No design at all, they just made an empty flashlight body and i guess they bought that LEP module somewhere else.
But they were the first to do a really small LEP flashlight, and it works (although some better than others), and it turned out a better throwing flashlight than anything before in this size. Lumintop gets my credits for that.
And now a better one must be made.
Wise words djozz.
If that’s all it takes, who wants to start a BLF LEP and take on designing a driver? 
… and give it a sliding bezel to make it a zoomie! A true zoomie with a wide enough flood mode for close-in use.
A simple buck boost driver could be like this :
Monomode, capable of 2A output at 4Vout (likely too much for the module used in the Ant Man, here it’s set at 1A), no thermal regulation (no MCU), but it does have LVP thanks to the R1/R2 divider on the Enable pin. We could do without the 2.5V voltage reference by connecting R3 to VOUT, but the current will be quite approximate due to the Vf variation, the LDO used for the VREF is quite cheap anyway.
By some miracle this BB converter is still in stock at Mouser and is actually slightly better than the one used in the lume1 driver.
With this it should still have a soft start at startup, the issue with the constant current topology I’m using in my other drivers is that it has the tendency to spike at startup, which apparently isn’t good for laser diodes.
For a more powerfull driver for driving a LEP module like this:
I would do something different since commonly available BB converter are not powerfull enough for 3A@4V, I would use a boost converter with a constant voltage output >= Vf and use a constant current linear regulation after that, since the Vf will be high even in thermal step-down the efficiency should remain good. A boost converter with passthrough would be used so that it just outputs Vin when Vin>Vout (e.g. when the battery is full).
The soft start function would also work properly in this case.
Hey folks, I rarely find myself on BLF, so just saw this thread.
My copy of AntMan does indeed produce 205lm at turn on with H10 battery, and around 100kcd inside at 6m. I measured 1.45A with clamp meter.
I’ll happily answer any questions I can.
is the difference due to variations in test equipment,
instead of variations between the lights themselves?
Ali don’t think so because I am using Texas_Ace’s calibrated lumen tube, and my throw measurements seem to support the higher lumens.
For throw I was using the LX1330B meter that is well regarded around here. I’m not saying there isn’t something I haven’t noticed, but I did run the tests multiple times now (4 times at different times).
I think there is just a laser lottery going on, and apparently I won this time. (In the past I always get under performing lasers, Welltool W4 that under performed, L2K that under performed), so good for me! If you look at the numbers that chibiM and djozz are posting, even they are considerably different.
My LX1330B luxmeter does not give greatly different numbers for LEP light sources than my expensive ones, I have not written it down but what I remember it is just a few percent.
Congratulations Cheule with the best Ant Man sofar reported! 
So id30209 got 0.4A, Djozz 0.9A and Cheule 1.45A, that’s some pretty wild forward voltage variation with these laser diodes.
Agreed. I wouldn’t believe it if it weren’t for the fact that I’ve seen this same exact variance with hand held green lasers for over two decades. At my astronomy shop we would order a case of 50 of them, and about 10 would be way brighter than the rest. My manager would have me sort them and charge different prices.
There may be something else going on here, but my money is just of diode variance.
The different outputs people are getting might be due to the choice of battery.
Note that Cheule said he is using an H10 battery.
The Vapcell H10 is the highest-current 14500 around. On a full charge, in my FWAA it produces noticeably brighter output than any other 14500 cell I’ve tried.
I used a H10 battery too, not sure about ChibiM.
LEP lights use a laser type that runs at a voltage (around 4V, significantly higher than most of modern white leds) that is on the edge of what li-ions will deliver even when full, so a small Vf difference already results in a large current/power difference. It also means that a more capable 14500 will make a relatively large difference. I have not done a proper measurement but I think that even my two H10 batteries give a different output in the Ant Man.
Maybe not. The delta is just about 1A, and if we simplify it down, an H10 has an IR of 70mOhm, so that’s only ~70mV variation. I’m dying for someone to get this thing onto a power supply
Does the Ant Man user a liner driver? No boost circuit?
It seems there is nothing much going on except maybe a reverse polarity diode. Definitely no switching circuits
Could you power it with 2 14250 cells in series to get the voltage high enough for more current or would that kill it?
From the look of id30209’s pictures, there is no driver, just a diode that looks wired in parallel, probably a TVS diode, which will likely die from overcurrent and then RIP the laser diode.
I actually went and quickly traced the single mode buck-boost driver mentionned before :
With guesstimated dimensions, ID30209 should be able to give me the measurements and I’ll order it with my next PCB order.
Cool stuff thefreeman! That driver could unleash the real potential of this LEP light.