The difference from before and after this mod, is quite remarkable. These are two pics that don’t really do justice,but you can see the tighter beam profile. This is with an R100 resistor mod with potted driver and dedomed emitter. A side effect that is welcome but I don’t understand is the low mode is dimmer after the mod. Low is way too high so this is nice but I don’t know why.
These pics are taken with my iPhone pointed at the ceiling on the floor with the flashlight. I have a camera but I don’t know what settings to use for these kinds of pictures anyway. If someone can point me to a tutorial I can try to get better pics.
This one is al the way focused: to the eye it is night and day brighter. Bigger difference than can be seen in these photos.
This one is zoomed out in low mode till they appeared to be the same size. Maybe the dedomed version looks dimmer because it was zoomed out farther to match the size of the stock beam size?
Has anyone tried the supfire a5? I ordered a cheap one on eBay and it’s not the same quality as the a2. I’m wondering if these eBay offers are for fakes. Anyone order from a reputable seller?
My Supfire A2 is modded now almost as far as I can get it with the stock driver:
-led=dedomed XP-G2 S3 3D (good old type) on 16mm Noctigon
-blackened everything around the led with flat heater paint
-two battery springs replaced by carobronze springs (Fasttech), both soldered into the brass cup, the driver spring soldered to the driver as well.
-transferred the led+ wire to the other side of the R500 resistors.
-did not replace or shorten the ledwires, may do that at some point but I read above that the driver has its limits.
On a fresh 30Q (30 seconds) I got 3.25A,680lumen,over40kcd throw, so the led is well driven.
If only Supfire had done the PWM and mode spacing right….
Ok. Good news. The charger is working correctly. I temeasured using my multimeter and got a reading of .65amps of charge current. My benchtop meter must be incorrect. I just installed it all and started using it so there may need to be some calibration to get it to read properly.
Here you can see the unit is charging and able to light up the led at the same time. Note the power reading of 9watts. This is incorrect. It charges at a bit over 2watts or .65amps.
The red indicator leds flash until the battery is fully charged.
It draws 50ua of power in standby mode. Not bad!
I mixed up some thermal adhesive to stabilize the piggybacked driver and add a small thermal path to the host.
I now have a supfire a2 zoomie with charging running Narsil and pulling 4.3amps from the dedomed origional led:) There was some uneccessary complicity to this build but I learned much. Most of all it looks as if our driver boards can run parallel to an onboard charging circuit!
Great job, I’m jealous of your A2 performance, I like that flashlight a lot, apart from the stock driver! Does this free me from measuring the A2’s charge current, as you already found it to be 0.65A ?
I think that you are right about the desirability of a BLF driver with inbuilt charging, but using it will always require a suitable host with microUSB opening, or else you will have to create one. It will never be an easy mod and at this moment only few lights are suitable.
I am still searching for a replacement for the Aleto N8 zoomie. It’s the shortest 18650 zoomie I could find and had decent build quality including heatsink fins. Unfortunately, I only found one supplier of it ever (on Ebay) and haven’t seen it for sale in well over a year. Wish I’d bought more of them at the time as it made a great mod host.
Supfire A2 looks beautiful and well-made, but it is far too long. I’m not sure how much length I could shave off by modding it and still have a decent looking light.
I did a simple check of the Supfire A2’s internal charging of a Samsung 30Q battery. I started with a fairly empty battery, at 3.32V, and just recorded the current delivered by the USB source and every now and then I took the battery out, measured the resting voltage, put it in again and resumed charging (this could be the cause of the not so perfect curve). At the end, with the battery at 4.19V resting voltage, the current instantly dropped from 80mA to zero, that looks like the charger actively stopped charging which is a good sign I checked: upon disconnecting the USB and reconnecting it, the flashlight started charging again at 80mA and after a few minutes dropped to zero again.
I did not check if it is the 100% correct charging curve but it looks not bad to me.
Ever since djozz mentioned bypassing the resistors without trouble, that is what I have been doing with good results. Actually I’ve just been removing the r500s all together to keep as spare parts.
The only concern I’ve had after increasing the current is the size of the trace going from the fet to led. It’s quite small, so I’ve been jumping over from fet out to led with a small 26awg wire. You can see a photo earlier in the thread. However, this may be totally unnessisary as djozz hasn’t had problems. It just gives me peace of mind.
My mod thusfar was always with dedomed XP-G2, the right one, together with ‘streamlining’ the current path, gave just over 50 kcd.
(here I wrote about the mod)
Amazing mod.
