Bear with me on this question…….
If you had a light that would run longer with Sanyo 2600mah batteries than Panasonic 3100 or 3400mah(or any other high mah brand) due to sag, would it be the same outcome if you could add extra batteries.
Example, if a TK75 would run longer with 2600s than 31/3400s, then you added 1 or even 2 extensions, making it 8 or 12 batteries, would the 2600s still run longer, or would the extra batteries curing the lag problem, make the Panasonics run longer?
I’m not sure how the extension stacking in the TK-75 works, but my guess is that it is in series. The only logical way to think about it is that part of the TK-75’s cost is in a fancy driver that compensates for the increased voltage by drawing less current. If that’s the case then yes, adding more batteries in series will reduce current draw because the increased voltage keeps the total wattage flowing through the driver consistent. Reduced current draw means less sag, so my estimation is yes, unless Fenix found a way to make subsequent battery extensions run in parallel.
All I know is that the battery extensions increase run time, so I figured the battery holders must be running parallel with each other. Since I only got back into flashlights a few months ago, so many things have changed it’s hard to catch back up. A lot of what people say is almost Greek to me…
To add, I’m going to ask Fenix and find out…
On Fenix’s website it says that they advise not using more than three extensions on a TK-75. I think that the light simply has a tolerance of 17-18v input and the driver is designed to accommodate for that, it maintains consistent wattage to the emitters.
You could just test it. Buy the extension and take a tailcap current measurement.
If its more then one battery running in a light, then the 3100/3400 will run longer because the voltage of the batteries is higher then the LEDs so the light will have whats called a buck driver, circuitry that reduces battery voltage to LED voltage and the more power there is to work with the longer the light shines, there would be no voltage sag problem at all.
If its one battery, them every budget light i know of will have a linear driver, as the voltage falls below whats needed to drive the LED at full circuit power the light will get dimmer until the battery dies or the circuitry shuts off the light because the battery is nearly empty
In this case the 3100/3400 will still last longer because as less power per second is used the light will be dimmer and drawing less power (a graph would be helpful but i don’t have one to show you)
The 2600 will maintain full power longer and will have higher brightness during its operating time but will still dim as its draining to empty, the 3100/3400 will start out just as bright but will start dimming sooner and run dimmer for longer then the 2600.
If a light is very well regulated, then doesn’t the only thing that matters about a battery is it’s mAhs?
A Panasonic 3400 will run longer than a 3100 or a 2600 with the same results and regulated brightness, as long as the light is well regulated?
Not necessarily. Well-regulated drivers will simply supply a consistent wattage to the emitter, and accomplish this by drawing more amps as voltage sags.
Batteries inherently have a phenomenon known as voltage sag wherein a load placed upon the cell causes an immediate dropoff in voltage, it is not a linear line or shallow curve, just look at the battery reviews HKJ does here. The Panasonic 3400mAh cells, especially the protected ones, have a fair bit of internal resistance and as such, when a heavy load is placed on them, the voltage of the battery sags quite a bit. And the voltage will eventually either drop below the driver’s minimum input or low enough to trip the protection circuit in the battery and it goes dead. Batteries without protection circuits or that have a lower mAh capacity tend to have less voltage sag.
The thing is, when you have more than one cell in series, it cuts the amperage draw the driver places on the batteries because the voltage is increased. If you have one 4v battery and the driver is programmed to give the emitter 16 watts, it’s going to draw 4 amps from the cell, a very heavy load. However, if you have two such cells in series, the input voltage is now 8v and the driver need draw only 2A to maintain 16 Watts to the emitter. As such, stacking batteries in series reduces the current draw and prolongs battery life.
What a great post, thanks.
I did buy the protected Sanyo 2600s for my first two 18650 lights, not because I understood completely, but because of posts like yours that indicated that they fit for those applications.
Does this apply to a PD35, which is a light that I am buying now for EDC, is the protected 3400 the choice for it since it is a single battery and well regulated?
Is the protected 3100 better, or would my Sanyo 2600 still be either THE or almost THE choice (a darned good choice, best value, or whatever)?
…personally i am buying Sanyo UR18650F 18650 2600mAh from FT because they are always “FRESH” and i mean really fresh (if i read the codes right R20B means they are made R= year 2013 and 20= week 20 which is middle of may and i received them in july and another shipment was last week and the code was R29B which means middle of july 2013…that’s amazing…i would say that most of us are eating food that is older than these sanyo cells
Nice, but that has nothing to do with the cell's performance, capacity, or lifespan after it's put into use.
Very True!
Sanyo 2600 has the edge on max performance on single cell flashlights only, but for multicell applications, where the voltage sag is less important (due to the sum of cells voltage being anyway higher than emitter Vf), the max performance is the same with good brand name cells (Sanyo, Panasonic, Samsung, etc) and the larger capacity cells always make for longer runtime.
I did some testing recently with a light running 2 x 18650’s, this driver, FT Sanyo 2600 protected batteries and Soshine 3100 protected batteries. The Sanyo batteries lasted 99 minutes before the light went dim and then off. Thinking the low voltage protection on the driver was faulty I then tested the Soshines. They lasted 117 minutes before the light started flashing the way the low voltage protection was meant to work. I would suggest that the Sanyos had drained very quickly not tripping the low battery warning on the driver and instead just turning of.
Protection really does two things:
-It prevents over-discharging of the cell on lights with no minimum voltage (Li-ions should not be discharged to less than 2.6v)
-It shuts the battery down if the voltage difference between cells in series becomes too great (to prevent reverse-charging of one cell to another).
Single cell-lights don’t need protected batteries. In the case of Convoy lights they’re designed only to accept unprotected cells. I’m also not aware of the discharge curves for either battery but to put out 450 ANSI lumens I think the PD35 would only need to draw 1.5 amperes, much less than what the TK-75 draws on high. I think it wouldn’t matter too much. The TK75 is a Dodge Viper, you buy it knowing the fuel economy is going to suck no matter what. The PD35 strikes me as a Subaru WRX STi, the fuel economy depends on how you use it.
As a Subaru owner I know the fuel economy sucks, no matter how you drive it. I would expect a Viper to be no worse 
For those interested, the battery extensions are running in parallel according to the reply I received from Fenix.
I want to know how the heck they do that and keep the # of extensions flexible. I still think you should do a tailcap current test after buying one. I’m willing to bet that the customer service people are not engineers. It just makes much more sense to have a [relatively] simple boost driver instead of having to specially design a system of electrical paths and contacts for running 1-4 battery caddies and keeping them all in parallel.
The only way for it to work longer, is to run in paralel. Btw, aren’t the cells 2s2p, which makes 8.4V?
Yes. And that isn’t the only way to get it to run longer. How many lights can be run 1 cell or 2 cell, and still have the same output? As I said earlier, the way that makes the most sense given the configuration is to have a boost driver with a low minimum input and a high maximum input.
But you run those in series, only voltage get higher, capacity stays the same.