What did you mod today?

A teaser for now until I can write a proper post. I managed to short 2 BeCu springs so I dont have enough energy after “repairs” to go into details tonight!

Lucky dip… :question:

Looks interesting contactcr

This is a special one, nearly all of the parts were made possible through other forum members in one way or another.

The light that BLF built:

Host Sofirn C8F 21700 (host version avail. thanks to BLF)
Spacer Prototype AL spacer from kiriba-ru
MCPCB LED4Power 3535 ANNA6
Driver Lexel’s 21mm C8F version TA FET+N+1
includes more clearance for the retaining ring
Optic Ledil ANNA-40-7-S
Springs BlueSword’s BeCu spring
LEDs 3x LH351D 4000K 90CRI from AEDe group buy
3x LH351D 5000K 90CRI from DigiKey

Some quick measurements:

18650 VTC5D
21.x amps
5,693lm at 0s

21700 VTC6A
5,851lm at 0s

21700 30T
6,479lm at 0s
22.5 amps at ~30s

Pictures:

Sloppy re-flow I had to do twice cause I used way too much solder paste. It’s hard to “tap” excess solder from 6 LEDs, 2 pads, 1 jumper in a reasonable amount of time. Note I jumpered the FET and left off the LD-x4 specific parts (temp sensor, FET, JST connector)

The C8F usually has a small tapped screw hole in the middle which holds the reflector and MCPCB to shelf from underneath. I drilled it out with a 15/64 bit, I could have used smaller but I left it big in case I want to add larger LED wires or use the signal wires if LED4Power releases and e-switch driver some day. In this picture I have a small blue spring but I ended up turning it upside down cause I think it’s too tight to prevent shorts this way.

Don’t forget your TWO layers of thermal paste. You could use thermal adhesive but I have no balls. Also, the spacer has some play so you would want the whole stack to be centered first if you went this route. You could also hold it like this if you have trouble soldering the LED wires but I didn’t bother. Those coated pads + a big honkin chisel tip will do the job.

Since i’m using a 7 LED optic on this 6 LED board the wire placement is something you need to fiddle with. The unused center TIR caresses your wires lovingly.

In order to seat the optic all the way down the edges press right up against the head of the light. If I remove the bezel again I will probably have to wedge the TIR out with a screwdriver. If that’s the case I think I will sand the TIR slightly to give it more room. It’s terrifying screwing on the bezel snug. Everything rotates some and I worry I will either crush something or bunch up the LED wires. I did it anyway, YOLO. It survived.

I did some tint mixing since I didn’t have 6 of the LEDs I wanted to use. The beam is nice, there is a defined and very even hotspot. You would never know the middle LED was not used. I suspected as much since Ledil makes this same optic with 3~~, 4~~, 5~~, 6~~ and 7-lens versions.

I had zero issues testing with various 18650’s but when I moved to 21700’s I ended up shorting something and melting a blue spring in seconds. This is when I put a new one on but this time upside down. I think my tail spring is too long and stiff also so I will play with cell spacing and refine this.

Heat doesn’t seem to be a big issue and not using DTP for the MCPCB seems fine as well. Further testing needed here as I only tested turbo in ~1m increments. The light never got too hot to hold even after several 30s-1m turbo runs with a 30T. There is a lot of mass in this thing after all.

When I first saw that picture, I couldn’t tell what it was. I was thinking maybe it was a bowl of greens and he’d taken a picture of his lunch! :wink: Especially since the previously posted picture was of some french fries.

Box of washers I ordered arrived much faster than anticipated. Guess it’s time to start modding. :disappointed:

Goal is to:

  • file down the bottom of the washers so the bases are smaller and the washers don’t stick up so much,
  • expand the interior hole by inserting long-nosed pliers and then expanding the jaws, and
  • glue them onto side-switch flashlights around the button with Arctic Alumina to help reduce the chance of accidental pocket activation.

I ordered #14 nickel-plated brass finishing washers. Hopefully that’s what’s in the box I got (box just says “nickel”). I definitely want nickel-plated brass as the brass is easy to file and work with, while the nickel plating is durable and looks great.

Lights I have in mind for this mod are the Emisar D4 and Fireflies E07. I’ve previously done this mod using #10 nickel plated brass finishing washers on a DQG 18650 Tiny III and got fairly good results. Looks classy and worked well.

Biggest disadvantage of this mod is that if the head of the light heats up around the switch, the metal finishing washer filled with thermal adhesive is going to heat up too. This can make for ouchy on your thumb when trying to turn off the light if the ring gets too hot. A plastic finishing washer would probably be more functional but would look much, MUCH worse.

