I’m surprised your MF01 look that much brighter than the ROT66 and even more surprised that it is so much more on the rosy side of the BBL than even the Cree 3A. I have two MF01, one stock and one with Lexel’s buck driver. The 219C tint on both is on the greenish side of the BBL. One is significantly greener than the other.
The MF01 renders colors the best in your comparison. I think you won the tint lottery with the 219C.
Does this concur with the the notes I quoted from the Q8 (FET+1)? I can’t really tell.
That first step down looks a bit larger than 100% to 55. Maybe more like 100 to 35%? Does that seem correct?
Sorry, all that code looks Greek to me. Lol. So does it look like the ROT66 is using their own tweaked version of NarsilM v1.2?
Is this correct? In mode 4:
Level 1 is set to 100% of channel 1.
Level 2 is set to 100% of channel 1 and 2.
Level 3 is set to 100% of channel 1, 2 and 50% of the FET.
Level 4 is set to 100% of channel 1, 2 and the FET.
This kind of makes sense to me. I hope it’s correct.
I believe the spacing in brightness levels will vary between the 219C and the xpl due to the Nichia version using 7 x 7135 and the Cree version using 14 x 7135. Is this correct?
Do you know how to set your own temperature limit? Here is a link to a video I made showing how to do it. Make sure to start with a room temp light inside and out.
I’m sorta glad that the 700ish lumen mode will probably be the one I use most, because it will be less rosy compared to the max. Those 9080 nichias driven to their limit are pretty damn pink. Although I might think differently once I get mine.
The driver and configurations on ROT66 (FET+1+6, 3 channels) is different from Q8 (FET+1, 2 channels), so sometimes you cannot simply apply what you observed from Q8 to ROT66. The thermal step-down in ramping in original NarsilM periodically checks if the temperature limit is exceeded, and if yes, it will reduce 1/6 = ~16.7% of the current output level (output levels are not guaranteed to be linear though). And the lower bound of this step-down is the max of the 7135s, i.e., on ROT66 it is (1 + 6) = 7 x 7135. That’s why I said “step-down to the highest regulated brightness level”.
The default mode-sets defined in the code Lexel posted are slightly different from the original NarsilM v1.2, so they did tweak something, and I wish they could release that modified version (they should, for license compliance).
The code from Lexel:
PROGMEM const byte mode7135Set4 = { 25, 255, 255, 0};
PROGMEM const byte mode7135sSet4 ={ 0, 0, 255, 0};
PROGMEM const byte modeFetSet4 = { 0, 0, 0, 255};
Put them in a table:
table(table#posts).
|Channel| #1 (7135 x1)| #2 (7135 x6)| #3 (FET)|
|Level #1|25|0|0|
|Level #2|255|0|0|
|Level #3|255|255|0|
|Level #4|0|0|255|
255 means 100% output of a channel. 25 means 25/255 = ~10% output of that channel. Hope this is much clearer.
Sorry, I have no idea how many 7135s are used in the XP-L HI ROT66.
The Nichia version is using 7 and the Cree version is using 14 according to Lexel.
Okay so I think I’m getting it.
Level 1 is set to 10% of channel 1.
Level 2 is set to 100% of channel 1.
Level 3 is set to 100% of channel 1 and channel 2.
Level 4 is set to 100% of the FET (channel 3).
That’s correct?
Doesn’t the highest level in other versions of NarsilM add the 7135 channels to the FET? In this version it turns them off and uses only the FET?
No.
When all 3 channels are at 100%, the output is actually a bit lower than the FET by itself. So for the very highest level, most lights run FET-only. The 7135 chips would only drag it down.
Tom and I got a copy of the code today. It’ll be up soon.
Okay, I must be remembering it wrong. Thanks.
Here’s the output behavior on a scope. Gif from low to max FET. Horizontal lines are 100% 1x7135 and (1+6)x7135. The latter is almost impossible to activate on ramp mode so had to swap to mode config for that, but the 1x7135 I managed to do a couple of times.
Interesting. 7 x 7135 on the Nichia is right on the spot, the max temperature is ~43C in my runtime test. Curious about the brightness and stabilized temperature when 14 x 7135 are installed for the Cree version.
Exactly.
Sounds good, looking forward to it. 
That’s fairly clear to demonstrate how the modes are achieved. Well done, maukka :+1:
For me, nope. For us non EE, it’s just squiggly lines. Lol
Just like in the code shared previously, you can visually see that the low levels PWM between zero and 1x7135 at 16 kHz. 100% 1x7135 is of course without PWM. Mid modes above the single 7135 are PWM’d between that and the 6x7135 channel at 16 kHz. (1+6)x7135 is again without PWM. Then it PWMs at 32 kHz between the 1+6 and FET and finally transitions to full FET.
On a side note, at some point I’d like to reduce that 32 kHz to 16 kHz, but the attiny85’s second PWM counter works differently than the first. I’m not entirely sure it can run at the same speed.
(pins 5 and 6 share PWM counter 1, pin 2 or 3 uses PWM counter 2)
It looks like the Mid modes are PWM’d between 1x7135 and (1+6)x7135 (i.e., the 1x7135 channel is always 100%), aren’t they?
Oh yeah, exactly like that.
Yes, the 1x7135 power channel stays on, steady, when the Nx7135 channel is active. And both stay on, steady, when the FET is active… Except for the very last ramp step, where it’s FET-only.
Nearly all multichannel lights since the BLF-A6 have done things this way, because nobody has found a better solution yet without major hardware changes.
it would be possible in the ramping table when the FET gets active after like 25% FET slowly decrease the AMC output level, but in fact of having them on reduces PWM on the light output and different PWM frequencies dont interfere with each other