Okay, I have a SP36 now. My neighbor is a good person, and hand-delivered the package they received by accident. The label on the package is correct; my local delivery agent is just sloppy.
So. I have a SP36 sample. Itās a little banged up in ways which werenāt caused by shipping, but the important part is, the driver isnāt glued. It actually falls right out if I point the head upward without the body tube installed. So this should be a good host to use for development purposes.
I have Anduril on it, working pretty well. I havenāt tested the thermal regulation yet, but other functions appear to work.
The charger works too. If I connect it to a USB power bank, it charges at 1.5A. The flashlight still functions while this is happening, but the buttonās charging LEDs override the driverās normal indicator LED functions. Thatās fine though. The light can actually be left in battcheck mode while charging, to get realtime progress information (ish; reported voltage is the charging level, not the cellās resting level).
It can also run entirely off USB power, with no cells installed. It just makes the button LED behavior change; when itās running at 5V it has no low mode for the button LEDs, and instead just goes smoothly from green to red as the main emitter power goes from moon to 350mA. And battcheck displays 4.2 to 4.4V in this mode.
Itās running the experimental SP36 build of Anduril I made a few weeks ago. I took some guesses about the light and the driver, compiled a version I thought might run on it, and ā¦ it works. So I mostly just need to make sure the calibration is correct, like for the voltage readings, thermal regulation, and the ramp shape. And then it should be good to go for production purposes.
It aims to be a smaller Q8 with built-in charging over USB-C, and that is a good description of it.
As a side note, the SP36 button LEDs are in low mode here. On high mode, its button is the brightest of the bunch. Easily bright enough to use as a firefly mode or a low moon. However, I wouldnāt recommend leaving it on the high mode, due to the amount of power it uses. Itās almost as much as moon mode. I measured the following:
At 4.2V:
Moon mode: 1.96 mA
Button LED (high): 1.54 mA
Button LED (low): 0.08 mA
At 3.3V:
Moon mode: 1.32 mA
Button LED (high): 0.78 mA
Button LED (low): 0.05 mA
FWIW, the estimated runtimes with 3x3500 mAh cells are:
Moon mode: ~9 months
Button LED (high): ~13 months
Button LED (blinking): ~8 years
Button LED (low): ~18 years
So Iām setting the default to low during āoffā mode, and blinking during lockout.
With the driver loose like this, something got shorted while putting the battery tube onā¦ and it turned the red LED+ wire into ash. But that was the only part damaged, so it was easy to fix.
So, although it would be nice if the driver was screwed in like the Q8ā¦ without that, the driver glue is kind of necessary. Hopefully the next version can replace the glue with screws. Until then though, donāt expect to do any modding without doing like djozz did earlier in this thread. It looked like quite a bit of effort to get the driver out.
Thanks ToyKeeper for the work on the SP36. I think the SP36 is worth developing well because Sofirn already made a good start with designing it as it is, it has a nice size-performance ratio (better than the Q8 for me) and personally I like reflector beams better than TIR beams (and reflectors look a lot better too ) . They could have made the design a bit more exciting (the ROT66 has that extra glamour) but because of the plain design it feels more like a working light to me, while Iām afraid to use my ROT66 in fear of scratching it :innocent: .
Now that I know how to do it, it is pretty straightforward, the main obstacle is that you need a pretty powerful solder iron (mine is 80W with a quality chisel-style tip) to attach the handle to the brass ring. My driver btw clamps well in place when pushed back into the head.
Btw, with a good mains USB charger plus short USB->USB-C cable (both Blitzwolf), the charging speed that I got was 1.95A, so the 2A charging that I was told by Barry it should have is met if the conditions are good.
I let the SP36 charger top off some cells. It runs at about 0.7 A when plugged into my USB hub, but thatās not surprising. Just slow, when divided between three cells. Anyway, it terminated right at 4.20V. I donāt know if this is normal or if I was just lucky to get one with an accurate chip, but itās nice that it did exactly what itās supposed to.
Samsung 25S will deliver the most power for the longest time, at a better price typically. There is a new 24S that might best it if it can be found. Either of these will outperform the Sony VTC5A which itself outperforms the 30Q or VTC6.
It seems that there were just two samples made of the SP36 and I have one and ToyKeeper has the other. So Sofirn had no SP36 samples themselves to make pictures of for use on the website. By request of Barry I made new pictures of my SP36 sample this morning and sent them to him, but it appears that they already made the page with my old BLF pics.
I was a bit upset when, the day after posting my D4S review, I found Neal using the same pictures to sell it.
Have been considering some sort of watermark system ever since then, if I can find an easy way to do it which doesnāt require manually editing each image and doesnāt significantly reduce image quality. Will probably just need an ImageMagick script of some sort, but I havenāt done it yet.