Tec' ing temps?

[andy_con]

im looking at tec'ing a quad red setup, i want to tec the entire base plate to keep the mirror mounts stable.

anyone got any recommendations on what temp to tec to?

currently the baseplate is 22.5 degrees? i was wondering if 15 degrees is to cool and might cause condensation?

i know tec'ing is hard due to the different seasons and different room temps but....

[Doc]

Whether condensation will form depends on the relative humidity and temperature of the surrounding air.

Given that the humidity is usually around 90% over here I'd say that you will without doubt hit the dew point.

This could be avoided by air sealing the whole shebang with a dessicant pack inside of course.

[andy_con]

sealing isnt an option, so ill just have to raise the temp.

the max i can go is 20 degrees

[Doc]

You'll probably be ok at that.

For 27°C room temperature at a relative humidity of 60% (average club with a/c) the dew point would be 18.5°C.

[andy_con]

20 degrees it is then, open cans can take like 60-70 degrees so 20 should be fine

[The_Doctor]

I've thought on this a lot, and I think there are two main methods, one is to cool lightly and generally, the other is to cool hard in a sealed system, locally.

You've just covered the general light cooling scheme so I'll describe the other. First, 'sealing' might not have to be total, you only need to exclude bulk ingress of humidity. Second, you can cool the diodes directly, to well below freezing. You might have bad results with some diodes, but I bet most will do -40, no problem. (I love that number, no scale conversion needed for that one). Water will condense onto the mount but if there is very little vapour in the chamber it will sublime off any surface even if it's iced there, so long as there is a colder one nearby to migrate to. Examine the inside of any freezer if you are in doubt, as this is true even when the two regions are part of the same aluminium sheet! So if the diodes are running, the light energy alone will very slowly and gently 'blast' a clear path through while the diode mount temperature drops. Lenses absorb enough energy that their local heat will prevent icing on them unless the chamber has saturated air in it, and even then it will start to reduce one things settle. For best results, keep the optics chamber small, and at least try to seal it like a well-fitting door or window seal, it doesn't have to be hermetic because pressure difference between inside and outside will settle and not change a lot.

Mounting each diode on its own tiny TEC is best, for several reasons: It's going to be a higher thermopile density, so stronger cooling for a given volume, TEC's are HUNGRY beasts, so keep them small and housetrained, a type meant for sensors, something about 1 cm by 1.5 cm or less is good if you can make a mount small enough. Strong cooling should be kept to tiny regions to reduce errors. Maybe even run an extra TEC purely as a 'getter' for the humidity, and start that one several minutes before the others if you're paranoid about humidity in there. If you cool the entire combiner system you'll have ice on your mirrors, and thermal stresses and inaccuracies getting into things, better to keep those at ambient temperatures along with the baseplate. Even if the 'seal' is just a narrowing of any possible gaps, you can expect the chamber to remain ice free for a couple of DAYS, that would be a hell of a gig. Just open the hatch to let it dry out thoroughly afterwards. A scheme like that might be a LOT easier to handle that perfecting a seal that would impress NASA. Just use A4 stainless, and other metals not likely to suffer. Keep it free of dust so you don't get particles that could react with condensed water, though if you do it right it will spend 99.9% of its life dry anyway.

[sean]

I have a friend in Tokyo who was using a diode module TECed to 18 degrees

Within 10 seconds after power-on, he got condensation, he said.

so even 20 degrees is still critical I guess.

[shrad]

if you want to seal it, you might consider filling the box with a neutral gas

small inlet and outlet copper pipes, you leave them open, find a way to connect a CO2 or N2O cartridge to the inlet (cheap pressure bottles where you rip off the cartridge port), flush gas into the appartus, then bend the inlet and outlet pipes to soft seal the box

CO2 and N2O cartridges are easy to find in softball shops or food shops, and these gases are neutral and will prevent condensation as they push moisted air outside the box

and voila, cheap cryogenic setup for expensive diodes (you might then try to freeze 658nm diode to 642nm ones )

[andy_con]

thanks for all your replys guys, even if some were a little extreme!

i think ill have to just do some testing at the moment, if i run into any problems ill have to rework things.

the plan is to cool the base plate and the temp sensor will be attached to the base plate. so the diode holders will be a few degrees warmer.