A really tough 445nm! Burned and fixed!

Hi!
This is my first post here I didn't check all the forums if anyone ever did this, but I thought you will be interested in what I did

A got my 445nm diode about six months ago but didn't find time to build a laser an it took quiet a while to get all the necessary stuff (bought an aluminum heat sink from ebay, aixiz housing with a glass lens, electronics, batteries, charger, plastic housing...etc)

Well last week I finally started doing some testing. For my laser diodes I always build linear drivers (a constant current circuit) with an opamp (LM358) and an adequate transistor. This time I used a TIP112, which is a NPN darlington type (so only one is enough to amplify the signal from opamp). A also installed a very small smd temperature sensor (MCP9700) on the back of the diode (glued with arctic silver thermal adhesive) to estimate the diode temperature and show it on a led bar. This is a very non-precise way of measuring, the sensor will always be cooler that the diode, probably a better and a lot more expensive way would be to install a thermocouple.

So...as I am using 2x18650 to power the laser, I get quite a lot of thermal losses on the TIP112 transistor, about 4W @ 1A. Good heat sinking is needed. TIP112 has it's collector connected to the housing and consequently the heat sink body is also collector connection. As the transistor is NPN, the driver is a current sink, the laser diode is connected directly to the battery + an then to the collector of the TIP112. I was messing something around and made a BIG mistake, with the battery negative I touched the TIP112 heat sink. At that moment the laser diode had full 8.4V over it, not good...a blink of blue light and it's dead.

I curiously opened the diode housing and made a few macros...and WOW the really tiny wires connecting the laser chip to leads the have melted!
I was really wondering if the laser chip is still working, so I tried my best to fix the connection, got a solder with temperature regulation and a very thin tip. After a hard day when all of the things went wrong, at 1.00am I thought...what the hell another thing can't go wrong, I should try to solder it. After some concentration and precise and gently alignment of melted wires I just barely touched the tiny wires with the solder, checked, had to reposition one wire, re solder and voila...finished.

Did some macros of the soldered/fixed laser diode. Everything looked ok, so I tested it, but only at 5 and 30mA (no heat sinking). The fact that the diode is working like a led didn't surprise me that much. But what about lasing? I installed the diode to the aixiz housing, mounted the heat sink and unbelievably the diode LASES! Just by inspecting the beam profile I can't see any damage. To check for any power degradation I still have to visit my friend, who has a high precision laser power meter at work.

Photos are on the way, I just found out that I can't upload larger that 800x600, I need to resize them.

Here are the photos of the opened laser diode:
First two pictures present the diode in burned state, the next three the solder fix and the last two emission at 5mA and at 30mA

And some laser action!

Nice Repair job... I'm just wondering if the LD would get hot
enough to melt the solder when running at high Currents...

What did you use to take the Macro Pics...???

BTW... Welcome to the Forum...:yh:


Jerry

Pretty amazing stuff Sammo. I think it will be a few years before I have the soldering skills to tackle something this precise. I haven't seen a fix like this since I've started browsing threw this forum. Great work with the macro photos.

Incredible Soldering!!! :yh:
would have been a giant blob of solder if I tried that.

Thanks for your replies.

I don't think the temperature can increase so high to melt the solder, thermal conductivity of those small wires is very small, would probably fry the laser chip long before melting solder. Diode also has now the advantage of better cooling because it's open type
I'm only afraid that any shocks might break those little thin wires because the solder is quite heavy (relatively of course).

For the macro setup I'm using a 35mm Carl Zeiss Flektogon (old M42) lens mounted on a macro extender, about 8" away from a Canon 350D.

sammo said:

Thanks for your replies.

I don't think the temperature can increase so high to melt the solder, thermal conductivity of those small wires is very small, would probably fry the laser chip long before melting solder. Diode also has now the advantage of better cooling because it's open type
I'm only afraid that any shocks might break those little thin wires because the solder is quite heavy (relatively of course).

For the macro setup I'm using a 35mm Carl Zeiss Flektogon (old M42) lens mounted on a macro extender, about 8" away from a Canon 350D.

Click to expand...

Thanks for the added info...
Yeah... with the LD open it would take quite a lot of heat to creep
up the small wires to melt the solder...:beer:

Jerry

Awesome photos and repair, Sammo!

+1

Just when I thought soldering broken pins to the diode was cool.

+1

Holy crap - I was almost ready to call photoshop fake on these photos!

That's amazing

How did you even open the diode can? (I wonder if doing that to a normal diode would increase the optical output ?)

Sorry

BTW...that red sphere on the photo is a ruby crystal (and it's put on a sapphire ring ). It fluoresces really beautifully, especially because of so much contrast between blue and deep red.

Very impressive fix :beer:

+1 for you

rhd said:

How did you even open the diode can? (I wonder if doing that to a normal diode would increase the optical output ?)

Click to expand...

I opened it with an OLFA cutter. I carefully cut the can all the way around, using not quite fast moves and applying little force, but at first didn't actually cut through it, just to make the material thinner for further cutting. Then I started to cut through at the part where the block (that holds the chip) of diode housing material is closest to the can, in order not to damage the chip and leads, lets call this the top of the diode if we are looking at it face up.
Then I cut also the sides, but left the bottom just thinned (if I did cut the bottom side I'd probably damage the diode, cause it's hard to know how deep are you with the knife and at that point I have never seed the inside of it...).
At this point i just put the Olfa at the top of the cut part of the can...and carefully "pry-opened" it towards the bottom, over the top of the laser chip. This way I avoided manipulating anything in the part of the diode where the chip and the leads are.

The chip is really amazing .. it holds better then the wires !

DTR inserted the diode into the chuck of a drill and held a knife on the edge of the can to open it.

Nice work BTW. I think thats a first here.