The "Other" Ebay US clone Sky Ray King - Mini review and tear down.

[Old-Lumens]

@Pavithra_uk

@Old-Lumens

@Pavithra_uk

wow 11A after mod ! this is a awesome Don’t throw away that stock driver. Its true current limited driver. uses AMC7136 chip with external FETs According to AMC7136 “datasheet”:http://file.yizimg.com/937/2007102710183541.pdf its current limit feedback voltage is 50mV So, in this driver, there are 3 x 0.1-ohm sense resistors per channel give 0.033-ohm Then, programmed drive current 50mV/33mOhm = 1.515A to each LED. total 4.5AMPs this is how you got 4.5A reading. mod is simple if you have that low valve resistors. to get 3A current 50/3 = 16 (0.016 ohms) add another 3x 0.1-ohm resistors to each channel. total 9 resistors required.

So if I wanted to have 2 amps to Nichia leds, then 50/2 + 25, so 0.025 ohms for each resistor?

yes! 50/2 = 25 (0.025 ohms) To get 0.025 you need one additional resistor to each channel. then 0.1/4 = 0.025ohms do you have 0.1 ohm resistors ?

I don’t think I do, but I will have to check.

[Pavithra_uk]

@Old-Lumens

@Pavithra_uk

@Old-Lumens

@Pavithra_uk

wow 11A after mod ! this is a awesome Don’t throw away that stock driver. Its true current limited driver. uses AMC7136 chip with external FETs According to AMC7136 datasheet its current limit feedback voltage is 50mV So, in this driver, there are 3 x 0.1-ohm sense resistors per channel give 0.033-ohm Then, programmed drive current 50mV/33mOhm = 1.515A to each LED. total 4.5AMPs this is how you got 4.5A reading. mod is simple if you have that low valve resistors. to get 3A current 50/3 = 16 (0.016 ohms) add another 3x 0.1-ohm resistors to each channel. total 9 resistors required.

So if I wanted to have 2 amps to Nichia leds, then 50/2 + 25, so 0.025 ohms for each resistor?

yes! 50/2 = 25 (0.025 ohms) To get 0.025 you need one additional resistor to each channel. then 0.1/4 = 0.025ohms do you have 0.1 ohm resistors ?

I don’t think I do, but I will have to check.

I have around 20-30 pcs… let me know if you need. but it will take time to reach you :email:

[Old-Lumens]

Or I could just use one single .025 resistor? Digikey has those. They look to be 1206 size. 3mm x 1.6mm

Thank you for the kind offer, but Digikey has most everything and I can get them in a day or two.

[Pavithra_uk]

yes you can use single 0.025 ohm. Remove all exiting 0.1 resistors

[texaspyro]

@Old-Lumens

Of course that’s all totally subjective like the best flavor ice cream.

That ain’t subjective… everybody with at least half a taste bud knows it’s chocolate… :party:

[texaspyro]

@Old-Lumens

Digikey has most everything and I can get them in a day or two.

So does Mouser… and being local you will have them the next day. I once ordered something from them in the morning and it was here by the afternoon.

[comfychair]

@Pavithra_uk

yes you can use single 0.025 ohm. Remove all exiting 0.1 resistors

I had one of those drivers that had this setup stock:

[Pavithra_uk]

@comfychair

@Pavithra_uk

yes you can use single 0.025 ohm. Remove all exiting 0.1 resistors

I had one of those drivers that had this setup stock:

even 0 ohm resistors have very little resistance, they put 000-ohm on this driver.
did you measured drive currents?

See L3 channel…
there are 4 resistors. because that channel under driven due to long track between FET & sense resistors.
this is why they put another resistor to balance. old buck driver SRK also have that issue.

[Old-Lumens]

@Pavithra_uk

@comfychair

@Pavithra_uk

yes you can use single 0.025 ohm. Remove all exiting 0.1 resistors

I had one of those drivers that had this setup stock:

even 0 ohm resistors have very little resistance, they put 000-ohm on this driver. did you measured drive currents? See L3 channel… there are 4 resistors. because that channel under driven due to long track between FET & sense resistors. this is why they put another resistor to balance. old buck driver SRK also have that issue.

Interesting, so maybe it would be good to check each channel’s output before signing off on it. Hope I remember to do that on the bench before assembly.

[comfychair]

That one was not well balanced at all, stock it was like this:

L1+/L1- 2.20A

L2+/L2- 2.60A

L3+/L3- 2.08A

Power from 4x 18650 in parallel, checking output one channel at a time on the same LED (so it wasn't due to different LEDs having different Vfs causing the skew).

edit: and all 3 channels had the same 3x 1R0 + 1x 000

[Pavithra_uk]

L3 channel has long track between resistors and FET. make it reduced output. to improve its current, solder thick copper wire to it.
other two channel difference is due to these sense resistors not 1% or 5% tolerance.

[comfychair]

So it needs a jumper from the L3 sense resistors to the bottom left pin on the 7136? Would you leave the original PCB trace intact or sever it (would it make a difference)? L1 & L3 are good as-is, correct? And then mess around with the resistors to get the target output & balance the 3 sectors.

[kralyevski]

@comfychair

If 11 amps and 6 amps look the same, and 6 amps and 4 amps looks the same, and 4 amps and 2.5 amps look the same, why not just go all the way and run no amps at all? It should look just as bright, right? And battery life would be excellent. Won’t burn your hand, either. :party:

i think this comment says it all… :party: it is so true……i couldn’t have said it better myself

[Pavithra_uk]

XML rated maximum drive current is 3A but some noticeable output improvement over 3A too. but after 6-7A difference is very small. so 6A to 11A no improvement at all. may be reduced output due to higher junction temperature.

[Pavithra_uk]

@comfychair

So it needs a jumper from the L3 sense resistors to the bottom left pin on the 7136? Would you leave the original PCB trace intact or sever it (would it make a difference)? L1 & L3 are good as-is, correct? And then mess around with the resistors to get the target output & balance the 3 sectors.

yes. disconnect original trace between FET (s pin) to 7136 left pin.

or

sense resistor to left pin of FET (Source pin) then no need to disconnect original trace.

[comfychair]

11A into 3 LEDs in parallel is only 3.6A each. And the old rules-of-thumb are out the window with the LEDs on copper.

[kralyevski]

@comfychair

11A into 3 LEDs in parallel is only 3.6A each…

i like the sound of that “ONLY” 3.6A…that’s the spirit…if we wanted light with “ENOUGH” lumens we wouldn’t be on BLF in the first place…

[Old-Lumens]

But, 11 amps is out of this driver,

Not this driver,

I Switched Drivers. I used the driver from the Securitylng in this light and I will be using the driver from this light into the Securitylng.

Just in case we all aren't on the same page.

[texaspyro]

@Old-Lumens

Justin case we all aren’t on the same page.

If we weren’t all on the same page, we wouldn’t all be reading this… J)

[Old-Lumens]

Modded the driver with R025 resistors and wire on the L3 run.

5+ amps overall, as a result. Didn't check each zone.