honestly, me too! Now that I have the beam centered and dialed in this is probably my favorite thrower. For size, ease-of-use, throw, and heat management plus battery life, I don't think this can be beat. price to, because I want to put one in a K-1 but that is twice as much money. I'll be interested to see if Astrolux has the beam centered well or if they use that stock centering ring which will not work properly
Here you go:
1. My first Emisar D4 lighted metal switch mod
2. Updated Emisar lighted metal switch with better lighting
Here’s a post that provides extra detail on how to make the metal button. This one is not lighted, but the process is the same. For the lighted version use a slightly smaller button so the silicone has a gap at the edge. And it is necessary to add a raised switchboot retaining ring since the stock ring is too flat:
3. Metal switch mod with no light, but more detailed instructions
Note that if you have the raised-ring Emisar D4v2 you may be out of luck. You can shave the boot down and glue a metal disk on top as-is, but at least on mine, I found that retaining ring impossible to remove.
Thanks for the links Firelight2 
I am not clear about the firmware of this driver. There are two, one cannot be turned off from the rear switch and the other cannot be turned on from the rear switch? isn’t there one that can be turned on and off from the back switch?
https://www.mtnelectronics.com
One basically ignores the rear switch. When the rear switch is on, the light has power. You can turn it off and on, whatever. Turn the rear switch off and power is physically disconnected. Light goes off no matter what.
The “turbo on” option just tells the driver “when you receive power the first time, go straight to turbo”. Then you can turn the light “off” with the e-switch and use it like normal. But if you turn the tail off, the light is still off no matter what else is going on, because it still disrupts power physically.
good explanation thank you very much, I think I would choose the “turbo-on”
the standard firmware is like that of fx30 convoy? if you turn off and on again the rear switch turns on in the last mode where did you leave it?
Both get turned off from the rear switch - turning off the rear switch cuts power completely off, so no power to the driver, no power to the LED.
The only difference between the two options is how the light behaves in turning it on from the rear switch - as he says:
understood, thank you very much! yes I will definitely ask for “turbo-on”
I was having problems with my quad E17A S8 with a Convoy SST40 driver. Every once and awhile the driver would reset to mode group 1 and memory mode.
I was thinking the driver was the issue and was planning on replacing it, though not looking forward to taking the head apart. Then I thought to swap the tail out with a non-lighted switch and sure enough that resolved the issue.
The driver says it works with lighted switches and all my others (10 in total) work perfectly fine with them. But it appears that since this was a green lighted switch, the 0.03 mA current was too low.
So I dropped the resistors a level which raised the current to 0.10 mA. And to reduce the brightness, I put some blue permanent marker on the 0805 SMD LEDs. It worked!
Swapped 219B sw30 into my 2019 RRT-01. Super easy swap, and I already had a good Kaidomain reflector and gasket prepared in advance thanks to some helpful threads around here. No pictures; the head came right open since I’d defeated the thread locker in a previous swap. I ordered some 12mm MCPCBs at some point (they came with XP-G2s installed on them) for both of these mods. The reflows were routine.
My VR 10 Ti+ was more challenging. To begin with, it required significantly more disassembly - not just the bezel, but the next piece down had to come off. I referenced this thread (from the other forum) for disassembly. I was worried about the reflector. The Kaidomain model for the RRT01 does not fit. There’s possibly an alternate but it would require heavy physical modification to even try it, and I don’t have it on hand. However, I test-fit the reflector and the focus looks okay - that is, I don’t think it can be worse than the stock focus with the XM-L2 (horrible, ringy beam, and like 6000K if I had to guess). I had to fashion a centering gasket adapter - I took a simple ring-style gasket and filed the outside slightly and reamed the inside significantly such that it fit inside the bottom end of the existing gasket, and sat around the 219B sw45k. Next was the grease - since I was at the point of disassembly where I could simply lift the control ring off the head, I stripped the existing grease out. The light was beginning to bind if it was left for more than a few days without use - I had to be careful to put it away at a “middle” setting so I could twist the ring without unscrewing the head instead. I replaced it with a thicker synthetic lube which I hope holds up - but if it doesn’t, I’ll order some of that Nyogel that JonSlider seems to like so much. The beam looked good in the reflector when I was done - for about 5 seconds before I let the smoke out . So I guess that’s my next pressing project… driver repair. Worst case I’ll have to scrounge up a used aluminum model (V10R or V11R or something) for a donor driver.
Today I made an EDC to fit the new F14 18350 battery from Vapcell.
Since the F14 retains its great capacity up to 3A but not higher (1300mAh@3A, HKJ’s test here), a 8x7135 current regulated driver is needed, while all my current 18350 EDC’s have direct drivers (mostly BLF-A6 versions). So a new mod was required.
I used a Jaxman E2L host, the latest 8x7135 driver from Simon (link), three oldschool Nichia 219B V1 SW40 R9080 D200 leds from Clemence, a ring-shaped tail-PCP from Oshpark slightly sanded smaller to fit the tail cavity (a pilotdog68 design, this one) with 6 cyan 485nm 0603 leds (each led with a 1.5 kOhm resistor in series).
The driver is set to mode group 1.2 : 0.1–1–10–35–100%, no memory.
The output on 100% (3.04 A) is 560 lumen, the tint is 3900K and very close to the BBL, visually it looks very neutral with no hints of green or yellow. CRI is as expected above 95 on all brightness settings, with R9 between 80 and 90. The tail leds are not ultra-bright but well visible in dim surroundings and they draw 0.18mA current which should be good for over 300 days on a battery charge. The emitted wavelength is exactly 500nm (cyan leds emit a higher than listed wavelength when at very low currents).
One remark about using a lighted switch in combination with this driver: Simon mentions on his website that this driver supports a lighted switch, without having to add a bleeder resistor. Although this is correct, and the user interface is stable with tail leds functioning, the driver struggles to provide enough voltage to light up blue or even green tail leds, red is fine. You can compensate some of this with lower than usual value resistors in the tail but blue never becomes bright. I think that some battery voltage is eaten on the driver before feeding the tail leds.
Two mods this last week:
Swapped driver in early-version Eagletac D25A Ti
Modified clip for Acebeam TK17 Ti
Exactly where they came from and they are good ones.
I’ve been tempted to make an E2L-based 18350 EDC myself, just to use a smart-tail board I have from gchart.
Wow, those are really good looking hosts! Look to be 26650, so one could use 18650, 21700 and 26650’s.
Today I swapped drivers between a Helius Beta VI penlight and a Lumintop IYP 365, changed out the XP-G2 in my Lumapower Incendio V3 for a Nichia 219B, and turned a cheap 3xaaa light into an 18650 triple. Can’t wait to get my parts in I ordered from various websites for other plans in the works.
That is very cool. I’m a flatworm with computers so flashing chips is out of the question, which I am ok with except for gchart’s smart tails, I’d love to be able to use those.
Quick mod,45°pebbled lens (right) on C01R
Bought here
http://www.aliexpress.com/item/32989679045.html
I just rebuilt two Polarion packs and one of them is fine (14.4V), but the other one shows this exact issue (only 5.55V between P+ and P-). It’s probably noteworthy that the PCBs were different and the one that works looks older…
The voltage between V+ and V- is 14.4V, so the pack is correctly built. Balancing wires are showing correct 3.6, 7.2, and 10.8V, respectively.
I’m afraid they have built in a “safety” program that kills the output after seeing a severe undervoltage (or after some amount of charge cycles? they repeatedly mention that the battery is good for 400 cycles only…).
Does anyone know how to reset that?