Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

Buy Site Supporter Role (remove some ads) | LPF Donations

Links below open in new window

FrozenGate by Avery

*PICS ADDED!!* Sipik 98 4 Amp Flashlight Mod!

Joined
Jan 29, 2012
Messages
3,163
Points
113
Just thought Id give a quick run down and mini tutorial of my attempt to mod my Sipik 98...:D

Well I started with an 18 mm thick copper rod drilled a couple of holes, cut it to length and inserted, or rather hammered ;), it into the pill. Had to widen and drill out the pill a little first however. Be careful drilling copper; its quite soft and sticky and I ended up breaking 2 drill bits...I found the best way, was to go very slowly and clean the swarf from the bit every time it appeared.

Re-flowing

Re-flowing is simply adding solder paste between two components to be soldered and heating. A thin layer of solder paste was applied to the top of the copper rod and a copper star (from BLF) placed on top. The rod was heated from the base with a heat gun, until solder paste started to bubble and change from milky to clear. I actually let it heat a couple of seconds past this stage and moved the copper bar to a block of Aluminium to cool.

I used solder because it is actually much better than any heat sink compound or grease at transferring the heat...(Solder is around 70 Wm-1K-1, where as most heat sink compounds are around 3 Wm-1K-1!!!);)

The paste was applied to the contacts of the sink pad with a pin and the XM-L2 T6 chip was carefully aligned on the star and the rod + star again heated from underneath. You dont have to worry too much about alignment as the the solder paste sucks the emitter down onto the board when it starts to bubble.

Unfortunately, I put it all together before finding out that the LED wasn't working...So I had to reheat the whole assembly and carefully remove the emitter from the star. I say carefully, but I actually dropped it after I got it off the star and spent 1/2 hour looking for it on the floor. Id actually given up on that emitter and decided to sweep up and low and behold I found it, along with a lot of dust, turnings and crud... :tinfoil:

So re-flow attempt 2! After cleaning up the emitter, I added a load of flux and a tiny bit more paste to the sink pad contacts and tried once more..I tested the emitter as soon as it had cooled and this time it had worked!

Assembly

I cut a piece of circuit board and glued it to the base of the rod. A spring was attached as a positive contact and the contact leads soldered. I picked up the negative path directly from the rod under the star. You will notice two holes drilled in the board instead of using the sides of the star for the wires. I had to do this as there was no room to drill holes at the side of the board... Anyway after a lot of trial and error (sanding, hammering, soldering and swearing) I got it all up and running! I was just tightening the pill and was one step away from finishing, when I slipped and my nail sliced through the dome...:wtf: Again and amazingly the emitter was still working, but the output had an ugly artifact.

Dedoming

The next day I decided to dedome it; my first attempt. A sharp razor blade was used to carefully cut away at the dome down to about 1 mm from the emitter surface. Here I was actually attempting a "partial dedome".;) I had read on BLF that one needs to polish it with a dremel after cutting and so tried that, but that didn't work so well as the dome actually became more opaque? :thinking: Deciding to test it anyway, I turned it on and the emitter began to smoke. Sh*t. I quickly turned it off and had a nice burnt spot in the centre of the dome. (need a pulling out hair emoticon here...:crackup:) I guess a small piece of dremel-wheel-cloth or dust got on there somehow. Well the only thing left was to completely dedome it, so I heated it to (Im guessing 80 to 100 deg C) and carefully picked at it with the razor blade until it all peeled off. That bit worked really well and surprisingly the bond wires were still intact! :D

Result

After the dedoming the colour shifted to a yellow/white from a blue/white and is actually quite appealing. The difference in colour is only really noticeable however, if you compare it side by side to another domed emitter, otherwise it actually looks white.

Its direct driven ie no driver, but due to the resistance of the current path, the diode is only pulling around 4 A!! Dont worry, that really is plenty and in the end I decided not to modify the switch or improve the contacts...Maybe I will later though! :eg: The light gets quite warm after a couple of mins so its not a practical work light, but works well as an intermittent search light or just a light to wow people as its ridiculously bright and great fun! :)

With the removal of the dome, the throw has also increased dramatically and I can now get a much smaller dot in throw mode!

