Ordered 3 drivers, so hopefully I’ll be able to get at least one running. Thanks everyone for all the tips and advice. If anyone wanted to do a video of the driver install, that would also be awesome. 
lume1 live stream this weekend! come create bonds while I create bridges.
smash the like and subscribe.
(in all serious I considered doing a video but I feel like I would have to edit it down a lot)
Dangit… now I think I want to swap the LEDs in my LM10 with Lume1 driver. I’m not looking forward to soldering all those tiny wires again. 
Recently I pulled out my bare aluminum FW3A. It has 4000K SST20 in it of a superior tint bin to the 4000K in my LM10. No green at all. Noticeably better than the tint bin in my LM10. Also that bare aluminum FW3A isn’t practical. It gets burning hot too fast. Now I want to swap the LEDs in both these lights.
Grrrrr…… the effort required to try to get a “perfect” light.
Did you try swapping just the lenses first? If it’s that noticeable it’s probably not just the lens but worth a shot.
Both lights are using 10507 optics with Sapphire lenses that I swapped in afterwards.
Actually the LM10 doesn’t look that bad. Tint still looks way less green than that in my Emisar D4 with SST-20.
But when put side-by-side, the tint of the SST-20 in my bare FW3A is noticeably better than that in all my other SST-20 lights. Even at low-power, it’s no greener than the tint in my high-CRI 4000K Nichia 219C light.
PM me values you need. I was too cheap to pay for a “book” so it might take a full day to find it. LOL
One of my FW3A lights is the first gen with a flat spot on the driver and non-symmetrical solder pads on the mcpcb. I was not thinking and decided to cut the aux board to make it fit the mcpcb and ended up cutting the trace for the red channel. Here is my repair:
Is there a chance that this driver will be available for purchase already installed in FW3A?
I don’t have great equipment to measure currents this small (or variable), but I was fooling around today and got idle currents of 62μA with the red aux leds on low, and ~35μA with them off, with peaks to around 250 when the MCU wakes up (which I wouldn’t really trust given the sample rate of the meter).
The temptation to bodge a current probe with a resistor and a differential probe is… large.
You saw the datasheet right? It may answer some of your curiosities
Yup! I’m just having fun (and confirming I can leave the aux LEDs on all the time).
FWIW, code for the Lume1 should be merged into fsm soon. I have it already merged in a dev branch, but I don’t have any hardware to test it on so I don’t know if it actually works. So, once it’s tested and confirmed to work, that branch can land. And then right after that, it should be easy to import into the anduril2 branch too.
The changes aren’t deep, but they touch on some parts which were completely rewritten since it was branched, so merging it has been a little more tricky than it would otherwise be. Not entirely sure if the external temperature sensor still works or if I broke it while adapting changes.
Will be available thoese kind of drivers for Lumintop EDC 18?
Just put the Lume1 in my FW1A XPL Hi (just driver, no aux board of course), and looking forward to seeing how the efficiency is. The emitter seems to be withstanding turbo even with 30Q. Don’t know if should be expected?
This is a practice run for using the driver for a W1 with a SMO reflector in the FW1A someone soon.
XP-L HI has a high forward voltage so even on direct drive (FET) it will withstand any battery basically (at 6A it’s 4.02V)
W1 is low fwd voltage and will surely be over driven/die on direct drive (at 6A only 3.5V)
If you value efficiency at all you will want to stick a large capacity battery in and disable FET for 3A max.
You’d be hard pressed to pop an xpl with a single cell, direct drive in a light. Vf at failure is usually >>4.1V
The W1 is easily overwhelmed though. Even an old low drain cell partially discharged can push it to the limit.
I’m not sure there’s a surefire way to get the w1 driven properly at ~5amps with the Lume1. Firmware might keep it from burning up but you probably won’t get maximum throw performance from it on turbo.
Honestly though, I’d recommend one of two things: switch to a W2 emitter or be happy with the 3A max of the regulated levels. The visual difference in throw between 3A and 4.5A in the W1 really, truly, isn’t worth the heat, especially in an FW1A. In fact, you might not be able to perceive it in the beam, which will get very indigo when pushed too far…
Just my 2c
-Cross post!
contractor - thanks for clarifying. And yes, the 30Q was the first thing I grabbed from my desk. Am now charging a Sanyo NCR18650GA to start proper use of the light.
Could you direct me towards info about disabling the FET please?
JaredM - I was contemplating the workaround to avoid frying the W1 and was going to settle on just avoiding Turbo use. But perhaps can disable the FET as mentioned. Would the W2 not also be vulnerable to being fried with FET (although less so than W1)?
Although I will get a lot enjoyment from the 3 Lume1’s that I have bought, I do wonder if the next step could be a 5A constant current driver for the FWxx. Perhaps there’s a reason loneoceans settled at 3A cc.
That would be a custom build from the source posted by ToyKeeper a few posts before this one.
However, maybe disabling the FET could be a good use of one of these unused external jumpers?
Heat management, space limitations, and cost are likely the answer… the first being most important imo. There’s little point to create a driver that can sustain 5A in a light that can only sustain 2-3. I don’t think white flats in an FW1A was considered a popular configuration either. Of course loneoceans can fill in the blanks there…
Two ways of disabling the FET are via firmware or physical desoldering. On this driver, is only recommended the first option.
The W2.2 is available from Simon/convoy/AE and likely can take turbo from a high capacity cell and thin/longer emitter wires. It is just as sensitive, but the higher current allows the resistance in the light/connections to burn off more voltage… with an older high capacity cell used for a minute at 3A prior, I bet it’s just about right.
Great idea