Quark Moonlight Runtime Test

[wantsusa]

Yeah this is great, thanks guys for doing this :twothumbs

Now if we could just come up with a huge spreadsheet that could list each regularly manufactured flashlight with runtimes with alkies, nimh, li-ion, lithium at each of the different power settings ROFL.

Then we could all just look and see what lumens for what runtimes for what type of battery, and bingo come up with a list of what we want to buy!

 

[AnAppleSnail]

It appears it is a hardware glitch, and consumes more power than moon or low modes. I would not recommend it. Also, I now have to look up this programming mode, I thought I saw it mentioned a long time ago, something about 200 quick power cycles? I saw it mentioned in the marketplace thread.

There's another light out there with a "deluxe" programmable and "lame" non-programmable, where you get the programmable UI with a ludicrous number of on/off cycles...but I hadn't heard of this with the Quark regular.

 

[Zendude]

Hey folks,

My tests for this batch are done.

NW Quark MiNi AA: 24-28hrs (with duraloop) .89V

Ti Quark MiNi AA: 36hrs (with duracell) .69V

Nitecore D10 GDP ~48hrs (with duraloop) .86V

I gotta say that I'm pretty disappointed in the results.:sigh:

I just started my ZL lights and I'm retesting the NW Quark MiNi AA(all with Eneloops).

Edit: @rookiedaddy: I don't know how I missed that thread! As usual, I'm one step behind! I won't bother retesting the mini. I'm still disappointed with the alkie test. It was the cell that came with the light.

 

[Darvis]

Sorry for the delay folks, my DSL was down all morning.

For starters, we're at hour 25.5 and there is no change for the three Quarks...

The Peak went an impressive 22.5 hours with no change in brightness compared to hour 1, however, at ~ hour 23, the light began its decline in brightness, following the Eneloops NIMH discharge curve. At hour 25.5, it's still brighter than the Quarks by about a bit more than half, so there is still time to go before it matches the moonlight brightness level. I'm calling this 50% brightness for the Peak.

Compare this to the 11-12 run of the peak on a AAA before it began to decline and until it was completely done at hour 22ish. This light continues to impress, especially mated to the Arc AA adapter. The #0 and Sub Zero should be all the more impressive!

Catch you all a little later on!

D

 

[wyager]

There's another light out there with a "deluxe" programmable and "lame" non-programmable, where you get the programmable UI with a ludicrous number of on/off cycles...but I hadn't heard of this with the Quark regular.

Thanks, I looked it up yesterday. I thought it was funny how they charged you more for the exact same light, with the exact same amount of code

 

[Darvis]

Hour 27 brings the peak to Quark moonlight level as it continues to decline, I imagine it will run until about hour 28 before it drops below.

The Quarks are as they were....

Will post back in a few hours

 

[Beacon of Light]

Thanks for testing the Peak with the AA body. How many hours will it need to be, to compare to 3x the runtime of the AAA used in the regular test? It seems like it may fall short of 3 times the runtime on AAA.

 

[Darvis]

It would need to go ~39 hours for a true tripling of hours (runtime to 50%) but I'm not sure it's a simple tripling to get there. The AAA's are 800 mah and the AA's are 2000 mah, so it's less than triple the reserves to start. That and I believe it really comes down the discharge curve of the battery and the VF of the led.

I had done the math based on Peak's specs and thought it would at least run 38 hours to half brightness; waaayyyy off I know!!! I now know that I 100% messed the math up when calculating for the AA runtimes. I at least know the AAA math is right as both your tests and mine verified the results I had gotten when running the pure numbers.

If I look at how the math worked for the AAA, I was getting 13 hours to 50% and this combination absolutely went 25+ hours before it hit that point. I think this is more realistic as I imagine that we're really dealing with a doubling of runtime when going AA and not a tripling. In other words, I think the Peak is dead on based on the specs published by Curt on the RMSK website.

