Another "Quark HV vs. LV and Li-Ion" Post (and Questions!)

I know this has been beat to death, but I want to summarize the gems of information I've gleaned from the dozens of threads already related to this on CPF. There are some holes in some of these statements; I just don't know enough yet.

When a single AW RCR123A/14500/17670 cell is used with a Quark HV or LV head:

• The HV head will run regulated for about ~10 minutes (RCR123A) to ~20 minutes (17670) on Max, then go into direct drive and slowly fade away.
• The LV head will run regulated ____ times longer on Max, but end abruptly.
• The HV circuit is buck-only, which makes if more efficient. This means that the lower levels will run longer with the HV head (True?).
• The LV head will always trip the circuit protection of the cell (2.75V?); the HV head will simply cut off when the cell reaches below ____ V. It's genarally safer to have the HV head cut off, than have the LV head trip the protection circuit (True?).
• The HV head will lose turbo when the voltage drops below ____ V.

Thanks!

I think I see what you're trying to do, but you may need to try a slightly different approach using the different current draw for the different emitters (remember we're dealing with both the XP-G2 and the XM-L) unless of course they are all driven the same regardless of emitter.

Basically, using a single lithium ion cell in the high voltage head, the light will run in regulation until the voltage (under load) of the cell drops below the critical point where the driver can continue delivering the regulated current to the LED, and then taper off until the reaching the low voltage cutoff of the cell. The run time will depend on the amount of energy the cell can provide before reaching that critical voltage point, and will vary depending on the capacity of the cell (14500/16340 vs 17500). It's worth noting that both this critical voltage, and the current draw when in regulation will depend on the driver which I suspect varies between the XM-L and XP-G2 based lights.

I don't know enough about the exact run time figures, but for the low voltage head, you can assume the light will run in regulation until the low voltage cutoff of the light kicks in. The run time here is a function of the current the driver is drawing and the capacity of the cells but should be predictable if you know both these things. Again, the current changes depending on the driver used which, again, I suspect depends on the LED.

These are just high level assertions and it looks like you've created this thread to gather actual, hard, measured data so I apologize if this hasn't been useful. Good luck on your Quark Compendium.

Cheers!

GeoBruin said:

I think I see what you're trying to do, but you may need to try a slightly different approach using the different current draw for the different emitters (remember we're dealing with both the XP-G2 and the XM-L) unless of course they are all driven the same regardless of emitter.

Click to expand...

Whoops, assume they're all XM-L.

GeoBruin said:

I don't know enough about the exact run time figures, but for the low voltage head, you can assume the light will run in regulation until the low voltage cutoff of the light kicks in. The run time here is a function of the current the driver is drawing and the capacity of the cells but should be predictable if you know both these things. Again, the current changes depending on the driver used which, again, I suspect depends on the LED.

Click to expand...

On the LV head, won't the cell's circuit protection kick-in first since the circuit will draw down to 0.9V?

GeoBruin said:

These are just high level assertions and it looks like you've created this thread to gather actual, hard, measured data so I apologize if this hasn't been useful. Good luck on your Quark Compendium.

Cheers!

Click to expand...

Thanks! My Quark Compendium's purpose is twofold: 1) I want to synthesize some real-world expectations from my summary, since almost all the data I've seen relates to the HV and LV heads running on High or Max.--two settings I use the least; and 2) I want to know which head to grab for various scenarios besides the "LV head can use just about any battery" option (which I do use and enjoy). So in other words, I need a McFly to do my homework for me .

Zeyeman said:

When a single AW RCR123A/14500/17670 cell is used with a Quark HV or LV head:

• The HV head will run regulated for about ~10 minutes (RCR123A) to ~20 minutes (17670) on Max, then go into direct drive and slowly fade away.
• The LV head will run regulated ____ times longer on Max, but end abruptly.
• The HV circuit is buck-only, which makes if more efficient. This means that the lower levels will run longer with the HV head (True?).
• The LV head will always trip the circuit protection of the cell (2.75V?); the HV head will simply cut off when the cell reaches below ____ V. It's genarally safer to have the HV head cut off, than have the LV head trip the protection circuit (True?).
• The HV head will lose turbo when the voltage drops below ____ V.

Thanks!

Click to expand...

Well I don't do typically runtime tests with Li-ions in my lights but....

- Selfbuilt tested the 123-2 X on a 17670 and 2x16340s and got 1:13 and 0:57, to 50% respectively... I would assume at least 0:25 on a single 16340.

http://www.candlepowerforums.com/vb/showthread.php?321913

- Yes, the LV will suddenly trip a Li-ions protection circuit while on Turbo. I did do this once... my Eagletac 14500 metered a resting 3.4V immediately upon reset by the charger.

- While it is true a buck driver is more efficient than boost driver, my understanding (I'm not an expert, someone may correct me) of these lights is that both the HV and LV heads will be using their buck drivers to the same extent with a single Li-ion and should be equally efficient. What you may be confusing is that a 2,300 mah NiMh vs a 750 mah Li-ion (both having the same watt-hrs energy), the NiMh should have a shorter runtime at a given lumen level due to it operating under a less efficient boost driver.

- I haven't run my HV head on Li-ions to any sort of cut-off yet. I will say, however, that the HV head will run to around 2.5V or so on moonlight and low modes (I've run half dead 2xNiMh in my Turbo X). But I suspect, the higher modes would cut out at 3V giving you a sort of LV warning before tripping the battery protection. Yes, it's better for the light loose the upper modes than for a battery to trip it's protection. For NiMhs, l tend to compare Turbo and high - when there's no difference, I know I need to recharge.

It sounds like you're concerned about the safety thing. Keep in mind the danger of explosion is really only on the charging side of the equation, not the discharge side. So as long as you monitor state of charge before and after charging a Li-ion, and keep it in the normal operating range, you are safe. Explosions with lithium batteries while in a flashlight are pretty much confined to multi-cell configs where one cell depletes before the other(s). When that happens, V goes past Zero on the weak cell and it starts CHARGING in reverse polarity and the potential for BOOM. The greater power of the stronger cell(s) the more power behind the charge and bigger the boom.

Hope that helps.

"On the LV head, won't the cell's circuit protection kick-in first since the circuit will draw down to 0.9V?"


Yes, of course you're right. I meant the HV head. Sorry.

I read a thread at one point that stated that a Li-Ion voltage protection MAY not trip under low current draws. Personally, I prefer to run NiMh on my LV heads that use a prolonged (more than a week) moonlight/low mode. My tactical heads use Li-Ions that switch between Med/Max. My "Pro" interface heads get voltage checked fairly frequently.