My 'Daedal Circuit' Confusions

[Benm]

Also: I always solder a 22-26 gauge solid wire to too fat or skinny components leads so to not ruin the plugboard,

Good idea really, especially the fat leads on things like LM317's and power transistors can wreck havoc on the breadboard contacts. I usually just cram them in anyways, but if you want your board to last, solder thinner wires on

Even though it did not die right away, chances are good the diode was doomed from too much current. Also bad breadboard connections would not help either.

I would bet on optical failure due to excess light in this case, not any thermal problem since it wasn't warm at all, and died abruptly.

I am not sure what circuit you are using exactly, but if its the one with the LM317 and a resistor between Vout and Vadjust, a severed connection results in NO current output, not in overload.

I use 1ohms .5watt ohm resistor as cheap a fuse. If you goof up wiring, your just loose a $.02 resistor. heat, smoke, poof!

Don't.. resistors are not fuses. They are horribly slow to burn out from over current (except a full short resulting in >>10x rated power). Worse, in rare cases some resistors actually short out completely instead of going open circuit. Get a fast fuse if you must!

 

[Daedal]

Comidt... I'm so sorry you had to join the club of diode killer... but hey... on the bright side... we're nice folks here

Practice makes perfect, I'm sorry you had to lose the diode, but I would say that you never really learn until you break some.

The problem was that the leads were connected to the wrong side of the resistor. The schematic as well as my picture show that the LD wire is fed from the Adj side of the resistor. The way I avoid confusion on this one is that the LM317 has a Vout line and an Adj line. I'm not regulating voltage in this circuit, so I don't use the Vout

The other thing is, you gave the diode too much current from before. It's a miracle it survived this long. Although, maybe you were measuring current at the source rather than at the LD.

This is only a practice session... keep it as a learning experience. It's only $17.50 Next time put in 3 or 4 1N400X diodes to simulate the LD and check the current right at the output lead rather than at the input. The output lead is between the resistor and the Adj line. The capacitor should be soldered to the LD. The voltage going in should be tested prior to putting it in every time, unless you are regulating it.

My speculation in terms of the diode's death is that it suffered from Catastrophic Optical Damage (COD) which is a resultant of too much light in the crystals and no where to dissipate it all efficiently enough. What happened there (proven by the fact that there is still light) can be better understood here... http://en.wikipedia.org/wiki/Catastrophic_optical_damage

GL Comidt, and I hope this doesn't put you down too much. It's part of the fun

--DDL

 

[Gazoo]

I would bet on optical failure due to excess light in this case, not any thermal problem since it wasn't warm at all, and died abruptly.

And excess current causes excess light which damages the mirrored facets of the resonant cavity. It did not exactly die, it turned into an LED, which supports your theory of an optical catastrophe. But I believe the damage was caused by excess current. Running too much current and/or longer duty cycles will cause this to happen, and it might not happen right away as the damage can accumulate over time.

 

[Daedal]

Agreed Gazoo... the damage can compound over years of use. You can turn it on for 5 seconds at 500mA, then leave it for 10 years, then turn it on again and it works, and 5 seconds later it dies... such is the life of a LD...

--DDL

 

[Gazoo]

I'm not regulating current in this circuit, so I don't use the Vout
--DDL

Daedal,
I think you meant to say you are not regulating voltage in this circuit... are you testing me again ;D

 

[Daedal]

Oops.. :-?

Thanx Gazoo ;D
DDL

 

[laserrod]

Comidt,
Hey, I just wanted to be helpfull.
This Forum has so many smart helpfull members that are really cool! It's just unreal!
I was trying to share the load of the highest level members who always so helpfull.
But everything I wrote you was fact. I know this stuff! Been there, done it!

Disclaimer: the 1 ohm resistors fuse cheat, must be those cheap carbon film enamel coating, shinny light brown,(Brown, black,gold, gold, color bands). They don't short out! But they won't protect LD's! But that's just nesascery, just tring to help U measure current with a safty effect.
They won't explode quickly under 1 amp, but fuses are slower.
Thanks, to member Benn, who wrote resistors will short out. I must say a wirewound resistor or else will short out and not act like a quick fuse sometimes.
Your mileage will very. Ha Ha ;D ;D

You're in the best off hands here, really!

Best wishes,
DGM [smiley=beer.gif]

 

[Hemlock_Mike]

Gazoo --

Another bad connection -- I found out that the open can LD's Won't take 1.4 AMPS It's still a great LED

Another one bites the dust -----

Mike

 

[Gazoo]

Mike-

ROFLMAF.... ;D ;D ;D

1.4 amps....oops... ;D Those bad connections are killers.

 

[JES]

Gazoo --

Another bad connection -- I found out that the open can LD's Won't take 1.4 AMPS    It's still a great LED  

Another one bites the dust -----

Mike

Imagines Mike cranking Dr. Frankenstein's lightning rods through his ceiling to generate the 1.21 gigawatts needed for his next laser experiment.

 

[Gazoo]

He already tried that to power his Co2 laser...it didn't work.

 

[Comidt]

Thanks All
You Guys are great.
@laserrod, I know you're trying to be helpful, I didn't say you did something bad to me or anything and I appreciate EVERYBODIES comment, because I'm sure most of you are a lot more knowledgeable than me.

So what must I change next time??
Could another problem have been when I tried to measure the currentof the LD, I removed the + lead which was connected to the capacitor, when I reattachedf it, could it have killed it then?

When I put in my new diode what must I do instead?

Thanks All

 

[Gazoo]

Now you definitely know how the circuit is supposed to be hooked up. Stop using the breadboard and get a PC board and a decent trim pot. Use the silicon diodes to test your circuit before hooking up the diode. Hooking up the meter most likely was not the cause of the problem.......it was too much current. But I have test leads with alligator clips hooked from my meter to the circuit, so the meter is connected before I apply power.

 

[SenKat]

So what must I change next time??
Could another problem have been when I tried to measure the currentof the LD, I removed the + lead which was connected to the capacitor, when I reattachedf it, could it have killed it then?

In short - YES...When you are measuring current like that, you cannot break the connection until the power has been terminated - otherwise, you can and will most likely create a short circuit situation, and that can make the diode go poof almost instantly ! Now - that MAY not have been the issue this time, but it will be next time unless you kill the power first, then disconnect the meter....I have murdered some unsspecting diodes this way

 

[Benm]

Could another problem have been when I tried to measure the currentof the LD, I removed the + lead which was connected to the capacitor, when I reattachedf it, could it have killed it then?

Possibly. Personally, i would not recommend using a capacitor at the output of a current source, since it will charge to input voltage when the load is disconnected. I doubt a few uF cap would store enough energy at say 5 volts to destroy the LD when you connect it, but it is a possibility.

Soldering the cap directly onto the diode as was suggested before would be much better.

However, LM317's do require an output capacitor fairly close to the device to guarantee stable operation - so these is little you can practically do in this case.

 

[Comidt]

OK, So I still don't know what I can do to stop this problem from happenning to my next diode.
Right Now, I don't know if it's sill gonna do the same thing.