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

LPF Donation via Stripe | LPF Donation - Other Methods

Links below open in new window

ArcticMyst Security by Avery

Broken/faulty diodes wanted

Joined
May 29, 2010
Messages
88
Points
0
Hi all.
Upon reading the articles suggesting that broken 445's could be repaired, it occurs to me that I have some 35 gauge solder here.

If anyone has one or more broken diodes which are open circuit, I would be interested in trying the bond wire repair technique on them.

Thanks!
-A
 





Joined
Jan 14, 2011
Messages
3,816
Points
63
Would you just be asking for these diodes and then sending them back upon repair (and test) for a nominal fee?
 

rhd

0
Joined
Dec 7, 2010
Messages
8,475
Points
0
I have a handful of reds, 445s, and IRs that have fried in the process of testing different drivers. Mostly 445s. I echo Wolfman's question though.
 
Joined
May 29, 2010
Messages
88
Points
0
Hi,

Yes, in the highly unlikely event this little scheme works I will send the repaired decanned diodes back for the cost of the postage.

My cunning plan involves an improvised 3-D microscope setup using two small 1.3MP webcams with retasked scanner lenses as they have excellent optical properties when used in this way.

As others have managed to repair broken bond wires it might be possible to improve on this with an optimised jig.

Not sure but it looks like using fine leaded solder is the secret sauce, I've made UV LED arrays with 0403 diodes by hand before.
Had 99% success, only one failed and this was easily fixed.

-A
 
Joined
Aug 4, 2010
Messages
1,249
Points
0
Please let us know if you get this working. That would be a HUGE success
 
Joined
Dec 27, 2009
Messages
1,321
Points
0
improvised 3-D microscope setup using two small 1.3MP webcams with retasked scanner lenses as they have excellent optical properties when used in this way.
Please grace us with pictures of your setup when you build it! :D
 
Joined
Sep 16, 2007
Messages
3,658
Points
113
If a diode is killed by catastrophic optical damage (COD) it cannot be repaired in this way.
Only a diode that is an open circuit has a chance of being revived. So, if your diode becomes an "LED" (which is likely COD) you're out of luck. If the diode produces no light at all and draws no current at all, that means that it is possible that the diode failure could be from the tiny wires melting, in which case it may be repaired.
Just in case that wasn't clear.

Good luck on this and I'll look forward to seeing the results.
 
Joined
May 29, 2010
Messages
88
Points
0
Cool, thanks for the support guys.

I will be posting pictures as soon as its working, as I have a few things which need a 3-D setup to finish.

Interestingly I found that it is possible that some "failed" diodes are actually repairable.
I have noticed that PS3 diodes sometimes develop whiskers between blue and red or blue and infrared anodes, have managed to fix this on one by carefully applying 1.2V across them.

Stupidly I didn't take a movie of the repair process under high magnification, it should look interesting having that whisker fuse open.

-A

"Bother!" said Pooh, as his GPS failed to pick up any satellites...
 
Joined
Feb 19, 2009
Messages
1,807
Points
48
This could end up being a very interesting thread to follow :D

Definitely at least post a few pics of your setup please. The whole idea of this happening on a lot larger scale sounds very cool!
 
Joined
May 29, 2010
Messages
88
Points
0
One red burner fixed, the two 445?s I have are both reading 20 ohms in both directions so they are 100% dead sadly.

I also now have to repair my 405 burner after the positive wire snapped off during my attempts to fix a short inside the can...
Will keep you posted :)
laser2.jpg

laserdot.jpg


-A
 
Joined
May 29, 2010
Messages
88
Points
0
If a diode is killed by catastrophic optical damage (COD) it cannot be repaired in this way.
Only a diode that is an open circuit has a chance of being revived. So, if your diode becomes an "LED" (which is likely COD) you're out of luck. If the diode produces no light at all and draws no current at all, that means that it is possible that the diode failure could be from the tiny wires melting, in which case it may be repaired.
Just in case that wasn't clear.

Good luck on this and I'll look forward to seeing the results.

Um, no. I just revived two totally dead (20 ohms in both directions) 405 burner diodes using a technique based on reverse pulsed bias.
Used 9V lithium (maybe important, there seems to be a current limit on these) applied in reverse with very short pulse cycle.
Seems that it "blows out" the short if the pulse is large and narrow enough, leading to a working laser at reduced power as long as the bond wires also don't blow during this treatment.
Unfortunately one diode failed shortly after when trying to increase power from LED-like levels by more zapping.
I would recommend the use of a relatively low voltage say 1.5V at several hundred mA for about 0.5us as this would reduce the chances of a blown bond wire.

The way to tell if it is working is to measure the reverse current. If the short is gone then the current will drop sharply down to 10mA or so with 100 ohm series resistor in the reverse direction from >100mA.
Just to prove that this isn't a joke, I include beam shots and power data from the diode which survived.

Used a PS3 diode as the beam power sensor, and got the following.

14.31 mA on diode at 40mA power in
>20mA on diode at 60mA power in

Used an attenuator consisting of polarising sheet in beam path.

0.59mA at 40mA in
0.88mA at 59mA in
0.98mA at 67mA in
Offset 0.12mA (quiescent)

Additional observations during this process.
The diode will get HOT! I think this is probably normal, as all the energy is being dissipated within the shorted area.
I also noticed that signs of a potentially recoverable diode are that it will read slightly different currents in forward and reverse bias, a permanently fried diode will not do this and read more like 3 ohms or less.

It appears that this effect is unique to GaN devices, I have never noticed anything like this with any other laser.
I also had a PS3 diode once with a big hole in the beam profile which recovered some power when I zapped it with 60mA short pulses.

Update:- Grr, shorted again. I think it is probably toast this time.
I did however manage to solder on new bond wires to the 63133, not much output though.
Comparing with a known good open can burner diode from a Sony laptop DVD writer it is obviously LED'd.

In retrospect I really should have posted a beam profile etc.
I am in the process of hacking a cheap Logitech webcam for this purpose, seems that putting sanded plexi on top of the chip works well.
So does that "broken" B/W CMOS camera as it clearly has sensitivity to laser light but not room light.

EDIT: Modified technique works with a 1mW diode from a broken red laser pointer.
This one was working then it went 59 ohms so I tried the same trick.
Works fine now, so far no sign of damage.

EDIT 2:- I need fried diodeage peoples.
send 'em to Mandoline Cottage, Port Soif Lane, GY6 8AH labelled "Toasted diodes"..
I will be sure to post results and if the diodes are reliable they will be sent back.





-A
 

Attachments

  • uvlaser.jpg
    uvlaser.jpg
    14.7 KB · Views: 106
  • laser34.jpg
    laser34.jpg
    28.8 KB · Views: 108
  • laser23.jpg
    laser23.jpg
    20.2 KB · Views: 89
Last edited:




Top