Typical flashlight efficiency question

Hi !

I've done some tests with my ArmyTek Dobermann Pro XHP35 (Warm White).

The lamp stays at a constant brightness level for around 92 minutes on the Turbo 1 level with a fully charged cell.

The turbo 1 is supposed to be 800 lumens with the Cool White led. In a review it has been measured to be around 650 lumens with the Warm White version (mine).

My cell is a Panasonic 3400 mAh with protection PCB, it's a 3 year old cell but it has done less than 50 cycles.

650 lumens, 92 minutes, 3400 mAh Panasonic, XHP35 HI warm (probably a D4 or E2 bin, it's a pinky 4000k light).

What kind of efficiency level is this ?

I know that nothing conclusive can be done but maybe an estimation ?

You would basically have to do a run time test and compare the results to each other. For those that may not be aware, I would consider efficiency as lumen x time, or Lumen-minutes. It really doesn't mean anything until you compare to other lights. Or against different levels of the same flashlight.

So what I have done is run time tests with a data logging light meter and a home made light box. The data logging light meter takes data at a certain rate (I usually use once very 4 seconds) and I keep it going until the battery dies or is tripped in the case of a protected lithium ion. Then with the data, I graph the output over time until the battery trips or is discharged. The light box isn't calibrated to anything but since I use the same light box with each test, it then becomes a constant variable that doesn't change. If I use the same battery fully charged with each test, then the results that I get simply represent the efficiency between tests. Basically efficiency is the area under the curve in a run time test.

The data logging light meter displays lux. Since I have done so many run time tests, I have equated this lux in my light box to lumens after verifying a number of manufacturers lumen numbers to the lux reading that I get in my light box. It's not precise, but good enough.

Comparing the graphs with different lights lets me know which have better efficiencies.......which when comparing different lights with the same LED is really mostly a function of the electronic driver. Or when some light models came out with a gen 2 LED, I could measure approximately a 15% to 20% gain in efficiency........mostly because of the LED. When I have changed out an LED on a particular flashlight to a newer model or a higher bin, I then can see the difference attributed to that LED simply because everything else stayed the same.

NCR18650B has an energy capacity of about 11Wh at 1.5 amps which would roughly correspond to the discharge time of 92 minutes.

Assuming constant output (which Armyteks do have) at 650 lumens as you say that would equal 650lm*1.533h=996 lumen*hours.

996 lumen*hours / 11 watt*hours = 90.5 lm/W

That's quite a typical efficacy at that output level for an XHP35 HI based light taking into account losses in contacs, cables, driver and optics. Just a bit behind the Zebralight SC600w HI which is considered excellent regarding efficiency.

Thank you. i'll try a ceiling bounce test soon with both Turbo 1 and Turbo 2 modes.