Without modifications you will face problems with this driver. At higher currents (more than about 8 Amperes) it resets itself, and you won’t be able to use the highest mode. With 219C triples you can easily see 15 Amperes and more, depending on the cells.
I think you can use a 105 driver too and just stack all 7135s with one extra, because the Nichia 219C has a low Vf.
But maybe 6 Amperes is not enough for you…?
Like was said you will need a better driver. The Texas Avengers have updated components to handle the voltage spikes and handle loads of current (I have personally had one at just under 20A without an issue).
As far as the host goes, just pick which ever is cheaper or the color you prefer. They all use the same 1288 switch, bypass the spring and it will handle the power.
Your not forgetting the spacer you need for a triple correct?
Are you sure it’s the Banggood Blf X6 and not the A6? Apart from the MCU (Attiny25 vs. Attiny13A) they are identical but the Attiny25 is more sensitive. Are you using Bistro? Bistro does ramping to achieve a smooth transition between the modes, if you disable ramping you might get a better reliability of the driver.
I have used the old hardware with the tiny25 on an XP-G3 @ ~8A and it work fine but I tried the same driver with a triple setup and it freaks out and resets when you try to go into turbo. I am guessing the limit is closer to 10A or so, at least in my case it is.
The inductance of the wiring/conductors to the LEDs also makes a difference. Higher amps and higher inductance both makes the FET switching ‘spikes’ worse. Probably why it is more obvious on a triple where we have both high amps and more convoluted emitter conductors.
Even in an ‘average’ setup the MCU is getting some ugly stuff on its supply pins with these drivers. I am surprised it works as well as it do (or even at all!) and that there are not more damaged MCUs.
I had problems with the BLF-X6 driver in triples, but all BLF-A6 drivers thusfar worked fine.
Is there a way to ‘fix’ existing BLF-X6 drivers from Banggood by adding or swapping components? (I can not flash software). Because I do like the UI very much.
Easiest way is to stack another 10 uF capacitor on top of the existing one (at the red wire, next to the diode). This helped in all cases where I had this problem (currents up to 16 Amperes), and there have been quite a few, since I am using this driver (with various modifications) for most of my builds.
For my SRK 7x which draws up to 28 Amperes I additionally installed a 4.7 Ohm resistor in series to the schottky diode (proposal of DEL). I am sure it helps but I don’t know if it’s really required in this light, since I didn’t test it without this resistor.
Inserting a 2.2 ohm - 10 ohm resistor in between cell+ and the C1/D1 junction will fix the problem. (Stacking C1 is a brute-force method that fixes the symptom, rather than the problem. Tantalum is not great for this, too high ESR.)
You still want to feed the emitter directly from the cell+. IIRC the layout of the X6 driver makes this tricky, but I am sure you will find a way .
(On some drivers TomE’s ‘tombstone’ technique works well. Basically you build a little pyramid with Rnew+D1+C1 placed vertically, their junction up in the air.)
Thanks DEL, I will do the trick in a problem triple and make a nice picture of the result
And report if it works: currently the light works fine on a high drain Efest IMR 18500, but not on a Sanyo or NCR18500A.
The OP has a link to 219C 92cri LED’s on a triple PCB with optic that is a great deal or if you want a tighter binned version Clemence is running a GB for some 3 step binned 219c’s as well.
.