Peak AAA Run-time results

[Beacon of Light]

Hi guys some of you may have followed the AAA run-time test I did in three parts in the main LED Flashlights forum. I figured I'd post it here. All lights used Duraloops (Duracell NiMH rechargeable LSD batteries) with a refresh cycle on my Lacrosse charger. Most were slightly higher than 800mAh off the charger, but all cells were fairly consistent.

Peak Matterhorn 3-LED SP 11+ hours @ .99v (light dim)
Peak Eiger #3 15+ hours @ 1.14v (light dim)
Peak Matterhorn 3-LED HP 15+ hours @ 1.16v (light dim)
Peak Matterhorn 1-LED HP 16+ hours @ 1.16v (light dim)
Peak Matterhorn 1-LED SP 37+ hours @ 1.07v (light dim)
Peak Eiger #0 37+ hours @ 1.10v (light dim)
Peak Eiger #Subzero 37+ hours @ 1.11v (light dim)

I think after the test I have more questions than I did before I ran the test. Here goes and feel free to chime in if you have any thoughts or answers.

1) Why would the High Power Matterhorn 3-LED run longer than the Standard Power Matterhorn 3-LED by quite a big margin? Also the HP light was brighter than the SP for a majority of the time as well. The remaining voltage in the HP was also greater. I spoke to Bob @ RMSK and he just said due to LED variances. I just figured this is too big of a difference to just say LED lottery. Any thoughts?

2) Why would the older Matterhorn Standard Power last just as long as the Eiger #0 and Eiger #Subzero? The new Eigers with XP-Gs are supposed to be more efficient but that wasn't realized by this run-time test. Any explanations on this?

3) The differences between the Eigers themselves where I was expecting much higher run-time with the #Subzero and I didn't see that at all. The #Subzero was dimmer than the #0 which was expected but I figured the #Subzero with a 80ohm resistor (compared to the regular #0 with a 56ohm resistor) would have approx. 1 1/2 times the run-time of the #0, or don't resistors work that way?

Thanks and feel free to ask questions as I have learned a lot just doing the test.

 

[Dances with Flashlight]

SP Peaks were "Super Power" and not "Standard Power". But this explains only some of the apparent anomalies. Perhaps the others really are attributable to LED variances (this seems more likely than ghosts in the light machines).

 

[Beacon of Light]

Dances with Flashlight, are you sure? I know you mentioned that in the other thread, so when I talked to Bob @ RMSK about it he said the unstamped heads were standard power, he never said anything about there being a Super Power or a Xtra Long Runtime (I'd still be wanting one of those XLRs if they exist for obvious reasons).

Also if I were to accept the unmarked heads were "Super Power" then the results of the two 1-LED Matterhorns is too great to dismiss as variances in LEDs as the SP racked up 37+ hours where the HP ended at 16 hours. That comparison between the Matterhorn 1-LED SP and HP in the one that does make sense as I'd expect the "Standard Power" would get at least double the run-time of the High Power version.

Also just putting in a new battery in both the 1 and 3 LED versions the HP is much brighter so that would negate the idea that SP=Super Power, no?

 

[DUMDUM]

Beacon the older ones I sold you the unmarked heads were the standard power level back then. They did not even have the super power at that time
Bob

 

[Beacon of Light]

Right Bob, I know you told me that when we talked. I still can't get over the fact the 3-LED HP got 3 more hours of run-time than the SP and was brighter for much of the duration of the test, AND had more reserve voltage left (.17v more than the Matterhorn 3-LED SP!!!) I still can't explain how this would be possible. I know you said variances in LEDs of the same type but this is a huge difference in what should be the opposite with the SP being the longer lasting light.

 

[Dances with Flashlight]

I don't recall seeing any of those unmarked Matty heads, so no doubt Bob is right about standard power. One of my first Peaks is an older Matty marked SP - and it really is Super.

Not sure that any Matty's were made up in the XLR, but the second or third Peak light I bought was a Pacific AA XLR that was very long-running and very sweet - until ripped off by one of my brothers. Just another reminder about other Peaks at #0 or SubZero: they can be had in the El Cap or in the Logan, and no doubt others as well. Other options for long-running Peaks would be on the 2XAAA Eiger or 2XAA Caribbean bodies. With alkalines, runtimes should be quite adequate for most anyone in most any situation. The downside to all this, of course, is that these ultra-low level lights aren't very bright, and for old, vision-impaired guys like me their uses are pretty limited. I expect that (because of the great reviews you and Scout24 have done) I'll have to pick up an El Cap head in #0 or #1 for long duration emergencies (on a 2XAA Caribbean body it should be bright enough for me), but will probably keep the light loaded with a good old #8 for routine use.

