GB: PL520 murder fund

[Blord]

It stopped outputting light at 760mA. I drove it up to 1A after that and it didn't die, still outputs the same when I turn the current back down. This is the most resilient diode I have ever come across.

From the data of the first diode I would say a cap of 350mA for these diodes.

I think to kill it I will have to see what kind of wavelengths it makes at 100°C

Did you push the current graduate to 350mA ? I want to see if the diode is still doing well after pushing on/off 350mA many times.

 

[ARG]

I don't think any of us have enough research on them to even know the direct effects of high current on the lifetime. They seem resilient but until we have a lot more runtimes on them I don't see it being possible to make a very good guess. Based on our driving of other diodes we typically get away with double currents and the diodes appear to survive quite well.

The diodes will just degrade faster, meaning after 10h of use they will output a little less.

Maybe that's what I should do to kill it, a 50h run time taking readings every 5h at 350mA to see how badly they degrade at high currents?

I am thinking about driving mine at 300mA, does this sound reasonable for decent lifetime?

The 200mA cap is probably for 10,000h of run time. For use in handhelds I think 300mA would be fine.

Did you push the current graduate to 350mA ? I want to see if the diode is still doing well after pushing on/off 350mA many times.

Graduate? I pushed it to 1A without any adverse effects, if that's what you mean.
However, it didn't lase past 760mA

 

[Blord]

I bet 120mW 520nm green is plenty bright enough to impress.
Do you have any torture methods left for the diodes ?

 

[Trevor]

I plotted the power output and efficiency against the current, for science.

Trevor

 

[ApexProxy]

^ thanks Trevor.

The diodes will just degrade faster, meaning after 10h of use they will output a little less.

Maybe that's what I should do to kill it, a 50h run time taking readings every 5h at 350mA to see how badly they degrade at high currents?

I completely agree with this method of torture. If you can maintain the heatsink temp at a fairly normal near-ambient temp it would likely give an even better idea of the effects in a typical situation.

 

[ARG]

I bet 120mW 520nm green is plenty bright enough to impress.
Do you have any torture methods left for the diodes ?

I could heat, freeze or run them for long times at high currents.
That or I could just put 5A into it and see what happens

I completely agree with this method of torture. If you can maintain the heatsink temp at a fairly normal near-ambient temp it would likely give an even better idea of the effects in a typical situation.

I think that's the best idea at the moment, since it gives us some useful data. Running these diodes at sub-zero temperatures or temperatures too hot to touch is impractical in real applications.

 

[Speedy78]

If you cant kill it with normal tests I recomend moving the whole experiment into your freezer

 

[ApexProxy]

I'm still going to back the idea of gathering more useful data. A long run test at a high but reasonable current like 350mA would give the most useful data out of those tests. Hell, what about a 48hr run at 350mA and then a 48hr run at 450mA. That way you can measure degradation even better. Might as well gather as much data as we can on how the diode performs in situations that we will most likely expose it to.

Its fun to run crazy off the wall tests on things and see how they perform and stuff but it really isn't all that useful if the tests are too far from what the diodes will even be exposed to. That's just my take on it.

 

[IsaacT]

^ Agreed....as someone who might be running theirs at 350mA, I agree that data would be highly beneficial. Most likely though, I will stick to 300mA.

EDIT: Couple Driver related questions that anyone who buys one of these diodes will face:

1. Will a BlitzBuck V3.1 Round Adjustable Driver work for these in a 2 Cell build? Or would you need more Voltage than 2 Cells?
2. Are there any drivers being designed for these?
----------A. Microboost: Everyone is out of stock.
----------B. BenBoosts: Apparent Current Issues and no longer carried by Mohrenberg.
----------C. X-Boosts: From what I gather, Lazeerer is out of these or in VERY low supply.

----------D. Flexdrives: Would these work? From what I understand it is a Boost driver so it would work with a 1 cell build. Hacylon.com offers them but is that up to date? And it asks for a current setting, can I pick whatever I want or is there a range of them?

I think that is it for now, any and all advice would be great.

 

[Speedy78]

I agree as well with useful information based on everyday parameters. But after all that is done and its still alive, why not see what we can do with it? haha

 

[ApexProxy]

I'm going to be running mine on a X-Boost I have coming on the way. Lazeerer told me that it can handle the required Vf just fine.

 

[IsaacT]

I just bought some microboosts so I am set!!!! Big shout out to flaminpyro for saving my life on that one!

 

[Wolfman29]

Just for those of you who want an answer to the BlitzBuck question: I wouldn't use it for these. The BlitzBucks are good for robust diodes, but I wouldn't want to risk anything for a $120 diode. It is actually for this very reason that I designed the BlitzLinear - those things are the most stable and have the cleanest output of any driver on the market.

 

[ApexProxy]

IMO, for use on these diodes linear is just not that practical for a lot of hosts since the Vf(6.5-7V) of the diode would require 3 cells unless you don't mind your full power runtimes being shortened. Also, these diodes have already proven to be shockingly resilient.

 

[Wolfman29]

You won't need 3 cells for he BlitzLinear It has a V-drop of .3V, so you have a lot of overhead even with 2 cells.

 

[ApexProxy]

1.6V of overhead for the diode ARG tested. That means your cells will run to roughly 3.4V before losing output. Also, my diode runs at 6.9V at 300mA which limits my overhead down to 1.2V.