I’ve done quite a bit of searching but it seems I have not been able to find much that helps with this.
(Or, at least not that I’m recognizing)
I have a 3x XM-L2 torch with a magnetic sliding switch.
The sliding switch gives about 22 steps of dimming.
The torch uses 2x 18650 cells, and a tail reading on max output is about 2.33A.
My very possibly bad math tells me the LEDs are getting about 1.5A ?
I’d like to drive the LEDs at a higher current than that (basically, get them close to their max rated current…3A or so)
The driver seems to be using a 12F675PIC.
Here is a picture of it:
The PCB seems to be 24mm… the outer ring seems to be 25mm.
It is only a single PCB with the underside containing the battery spring + the magnetic sensor.
What recommendations would you provide in order to drive the 3x LEDs at a higher current ?
Resistor mod ? Cap mod ?
I guess a completely new driver would be an option, but I’d rather not lose the magnetic switch.
This is a diving light and I’d rather not have to twist the tail cap to operate it under water.
Thank you very much in advance for any help you guys can provide.
Can you provide a few more pic’s so we’re able to see the entire board (dosent have to be the whole board in a single shot, thats a good pic, can you get one from the other way so we can see around the toroid (big yellow wire wound thing)
Where do the batt+ and - wires hook up? Are the red and black wires pictured LED out or Batt in?
There arnt any drivers for mag control lights so your going to have to hope it can be resistor modded, we can help you with that once we see the pic’s. What exact light is this?
Welcome to BLF! I see you recognize the PN and/or other markings as a PIC part, are you a PIC guy? There arnt many of us here, mostly AVR but more of us are starting to switch over / incorporate PIC uC’s instead of AVR. I’ve pretty much switched exclusivle to PIC
Those 2 resistors seem to be in parallel, and connect the drain pin of the mosfet to the negative LED lead.
The source pin on the mosfet in turn seems to be connected to batt - (chassis/ring).
That seems like its the ones to try then. Do you have other resistors you can stack? you can use an online resistor calculator to see what adding different values will do to the total resistance so you’re not just guessing.
I measured those parallel resistors and they seem to be 0.2 Ohms.
The other two R200 resistors on the board are also in parallel, and also read 0.2 Ohms.
Im confused now…How is it possible that those 2 parallel resistors can dissipate the current in the XML circuit without burning up ?
The negative lead from the LED is connected to those two resistors, and they are in turn connected to the drain on Q2.
The negative LED lead also connects to a 6.5k resistor going to pin one of the 5305DA chip.
How are those resistors dissipating the current driven through the LED ?
I would try stacking something like R500 on top of the R200s (.2 .2 .5) but I’m crazy, I want to see the light explode. It should lower the resistance to whatever the calc says. See if that increases the amps through the tail. The resistance is so low a pencil won’t work too good. You can be more conservative your just looking for an increase. After you figure that out you are close to blinding some fish.
What kind of diving you doing anyways in las vegas… Up and down or ocean
Well, I added 0.327 Ohms of parallel resistance to those two r200 resistors…
But it did not seem to make any difference in the current draw from the batteries.
?
Diving:
In Las Vegas we go to Lake Mead and Lake Mohave.
Not too bad.
But we also go to wreck alley in San Diego, as well as other cool spots.
Maybe…
But, I don’t think the current is being limited by the wiring on this.
Even 26awg wiring should be able to handle 3-4 amps on these short cable lengths.
Specially for a short duration to get a reading.
Yes, maybe if I manage to increase the drive current, I’ll improve the wiring.
Hehe that’s true… I like watching those guys who do the kelp runs at night. I think they have big cahonees, they swim around and see grouper bigger than sharks. Ocean got way too many sharks.
I’d like to see the catfish down at the bottom of those lakes though. In Georgia there was some monsters but they rarely bite. They do noodling over there
Re: diving
I’ve done those dives with the Goliath grouper in Key Largo.
And kelp forest in San Diego.
And seen catfish at the bottom of Lake Mead and Mohave.
Plenty of night dives too.
I just got one of these in for review the other day. I’ll see what I can do for this driver and let you know. It’s a pretty cool light really. I’m sort of surprised more people are not playing around with them. It’s like a longer C8 in size and the dimming works nice.
Those resistors are what we call sense resistors. Maybe take a look at the first 3 paragraphs here for a better understanding: http://www.mouser.com/pdfDocs/Ohmite_CSResistors.pdf The diagrams in that document do not accurately depict what is going on here. Instead, a boost controller chip monitors the sense resistors to determine output current. The boost controller modifies it’s behavior in order to maintain the set current.
Since this is a boost driver, it should give max current despite resistive losses on the output wiring - until we hit the output voltage limit. That said, it wouldn’t hurt to upgrade it while you’ve got the thing apart. And who knows, maybe that’s what you’re hitting - the output voltage limit!
I don’t know this for sure, in fact it’s pure speculation, but I’m wondering about that second set of sense resistors. Maybe the second, unmodified bank is input current sensing. You mentioned that that decreasing the sense resistance on the output bank didn’t do anything, right? If so, maybe the problem is that the driver is limiting based on input current. That seems kind of unlikely though. If you switch from mostly depleted cells to fresh cells does the light get brighter?
The 5305DA chip might be the boost controller (I don’t see any other likely suspects in your photos, unless there is something hidden under the inductor next to the U1 marking). I can’t find a datasheet on it.
Better, less obstructed pictures would be helpful. Or just more pics from different angles.
Re: current sensing resistors
I get that, but they are in-line with the load, so they are getting the full drive current through them.
The PDF you linked shows some example resistors, and all the SMT ones have very low power ratings.
If this 3x XML circuit is driven at, say, 6 volts/1.5 amps, thats 9 watts, divide that by the two resistors and each is still dissipating over 4 watts. These SMT resistors would fry, no ?
Or are these tiny SMTs really rated for that much power ?
I think I’m still missing something here.
I haven’t tested that specifically, but I think the brightness doesn’t change much as long as the batt voltage is within the allowed range (8.2-6.0v ??)
That other pair of R200 resistors are connected in parallel from batt- to pins 1,2, and 3 of the 4410 chip.
Pins 5, 6, 7, and 8 of that same chip are connected to batt+, and pin 4 is connected to pin 5 on the 5305DA.
The magnetic sensor (49E3E1) is connected to the 5305DA chip (1 of the pins), and also to the PIC chip (the other 2 pins).
Yep same driver. Slightly different light. Mine is a Ultrafire UF-450 and it has a little different head. XM-L2’s in it. I’m going to see if I can find a source for the reflectors in these. They would work fantastic in C8’s or C12’s.
I’m afraid I won’t be any help on this though. I managed to kill mine putting it back together. Shorted it somehow.
Think what C is saying is use a calc 
