The one with 6 of these LEDs on them was a XM-L board, the one with 8 (in the light) and the bare one were XP boards.
Some of them have that long centre pad (you say āmid strokeā), which is ideal for this purpose.
I just sanded off the masking paint.
:+1: Exactly.
They have practically no tint shift, so everything is the same tint, both in front of you and to the sides.
My other headlamp (which is starting to wear down a bit) has a single 219C and does have a little more yellow on the outer perimeter, even with diffusing optics.
Yeah, it looks like it, but itās not so.
Thereās just very little space left when you put 8 of them on it.
In the picture with 6 on it you can see better.
So i soldered a jumper over the outer pads, where the red wire is soldered on to.
low, high, off. e-switch.
Long press for ramping. First long press increases, second long press decreases output, would be better if it decreased the first long press.
No ramping memory.
double click for strobe.
Not perfect, but no PWM. :+1:
Well, i use a headlamp daily.
Together with my reading glasses it makes me see very well what iām doing.
With external lighting thereās always shadows cast on the things iām working on / with.
Iāve found a lens that fits (after some filing) too, to finish it off, but havenāt found a proper way to hold it in place yet.
Very nice headlight build. I also love hi cri mule 18650 headlights. Iāve always used nichias. Iām not familiar with the emitters your using. They are cheap enough to experiment with some. Thanks for bringing these to my attention. What would you guesstimate the lumens on high is?
Iām not sure, but the little LED goes green when itās fully charged.
Good question though.
Iāll have to (finally !) replace the fuse in my DMM and test this.
I really donāt know thatā¦
Those LEDs are rated 0.5 Watts and 47 Lumen, but maybe itās lower because these are high CRI.
But then again iām using them on lower power than 0.5 Watts each, so they probably work a little more efficient, especially on the low mode (which seems to be without PWM).
I found a useless double convex PMMA (i think) lens that i made to fit in the opening.
I sanded one side flat in one direction so that it spreads the light a little horizontally.
I thought iād try that and it works rather well.
Yeah, i painted the bare aluminum circumference black again too.
It got damaged when i took out the original LED board so i sanded it.
Looks neat huh?
And then the bad newsā¦ :partying_face:
I enhanced some things in the picture.
On the right, the charging PCB.
Iāll just leave that as is.
Maybe someone can tell what kind of circuit it is and if it shuts off when fully charged or if it āfloatsā on 4.2 Volts.
On the left, the driver board, the e-switch and some peripheral parts (2 Rs and 1 C) are on the other side.
So basically the unmarked IC does the UI and regulates the 2 tiny ā00A8Cā FETs.
Their output goes to 3 x R680 in parallel (0.23 Ohm) to the LEDs.
So thereās no current regulationā¦.
ā¦.so i took the FETs and the resistors off.
I want to replace those parts with those 3 chips there, those are CN5611 current regulators, like variable 7135s.
Bought a bunch some time ago, and i think i can drive them with the IC output (connect to pin 4 āISETā via a yet to be determined resistor), instead of the FETs.
They can do 30 to 800 mA each, but thereās a considerable voltage drop above 400 mA, so i plan to use 3 of them in parallel, and 1200 mA total would be quite enough.
The CN5611 has a Voltage drop of around 0.2 Volts at 400mA.
The challenge is getting them on the board with some sort of thermal path and working within the desired range.
It would be 90 to 1200 mA when all works out well.
1200mA is not much, but itās enough for its purposes and otherwise it will get hot and itās around the maximum rated power of the Osram LEDs.