TEC cooled Diode testing station, suggestions?

[FINISHED] TEC cooled Diode testing station

EDIT: 7/3/2014

So, I'm not sure how active this forum is still, but I miss it nonetheless so I'm (trying to be ) back. I never posted the results of this project, so here they are:

Laser Diode Testing Station | IEEE @ UCSB

Basically, it works. If someone were to compile it into a more portable setting, I'm sure it would be a very profitable and helpful venture. I assembled everything for a minimal price, and with a little expertise you can circumvent the need for the arduino and stick a microcontroller chip on there yourself.

Best,

amk
------------------------------------
So Im putting together my long awaited (by myself =P) Diode testing station.

So far I've taken some ideas from different projects, including Jayrob's TEC icebox.

Parts I've ordered so far:

• 91.2W TEC (Peltier) 40mm
• Aluminum Heatsink for the peltier
• Centrifugal Blower, same one Jayrob uses in his TEC Icebox. For the Peltier heatsink
• 5v Voltage regulators
• Some assorted small heatsinks for various parts that might need sinkin
• I posted on flaminpyro's thread to see if he could make me a block similar to the meridith mounting block, only with a set screw to fit aixis modules, and 40x20x40mm dimensions.
• Fleabay Ammeter for measuring diode current draw, LCD style
• Fleabay Voltmeter for measuring Voltage across working diode

So I haven't actually worked with TEC before, just read about it. Anything I might need to know about regulating/running/rules of thumb?

What kinds of conductive adhesive do I want to use? Arctic silver adhesive that Jayrob uses in his TEC Icebox? From radioshack I presume?

also, what is best to smear around the aixis module that is about to be tested before inserting it into the block, in your opinion? I'm personally not a big fan of arctic silver.

Also, do I want like a sheet of copper on top of the peltier and then stick the mounting block onto the copper? Or stick the mounting block straight onto the peltier itself?

As for the LPM, I will be using the old Knimrod DIY Thermal LPM with graphing capabilities. I will most likely just have the setup and the LPM head all in a box. with access only to the mounting block setup. LCD display of the power reading as well as computer output.

Please feed me suggestions,

AMK

When I use them, I just use a bit of thermal paste between the TEC and a heat sink.

For what are you using the 5V regulators? These take 12V or so.

Well you don't want to run the TEC full blast, especially not one that big or your diode will be frosting over with condensation.

How bout a clamshell approach to heatsinking the aixiz? You could leave the bottom half attached to the tec and just bolt the top half on top of the aixiz, or just stick something heavy on top of it. I dunno how many diodes you'd be going through, but if its 10's per day, clamshell might be better.

Besides, then you'd have more contact area between aixiz and heatsink, mathematically (though in practice, I reckon the tolerances are close enough that the aixiz still touches 1/3 of the inner surface with the set screw)

BShanahan14rulz said:

How bout a clamshell approach to heatsinking the aixiz? You could leave the bottom half attached to the tec and just bolt the top half on top of the aixiz, or just stick something heavy on top of it. I dunno how many diodes you'd be going through, but if its 10's per day, clamshell might be better.

Besides, then you'd have more contact area between aixiz and heatsink, mathematically (though in practice, I reckon the tolerances are close enough that the aixiz still touches 1/3 of the inner surface with the set screw)

Click to expand...

Hmm, so you are saying have the mounting block 40mmx20mmx~15mm, and have part of the aixis module touching the peltier itself? Am I understanding it right?

Also yea, I had read in a few places that I don't want the diode to go super cold especially really fast because condensation might damage the diode itself. That's one of the reasons why I thought 5v would be enough, but if I am wrong I could always just use the regulator on the driver (I think I might use a flexdrive, if not either an AMKdrive or a rckstr). I don't really need/want compact though, I want something stable and easy to adjust current. I think I might just have the standard AMKdrive (317) with a rather large pot attached to a plastic knob that sticks out of the case. This way I can easily adjust mA without a screwdriver while testing.

Also, would one of those LCD's on ebay hooked up in series with the batteries be workable so I can see mA on the fly when adjusting, without the need for a multimeter?