If that's the case, I would see the #0 easily running right around 36 hours to 50% on a AA Eneloop... I have no data for the Sub Zero.

All that aside, @ 28.75 hours, the Peak is still as bright as the Quarks!!!! I now feel the Peak will go at least 30 hours before dropping lower than the Quarks.

No change for the Quarks.

The final results for the Peak XP-G #1 Eiger will probably look something like this:

25.5 hours to 50% (confirmed)
~30+ hours to less than Quark moon mode (tbd)
???? to tritium brightness, and thus the end for the Peak (tbd)

 

[Beacon of Light]

Darvis, did you read my results on the Eiger #0 and #Subzero? I am puzzled why they both had pretty much the same runtime when I ended it (once it got dimmer than Quark brightness). The voltage wasn't particularly favoring the #Subzero either I think it was .1v more on the #Subzero than the #0. The #0 has a 56ohm resistor and the #Subzero has a 80ohm resistor according to Mark @ Peak. I was thinking that the resistor would not only reduce the lumens much lower (which it did), but also run the LED at a lower drive current so it would extend runtime, but I didn't see that benefit. Any idea of why this would be?

 

[Darvis]

I did see your results and I honestly do not know the answer, but my guess would be something akin to resistance having more effect on the brightness of the bulb and not the equivalent effect on runtime. I would think the battery is supplying as much power to the resistor in both cases, and that does not, somehow translate to increased potential as much as it does decreased voltage to the LED. I guess what I'm getting at is maybe it has an exponential effect on brightness, but not on runtime, so there's no apparent 1:1 correlation that we see? Maybe the circuits are less efficient as the resistance increases????

But am no electrical engineer, so that is my SWAG!!! Maybe someone can explain it? That would be good to know!

 

[ama230]

Cree 5mm Vfmin = 1.7V Imin= 1micro amp

#0: 56ohm @ 3.4v = 60mA, then the eneloop takes a poop @ under 1v due to it not being within the threshold voltage. This transistor wants to have the higher current to keep the gate open. The lower the current the transistor wants to shut prematurely. It does make logical sense that the bigger resistor would deliver longer performance but this applies for direct drive. Its in the transistors characteristics to use a higher drive current.


subzero: 80ohm @ 3.4V = 42mA, this should last longer but with the threshold voltage of an eneloop its not within a range that the eneloop wants to be and this is why you see it not last as long. Also you are dissipating power rather than using it as its is driving it at a lower intensity but not necessarily going to give you a longer runtime. A better transistor and better matched capacitor and bridge resistor is going to give you a better handle on runtime. It sounds as if the transistor loves the lower resistor for this setup.

anybody else have an idea as it is weird...

hope this helps BOL

 

[PeaceOfMind]

Darvis, did you read my results on the Eiger #0 and #Subzero? I am puzzled why they both had pretty much the same runtime when I ended it (once it got dimmer than Quark brightness). The voltage wasn't particularly favoring the #Subzero either I think it was .1v more on the #Subzero than the #0. The #0 has a 56ohm resistor and the #Subzero has a 80ohm resistor according to Mark @ Peak. I was thinking that the resistor would not only reduce the lumens much lower (which it did), but also run the LED at a lower drive current so it would extend runtime, but I didn't see that benefit. Any idea of why this would be?

This may have to do with the fact that the forward voltage (Vf)/forward current (If) curve of an LED is very non-linear. At low voltages, changing Vf has decreasing effect on If, so you have likely entered a region of very diminishing returns in terms of runtime. If the peak circuit is setup like I'm picturing then you may be in a region of LED performance where the drop in voltage caused by the resistor creates very little change in the current through the LED. The voltage drop will make the LED dimmer, but there'd only be a very small decrease in current due to the LED's properties. If you look at the Vf/If curve for any LED you'll see what I mean about the lack of change within lower voltage regions.

Just guessing, as I don't know the specifics of how the circuit is setup or the specific Vf/If curve of the LED used, etc.