All in all a fun if not chaotic build, but it worked out well in the end! :yh:


If anyone wants any more details on dedoming or re-flowing emitters just let me know! Hopefully I can also advise you on how not to do things!:crackup:

Pics below! :beer:
 
Last edited:





Re: Sipik 98 Mod!

Here's the completed light. The white PTFE disc that originally covered the star has been removed as it was not needed and actually reduced the output in flood!

7901-1305-sipikmod-002.jpg


This is without the aspheric lens and here you can see the copper sinkpad. :) The silver under the pad is solder as the pad was soldered to the copper rod.

7902-1305-sipikmod-003.jpg


This is the contact board, just crudely cut from a bit of circuit board and the solid copper rod can be seen beneath.

7903-1305-sipikmod-007.jpg


On...:p

7904-1305-sipikmod-009.jpg


Here I was trying to get an indoor shot of the beam, but it was a little difficult as the entire room lit up...:whistle:

7905-1305-sipik-mod-006.jpg


Similar problem here too! I was hoping it would only light up the other room...;)

7906-1305-sipik-mod-008.jpg


The last two pics are quick out door shots. In the first you can see how easily it lights up the top floors of the buildings at least 250 m away! Ill have to take some more pics, but I think it may throw (at a guess:thinking:) around 500 m?

7907-1305-sipik-mod-009.jpg


7908-1305-sipik-mod-010.jpg


Hope you liked the pics! :beer:
 
Last edited:
Re: Sipik 98 Mod!

Thanks for the write up!

I've heard that some dip the LED's in gasoline for a couple hours to soften the dome and then had some success popping it off easily.

Can't wait for pics! +1

but I actually dropped it after I got it off the star and spent 1/2 hour looking for it on the floor.

:crackup:
 
Last edited:
Re: *PICS ADDED!!* Sipik 98 Mod!

Cheers Ryansoh! Havent heard of the gasoline trick, interesting. Just dont turn it on before its dried! :tinfoil: ;)

Have you tried dedoming yet? I think the way I inadvertently ended up doing it worked quite well. Cutting down to 1 mm from the base meant that the remaining dome layer material was very thin and flexible. It literally required a couple of seconds from the heat gun to become soft and pliable and it just peeled off! I guess the thicker the dome the more rigid and harder to remove it would be? :beer:
 
Last edited:
Re: *PICS ADDED!!* Sipik 98 Mod!

Thanks for the pics! I really like this one:

7905-1305-sipik-mod-006.jpg


It's pretty neat to be able to see the beam in the room.

I've accidentally dedomed some XP-G's while installing them. They just slipped off.
I haven't dedomed XM-L's yet, but I'll try your heatgun method, as I don't think I can easily access gasoline. :D :beer:

Edit: Thanks for the rep! :D
 
Last edited:
Ok Ive edited the block of text above so its not so daunting! ;) Ive also divided it up into sections and modified the text so people can find info more easily; more specifically info on re-flowing and dedoming, that might be of interest. :beer:
 
Last edited:
If it is direct driven than I guess it pulls max amps out of the battery?
(does the led itself act as a resistor?) How many amps can 1 18650 put out anyways?

I might put a heatsink in mine because all the reviews on manafont say their light got so hot so fast, and melted all the contacts. I will prob leave the board so I can use low mode w/o burning it up..

I got two sipik sk98's :D (how long does manafont take?)
 
If it is direct driven than I guess it pulls max amps out of the battery?
(does the led itself act as a resistor?) How many amps can 1 18650 put out anyways?

I might put a heatsink in mine because all the reviews on manafont say their light got so hot so fast, and melted all the contacts. I will prob leave the board so I can use low mode w/o burning it up..

I got two sipik sk98's :D (how long does manafont take?)

Depends on the resistance of the circuit ie switch connections etc...I use either an AW or a panasonic 18650 both can easily put out over 6 A.

Before the mod I really didnt have any problems with over heating. Yes it got warm, but only after a few mins on high and it definitely didnt get hot enough to melt anything. One thing I would suggest however, is that you heat sink the star when it arrives. It was the first thing I did when I got it. If you take off the white plastic cover over the star you'll find that it doesnt have any heatsink compound (grease or thermal epoxy) between it and the pill. If you dont make a proper thermal connection then I suppose the star could get hot enough to melt the insulation around the wires to the LED.

