Cooling LPC-815 Diodes

I have run some initial experiments with a LPC-815 diode mounted in a heat sink which is in turn cooled by a recirculating chiller. By simultaneously measuring the output and the wavelength at several temperatures, I was able to learn a few things:

1. The wavelength shift with temperature is in fact pretty close to the published redshift of 0.38nm/C for temperatures between 18C and 63C, but is closer to .25nm/C blue shift from 18C down to -10C.
2. The power output in mw/ma is only aprox. 5% better at -10C than at 18C (room temp.)
3. The max. sustained(hours) power was tested at -10C and was 415 mw at 700 ma. As the current was increased to 800 ma the wavelength steadily increased up to 6 nm above baseline (aprox 660nm), but the power remained flat at 415mw. Then at 810ma the diode failed.

I plan to rerun this test within the next week with a TEC between the "water" block and the heat sink to get to significantly lower temperatures. Is anyone aware if this kind of testing has been done already or if other diodes, especially the 445nm diodes, have been looked at.

:beer:
I always love hard data that result from experiments.
I'd love to see pics, too.
+rep for you.

What's your equipment to measure wavelength?

How close to the diode did you measure the temperature?

Temperature shifting the wavelength of diodes is neat, I've gotten some red diodes pretty orange by cooling them with a 2 stage cascade refrigeration system to -100F but I don't have a spectrometer so I don't know how far I actually shifted the wavelength

I don't have a convenient camera, but Ill work on that over the next few days. The set up is pretty straight forward. The diode sits in a brass axiz press fit host and is then screwed into the 11mmx0.5mm thread that I sank aprox 5mm into the 12mm Al plate. The thermistor sits aprox. 4mm lateral to the diode and like the diode aprox 25mm from the closest edge of the "water" cooling block. The beam passes through an axiz glass lens which like the diode I obtained from Hightechdealz

I sent the beam through a 50/50 metalized beam splitter (from Ross optical) then to an Optic 300um slit monochronometer as well as to an Ophir 0-150 W power meter. The cooler was a Julabo 450 recirculating chiller and power came from a Bk analog 3A DC power supply.

Awesome setup! Please keep Us abreast of Your testing. Surprising that you hit 800ma before the LOC failed.

Interesting.. Though this diode seems to be fairly inefficient compared to the last lpc-815 build I did. Mine was doing 330mW at 500mA. That build has been alive for months now with no noticeable degradation. It, and the information in this post has made me wonder if the power handling capabilities of these has improved and nobody's pushed any of them recently to have noticed.

The last of the 3 LPC-815 diodes that I received from Hightechdealz is now installed into the heat sink with a 16 V/22Amp 62mmx62mm TEC sandwiched between the heat sink and the "water" cooling block. Today I will run this diode as I did the last, but will extend the low temp. much further. I'll note any differences between these two at similar settings, but I"l try to avoid killing this one as it might be a few days before I get any more.

Those are pretty cool results!

I wonder why the blueshift is much lower below ambient temperature though. Where are you measuring the temperature, at the heatsink, the diode or the actual laser die?

Afaik the voltage drop should change linearly with temperature over a fairly wide range, and i suspect the problem might be the actual dye temperature. You describe how the wavelength goes back up at increased current levels at -10 degrees, and the only reason it should, is that the dye is heating up.

The temp. is measured in the bulk of the heat sink about 4mm lateral to the module which is screwed tightly into the heat sink with thermal compound. Nevertheless, the temp. is measured the same way above and below ambient. I don"t mean to imply that something magical happens at room temp. these are trends that when averaged seem to show a greater change in wavelength/C at higher temps. I'll post actual data as soon as I get some. My last LPC-815 either got cooked in shipment or I did it when I installed it. The tests will have to wait a few days for a few more diodes. Damn.

Alright, perhaps charting them out would make it more obvious

As far as measuring the actual dye temperature, both the wavelength emitted and forward voltage at a give current would actually be good measurements of that.

Thanks for the advice. The voltage at each temp and at each current is easy to provide. I'll just post a table. Incidentally, I'll be doing the same set of tests for the 445nm diodes I expect from DTR in a few days.

I don't have a convenient camera, but Ill work on that over the next few days. The set up is pretty straight forward. The diode sits in a brass axiz press fit host and is then screwed into the 11mmx0.5mm thread that I sank aprox 5mm into the 12mm Al plate. The thermistor sits aprox. 4mm lateral to the diode and like the diode aprox 25mm from the closest edge of the "water" cooling block. The beam passes through an axiz glass lens which like the diode I obtained from Hightechdealz

Click to expand...

I went ahead and tested the cooling system prior to threading in the diode and the thermistor from my Fluke and my Taylor panel mount and both pegged at -40C. Any one know a good (inexpensive) source for a thermometer that will reach -100C or more?
Thanks

A thermocouple and an arduino could go that low.

I just blow on my diodes when they get hot. Who needs a heatsink?

Arduino?