Sky Ray King Burn-in test

LED Flashlights – General Info
1

My King (from FastTech) is the same as mhanlen’s and RaceR86’s (at least driver is).
Measured ~6.7A with fully charged batteries. Some say it’s a good driver, some that it’s not. So the question is: how long, how bright and how hot will King shine?

Test ingredients

  • 4x TrustFire 3000 mAh ~10Ah
  • voltmeter
  • alcohol thermometer
  • Samsung Note II as luxmeter

Since I don’t have thermal cam nor IR thermometer, I used alcohol thermometer to measure temperature of the head near the switch (not the most accurate method, but my fingers think it’s pretty close).
Did 2 runs: one with small cooling fan and one without. Ambient temperature 21°C (70F). Took manual readings at 1 minute intervals.

RELATIVE BRIGHTNESS
As long as input voltage is above ~3.3V, light output is constant. Below ~3.3V it goes to DD.
Brightness drop due to temperature is exactly as shown by Cree.

VOLTAGE
Cold batteries seem to have a constant voltage drop.
Warm batteries on the other hand, seems to love heat, alot.

TEMPERATURE
As long as driver is in regulation, heat will build up. I guess that temperature of the LEDs is probably around 25°C (45F) higher then head.

LOW BATTERY WARNING?
At around 2.93V one of the LEDs starts to flicker and turns off. (cooled at 138 min ~11%lux)
At around 2.87V second one starts to flicker and turns off. (cooled at 156min ~5%lux)
At around 2.86V last LED starts to flicker and probably turns off too at some point (cooled at 210min ~0.5%lux and still going)

RaceR86 measured emitter currents at LED1 2.7A, LED2 2.6A, LED3 2.2A. If we look at the I/U graph, we’ll see that Vf difference between LED1-LED2 is 0.015V and LED2-LED3 is 0.06V (same as my turn off points).

So my guess is that it is indeed a low battery warning.

CONCLUSION
Sky Ray King is one hot light. Sitting without air circulation gives lowish runtime but high brightness overall. Good air circulation (like running or cycling) gives longer runtime but brightness drops to soon. As long as there is at least some air circulating around, resulting brightness and runtime shoud be balanced.

FEW NOTES

  • estimated 2550 (2300 OTF) lumens (XML U2 CW)

  • despite being CW, color rendering is superb

  • some sharp edges inside the body tube (can damage battery wrapping)

  • no centering rings (two perfectly centered, one very little off center)

  • one the first run/heat up there was a little fog on the lens

  • low quality lenyard

  • low mode is ~4.5%

  • PWM on low is not visible nor audible to me, but can be seen on camera (tried quick movements + riding a bike)

    P.S. MOD
    When I ordered the King, I knew that charging 4 batteries with a single bay charger will be PITA. Charging one battery takes about 3.5 hours on my 1.5A TR-005 (starts with 1.3A and quickly falls). So I was thinking to put them into a 1S4P battery holder and wire it to the charger, but that would still take at least 14 hours. When King came, I noticed that there is a sufficient space between cells to put something through. Went through my toolbox, and the only thing that would fit perfectly was antenna jack. So I drilled the board, cut jack down so tailcap can be screwed and replaced oring with smaller one. Then made a cable with USB on the one end and antenna jack on the other. It’s not perfectly made, but it serves the purpose.

    Now I can charge on any USB, but prefere my phone charger.

    It draws 2.4A from the start, at 4.13V it’s down to 0.7A. USB powered port on my PC starts at 1.4A, at 4.13V it’s down to 0.4A
    The only “problem” so far is topping batteries to 100%. USB is 5V and protection on batteries stops charging when voltage goes above 4.20V.

2

Depending upon the protection circuit on the cells for charging shut-off is not a good idea. Those circuits can be of rather iffy accuracy and reliability. Also, they do permit a proper constant current / contant voltage charge profile.

3

Texas means “they do not permit a proper constant current/constant voltage charge profile.”
That’s a great mod, if you are OK with charging inside the light (I’m a scardy-cat), Get a 4.2V supply and it will be better/safer.

4

Same one I ordered but have not the nerve or equipment to find info contained herein. THANX!!

5

Thanks for posting. Some really nice and interesting results there! :slight_smile:
I never ran mine out of power. Nice to know what will happen.
Funny to see that you could relate the various LED cut-off to my measurements. :slight_smile:

Id just like to add.
There are several of these drivers that are similar looking, but have different parts.
Relic38, Comfychair and me, have both got the same looking driver circuit but with slightly different components (and different output). Relic38 seemed to have other resistors (and lower output). Comfychair seemed to have same resistors, but different FETs (not the same numbers as mine) His had about 7 amps total to the emitters.

The tailcap reading I posted on mine right after I got it was too low btw. I did not have enough hands to get a stable reading with 4 cells, so I just used a single Panasonic 18650PD (protected) to make it easy. I should have said that, but im not used to seeing 7 amps or more on SRKs, so I just assumed it was able the show the full current at that point.
Judging by the emitter currents, my tailcap reading should have been more like 7,5 amps (if not more).

Based on the inconsistency, even from same supplier at times, I would suspect that these driver circuits are generally somewhere between 5,6-7,5 amps depending on the parts. All seem to have some differences between the output to the different LEDs. Some would probably consider that a bad thing, but then again. It seems to function as a decent battery warning.

6

Getting a 4.2V CC/CV charger that can deliver more than 2A (aka hobby charger) would be great. But for now I think I don’t need it (have only 12 lithium batteries so far) and it’s a bit over my budget for now.

As far as safety goes, I’m periodically checking voltage so I’m not depending upon protection circuit.

As semi-CC charging, starting with ~0.25C and ending with ~0.07C, when protection trips at 4.2V, batteries are at >90% capacity. For the final 10% I found a simple solution: just add 5 meter extension USB cable (additional 1Ω) bringing current down to ~0.02C.

For tailcap reading, it is quite possible that 6.7A my DMM is reading is low (my leads are 0.3Ω).