Dead LOC, any ideas why?

[spectral532]

Recently, I purchased a LPC-815 sled from HighTechDealz. When it got here I soldered a few wires to the ribbon cable and powered the diode up. It worked fine. I didn't have the tools to extract it at the time, so I put it back in the anti-static bag and left it alone. The next day I try powering it up, and it's an instant LED, not even a flash . The diode is still in the sled and I haven't even tried extracting it yet. The only thing I feel that might have gone wrong, is that I had the sled facing down on my desk when I turned it on, so maybe the light reflected back into the diode and killed it?

I'd like to figure out what went wrong so I dont kill the next one. Any help is appreciated.

 

[Eudaimonium]

I'm packing up LCC for ya after all, eh ?

What driver was used, current set to (resistor, or trimmer?), capacitor placement?

Some pictures could help!

 

[Greenmechanic]

It could have been a small piece of dust settled on the die.
I've had an LOC work fine for hours and hours, I bought a glass lens for it. Carefully blew off the focusing nut with clean air. Held it upside down while changing the lens.
Didn't last 15 seconds after that.
This has happened more then once.
Also your statement about placing it face down could be correct, if the desk is reflective. Highly reflective
My bet is on dust, although.:thinking:

 

[spectral532]

I'm packing up LCC for ya after all, eh ?

What driver was used, current set to (resistor, or trimmer?), capacitor placement?

Some pictures could help!

I used a DDL driver set at 400mA with a 10µF and 0.1µF capacitor paralleled. The capacitors were discharged before I connected the diode. I would take some pictures, but my camera doesn't have a good macro function.

Also your statement about placing it face down could be correct, if the desk is reflective. Highly reflective
My bet is on dust, although.:thinking:

I dont think it's dust, since the diode was still in the sled when it broke. There is no way dust could have gotten on the facet.

Maybe, just maybe I killed it when I desoldered the ribbon from the diode. But I only held the iron on it for 2 seconds max, so I dont think that is the cause either.

 

[Eudaimonium]

C'mon if desk reflection could kill the diode in a sled, they would not make DVDs so reflective, would they?

I'm betting on simply defective diode.
BTW, why 10 and 0.1 uF in parralel?
That one 0.1 uF is not doiing much good.

Is it possible that while it was sitting idle, the power could somehwo be applied to the driver, and feed the cap?

Or could the cap be disconnected somehow?

And you did not say, how did you set the current? If you did it with a potenciometer, it is very likely you moved it during connecting-disconnecting, increasing the current.
I always use fixed resistors, makes it so much easier.

 

[Deleted member 8382]

C'mon if desk reflection could kill the diode in a sled, they would not make DVDs so reflective, would they?

I'm betting on simply defective diode.
BTW, why 10 and 0.1 uF in parralel?
That one 0.1 uF is not doiing much good.

Is it possible that while it was sitting idle, the power could somehwo be applied to the driver, and feed the cap?

Or could the cap be disconnected somehow?

And you did not say, how did you set the current? If you did it with a potenciometer, it is very likely you moved it during connecting-disconnecting, increasing the current.
I always use fixed resistors, makes it so much easier.

I don't know at all if desk reflection can kill the diode or not, but remember that before the light emited by the diode hit the DVD it passes trough some optics, which I guess that block the relfections too

 

[Eudaimonium]

Not all.
Take a sled of any kind, remove IR, RED diodes and the sensor, and also remove the shortfocal aspheric lens above which DVD is.

Shine the (green preferably) laser from DVD's point, as if it were reflected from the disc.
You will notice that the beam will go out from all three holes almost equally, in great power, with not-as-low-as-expected transmissive losses.

 

[spectral532]

I'm betting on simply defective diode.
BTW, why 10 and 0.1 uF in parralel?
That one 0.1 uF is not doiing much good.

Is it possible that while it was sitting idle, the power could somehwo be applied to the driver, and feed the cap?

Or could the cap be disconnected somehow?

And you did not say, how did you set the current? If you did it with a potenciometer, it is very likely you moved it during connecting-disconnecting, increasing the current.
I always use fixed resistors, makes it so much easier.

I have a 10 µF electrolytic to suppress any large spikes, and a 0.1 µF mylar cap to catch any high-frequency spikes.

I had the capacitors shorted with a jumper while I connected the LOC, so I'm pretty sure they were discharged. The current was set with a pot, but I had verified that the current was 400 mA with the LD connected.

I'm not sure if it was a defective diode, since it was working when I first got it. It drew about 470mA @ 3.3v, so it seemed like an otherwise healthy diode.

 

[Eudaimonium]

I have a 10 µF electrolytic to suppress any large spikes, and a 0.1 µF mylar cap to catch any high-frequency spikes.

I had the capacitors shorted with a jumper while I connected the LOC, so I'm pretty sure they were discharged. The current was set with a pot, but I had verified that the current was 400 mA with the LD connected.

I'm not sure if it was a defective diode, since it was working when I first got it. It drew about 470mA @ 3.3v, so it seemed like an otherwise healthy diode.

Wow, 470 mA is waaay too much for a diode in it's original heatsink...
Only Kryton head or something will do when pushing these that far.

Dave has a Kryton LCC (not LOC) on 0.5 A, still kickin' !

 

[spectral532]

Wow, 470 mA is waaay too much for a diode in it's original heatsink...
Only Kryton head or something will do when pushing these that far.

Dave has a Kryton LCC (not LOC) on 0.5 A, still kickin' !

I made sure not to keep the diode on for more than 5 seconds at that power. During that time the heatsink didn't even get warm. I had maybe a total of 30 seconds of run-time on the diode before it died the next day.

I'm completely out of ideas, I guess I'll just have to hope for better luck when my next sled gets here.

 

[pullbangdead]

Why did you solder to the ribbon cable? Why not just solder to the pins?

I can't imagine a wire soldered to a ribbon cable would be that stable an arrangement, so if it disconnected at the solder joint for even the tiniest amount of time, the caps would've charged up and killed the thing.

 

[spectral532]

I soldered to the ribbon cable since it was easier than soldering to the pins at the time. I was planning on soldering wires directly to the diode after I had extracted it, but I wanted to check if it worked while in the sled, so I just soldered some wires to the ribbon cable.

 

[BlueAlien]

This same thing just happened to me tonight with 2 lpc-815's. Powered it up while still in the heat sink and it was very bright for a few tests then it suddenly became a LED. The second one I tried in the sled and it powered up for a few tests as well then LEded. I had them running on a rckstr set to 360mA running off of 5 AA batteries totaling 7.9 Volts. This is the last time I power one of these up outside of a module.

Now that I think about it, I had it all hooked up with gator clips. It probably was a charge from the non-stable connection, like pullbangdead said. Dam, I cant believe I just bought these and killed them. Makes me pissed at myself for throwing money down the drain from stupid mistakes.

Sorry to jack your thread, seems relevant.