It does seem possible that at different times in the past Peak used different resistors in its few classes of power levels, and this could explain some of the test results. These other questions are best left to Curt.

 

[Curt R]

We used three different 5 mm LEDs in the Matts, from SuperPowerLED to start with, and then the dash 27
and finally the dash 29. Each had a different output and drive requirements.

The Cree XPG-R5 LED is not more efficient than a good quality 5 mm LED at very low power drive currents,
the great advantage is far longer life at any drive and the ability to output up to 460 Lumens as apposed
to 7 to 10 Lumens. Also when you get down to very low drives, the higher ohm resistor is going to
consume more power than a lower ohm resistor. There is going to be a point where the law of diminishing
returns is going to apply.

The #0 LED is driven at 12 mA with a 1.5 volt battery, with heat-sinking that LED should last over
250,000 hours to 70% of initial brightness. Wake me when you are through with that test. A AAA Lithium
battery has twice the current density of the NiMH and should run twice as long. A 10440 Lithium-Ion AAA
should give 20 hours at twice the light output level as the NiMH.

Curt

 

[Beacon of Light]

The Cree XPG-R5 LED is not more efficient than a good quality 5 mm LED at very low power drive currents,
the great advantage is far longer life at any drive and the ability to output up to 460 Lumens as apposed
to 7 to 10 Lumens. Also when you get down to very low drives, the higher ohm resistor is going to
consume more power than a lower ohm resistor. There is going to be a point where the law of diminishing
returns is going to apply. Curt

Thanks for that Curt. That makes sense. So I guess below a #0 power level the law of diminishing returns apply in regards to run-time. I really like both the #0 and #Subzero but knowing they both have the same runtime, when I order a Logan, I think I'll just go for a Logan #0 if a #Subzero isn't going to achieve longer runtimes.

 

[defloyd77]

The Cree XPG-R5 LED is not more efficient than a good quality 5 mm LED at very low power drive currents,
the great advantage is far longer life at any drive and the ability to output up to 460 Lumens as apposed
to 7 to 10 Lumens.

I always wondered how 5mm LED's compared to XP-G's at lower drive levels, I would think the larger die size and more phosphor due to that large die would hurt performance at low levels more than a smaller die LED such as a 5mm. I kind of have a theory that the more efficient LED's get at normal drive levels, the more inefficient they will be when you want a low output such as .2 lumens. Maybe I'm just a little crazy, I don't know.

 

[Darvis]

This is great info, and I'm even more stoked that I was able to call Bob and get a few of those remaining unstamped Matty's!

Beacon, thanks again for running these tests; I too have learned much, especially when it comes to the capabilities of these little Eigers. I think its one of the most versitile and impressive little lights out there especially after seeing first-hand what it can do.

I know you like your runtimes, so I have to recommend you grab one of the Valiant Concepts AA adapters for the #0 light. Given the 47 hour run I got with an eneloop and my #1, I'd bet the #0 and a lithium primary would hit into the 60-70, maybe 80 hour area... That would be a formidable emergency light!

 

[Curt R]

Run time with the lower drive levels is dependent upon the watt energy density of the battery and the
conversion of the primary voltage source to that required by the LED. A 3 volt source will be better by about
20% than a 1.5 volt source and about 25% from a 1.2 volt battery. A typical 1.2 volt NiMH AAA has about
0.96 watts of energy. Divide .96 by the number of hours driven to get a reference. Use that reference to divide
the energy density of other batteries for an approximate run time.

AAA NiMH 1.2 volt = .96 watts
Alkaline AA = 4.2 watts
Alkaline AAA = 1.8 watts
Alkaline 2 x AAA = 3.6 watts
10440 Li-Ion - 1.9 watts
CR123A = 4.2 watts
Lithium AA = 4.5 watts
Super Alkaline AA = 4.8 watts
18650 Li-Ion = up to 10.8 watts
38120 Li-Ion 'M' 3.3 volt cell = 33 watts

Curt

 

[Ice]

@Curt R: I'm sure you meant Wh (WattHours) in your last post!

@Beacon of Light: Thanks for those interresting measurements!

@all:
As Curt said: Any resistor "consumes" energy (or better: converts electrical energy to heat), the question is: At what point will the energy consumed by the resistor be as large or larger than that consumed by the LED?

Now for some mathematics, if you don't mind.

If the Eiger #0 run for about 40 hours, this means that (with a good 1Ah-AAA-battery) there was an average current of about 0,025A.
At this current the 60Ohm-resistor consumes the power P = UI = R I^2 = 60Ohm * 0,0006A^2 = 0,036W.

If you multiply that with 40 Hours, you get a very interresting result:
The resistor consumes 0,036W*40h=1.4Wh, which is about ALL THE ENERGY the battery has!!![/I]

Thus with those resistors the LED itself consumes a negligible part of the energy!

