLY because of its unique ability to preserve dark-adapted vision... I would probably give it a 4 otherwise.

Your own mileage may vary.

Best wishes,
Bawko

 

[Atomic_Chicken]

Greetings!

Bawko,

No.

The 10 ohm resistor is not an additional resistance to the circuit.

The 10 ohm resistor replaces the 10 ohm resistance of the tailcap. To use my method, you REMOVE THE TAILCAP and put a 10 ohm 1/4 watt resistor in its place.

Both tailcap and a 10 ohm resistor have 10 ohms of resistance. So there is no additonal resistance added to the circuit with this method. The only difference is that with a 10 ohm resistor you can easily measure the voltage across it, whereas it is impossible to do that with the actual tailcap.

Clear?

Clear. I don't think it will make a lot of difference, again because of the 10 ohm vs. 12 ohm tailcap thought experiment I wrote about in my post - but I will re-do the measurements with the 10-ohm resistor as you suggest and post the results here sometime early next week.

Thanks again!

Best wishes,
Bawko

 

[js]

Bawko,

Yeah. It might be a small effect; it we might be surprised! Either way, I'm very interested to see the results.

I was talking on the phone with SilverFox when he told me about this thread, and I immediately brought up the resistance involved in the 400 mA range Fluke current measuring. I measured the ohms of my Fluke in that range and it was about 2 ohms. Which is twenty percent of ten ohms. So I took twenty percent of 110 mA and got 22 mA. 110 - 22 = 88 mA. Which is pretty close to your value, isn't it.

That isn't the proper way to figure that out, actually, as it ignores the resistance in the LED ring. It'd best be done with the voltage source, forward voltage of the LED ring, total resistance with and without the Fluke (and leads and clips and so on). But I'm lazy, so I did that back of the envelope calculation.

Anyway, interesting. I'm especially curious to see if the 10 ohm resistor method reverses the ranking of red and yellow-green.

 

[Atomic_Chicken]

Greetings!

I finally had time today to re-measure the 3 lights using js's method, and added one light to the mix. I received the White A2 modded with THC3 LEDs today, so I measured that light along with the original three.

Test method: First, I measured the battery voltage of the batteries I was going to use for all the tests. The pair came in at 5.95 Volts. Then, I selected a 10 Ohm 1% resistor and measured its actual value. The resistor I chose just happened to be exactly 10.0 Ohms, measured at the end of the alligator clips. Finally, I took each light, one at a time, and removed the tailcap and batteries. I then put the test batteries in the light, then shorted the negative end of the batteries to the flashlight body through the 10.0 ohm resistor (simulating the tailcap being at switch position 1). I measured the voltage, wrote it down, then replaced the flashlight's original batteries and tailcap. I repeated this procedure for all 4 lights, using the same set of test batteries for each measurement, then re-measured the first light to make certain that any battery drain from measuring the 4 lights wasn't giving an artificially low reading on the last light tested. The first light re-measured at the same voltage, proving that the batteries did not drain enough during testing to affect the final light's measurement.

Here are the results
===============

Red A2: 0.946 Volts (Calculated: 94.6 mA at 5.95 Volts)
Y/G A2: 1.125 Volts (Calculated: 112.5 mA at 5.95 Volts)
White (Stock) A2: 1.133 Volts (Calculated: 113.3 mA at 5.95 Volts)
White (THC3 Modded) A2: 0.728 Volts (Calculated: 72.8 mA at 5.95 Volts)

The results seem to mirror the original measurements quite closely, with the Red doing better than the Y/G or White, and not much power savings by the Y/G A2 when compared to the White A2. The real surprise here was the White A2 that I received today, which has been modded with Optoelectronics THC3 White LEDs. It came in with a lower measured power draw than even the Red A2. I don't know if the person who modded the light changed out the current limiting resistors of this light or if the THC3 LEDs just draw less current than Nichias... I'll disassemble that light in the next few days and see how its resistor values compare to the stock (Nichia) White A2.

Best wishes,
Bawko

 

[js]

Bawko,

How cool is that?!? Neat. You got about 110 mA, too. I was curious to see what you'd measure.

Nice work.

As for the THC3 LED's, I've been told by milkyspit that they have a higher forward votlage than the stock LED's, and I suspect that this explains the current draw difference.

Or it could be the resistors were changed out.

Your Fluke can measure the forward voltage of the LEDs; maybe you could add that to your measurement list?

Great stuff, Bawko! Thanks for doing this.

 

[Atomic_Chicken]

Greetings!

Bawko,

How cool is that?!? Neat. You got about 110 mA, too. I was curious to see what you'd measure.

Nice work.

As for the THC3 LED's, I've been told by milkyspit that they have a higher forward votlage than the stock LED's, and I suspect that this explains the current draw difference.

Or it could be the resistors were changed out.

Your Fluke can measure the forward voltage of the LEDs; maybe you could add that to your measurement list?

Great stuff, Bawko! Thanks for doing this.

Hey js... no problem! I enjoyed the measurement, and I like learning all I can about the tools I use.

I have an unmodded Blue A2 and a THC3 Modded Green A2 on the way, they should both be here sometime early next week. I also have a friend in town who owns an unmodded green A2 that I can borrow for an evening. Adding those 3 lights to the mix will provide current-drain figures for ALL versions of the A2 currently made by Surefire, as well as two modded THC3 LED A2's. I'll probably take a few beamshots as well... if I can find the time. Should be fun!

Best wishes,
Bawko