LED voltage/runtime

[Lost Time]

I've noticed that some of the P-60 drop-ins are good for 3v-9v, while others are good for 3v-12v. My question is in regards to runtime and output vs. voltage input.

I've read about buck circuitry, but I'm not exactly sure what that means, or if it even applies here.

Some of the drop-ins claim higher output depending on input voltage. Some claim no difference in output, just in runtime when more voltage (read: more batteries) are used. I'm just trying to understand the relationship between voltage and output. If a particular module will operate on 3v-9v, will you be better served showing it two larger Li-ion, or three smaller primaries? And does that depend upon the particular module and what type of circuitry it's using?

I've read that Fenix lights will give you the same output, regarless of voltage applied. So 2x123's will give you same output as 1x17670. How is that different from the LEDs that claim higher output with higher voltage input?

 

[Curious_character]

I've noticed that some of the P-60 drop-ins are good for 3v-9v, while others are good for 3v-12v. My question is in regards to runtime and output vs. voltage input.

I've read about buck circuitry, but I'm not exactly sure what that means, or if it even applies here.

A buck regulator is a switching regulator whose output voltage is less than the input voltage. The voltage required by an LED is roughly 3.7 volts, so lights designed for running from two lithium cells, for example, use a buck regulator. A boost regulator steps the voltage up instead of down, so lights using one or two NiMH cells have boost regulators. Buck-boost is necessary when the cell voltage is approximately equal to the LED voltage or where it might be either more or less. Buck-boost regulators are more complex and generally less efficient than the other two kinds.

Some of the drop-ins claim higher output depending on input voltage. Some claim no difference in output, just in runtime when more voltage (read: more batteries) are used. I'm just trying to understand the relationship between voltage and output. If a particular module will operate on 3v-9v, will you be better served showing it two larger Li-ion, or three smaller primaries? And does that depend upon the particular module and what type of circuitry it's using?

I've read that Fenix lights will give you the same output, regarless of voltage applied. So 2x123's will give you same output as 1x17670. How is that different from the LEDs that claim higher output with higher voltage input?

If a light has a buck regulator, it will provide constant current and therefore constant light output as long as the input voltage is sufficiently high, for example when running on two lithium or Li-ion cells. If you put a single Li-ion cell such as a 17670 or 18650 into the light, it might regulate for a little while when the cell voltage is sufficiently above the LED voltage. But then the regulator will quit working properly and the light output will drop as the battery voltage drops. If the regulator is a boost regulator, the light will regulate with one or two NiMH cells, but run in more-or-less direct drive mode if you put in a Li-ion cell. Only buck-boost regulators will maintain constant output over an input range that includes the LED voltage. They're relatively rare in flashlights, but a few like the Tiablo A8 have them.

When a light is regulated, the power out of the regulator stays pretty constant and the power in stays pretty constant. So as the cell voltage drops, the cell current increases. If a boost regulator drops out of regulation, say as a 17670 cell voltage gets too low for regulation, the rules change. Then the power to the LED, and hence its light level, drops as the battery voltage drops. The battery current also drops with its voltage when unregulated. This gives you more run time but at a lower light level. Finally, a buck-boost regulator will usually give you less run time than either a boost or buck due to its lower efficiency.

So how a light behaves with different battery types depends on what kind of regulator it has and also on how the regulator circuitry behaves when it's no longer able to regulate.

c_c

 

[Lost Time]

In the case of a buck regulator where the regulator is taking a voltage higher than the 3.7v needed, and is basically only passing what is needed along to the LED, will increasing the voltage simply increase runtime? In an extreme example, if you have a buck regulator in place which will happily run an LED on one li-ion, if you were to put 6 li-ion cells in it, would you just be increasing runtime x5? Or does the math not work that way? And is there a limit to how much the buck regulator can handle on the input side?

Then, in the example of the LED drop-ins that advertise higher output with higher applied voltage, (3v = xlumens, 6v = xlumens, etc.) does that mean they are running direct drive with no regulation?? Simply, the more voltage you apply, the higher the output (until the LED fails), and it has NO effect on runtime?

 

[Curious_character]

In the case of a buck regulator where the regulator is taking a voltage higher than the 3.7v needed, and is basically only passing what is needed along to the LED, will increasing the voltage simply increase runtime? In an extreme example, if you have a buck regulator in place which will happily run an LED on one li-ion, if you were to put 6 li-ion cells in it, would you just be increasing runtime x5? Or does the math not work that way?

When the regulator is regulating, run time is determined by the energy capacity of the battery and not its voltage. So two cells will run the light about twice as long as one cell of the same type, because two have twice the energy as one. If a regulator can regulate when using either a 17670 or two RCR123A Li-ion cells, the run time will be just a little longer for the 17670 because it has a little more total energy than two RCR123As. In practice, the efficiency of a regulator will change some with voltage, so you won't see exact proportionality.

And is there a limit to how much the buck regulator can handle on the input side?

Sure. All regulators have limits, both for the range they'll regulate and the highest voltage they'll take without destruction. It depends entirely on the individual design.

Then, in the example of the LED drop-ins that advertise higher output with higher applied voltage, (3v = xlumens, 6v = xlumens, etc.) does that mean they are running direct drive with no regulation??

Either that, or they're just not regulating very well. They might have some current limiting mechanism beyond direct direct drive which applies when not regulating.

Simply, the more voltage you apply, the higher the output (until the LED fails), and it has NO effect on runtime?

Actually, if you're running direct drive or something approximating it like a simple series resistor, the higher voltage you apply, the more current is drawn (which makes the light brighter). So if you have two batteries with the same energy content (e.g., one 17670 or two RCR123As), the run time will be much less with the higher voltage battery because the energy will be drained at a higher rate. You're trading brightness for run time. This is the usual trade you can make -- think of a cell's energy (voltage times current times time) as being a brightness-run time product. For a given amount of energy, you can have more brightness but only at the expense of less run time, or vice versa. If you double the battery energy, you can get twice the brightness at the same run time, or twice the run time at the same brightness -- or other similar combinations, depending on how the regulator is designed. This is, of course oversimplified -- the trade is also subject to differences in regulator efficiency and other factors. But you get the idea. For a given battery technology such as Li-ion, the amount of energy in a battery (which can consist of one or more cells) is approximately proportional to its size. Small, bright, long run time: pick any two.

c_c

 

[Lost Time]

c_c, thank you very much for taking the time to give me such a detailed response. Much appreicated.