I ordered and received the Lanshi C8-c from PriceAngels (SKU :94229). Price was US$10.39 after %5 off coupon.
Overall it is a cheap light at a cheap price, I'm neither amazed nor disappointed.
On arrival I disassembled it:
Decent machining - could be better
Nice finish - minor blemishes
Very tight spot - great thrower
full o-rings, glass lens, tail stands
plastic smooth reflector (but I think that's OK)
reverse clicky - a bit wimpy
XR-E emitter, unknown rank - Nice white color
Direct drive/PWM driver with selectable modes (with soldering iron)
No visible heatsink paste ooze around the star*
That last item bothered me, so I took apart the pill and there was a blob of paste between the pill and the star...Huh?!?!
When I wiped everything down I saw the problem. The flat recess in the pill where the star sits wasn't flat it was dished in, concave! The star was touching the pill only around the edge with a gap under the middle of it. So much for quality control... Sorry didn't get a picture before I flattened it with a small end mill in a drill press. Before reassembly I check out the driver board, one IC (controller) a diode, a capacitor, two resistors and a big transistor. No inductors - so no step-up or step-down. It had 5 modes, no memory - hi, med, lo, flash, SOS. The controller actually had four different outputs that could drive the transistor. I can't remember all the combinations but they all had a blink, flash or SOS mode. So I choose one that gave me hi, med, lo, slow flash, fast flash. There is no current regulation at all with this driver. The transistor either turns ON, connecting the LED straight to the battery or its OFF. Medium and Low are the same except the duty cycle is lower. Again sorry I didn't measure it. But the frequency of PWM on low was about 76Hz! You can almost count that by eye! A real driver is in its future.
The fact that the reflector is plastic doesn't bother me because the pill screws into the head giving a shorter path for heat to the outside surface. If it gets hot enough to melt the reflector, I have other things to worry about... However this arrangement causes other challenges with optic alignment and focusing. You win some, you lose some.
I don't know how common is the problem I described with the pill machining, but it is a show stopper if not corrected. This isn't a bad light it just needs some TLC. It all depends on what your expectations are.
-Crux
P.S. Just noticed the template for reviews - should I redo this in that format?
And the top side - The MEM2309 is a 4A mosfet, I couldn't find anything on the CX2807... (Sony?)
The row of pads (the jumper on the first) is where the various outputs can be selected to drive the transistor. Current limiting depends on LED Vf, cell voltage and gauge of the connecting wires!
Personally I don't think there is much that can be done with it.
This is have to be the simplest driver that I ever seen!
could I convert it to single-mode operation (currently has 5 modes, but it needs awful lot
of time to revert to HI mode or it has some kind of memory, not really sure!!)
is it one of the hated next mode memory drivers? you could probably solder a 500k resistor over the capacitor and have it starting on high if you turn it on more than 3 seconds after you’ve turned it off.
i’ve done that mod to some lights, used 220k-880k resistors.
ok I checked all my salvage smd resistors and unfortunately did not find anything close to 504
but just to clarify this procedure, I need to solder resistor right on top of that capacitor, right? Does it matter how it is turned?
I an not so familiar with electronic as you probably can see but I can solder no problem
P.S. Thanks
It doesn’t have to be a smd resistor, i used legged ones. If you use too low value on the resistor it can become difficult to change mode as you have to be very fast with the half-pressing. If you use a higher value it takes longer time to discharge the cap and it will take longer for the driver to start on high again.
It does not matter which way you solder the resistor, just make sure it’s connected to both side of the cap since the idea with the resistor is to discharge the cap and make the driver forget which mode it had before turning off the light.
I get the point
What is the actual function of that capacitor, what would happen if I completely remove it and just do a short circuit? Would I get 1 mode driver then, which is what I want!
I suppose that this capacitor is needed otherwise it would not be there.
I have another “extreme case” driver that I want to turn into 1-mode. This one “remembers” last user mode for hours (more than 4h) after power off and switches on in next mode, so annoying
Yeah, i guess you’d get an 1-mode driver if you remove the capacitor. Why would you rather have just one mode instead of a driver without memory that always starts in high?
Yes, but if you solder a resistor over the cap it always starts in high. It takes 3 seconds to reset the memory on my drivers. So 3 seconds after you’ve turned off the flashlight it starts in high again. But sure, if you never will use Med/low mode just remove or short the capacitor
This driver that I am trying to outsmart is from (almost) identical flashlight, I have a XM-L T6 version and driver puts out about 2,2A with unprotected Sony 18650 battery.
Problem with resistors is that i don’t have any at hand (managed to find two old ones which in series give me 476 ohms…and then realised that you suggested 500 KILO ohms ), I’ve been planning to get those 1000pcs pack (50 values 20 pcs each value) SMD, but never got to do it actually… ————————————————————
Update:
I removed capacitor and now it is working as 1-mode driver
but now it pulls only 1,2A :exmark: I could swear that I measured 2,1A - 2,2A on high with capacitor in place, hmmm…
maybe I should short circuited capacitor pads?
So I choose one that gave me hi, med, lo, slow flash, fast flash. There is no current regulation at all with this driver. The transistor either turns ON, connecting the LED straight to the battery or its OFF. Medium and Low are the same except the duty cycle is lower. Again sorry I didn't measure it. 
But the frequency of PWM on low was about 76Hz! You can almost count that by eye! A real driver is in its future.
but I can solder no problem 
