New overkill TEC heat-sink lab host advice

[Greenmechanic]

Nice little build you have there green, I really like the design.

Since a few have pointed out a TEC cooler would be un neccesery, and simply a large heatsink would be more suitable and a fan maybe the TEC is more hassle than it's worth.

I think moving the fan too the top would make more sense, but just too make it clear that would be sucking air out of the enclosure not blowing in, I think I want too keep the small screen in the design, it will make it a lot easier to set current etc if I want to swtich diodes around.

I've been searching around for some local CNC machinists without much luck so far, if anyone knows of anyone willing to do some one off products on the CNC please let me know.



I've been trying too work on some interior plans in between school work and have something going, I might post some sketches, since I don't really enjoy using CAD much and finding sketching easier and quicker.

Thanks for the ideas and help guys

Yes, since heat rises, putting it on the top would be the most effective.
You would need to have the air intake on the opposite side of the heat-sink. (air needs to flow around,or through it)
This means, bottom of unit, which means elevating it with "feet" of some kind.
Design wise, it makes more sense to put the fan in the back and air intake(s) in front.
Sucking from the bottom will pick up more dust.
Dust will settle down in the unit, if not covered up after every use.
The unit would have to be at least 1" off the surface for the fan to be effective.

 

[Hemlock_Mike]

You can cool the diode all you want but those tiny gold wires connecting to the die do have a current limit. If you look inside a low current fuse, you will see small wires. No amount of cooling will prevent a tiny wire from acting as a fuse.

HMike

 

[ossumguywill]

^Sure it will. Superconductors!

 

[Benm]

You can cool the diode all you want but those tiny gold wires connecting to the die do have a current limit. If you look inside a low current fuse, you will see small wires. No amount of cooling will prevent a tiny wire from acting as a fuse.

Is failure of the bond wires a common cause of death?

I've seen them melt in (RF) power transistors when operated (quite a bit) beyond manufacturer rated currents, but that takes more than a couple of amps to do. Surely there are several in parelel and they might also be thicker, but i doubt 500 mA or so would melt the bond wires in a laser diode.



For the heatsink with the little fan on the aixiz: How does it perform when you put it on its side, with room for air to flow in from left and right? Normally vertical fins are preferred, but the doubled flow area might offset that.

 

[Kenom]

OK, I have to interject a little bit of my knowledge here. We really need to take into account what another forum member has said about burning of optical fascets when taking these lasers way beyond what they are capable of outputting. You can push a laser diode as hard and as long as you want providing the bond wires stay intact, however, your burning the output mirror on the die of the laser diode. Hell even at 350mw on a loc there is going to be degredation... That's where that degredation comes from. It's not from the diode getting tired, or weaker, it's cause the basic fascets of the laser die are being burned.

 

[Asherz]

OK, I have to interject a little bit of my knowledge here. We really need to take into account what another forum member has said about burning of optical fascets when taking these lasers way beyond what they are capable of outputting. You can push a laser diode as hard and as long as you want providing the bond wires stay intact, however, your burning the output mirror on the die of the laser diode. Hell even at 350mw on a loc there is going to be degredation... That's where that degredation comes from. It's not from the diode getting tired, or weaker, it's cause the basic fascets of the laser die are being burned.

Thanks for the useful info, it's all good too know.

I'll definitely take this into account when I get round to ramping the current up on the diode. I'm not expecting it too last a whole lot of time to be honest, but for the price of LOC's now a days it's not a major issue if it does fail.

 

[Bluefan]

leting the fan blow or suck air through the module may add a lot of dust inside. Maybe you can put the module on the TEC and put the other side of the tec through the case to a heatsink. Now, the heatsink is outside of the case and the optics inside, and you can easily cool without "blowing" the optics inside.

 

[Gryphon]

^Sure it will. Superconductors!

I like your thinking