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

M140 diode broken?

definitly zombie sorry man. I am building lasers for almost a year now and i've killed like 5-6 diodes..But from each one you learn especially in the beginning. Just really sensitive succers.

If you want to take less risk build a 9mm laser. Those are more robust ;) But more expensive.

greetings,,
 
Last edited:





Yup, unfortunately laser diodes are quite sensitive. I somehow managed to LED a diode on its first run with a new bench PSU. It appears that it spiked when I turned it on and poof. Diode was fried before I ever even saw it run.... Oh well, lesson learned.
 
I have no money to buy another diode.
My monthly salary is only 900 dollars. I have to pay house, light, water, food ...
So imagine the effort I made to buy a DTR LED, driver and lens.

Someone donate me a diode???:whistle::whistle::whistle: LOL
 
If you can get the broken diode out intact I might be interested in donating you a few spare parts here (micros, etc) .. I have a need for LED'd diodes for my experiments in optical comms.

Seems that as long as it has contact to the diode it should be fine, O/C ones aren't.
PM me.
Also I have a few random BRD diodes here which might be fine, including a fairly recent 405 from PS3 Slim.

If you feel very lucky I have had some success with careful application of reverse pulse bias (under scientific conditions) to "blow out" partially shorted emitters so at least it recovers some power, if the diode measures less than 100 ohms in both directions this is often the case.
From the pictures you sent it looks like two of the three emitters are damaged.
This is a common failure mode of overloaded diodes with the centre emitter often blowing first due to heat.

Hope this helps, -A
 
Last edited:
fermintm said:
I have no money to buy another diode.
My monthly salary is only 900 dollars. I have to pay house, light, water, food ...
So imagine the effort I made to buy a DTR LED, driver and lens.

Someone donate me a diode???:whistle::whistle::whistle: LOL
Click to expand...

Well in the beginning i too fried 5 diodes total. So all i can say is that it is part of the hobby :p

greetings,,
 
Yup, LEDing a diode is not a good feeling but, you live and learn. I now know what not to do in order to reduce my risk of making LEDs. One big one is, always remove all power and short your driver before soldering it to the diode. Also for using a bench PSU I now always hook the diode up and make sure all connections are secure before I slowly dial up the power to avoid extreme instant diode killing spikes and it has worked perfect or me so far. I would like to keep my LED making record at one single diode if I can :P

Is there a single component I can put in-line or parallel to a diode on my PSU that I can use to help protect it from sudden tiny spikes?
 
You can try putting some ESD protection diodes in the circuit. Linear regulators (LM317 for example) are also good at preventing voltage spikes, etc. from making it to your diode. If you've got one that is your constant current driver, it should be pretty well protected so long as it remains connected.

Still, those are more to protect your already wired system from spikes from something like plugging in an input to the system. For most people, just ensuring that you short the leads before reconnecting wires, etc. is good enough.
 
I could always just use a regular driver but my intention is to connect a diode to my bench PSU and be able to use variable power and have something connected to the diode or on the lead wire to help further increase my protection. If such a simple component doesn't exist I can just be careful.
 
Hi..
I also would like to share my own tricks for avoiding LEDs.

Adding a parallel 470 ohm resistor and reverse biased blue LED, this means if there is a damaging voltage spike the LED will break down first protecting the diode.
It also allows high frequency operation as the resistor discharges the LED and diode capacitances.

Another good idea is a "crowbar" circuit with a simple sensitive gate thyristor, resistor and a fuse.
It can be put in series and set to crowbar at say 1.45A so at least if you make a mistake the fuse blows first.

For smaller diodes use a bank of blue LEDs in series with 47 ohm resistors.
Select these so that they draw the same current and voltage as the diode being tested and CLEARLY MARK THEM !! ie max "300mW" 250mA etc.
Mounting in an Aixiz sized casing is also wise, so you can adjust the circuit with a dummy diode before risking your expensive Bluray(s).

An audible beeper is also a good idea based on a 1381 4v1 and series diodes, set to go off if V>5.2V or I>250mA so you instantly know something is awry.
 
