RD: Interesting - moonlight levels are typically a problem, so they probably only use this output channel for very low output levels.
How do you know all this? Is this basic EE design stuff, or do you know this driver specifically? In my case, I'm a SW engineer knowing enough of EE design to get myself in trouble... I've actually learned more here on BLF and working with flashlight drivers than on the job. In the last few years, I haven't tinkered much at the low levels of MCU's. Our dept. is big into PIC controllers but I'm not involved much at all in them.
I'm doing heavy C# work in the panel Win PC's we use on our devices now.
All that I know about designs and driver circuits, I learned in a short time spent here on the BLF. I have some knowledge in electronics, so it’s interesting for me to study various drivers.
As for the driver SP70. Sofirn began to use a similar driver design (PIC - Main FET - Moonlight FET) in the C8F 21700 flashlight. A similar driver is also available in SP33v3. I don’t know why they used Moonlight FET, maybe they had difficulties, or maybe they went the simple way.
By the way, the moonlight channel does not use PWM, It works as ON / OFF.
I've built up FET only drivers before, using NarsilM, for high powered lights - 6V LED's with 20K to 40K lumens, and at the low levels for ramping, it gets very jumpy, erratic. I didn't really care too much - the lights were really for the wow effect on high output. For a production light though, I'm sure you have to have some low lumen output support.
Ah, thanks. This makes more sense now that I look at it. The bottom right had to use vias because it was surrounded by signal traces and cut off the ground plane all around it. I edited my original post to not confuse people.
I agree. It looks like this method uses this driver.
The moonlight channel is included in the full range of ramping and is used to support low lumen output.
For the experiment, I took the C8F 21700. At a low ramp level, the duty cycle of the main FET = 1.86. I unsoldered the moonlight resistor and got a very low output of 0.3-0.4 lumens at a low ramp level. If set the duty cycle of the main FET 2.0-2.1, it is possible to achieve an increase of 1-2 lumens without using the moonlight channel, but it is very difficult to set a low level, the brightness changes too fast.
We also have another problem, drivers may need to be calibrated, since different batches of electronic components can give different results.
And you are right. For mass production, it is easier and more stable to use the option with a separate moonlight channel to support low lumen output.
Thanks for the answer and photos. Refreshed my memory as I now recall learning about the LDO drivers a few years back.
I could also use this type of driver in smaller form (17-20mm) and for tail clicky. I’m pretty sure Adventure Sport and MTN both sell a zener modded 17mm FET/DD, anyone know who makes one with the LDO instead?
Or a 17-20mm constant current driver with LDO and high amp output?
Zener takes me back a few years. Think the parasitic drain with Zeners was too high for e-switch drivers. I reflow my own LDO drivers, but it's been a while. I would think Lexel sells them assembled, or would be willing if you ask him. MtnE doesn't have much in e-switch compatible drivers, and hasn't kept up with the latest. I think it's been years since Richard even posted on BLF, but I know he's still active with MtnE.
I do not know what diameter of the Convoy L6 driver. The diameter of the SP70 driver is 30mm but it has two bumps. You may need to grind them. The LED on the SP70 XHP70.2 is 6v.
If it fits, you could. It would at least be close on size.
I wouldn’t, though. The UI isn’t great and the FET is sub-par. You’ll also want to replace and/or bypass the spring if it comes with one.
Grab something from Lexel instead if you can.
Edit: Saw you said original L6, so XHP70. Nevermind the comments about springs/FET.
My L6 uses a 30mm driver. It’s 6v as it’s 2 cells in series. No boost drivers here. It should fit once you file off the 2 ears. The UI is fine for me in ramping mode. I don’t see a need to mod the double spring. The stock FET should keep you from burning up your led, so I would not bother changing it. The only thing I miss on the SP70 is the voltage readout that NarsilM has.
I have a newer model Convoy L6 70.2 that I got in January. It has the notches already so it is easy to remove the side switch. The twin Wainlight I got in March did not though.
Also I have an SP70 driver, it does fit the convoy but the wires to the side switch are too short for use in the L6/clones, so you will need to replace them with longer wires if you chose to go that route. Since you have the SP70 driver you know it has 4 cables to the side switch. Mine were all black so you need to keep track of what positions the new wires need to be. I know it fits, cause I tried it before getting a Lexel driver.
Its not a bad driver for basic ramping mode and memory (similar to C8F triple 21700), plus double click for turbo. The side switch PCB isn’t round like the Convoys, more like octagon shape, and tiny bit smaller so should fit through even without the notch. Mine was in the Convoy so already had the notch.
I have the Wainlight cousin as well (Brown), but that one has a Lexel driver with Anduril and lit side button (blue LED), with bypassed springs running around 6800lm start up (Texas Ace Lumen Tube). I haven’t changed the LED wires to 18ga yet. I think that will help bring it up past 7000lm. Using Golisi S43 gold cells.
I’ve already switched it over to the Sofirn Sp70 driver and it’s getting 3600 lumens instead of the previous 2850 so that’s nice. I ended up just re soldering the wires as it was less hassle than filing in notches.
The Sofirn TK70 gets 7200 lumens with bypassed springs using button top PLB 26650’s, the same ones Sofirn sells but my supplier added a button for me.
The L6 had some great potential. That’s the issue I have with Convoy is Simon isn’t into using drivers/firmware of the competition e.g., Anduril & NarsilM.
Cool - Yeah didn’t see this until now. Didn’t know if you had already done it. Glad it worked out. With the XHP 70.2 it should provide more output.
I have no idea what the bin is for the Convoy vs Wainlight, both are 5000K temp. My Convoy is now back to stock driver with sense resistor mod, and lit side switch running just a tad over 4500lm on Texas Ace lumen tube at start up. Pretty satisfied with that. Especially with stock LED wires, and spring bypass.
I think his was using the XHP 70 not the newer XHP 70.2 so output is a bit lower. Even with resistor mod it would go up only to about 3800lumen based on Vestureofblood’s posted output.
Not too familiar with the XHP 70 output vs 70.2 but could be the reason.
