I see, thank you for the clarification. I assumed the spacer was similar to the S2+. After seeing photos of the pill I see that it is just a disk that is needed.
1 Thank
I bought a cheapo bike light set on Amazon for my new used bike that I got for free from a neighbor who left it next to the road for the taking.
I chose the front light based on that it appeared it could be taken apart (screws in front and back) in order to modify it. And sure enough I was able to replace the respectable 4000K XP-G3 LED with a 2700K SST-20.
Why? I may have lost some lumens, but I feel the lower CCT is less annoying for oncoming people and I like how it looks. I have yet to see another bike light while riding around with an incandescent looking light like this.
8 Thanks
Seems like those two LEDs on either side of the board are covered by housing? Strange design 
Those 2 amber LEDs flash in the last 2 modes, which is what I use when riding at daytime. There is clear plastic on both sides of the light where they shine through. I think itās kind of a neat design.
5 Thanks
I have replaced the LEDs in an Astrolux MF01mini full copper.
I bought this lamp second-hand two years ago, but because of these Luminus/Sanan SST-20 LEDs I wasnāt really happy with it, so it just sat on the shelf. It has this copper heatsink for the AMCs on the driver:
It previously had SST-20 4000 K 95 CRI in it. Unfortunately, these were a bit green, and using the lamp at full power to make the tint pleasant canāt be right either.
Nichia 219BT-V1 sw45k have been installed:
Unfortunately, this was an extremely difficult reflow. Whether it was the LED board or the LEDs, I donāt know, but there were constant short circuits between the thermal pad and the anode/cathode, even though the amount of solder was kept to a minimum. The LEDs also did not center themselves, I had to do this manually. Weird, I never had such complications with 219BT before. So I had to reflow the LEDs one after the other (always checking continuity in between), which of course put a lot of thermal stress on them. Hopefully they didnāt suffer too muchā¦
But the tint is perfect, just rosy 4500 K.
And the lamp gets brutally hot very quickly. Just 10 seconds of turbo (from room temperature) is enough to notice the heat in the head.
Iāve often run the lamp on turbo and there have been no problems with the LEDs.
After 5 seconds with a full 40T battery, the MF01mini now reaches 3900 lm, immediately after switching on 4200 lm. The beam is simply perfect, no color distortions or annoying rings.
8 Thanks
Out of curiosity, is it possible the stock SST-20 were the B variant with the 3030 solder pad? It would explain the difficulties with centering and short circuits.
Iāve got an MT07 of my own, but fortunately it had already been swapped to 519a by a previous owner.
2 Thanks
The stock SST-20 were normal 3535 footprint.
I am pretty sure that the combination of the board with its tolerances of the footprints and the smaller contacts for anode and cathode on the 219BT-V1 are the cause for the trouble I had with this reflow.
But still, I had never any issues with Noctigon XP boards and these 219BT LEDsā¦
2012 RRT-01 w Domed 2700K 519a and Sapphire lens:
it was a nightmare mod
first I got some stray solder bits that shorted the stock 2012 driverā¦ the light would not turn on but got hot really fast.
I removed the mcpcb, redid the reflow, and cleaned everything, more than once,
I removed the driver and found a resistor doing a headstand (one end came loose). I resoldered the resistor. But the driver has permanent damage, and though it turns on when I close the battery compartment (w the control ring all the way off), the light would not turn off by the control ring, and would not rampā¦
I took a break, felt sorry for myself, made plans to buy a new driverā¦ but
Fortunately I had a spare 2019 driver, now installed. The light is working properly now. Im not thrilled by the 519a positive DUV, butā¦ its been a rough modding day, and Im going to just leave it alone now.
4 Thanks
Where can i buy springs to replace/add to the convoy 22mm boost driver positive terminal posts?
I have used the ones convoy sells with much work and suboptimal results. I am sure better ones exist, like the ones on the new 10A buck driver
Are these good for high current?
Based on the thickness of 0.6mm, I would say no.
The springs Simon sells are 0.9mm and rated for 3v 10A.
1 Thank
Asking Simon if he can sell them separately might be our best option.
The closest spring I could find is this one:
1 Thank
Another option might be Hankās beryllium copper springs, possibly the cylindical ones used in the boost drivers, though they would need to be specifically requested.
1 Thank
I swapped one of the Hank ones onto the 6V8A driver and the driver shortedā¦
Be very careful with springs on the high density switching drivers, a lot of them have a brass heatsink on the back electrically connected to GND.
3 Thanks
Yes they are so beautiful but are extremely hotš
Thank you!
1 Thank
Thanks, I will do this!
2 Thanks
RRT-01 mod to 3000K 219b
I thought this was my final answer for a warm option in this light:
but then I noticed that at 0.2 lumens there is severe flicker: (the 2019 driver is not happy w 3000K 219bā¦ darn)
aaargh!
I also tried an SFT-40 3000K and a 2700K 519a Domed. These are the three Opple 3 plots
2 Thanks
Thanks for the warning! I may just solder it on top of the post to get some springyness added.
I remember you said you wanted a domed emitter for that reflectorā¦ have you considered slicing a 519a? Maybe thereās a sweet spot where you slice off just enough to get your ideal duv but keep enough that the beam still looks good.