Stock output on high is about 1,8-1,9 to the emitter.
Adding 0,12R resistor did close to nothing on top of one R200 (where there are several of them) Added another one, little improvement. Added a wire. Still, just 2a out.
I have no idea have healthy that is.
Does anyone see a better way of improving output? If not, Ill just go back to stock…
Hi R86, I just went back and found this. That light runs on 1x18650 or 2xCR123A, correct? If so, I think it is a buck driver, which is basically DD when on 1x18650. The resistors will not do much because it us not regulating at that point.
You will need to improve the resistance path to get any gains here. Try bigger emitter wires and braiding any springs there are.
Yupp, it should be able to run on two CR123 according to specs. That should make it into a buck driver.
What you say about the resistors makes sense. For now, I have the driver slightly modded with resistors (no wire). But have not put it back in the light since Im waiting for some LEDs.
Wires to emitter are already fairly thick.
When testing the driver in a "bench setup" tail-cap was eliminated and emitter current was basically the same as assembled in a flashlight. That should mean that its close to nothing to see in current improvement by modifying the tailcap as far as I can tell.
The main thing that can be improved are the spring on the driver circuit. I have a feeling it will not do wonders though, but Ill try once I put the light properly together.
If its basically DD with a single cell, what is limiting the current to the emitter that much?
No electronic knowledge here, just an observation. The top right component in the 4th picture looks out of place. The bay label is ZD1 (does that stand for zener diode?). The bay boarder is covered by the component. That is another sign that it is probably out of place.
I paid 25$ for mine (some other instant kill sale). I just wanted a "premium" light with magnetic ring (never had a light with magnetic ring before). If you like under driven lights, that does not seem mod-friendly in terms of the driver circuit, sure. Its worth 20$.
I have a feeling I might sell it locally. Personally, If I could throw 5-10$ at someone and "upgrade" it to a XinTD, then I would do that.
When that is said, I might like it. I have not actually used it properly yet.. :p (lol). I just did what I normally did.
"Put it battery, does it work. yes. CW tint, no surprise there. That sucks".
"Tested tailcap.. Less than 2 amps.. that sucks"..
And then I took it apart.. :p I have various emitters incoming later this week, so then Ill actually start to use it soon... I have a feeling I will end up selling it though.. And I don't sell many lights... I have gotten a Fandyfire Padme, and the magnetic ring is much sweeter on that IMO.
As for your question about electronics/the component that is out of place. I noticed that, but I think its supposed to be like that. Its only got one "leg" in the middle. It probably have to be mounted a bit sideways. I just did not answer in hope of someone who actually knew what they were talking about answered.. ;)
Edit: And my light as a bit strange grey-looking color..
Despite that, I picked up the Archon M30A on the recent/ongoing sale. But only because it was 10$. Triple fatty AA hosts are generally quite expensive. 10$ for one with a Magnetic ring, that was too hard to say no too though. If that light is mediocre too, I had it with Archon lights, even when they are sold at 60% +++ discount.
Thanks R86. If drive current could not be increased, I was thinking of dropping a 219 H1 B10 in place of the xml.
It looks like that component is supposed to be connected to the 2 pads above it in the picture. I bet if you remove it, you will see connection points on both ends of the component. You will also have a better view of the board and what it was (or was not) connected to.
A Nicha 219 would not be wrong, output would be quite low compared to a CW XM-L though. You would need a 20mm (copper) mcpcb. The body is partially hollow. The emitter is nicely secured though. There is a metal ring that you screw on top of it. Should help make up for the hollow body.
Hopefully the driver would give the same output to the Nichia as the XM-L..
The comment you made about one leg flew right past me. It made realize the component has not connection points visible from the top on the two ends. I don’t know how I missed that.
So it looks like the component is 180 degrees out of place. There must be two legs under that big solder blob. I’m guessing those two legs would attach to the two unconnected pads above the component. There must be a pad for the center leg. That pad would currently be hidden under the component. The bay border mark make more sense that way too.
Ill take a guess here, but the "board" (when looking at all four sides) obviously have several unused places. My guess is that its being used for several different Archon lights. Some using different components. In the M20XL, they just mount that one component a bit sideways like that, otherwise it might make contact to something that its not supposed to get contact with. But what do I know... :p
The black wire on the left, where ist that going? I guess LED negative?
Another edit (last time I hope :P)
It’s a comparator based buck driver. The voltage above the current shunt goes to the non-inverting input on pin 3 of the comparator, the reference voltage to the inverting input on pin 2.
If you want to increase current, you have two options. You can change the current shunt resistor as you already tried, or you can change the reference voltage generated by the zener diode and R6-R9.
Should not be that hard to modify the resistor network, but I can’t figure out how the resistors actually are connected on the photo.
Such self oscillating curcuits can be a bit touchy, might very well be that changing the reference voltage is the better option - if you are still interested in increasing current.
relic38 might be right that it is not regulating anymore at that point thought.
Could try to disable the driver, maybe by bridging the current shunt resistors with a thick wire to make it DD. If nothing changes, the resistance of the whole build is to high.
We don’t know exactly what FET they used. It could easily have a quite high on resistance for example, especially since it is a PMOS.
No, looks fine to me. The footprint on the PCB was not made for this package type, but its connection makes sense.
I would say it is not even a zener but a standard diode in sot23 package whose forward voltage is used for reference.
But it’s not that easy to say what exactly is going on since I cannot see all traces. Maye it works better on my big screen at home.
That’s why I guess it’s not a zener but a normal diode. One would not get zener voltage the way it is installed, only the forward voltage.
But the ref voltage will probably be only 100mV or so, so there is nothing wrong with that. (In such a crude circuit, anyway. ;-))
(last time I hope :P)
