GB: PL520 murder fund

[Tw15t3r]

Same here. Most people would be doing standard room temp builds, so it'd be better to replicate the condition for results that would be of practical use.

EDIT: While you're testing within manufacturing specifications, would it be worth it to do a wavelength plot as well? Or would that take to long?

 

[Mohrenberg]

I would go with testing the wavelength first, and then plotting it till it pops. A lot of times you can see the max optimum current for a diode by seeing the shift in efficiency.

I would also go with regular heat sinking, so we know what to expect in our handheld builds.

 

[DragonSoulSong]

Well...it seems that I'm a little late to the proverbial party, thanks to the way that my time has worked out today, but in any case, payment sent. ^_^

 

[ApexProxy]

I agree on the room temperature handheld resembling setup for the tests.

Procedure I would do:
1. Plot output from threshold up to spec current (standard handheld ambient temp)
2. Plot wavelength from threshold up to spec current (standard handheld ambient temp) - overlay on to #1 plot
3. Plot output up to the efficiency curve (standard handheld ambient temp)
4. Plot wavelength up to the efficiency curve (standard handheld ambient temp) - overlay on to #3 plot
5. Retest output to spec current to verify no damage (standard handheld ambient temp)
6. Set the diode to the peak efficiency curve and plot the wavelength as heat is increased.
6. Plot output till death (standard handheld ambient temp)

 

[IsaacT]

Apex, I might add to set current to threshold and plot wavelength there. A lot of people may want to run these lower if it means a lower wavelength. (like MEEEEE )

 

[ApexProxy]

Threshold would be included in the plot up to spec, at least I would include it. Threshold current all the way to Spec current. That would give you the ability to see right where the wavelength shifts and where it would be worth running in each persons eyes.

 

[IsaacT]

Ah okay, then yes! Hail ApexProxy's suggestion! Huzzah

 

[ApexProxy]

updated to reflect the specification

 

[ARG]

I agree on the room temperature handheld resembling setup for the tests.

Procedure I would do:
1. Plot output from threshold up to spec current (standard handheld ambient temp)
2. Plot wavelength from threshold up to spec current (standard handheld ambient temp) - overlay on to #1 plot
3. Plot output up to the efficiency curve (standard handheld ambient temp)
4. Plot wavelength up to the efficiency curve (standard handheld ambient temp) - overlay on to #3 plot
5. Retest output to spec current to verify no damage (standard handheld ambient temp)
6. Set the diode to the peak efficiency curve and plot the wavelength as heat is increased.
6. Plot output till death (standard handheld ambient temp)

I like that idea, but no tests for wavelength when cold? That will have a bigger effect on the wavelength than the current I imagine.

It seems the popular vote is plotting out output till death at ambient temperatures, so that's what I will do.

 

[ARG]

That's the set-up I plan to use for testing. No set screw nonsense
I can use the heatsink (28g) as a standalone to simulate a handheld heatsink, and I can also use it with the 10mmx10mm TEC and the heatsink to chill or heat it & take a wavelength reading. Temperature readings will also be taken with a thermocouple taped onto the back of the diode with kapton tape.

 

[Tw15t3r]

Nice setup! It looks like you could definitely plot many graphs with that!

A Temp (x-axis) vs. Wavelength (y-axis) at low currents (to avoid stress) would be nice, since you've got a TEC setup!

 

[ARG]

Just need to drill some holes in the heatsink to clamp the TEC between the two heatsinks.

I am planning on doing a temp vs wavelength chart at 180mA and 90mA, maybe another at threshold current because I'm curious

 

[ApexProxy]

I didn't include super low temp wavelength readings mainly because its basically an artificial scenario that few would ever get and its not something anyone would ever get in a handheld and around here I'd imagine a vast majority of those diodes would end up as handhelds.

Thats one tiny TEC bro. What is its rated wattage?

 

[ARG]

Doesn't have to be big, diode only makes 1.5W of heat. The TEC is rated for 4W, I can fit two under there for 8W if need be.
I know Bloompyle is interested and I'm interested in a chilled reading, so I'll include that as well.
LPF isn't just for handhelds, though that is mostly what goes on here and I think a good chunk of these diodes will end up being in a lab style situation, and plus if a really low wavelength can be achieved people would want to know so they can chill to diode for the super low wavelength

Plus, the more data that is possible to acquire with one diode, the better

 

[Grix]

I'll give $10 towards the second diode.

 

[ApexProxy]

Doesn't have to be big, diode only makes 1.5W of heat. The TEC is rated for 4W, I can fit two under there for 8W if need be.
I know Bloompyle is interested and I'm interested in a chilled reading, so I'll include that as well.
LPF isn't just for handhelds, though that is mostly what goes on here and I think a good chunk of these diodes will end up being in a lab style situation, and plus if a really low wavelength can be achieved people would want to know so they can chill to diode for the super low wavelength

Plus, the more data that is possible to acquire with one diode, the better

True that. More tests is certainly not a bad thing.

BTW ARG, for say an M140, when diode running temperature is reduced what direction does the wavelength shift? I have heard opposing theories on it and figured if anyone knew it would be you.