I should have read the thread more carefully as I assumed this was about the sandwiched chip that steve-o was working with. It is a shame that the supply has now increased to $200 as blowing a few to learn their limits at $125 is at least a little like experimenting with an expensive diode. Anyway, I'll take a little different angle and say that coupling the heat out may not be that much of an issue. The interface of the crystals with the sapphire base is fixed by the manufacturer, but from here the heat (all the heat) will be removed by the bottom of the sapphire as the thermal conduction through the thermal epoxy within the module will be negligible in comparison. Now, if pumping with 5W and getting at least 1W of 532 out then your heat load through the base is 4W and given its dimensions of 7mm x 4.5mm then only 12W/cm^2 must be removed. This is easy. I'd use a thin layer of thermal grease, slide it around slightly to distribute it and after cleaning any smear that lies outside the interface then tab it in place with a few small buttons of thermal epoxy. This way if you want to remove it a razor to chip the epoxy or a small dab of methylene chloride to the tabs and 15 min. later your free again. And of course there is no pressure in this method.
The hot spot on these vanadate crystals can rise close to 500K and no damage occurs. Other than the internal epoxy softening and allowing the crystals to move out of alignment I don't think bulk warming will be too significant. Also a thermistor on the top will give at least some idea of what the temperature of the outer layer of the internal epoxy is doing. Of course thermal lensing and the effect on the beam quality as well as the conversion effeciency might become a problem well... that's the next hurdle.
Reply With Quote
