Looking forward to this. I’ve had to crank the temperature limit up quite high to stop the stock driver from thermal throttling simply due to the driver heating up, rather than the light itself becoming hot. At stock thermal config at top-ramp I was getting almost instant throttling even though the light itself was cold (yes the sensor is calibrated).
With low Vf emitters and a charged high current cell; by my maths there’s about 3W being dissipated in the driver itself at top-ramp, so it gets mighty toasty fast.
The problem with having the temp limit at 70 degrees though is that eventually the light becomes too hot to hold.
Hopefully this driver dissipating <1W in the same circumstances, I can use a sensible thermal limit.
I hope everyone has been well. Here are some new developments for the lume1 project!
Lume1 Driver on Nealsgadgets?
As you may know, I was planning a small 50-100x group order of drivers due to the large interest in the Lume1 driver.
Because of this group order and the interest garnered, Neal from Nealsgadgets has reached out to me and he will be picking up this driver for production. This will likely result in much better prices than this small group order, and also with better shipping / logistics! Discussions and planning are currently underway, so please hang tight. The idea is to be able to purchase these drivers directly from www.nealsgadgets.com, and perhaps also as an option in future flashlights. This also means that the group order will no longer proceed forward, since you should be able to purchase them for better prices too!
The details are still being finalized, and the timeline and other logistics are still being worked out. I will post more detailed updates when they come.
Lume1 Driver Datasheet (Preliminary)
In addition, I managed to get around to document the driver in this datasheet:
I assume so, and that’s what I was thinking of doing for some other lights, the via for the switch ring seems to start here :
Probably a bit small for soldering a wire there, but there is a tiny bit of space up north without components (blue) to make yourself another via.
Or even better loneoceans could add a solder pad for the switch, unused for the FW3X, but would allow to use the driver for other lights where it fits.
It is possible to use this driver in flashlights with 22mm diameter driver cavities - just make sure that the shelf and retaining ring fit the driver and don't hit other components. Afterall, the Lume1-fw3x was designed specifically for FW3X flashlights only. thefreeman is right that you can solder the e-switch wire where the red arrow is drawn in the diagram. Perhaps more designs will be made for other flashlights in the future.
As for springs, yes, since Neal / Lumintop is fabricating the drivers, they will be able to use the same spring as the stock FW3A driver. It'll be nice if they sold the springs separately too, since I think they're the perfect size and profile for many small drivers. Individual drivers sold by Neal will of course come with a spring
Finally for AUX LED boards, I'm working with Neal to find the best way to get these produced. Most likely it'll be the RGB AUX boards - for the tri-LED boards, different LEDs will require different series resistors, and I'm not sure what the best solution is to sell that since there can be so many variants.
If I may I have one minor comment about your design. I’ve noticed that your pads for pogo key have switched places for mosi and miso against already standardized pattern. Look like you can easy fix this to be in line with other drivers on the market.
I would love to see another version of those efficiency plots done with a triple Nichia/Samsung/Luxeon/SST20. As you know, the single XM-L2 is the worst case for buck-mode efficiency and also and unlikely case for the application. Boost-mode might reach the full 3A @2.7V too?
You know… I just now noticed that there isn’t a flat notch on this driver to match the FWs driver cavity/ledge. Am I missing something here? Is it not needed? Or perhaps is this a feature that’s not on the copper model that caused a minor oversight in the pcb design?