Interesting idea. I have a number of blue LEDs around that are designed for 12v operation and have little unknown spec resistors inline with their positive leads. I am wondering how a backwards LED could protect the laser diode? I am actually in the process of learning more about electrical circuitry so things like this are quite useful for me.
 
You should always have a driver and not depend on the PSU directly unless you're absolutely sure it's designed for precise control. Precision PSUs are more expensive because of that precision control and safety, but you can get around that by building a driver that you know the characteristics of.

It appears that you're not sure about your PSU, so you need a driver.

For myself, whenever I want to have variable control of the current/power my laser diode receives, I just use a transistor-based, current-limiting circuit like this. Then what I do is hook up one of the voltage outputs of the PSU to control the transistor input of that circuit above, therefore regulating the current. I haven't friend any lasers using that circuit.
 
Interesting. Well, so far I haven't lost another diode. I was bored and did some voltage and amperage tests with the PSU and based just on my multimeter the displays are spot on and from what I can tell its pretty damn stable. Obviously my multimeter is not an oscilloscope but from what I am able to tell its at least more stable than I can test it to. I haven't lost another diode yet by being very careful but I think I might build a circuit like the one you linked me to.

My first diode loss was a combination of turning on and off the PSU with the diode hooked up and accidentally disconnecting a live lead and then, like a massive noob, I reconnected it without turning off the power. So now I turn on the psu with all controls bottomed out but the PSU on and already connected to a 12v LED and I then add in the laser diode parallel to the LED and then disconnect the LED and power up very slowly. My hopes are that with the LED soaking up any small power spikes before I hook up the laser diode I can be sure that the PSU is not storing power for a spike as soon as I connect it up.

I was planning on designing a little LED circuit that will have a large Cree LED and the tiny blue LED. The Cree would be on a switch so that I enable it while I am hooking things up and I disable it when I have everything connected. The little blue LED would stay connected because it uses a nominal amount of power. My hopes are that the Cree, being a very large LED, would be able to buffer quite a bit of power to avoid spikes on connect and the tiny LED would just act as a buffer always. I am not sure if this would work like I am thinking though. Thoughts?
 
Last edited:
To the OP;

What I found you can do to help a zombie diode not produce so much heat is to blow out the bad emitter completely.

What you do is take a 9 volt battery, and very quickly tap it reverse bias across your diode.

I was experimenting with a broken 405 nm diode that was LED'd... I found it would get extremely hot no matter which way I hooked up a battery (one way it would emit uncoherent light, the other, no light but still heat)

So I took a 9 volt battery, and hooked it up reverse. I saw a small puff, and then hooked it up how I had before; the diode would still emit the same amount of light, but the heat was greatly diminished, and it would no longer get hot when reverse biased.

So it may be worth a shot to try blowing out the bad emitter!
 
That is ONLY suggested if you are 100% beyond all doubt that the laser diode is messed up and LED'd and the issue isn't something else. Use it only as a last resort just to get the thing to emit some sort of light without creating a ton of heat. Don't expect it to be fixed.
 
Le Quack said:
To the OP;

What I found you can do to help a zombie diode not produce so much heat is to blow out the bad emitter completely.

What you do is take a 9 volt battery, and very quickly tap it reverse bias across your diode.

I was experimenting with a broken 405 nm diode that was LED'd... I found it would get extremely hot no matter which way I hooked up a battery (one way it would emit uncoherent light, the other, no light but still heat)

So I took a 9 volt battery, and hooked it up reverse. I saw a small puff, and then hooked it up how I had before; the diode would still emit the same amount of light, but the heat was greatly diminished, and it would no longer get hot when reverse biased.

So it may be worth a shot to try blowing out the bad emitter!
Click to expand...

Ha, there was I thinking I "invented" this technique.

It was discovered by accident when reading Sam's LaserFAQ and he mentions a method of fixing dead laser bars with a sharp blade.

I actually came up with it after a similar method worked on a defunct large area blue LED which had a near identical fault induced by poor current regulation.

I'd add a series resistor though as too much current WILL blow the bond wires, 100% fatal to the diode.
 
You must log in or register to reply here.


Back
Top