Also need actual design suggestions, what box/enclosure/name.

i did a quickie test yesterday with my tec and ddl driver set at 1500ma with 12v input.. it was way to cold for a diode so that is good news if i plan on using it with batteries. my tec killed one of my rkcstr's... also i plan on mounting all of my stuff on an aluminum base and mount the tec to the base near my module.
also i will first be having my tec's heatsink be only passive cooling.
michael

Just don't do what I have seen some members here do and use metal screws to sandwhich the TEC between the hot and cold plate Doing that means all your TEC power just gets wasted, which produces more heat than if you were just to run the diode on a block of metal.

Things said:

Just don't do what I have seen some members here do and use metal screws to sandwhich the TEC between the hot and cold plate Doing that means all your TEC power just gets wasted, which produces more heat than if you were just to run the diode on a block of metal.

Click to expand...

Rofl, and MAJOR typo in Origional Post, I meant Jayrob*'s ICE box, Not Kenom's >_<

fixing it now.

It's all good amk...

Hey I've built a new one since you have been around... I still have the first one. But I have a 12X one with FlexDrives now:
http://laserpointerforums.com/laser_pointer_forums_3/forum/showthread.php?t=37239

Jay, I had a question, In that 12x build did you have the meridith mounting block glued straight onto the peltier with arctic silver adhesive?

My peltier doesn't have that copper plate that yours has, is it safe to glue my mounting block straight onto the cold plate using some arctic silver adhesive?

Honestly, I'm no TEC expert but my experiment does work well...

And I would say yes, you need to use Arctic Silver thermal adhesive for the mounting block to the peltier.

Keep in mind that if you use a Meredith block, the hole is 1/2". I have 1/2" copper modules if you want to use an AixiZ threaded lens like my 405-G-1.

Not sure if you read it in my thread. But when I went to the 12X, I decided not to use a Meredith module because the Meredith is not a 'pressed in' diode. But the diode is only held to the module with spring tension.

So that's why I made a custom copper module for a pressed in diode. And I had to get 1/2" material to have it fit the Meredith block that was already built...

Hm.. I would do that, but the only problem is that then I have to either take diodes in an out of the copper module when I test them, or I would have to buy a bunch.

That's why I needed something that I could slide the aixis module into, because chances are 65% of the time the diode will stay in that aixis module afterwards.

Flaminpyro is "on the road" though... so If I can't get a quote for the copper block I think I might just do something with the meridith.

Is there a way to make like a ring that would fit into the Meridith mounting block so you can slip an aixis in there? But that would just screw around with heatsinking...

I donno, I need to put more thought into this.

Regards,

amk

I would say that it's better to make a block to fit an AixiZ module then...

HM, since I will need to be switching out diodes frequently, would it be workable to solder some of These onto the driver circuit then tin the Diode leads and plug these onto the diode? Or would that be too dangerous for the diode?

I'm not too sure how flexdrives are made, is there a capacitor in parallel with the diode? If there is one then possible quick disconnect/connect from one of these would fry the diode...

But if this is workable it would make everything cleaner, faster, and the driver wouldnt be in contact with a soldering iron a few times a day...


regards,

amk

D'oh or I could just solder the female plug to the driver board, and solder on/off the male plugs, that way I accomplish the same thing with less noob hazard...

Not sure if he still has any, but these would probably work good, with a TEC on a cpu heatsink with a fan, just moving aixiz modules in and out.

Prototype said:

Not sure if he still has any, but these would probably work good, with a TEC on a cpu heatsink with a fan, just moving aixiz modules in and out.

Click to expand...

Hmm, You are right...

If I can't get ahold of flaminpyro for the copper block, that would be a great alternative..

Or better yet, jay, if you have some spare time do you think I could buy a custom machined aluminum block off of you? (I'm preety sure you said you don't do copper, right?)

All I would need is basically a meridith module only made to pop in/out an aixis module, with a set screw on the top. Block no bigger than 40x20x40mm

Thanks in advance,

amk