Manafont took around 2 weeks to Europe.

Enjoy them theyre great little lights!!:D :beer:
 
this is just awesome! I had a ultrafire knock off that only pulled like 2.2A so I chaged it out with a new driver and it's now pulling 3.15A. I like this copper sink pad! might have to try this mod as well!
 
Cheers! Just to let you know depending on the batteries you use and the emitter type XML or XM-L2 you may be limited in current. Ryansoh3 just sent me a graph of the Vf for the XM-L2's, which are higher than their predecessors:

XML-2%2520vF.jpg


I decided what the heck and tried to drive the XM-L2 to the max by wiring the sink pad directly to a battery and all I could get was a tiny bit over 4 A...So I guess my batteries were sagging too much under the draw. I used a Panasonic NCR 18650A 3100 mAh (76 mOhm internal resistance) and an AW IMR 18650 2000 mAh (68 mOhm int resist), so they werent crappy batteries either.

One option might be to try running it with 2 x IMR 18350's and a resistor to limit the current to say 5 or maybe 5.5 A?..:thinking:

:beer:
 
Last edited:
Thanks, which bit are you referring to; the copper slug, the dedoming, reflow, loosing emitters on the floor(J/k lol), or the whole lot!;) My process evolved out of a series of mishaps :crackup:, but Id definitely be interested to see how you do it. If you like you could even post here so people have some good reference material all linked in one place! :beer:
 
Last edited:
ok so mod wasn't as in-depth as yours but has given me some ideas :D

Mine involves these parts:
https://www.fasttech.com/products/1206301
https://www.fasttech.com/products/1164600
https://www.fasttech.com/products/1127403
and one additional AMC chip.

don't mind the dust it's my EDC and I took the pictures really quick and didn't clean it up.
IMG_1375_zps35161448.jpg

Below I added a silver disk that they use in the fasttech C6 to isolate the reflector and the LED.
IMG_1376_zps0ae8fd0e.jpg

Below you can see where I placed the p60 Pill inside of the Driver cavity so I could fit my 17mm driver where the original 20mm driver was. This also add's heatsink to the overall pill. I've also added one AMC chip to the other side of the driver.
IMG_1377_zps86f9941d.jpg

My homemade Blue glow in the dark O-ring for an after glow effect.
IMG_1381_zps0f5ddca1.jpg

and a little picture of light output.
IMG_1382_zpscbf00ce5.jpg
 
@ Lazerbeak I just thought you had a similar process for one or a few of the steps and if so I was just hoping you'd share it. ;) Anyway the build was a series of disasters and Im really surprised the emitter survived the ordeal! I didnt mention it, but there is still a bit of drill bit stuck in the copper heat sink...:tinfoil: :crackup:

@Awillis great job there I especially like the silver disk and the GITD O-rings! Not actually sure if you can get a blue GITD tail cap cover, but did you change the green original to match?

Funnily enough I also had a go at the AMC driver mod and added 4 chips to the top side of a Kaidomain 3040 mA board. It took me about 3 hours to solder as it was so fiddly and difficult to get to the pins. Actually managed to get that light running at 4 amps too, with a 16 mm copper rod heat sink and with modes...I built it in the other light we discussed a while back. Seeing as Im limited to 4 A, I kind of wish Id done that driver mod to mine too, just so I would have modes... :can:;) Oh well next time...:beer:
 
Last edited:
Quick update.

I recharged all my 18650's to 4.2/4.21 V and tested the current at the tail cap. Here are the results:

1) Ultrafire 3600 mAh (fake I know)__2.35 A
2) Ultrafire 3600 mAh (fake I know)__2.8 A
3) NCR18650A 3100 Panasonic______3.8 A - Not too impressed with this one, internal resistance is 76 mOhm and its less than 1 year old...:(
4) NCR18650B 3400 Panasonic______4.2 A
5) NCR18650B 3400 Panasonic______4.25 A
and the winner!!

6) AW IMR 18650 2000 mAh________4.6 A! :)

So if run time is less important than power the AW wins hands down. :beer:
 
Last edited:


Back
Top