Really nice C8F mod! The ledil anna looks like a nice optic with good frontal coverage. Better than the optic used in the fireflies E07.

That’ll just cause plastic deformation and actually thicken the rest of the metal. When I make my own washers I make / enlarge the center hole with a step bit [handheld].
This was not a washer, it started off as a square I cut from a sheet of Cu. I drilled a small hole with a drill then mounted the step bit in the vise and went to town rotating the material around the stationary bit.

Yeah, lets see those D4 switch mods :smiley:
I am waiting on a replacement. I trashed the membrane.

It works.

As you expand the jaws, the washer bends into a bean shape. But then go around to the short part of the bean and expand again. Continue going around in a circle until the washer is circular. Yes, you get plastic deformation, but when you go back around in a circle the inner edge of the washer fails to fully return to its original position and stays closer to 90 degrees vertical (compared to 45 degrees before modding).

This technique easily opens the center hole up a couple mm when done with a 10mm #10 brass finishing washer. At least that was my experience when I used this same technique on two previous DQG Tiny IIIs.

Not sure drilling is a good idea. These washers are very thin. Drilling is likely to go through the side of the washer. Also, even if it worked, drilling would remove the beautiful nickel finish.

Fairly certain Turning Blue Chips is about to turn a pile of shavings into a bar that he will then make a light from. Converting “scrap” into a nice final product. :wink:

Are you sure that isn’t a bowl of veggies he’s about to convert into calories and biological waste products? :smiling_imp:

Oh, it’s aluminum shavings. I have too big a pile of those myself to not recognize them… even under the odd color liquid.

Did me a pintrest project with my son today. Accent lighting for his room.

Edit it didn’t work, here’s a link (Idk if you have to be logged in or just open it in a new tab or what, never tried to do a video on imgur before)
https://i.imgur.com/ZhGemH2.gifv

Not sure of the imgur video is gonna embed so here’s a photo jic…

“pencil mod” on the suspect capacitors on two Astrolux A01s — both now behave properly, starting on low, eliminating the next mode sometimes flakiness.

Super cool project CK :+1:

Finally got my Skilhunt H03 modded. Used a LH351D and Lexel’s newest 21mm driver, flashed Anduril and bypassed that weird tail (mostly). One thing I think people don’t mention for this mod is to make sure the optic presses against the PCB firmly if you switch to a 3535 LED. I sanded down an XP butterfly spacer flat and thin and now it presses against that and holds everything with good pressure.

I also had a bad re-flow the first try thanks to the damn BLF A6 star I tried to re-use. Gotta love seeing that blue light.

My man cave needs one as well Ck. :+1:

My man mousehole needs one too :+1:

Built my second S2+ triple yesterday. Headline specs:

Desert Tan S2+ Host
Triple Luxeon V 4000K on Led4Power MCPCB with MOSFET and NTC
LED4Power B4 driver
Carclo 10507 TIR optic
Kiriba-ru copper spacer
Both springs bypassed with 20AWG silicone wire
LG HG2 battery

My Nichia 219C triple (in the copper host in the photo) is the first torch I built from scratch, i.e. with a host and collecting all the parts together. I love it, and it’s a ridiculous little hot rod. But I felt I could go even further, so after some browsing of led4power.com’s site and these forums I saw that the Luxeon V seems like a very efficient and high-power emitter. I’m a huge fan of the S2+ so I decided my next build would be a Luxeon V triple.

The results are quite amazing. It is both brighter and throwier than my 219C triple, as you can see in the photos (same camera settings for the outdoor comparison shots, including the hilariously grainy high ISO setting that I didn’t notice on Lightroom Mobile!). I measured 19A with my clamp meter which is roughly the same as my 219C triple, but according to the tests posted here, a Luxeon V at 6 amps is 2000 lumens! I’m not sure if that really equates to 6K of real-life light output with three of them, but it is very impressive nonetheless.

I used the L4P B4 driver and their triple Luxeon V MCPCB with external MOSFET and temp sensor. This means 5 wires to the MCPCB rather than just 2 which made for some fiddly soldering, but it allows for much quicker, more granular temperature control, and more efficient heat dissipation thanks to the external FET. It’s quite a sophisticated little build and it works really well. You can set the temperature stepdown threshold wherever you please with the L4P driver, so I increased it a bit since the stock 65C was quite tame :slight_smile:

All in all, I’m extremely impressed with this build and the quality of L4P’s hardware. But… I’m wondering if I can go even better and brighter with the S2+ host…. What say you all?

I’m jealous of your soldering skills hcanning, those joints look immaculate! :open_mouth: