Thanks both of u for the links regarding krytox.
Here's the current draw measurements I did late last night for the
Quark AA Tactical... pewpew!
High 70 lumens, current to LED = 250ma (4sevens spec):Ikea (1500mah) @ 1.482v --- 1.26A
Powerex Imedion (2100mah) @ 1.280v --- 1.26A
AW14500 (750mah) @ 3.87v --- 250ma
Turbo 90 lumens, current to LED = 350ma (4sevens spec):Ikea @ 1.482v --- 1.27A
Powerex Imedion @ 1.280v --- 1.26A
AW14500 @ 3.87v --- 950ma
Medium 18 lumens, current to LED = 50ma (4sevens spec), sorry I only measured this with 14500:AW14500 @ 3.87v --- 30ma
**Measurements on Alkaline and NiMH should have higher current draw because of lower voltage hence boost circuit
I assume the buck and boost drivers appears to be intelligent. Driving the boosted power it wants and bucking appropriately for regulation. Noticed how it wants the same current at high and turbo modes for AA batts. I should measure on the other low, low modes. David from 4sevens explains his technology:
» Click to show Spoiler - click again to hide... «
Yes they are designed to take 3.6v (actually 4.2v when full). There is full buck/boost regulation.
Efficiency? Two part answer - Yes, our circuit is more efficient than PWM (as demostrated by the 30 day 1ma runtime) And Yes it's more efficient than a just a boost circuit which overdrives the LED for the first 10-20 minutes. But regulating both buck and boost allows efficient use of your battery. I always say, a flashlight that runs out of juice is WORSE than a rock when backpacking. You can always throw away the rock but you're forced to keep the useless flashlight for the rest of the hike until you find batteries. Efficiency = smart.
Second part of the answer. Buck/boost is slightly less efficient than say just buck or just boost. Effectively you're supporting to sets of drivers - each kicking in as needed. Boost, pushing voltage up as needed and Buck pushing voltage down as needed. We had this consideration when we designed the circuit and considered everything. 1) we wanted to fully support li-ion properly with buck/boost - not many manufacturers do this because it's a hassle and more expensive 2) since you're bucking for the first bit of the runtime, the little bit of efficiency hit (from supporting two circuits) is outweighed by the extra runtime from bucking the circuit instead of direct drive from boost only (which wastes lots of energy as well is dangerous for the LED).
Btw, I'm sure with 14500 current to the LED is more, like equivalent brightness to CR123^2

700ma? just assumptions.
Just for comparison, boost only circuit measurements for
Fenix LD10 I did earlier. I did for all modes except turbo, and HKJ from CPF did his own with NiMH.
Energizer Lithium Photo @ 1.70v:Low, med, high --- 90ma, 360ma, 2.3A
Eneloop:Low, med, high, turbo --- 90ma, 540ma, 1320ma, 2100ma.
Noticed how buck and boost circuit handle High and Turbo in comparison with boost only circuit. It will "squeeze" out current to drive the LED at constant rate until voltage drops low enough to sustain and the light starts to dim. Cheaper method but it works.
***corrected my statement***
Also, an
interesting note about Lithium
primaries behavior for handling high voltage draw. Have not done more research, yet.
David claims,
Quark lights have protection at 0.9v to protect NiMH. But not for LiIon.
Info here if you don't want to load from CPF:
» Click to show Spoiler - click again to hide... «
No voltage protection for li-ions.
I have to say at this juncture that UNPROTECTED li-ions are NEVER recommended. In any of our lights or others. It's not worth the risk.
You've read in the news about exploding cell phones - people getting their face burned - those are from unprotected aftermarket cells.
We don't even like to sell our protected ones. Folks, please becareful.
Don't use unprotected li-ions. Just don't do it!
----------
Quote:
Originally Posted by Egsise
Wut, I was talking of regular AA NiMH!
The question was: is there over discharge protection at 0.9V or not?
4sevens:
Sorry I was thinking li-ions. Yes there is protection.
Yes it should be very dim at 0.9v. I'm making a runtime test right now.
Yes it cuts off a 0.9v which should be safe for nimh. By the time it gets close to 0.9v it should be very noticeable already.
Enjoy your Quarks!
This post has been edited by pseudoblue: Sep 4 2009, 11:39 AM