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

Extracting Diodes From Nichia diode bank

I give it another year or two until the Chinese have for sale on flebay any color diode in excess of anything that N company is making, China has the biggest Osram factory in the world and they are already investing in a better substrate, the copper backed stressed lattice is a Chinese innovation, they assemble the media units with parts from N company, how long until they offer the whole thing for less with their parts? We wont even have to take them apart to buy the pieces. Besides single emitters are the end product, tapering up power and cutting numbers is just the stepping stones on the path to a big single emitter or we may be stripping the phosphor off 1 big led, at this point the cat is out of the bag, I'm not trying to be rude, but it's a bit late for walking tip toe.

Actually I would be happy paying 500-1000 dollars for 1 big SOB diode, it is all the 15 dollar ebay 303's causing trouble.
 
Last edited:





So I tried popping one out from behind and it did come out. The problem is it bent the diode frame and it is now deceased RIP. so I think if you made the tool to press it out that would be a viable means of extraction. They did a good job of making it tough to harvest these bad boys. But there is a way. That is for sure.
Does anyone know how hot let's say a NUBM44 diode is rite at the semiconductor chip when running at full power?
 
I don't think they like to get hot, most data sheets show a storage temp higher than running temp, but its no where near solder melting at 350 minimum.

I had thought of drilling 2 holes at the base and using a 2 prong fork tool that is xacto knife thin, actually I was thinking to cut a slot for the pins out of an xacto.

Tape across the can shoulders around the glass window but not touching the window.

Cut and peal/crack.

Thin razor xacto knife to wedge behind diode.

There are a lot of clues if you look at pics, some I don't see or can't find, but it's not heat.

Avoid heat.

Look at the pics in this thread, you see the hard way, but also clean backs.

http://laserpointerforums.com/f40/nubm41t1-interior-break-down-93525.html
 
Last edited:
Pi R Squared said:
That is correct, that's why we can't use the word Ca$10 in the forum, they don't want their products torn apart to extract laser diodes, they know about us and that's why these diode banks have been developed to be extraction resistant.

Alan
Click to expand...

Well, that didn't stop SyroPyro one little bit did it :)

 
I am so about to order a unit, but I have a different idea, it could be pretty neat, or I may have to harvest the bugger, either way my trigger finger is itching something awful.
 
I more I talk to people the more I think heat is the way to go. They were soldered in. And the operating temp of the diode is most likely higher than that of the solder. I'm going to cut two off the diode block and try heating it in a toaster oven. I will post my results.
 
Well, I believe they are being heated and punched out, I asked a guy who does this if it is possible whether vapors from the hot solder could contaminate the window of a NUBM44 and he said he checks the window with a microscope on each one he extracts to make sure they are clean. Whether he just meant he only checks the windows, or checks for what I was worried about, I don't know... likely though, he was answering my question as I framed it.

Maybe don't heat them enough to melt the solder, only enough to make it softer.
 
I had that same thought. If you made a punch that had a spot for the pins to go you might have enough surface to press it out without bending the diodes frame.
 
If they are soldered in, there is no way they could have soldered them in without having heated them enough to melt the solder! Still, if you don't need to use so much heat that the solder becomes molten to extract them, why heat them so much? You might need to do more trial and error to figure it out though, unfortunately, error is expensive :(
 
Last edited:
How about a hole saw, like a door knob hole cutter the width of the diode well, that part stuck to the back will be all you need to get off

EDIT:
Looking at the Chinese pulls they look heated, but a toaster oven, LMFAO oh man I'm cracking up.

It's likely not a hot install, it's likely hardening heat sink paste.


If you had an oval tool with pin holes drilled out you could try pressing it out, but I would avoid a lot of heat.

QQ20150425150836_zps09k935c9.jpg
 
Last edited:
Finding the rite temp is the key to this puzzle. A hole saw might work and that's a pretty good idea. The only thing is that the diodes are pretty close together and the kerf of the saw might be to much.
 
Even if you can use a small hole saw, you will still need to heat them enough to get the remaining piece off, I say heat the whole thing a minimum amount to be able to easily push them out. If you tried to remove the remaining ring from cutting it out, you would be more likely to overheat the diode, less conformity for heating that way.

Edit: I just looked at that photo you added, ouch, not much room to put anything in to punch them out, that is probably their main security feature to stop extractions, seems like you might need to find a way to remove more of the material on the back so so you can get a punch on them, unless you can machine something very small and precise the pins can't be damaged by it.
 
Last edited:
I think you are rite. I'm guessing they used some kind of solder paste. I'm going to look to see what the melting temp is on those types of solder. And I will start by heating it to just under that.
 
I bet a little tool could be made to push those diodes out. That and doing it extremely slow so it doesn't bend the metal in any way. I bet that's how DTR does it because he admitted to not using heat for extraction.
 
Last edited:
You could clamp it in the vise and take it down with an end mill. Just mill straight down those 3 runners the height of the diode backs.
All that will be left is a tab at the left and right.

Yes VG I was thinking to make that tool, oval the exact diameter with pin holes drilled in it and a block end to keep it perpendicular in the vise when pressing it out.
I like that better than heat or vibration of milling, I would press them forward.


NO NO NO.........NO HEAT! Heat may kill them or shorten lifespan.
 
Last edited:
I wouldn't believe anything someone says about how those diodes are extracted when they rely upon keeping the competition at bay.
 
You must log in or register to reply here.


Back
Top