 

[Darvis]

Folks, at jusssstt under 31 hours, the Peak is no longer as bright as moon mode.. sigh.
It ran from just brighter than moon to just dimmer than moon over the course of about 3.5 hours. Not too shabby!!!

I'll now track it to tritum level...

For those of you that count the Quark at moonlight level your lowest possible output level, you can expect the Peak Eiger XP-G #1 to run brighter at first, then match moon for 3.5 hours before dropping below the Quark over the course of almost 31 hours on an Eneloop AA. But this is a hard one to call, really, as the Eiger seems to have really hit a nice looonngg steady decline path. Even though it is a bit dimmer than the Quark, it's not dropping rapidly. In a survival situation, this light will continue to provide value even at this junction.

In fact, I just took it into a dark room and it is plenty bright... this light is far from out of the game!

 

[rookiedaddy]

...
Ti Quark MiNi AA: 36hrs (with duracell) .69V
...
Edit: @rookiedaddy: I don't know how I missed that thread! As usual, I'm one step behind! I won't bother retesting the mini. I'm still disappointed with the alkie test. It was the cell that came with the light.

Zendude, if the Duracell that came with your light is similar to what came with my WW Mini AA, then perhaps it explains the lower runtime. From this AA battery test thread, you can see that these made-in-China Duracell has lower capacity compare to those made-in-USA Duracell as tested by UserName.

I did the lo-mode Mini AA runtime test in Alkaline using Panasonic Evolta instead of the the included Duracell (although 4Sevens did says that the quoted runtimes are based on included battery :shrug due to...

1. Panasonic Evolta has consistently perform really well in low-drain application
2. These made-in-China Duracell Coppertop are leaky (it leaks even when still in plastic wrap , and I have seen it leak in those retail blister pack)
3. Compare my results with the AAA version of these made-in-China Duracell in the E01 runtime thread to Scout24's made-in-USA Duracell test shows that even the AAA version of these made-in-China Duracell has lower capacity than the US-made cells

 

[guiri]

First, let me say I love this test 'cause I'm one of those that is extremely intrigued by super long runtime on lights and as a matter of fact, I've started only buying lights that DO have a low or super low mode in addition to their other modes.

However (and I haven't read all the posts), aren't there lights that have claimed 30 day runtimes? Those are the ones that really tickle my fancy. I mean, if some of these are giving out after 30 hours, what's gonna happen with the 200 hour (claimed) lights?

George

 

[guiri]

Maelstrom G5
Moonlight: 0.2 lumens, 7.5 days

TK40
Low - 13 Lumens, 150 Hours

TA21
Mode 1 (4 lumens - 215 hrs)

JetBeam III M R2
Min Output: 2 Lumen, lasting for 200 hours

Quark AA
0.2 lumens for 10 days (1ma)

 

[Beacon of Light]

guiri, how about a 4Sevens Quark "Lunar Eclipse" model? I'm hoping they do something like that in AAA format. Super low out put/ super long run-time. Sorry for getting off topic Darvis.

 

[wyager]

guiri, how about a 4Sevens Quark "Lunar Eclipse" model? I'm hoping they do something like that in AAA format. Super low out put/ super long run-time. Sorry for getting off topic Darvis.

This would be really hard. Right now, they're either using 10 bit PWM or hardware dimming for moon mode, and if they're using 10 bit PWM you really can't go any lower than max output/1024 (aka .2 lumens), and if they're using some kind of hardware trick to give .2 lumens it would probably be difficult to add another low mode.


Also, at some point, isn't it more practical to use tritium illumination over a light? I know they have some that are designed as reading lights...

 

[guiri]

guiri, how about a 4Sevens Quark "Lunar Eclipse" model?

I don't know that one..

 

[wyager]

I don't know that one..

We were just talking about that, it's basically a hardware malfunction it would seem, it actually draws about 11x more power than moonlight mode IIRC.