So besides the actually darker light itself I see absolutely NO reason, to get a sub-#0-Eiger (or maybe actually even a #0-Eiger) at all!


Edit: I forgot to take the booster circuit into account. See the following posts! Thanks to Curt!

 

[Burgess]

to Ice --


Very interesting calculations !


:goodjob::thanks:
_

 

[Burgess]

to Beacon of Light --


That " Inova T-something " in yer' first photograph is actually an Inova Radiant.

 

[Curt R]

This is for '0' power output level light:

The measured current through the 56 ohm resistor and the LED is 12 mA. 0.012 x 0.012 x 56 = 0.008064 Watts,
times 40 hours = 0.32256 Watts used by the resistor, more or less.

Battery current draw from a 1.5 volt battery was 30 mA.
Voltage output from the circuit was 3.3 volts at 12 mA.
1.5 volts x 0.03 A = 0.045 W.
3.3 volts x 0.012 A = 0.0396 W.
0.0396 / 0.045 = 88 % efficiency. Not bad for a boost circuit.

One item normally overlooked in run time testing is the resistance build upon the surface contacts
of the battery. We have seen as much as a 20% increase in measured light output in Candela
when brand new batteries were cleaned. They may look good to the eye, but there can be a big
difference in performance.


Curt

 

[Ice]

Thanks for the information!

I didn't take the booster circuit into account, so actually the resistor itself uses about 0.35Wh, which is about 40% of the energy of a normal AAA-battery!
Still rather much if you ask me...

The 80Ohm resistor of the sub-#0-version consumes about 0.46Wh, which is about 50% of the energy of an AAA-battery. Mind you: 50% of the total energy wasted for direct heat production...


Now I understand why many flashlights use PWM instead of simple resistors... On the other hand I really hate those flickering and humming flashlights! Oh well, there seems to be no ideal solution...

 

[Curt R]

The Effective Series Resistance, (ESR), of a PWM circuit is also going to use energy as does a simple boost circuit.
All electrical components have both ESR and Effective Series Inductance, one effects power consumption
and the other operating frequency and sub harmonics within the circuit that could effect the stability of the
circuit, causing unnoticed power surges. Component placement in very compact boards that are required to fit
inside a flashlight are very critical for proper operation and stability. The analog compensation network that controls
the Integrated Circuit must be isolated from the power path side of the circuit. If this is not done correctly all
kinds of weird things can happen to the output.

PWM requires additional parts for operation and some sort of operator to circuit interface that is user friendly to
implement a variable output control. One of the major problems with PWM circuits is that at lower power outputs,
is the limited on time of the LED, this causes the flickering effect. This is quite noticeable when using a video recording
device or surveillance camera. The only time that I can see as an advantage of PWM is in the intermediate range
of output levels between full power and very low power for light output adjustment.

When we talk about a regulated circuit we must consider two different basic types of regulation. One is voltage
regulation and the other is current regulation. The circuit that we use in the small pocket lights is a voltage regulating
circuit. The output is regulated to be 3.3 volts as in the case of the Eiger. The resistor used to set different power
levels is known as a current limiting resistor. The LED in the circuit for all practical purposes is also a resistor, it just
happens to be called a light emitting diode or an LED, but to the circuit it is a resistor. The current flowing through
the output circuit to the LED is set by the voltage drop across the resistor and the LED in reference to the voltage
produced by the boost circuit. As the resistor and the LED heat up, their resistance values decrease and the current
flow increases producing more light.

In a current regulated circuit the current output is monitored by the voltage drop value across a sense resistor and
the output voltage is changed by varying the signal to the mosfet transistor. The voltage change limited to the
LED in turn restricts the current flow through the LED. This method changes the light output of the LED. In our
LEO lights we use that type of circuit, but we also can vary the output voltage with a voltage divider network, in our
applications we use a simple potentiometer. This allows a 0 to 100 % output control that is almost impossible to achieve
with a PWM circuit.

Curt

 

[infinitelight]

What is the lifetime of the white leds used in the three led models in the high power and ultra power mode?

 

[cave dave]

Dagnabit, see what you did, you made me order a Peak flashlight.

I only had to call twice to get someone on the phone. I ordered a Brass Eiger Pwr#2. This will be my first Brass light. They were out of stock of Neutrals and it will have to be custom built. Robyn said it should be built in about 10 days, but I expect it will be longer.

Beacon here is a picture I took for you on the Oregon coast:

Runtime: 100 yrs
Throw with Fresnel optics: Several miles
Output: more than yours!
:nana:

 

[Beacon of Light]

Awesome picture cave dave!