Laser Diode Question

This is a forum to share experiences and ideas about holography.
JohnFP

Laser Diode Question

Post by JohnFP » Fri Sep 03, 2010 11:15 am

Hey you diode guroos, got a question.

Here is a pic of my laser diode.

Now where it the most important place to cool, the top skinny part with the window or the bottom larger diameter area.

It seems to me as far as I can tell that the top part with the window is what houses the diode itself and the bottom is just a mount and possible a feeble attempt as a heat sink.

Or should I try to mount this such that the top and bottom parts are in contact with the heat sink, that is bore out a stepped hole that seats both the top and the bottom parts right against the walls of the bored out hole?
laserdiode.jpg
laserdiode.jpg (6.16 KiB) Viewed 3229 times

dave battin

Laser Diode Question

Post by dave battin » Fri Sep 03, 2010 11:31 am

hey John, the way i see it is to make as much contact on the shoulder as possible, i have been using basically a press fit housing which only makes contact on the largest diameter, and ive had no problems so far.................... :pray:

Joe Farina

Laser Diode Question

Post by Joe Farina » Fri Sep 03, 2010 12:52 pm

JohnFP wrote:It seems to me as far as I can tell that the top part with the window is what houses the diode itself and the bottom is just a mount and possible a feeble attempt as a heat sink.
I'm not a diode guru, John, but the opposite is true here. The base is where the heat sink should be attached to. The top part, the can (with the window) has poor thermal contact with the base. I'm working on a heat sink also, with Ahmet's help.

Press-fitting the rounded edge of the base into a soft metal like aluminum or copper might work well. I'm planning on using copper, which should arrive today. Ahmet tells me that good heat-sinking of these diodes is very, very important. I'm planning on using a finned aluminum heat sink measuring 6 X 10 X 2 inches. Then drill a hole so that the diode can rest on the front flat surface (and the leads can go out the back). Then add a thin piece of copper the same thickness as the base, with a hole drilled with a 7/32 inch drill bit, for press-fitting the 5.6mm diode. (5.6mm equals 0.22 inch, and 7/32 inch equals 0.2188 inch, so this might work well for press-fitting the diode, but I haven't tried it yet.) Insert/press fit the diode in the copper, then add a thin piece of aluminum about the same thickness as the depth of the can, and have 4 threaded holes in the assembly to hold it all together with screws. And of course use a liberal amount of thermal compound.

So there will be 3 areas of contact of the diode base to the assembly, the bottom of the diode (to the aluminum heatsink), the perimeter of the base (press-fitted into copper), and the top piece of aluminum, which presses down on the top surface of the diode base, and holds everything together. Ahmet mentioned that the most important "connection" is between the perimeter of the diode base and the surrounding metal (in my case, the copper) because that's the largest surface available on the diode base for thermal transfer.

Thieu

Laser Diode Question

Post by Thieu » Fri Sep 03, 2010 1:11 pm

Hi John,

This is an older diode I uncapped, held in a chuck of my lathe:
uncapped%20diode%202.JPG
uncapped%20diode%202.JPG (16.74 KiB) Viewed 3216 times
The gold plated part is all one piece, it's like a disk with a rectangular post on top, the diode rests on the front side of this post, it's the black block with another golden contact on top. This contact is connected to the pin on the right with a very thin gold wire. The bluish rectangle mounted directly onto the disk is a photodiode. It's base, like that of the laser diode is connected to the gold plated part. You can see the tiny wire connecting the diode to the pin at the left.

As you can see, the diode is in direct contact with the golden part, so the disk is indeed the most important part to cool. It's gold plated copper, so an excellent conductor of heat. No reason *not* to cool the cap, but it adds little cooling, not much compared to the base. The cap is made of steel, not that good a conductor. Plus the contact area is not that large. The cap is even thinner in reality than in the pic, because the top is burred from the sawing.

In the original beamer, the gold plated part is countersunk in a hole just as thick as the rim, and there's a thin metal plate on top to keep it in the hole. It's loosely press-fit, but not tight.

Btw, this pic is not of the blue diode, but the difference is only in the amount of bond wires, the length of the die and the omission of the photodiode.

Thieu

Laser Diode Question

Post by Thieu » Fri Sep 03, 2010 1:38 pm

The theory of heatsinks is pretty simple btw. Their performance is expressed in degrees C per watt. So if you got a component that produces 10 watt and you connect it well to a heatsink with a 2 C/W performance, the heatsink's temperature will rise 10x2=20 degrees above ambient. If you run this diode at 250 mA it will generate about 1.5 watt. To get an impression of the performance of commercial heatsinks, you can check this site:

http://nl.farnell.com/jsp/search/browse ... allpartial

You can choose one of the many commercially sold heatsinks and look up its performance to get an impression of the performance of your own.

Be aware though that this is the temperature of the heatsink, the temperature of the diode housing will be higher, and the temperature of the die even higher. Every boundary has its own thermal resistance and corresponding temperature difference. 1.5 Watt is not that hard to get rid off, even with a small 10 C/W heatsink the temperature will rise only about 15 degrees. At 1 A diode current, things are different, 6 watt would mean a diff of 60 degrees, so the heatsink temp would be almost 80 C which would likely kill the diode.

Forced air flow around a heatsink improves its performance *a lot*. That's how they can get away with producing >100 Watt of power in a relatively small heatsink in the beamer.

JohnFP

Laser Diode Question

Post by JohnFP » Fri Sep 03, 2010 6:38 pm

Wow, unbelievably, great, fantastic information. Thanks all. I love this forum!!!!!!!!!!

Now I know exactly what I need to do.

First of all, get a reem the that is flat and the diameter of the base.Then another that is the size of the cap.Reem the base size to the end of the sink. Ream the base size such that it is to the bottom of the base and is level with the sink and put a plate over it. Exactly the way I bought it but reverse the way it is facing such that I can use it without a lens and thus the diverging beam will not hit the fins.

Again, grazie mille.

Looks like i am entering a whole new field.

Which reminds me, I go to get those Ion lasers shipped out soon.

I wish i know someone in the area that was confortable with packaging them up for oversees shipping. I am just so scared on doing that properly.

Colin Kaminski

Laser Diode Question

Post by Colin Kaminski » Sat Sep 04, 2010 2:41 am

Reams are very accurate. So, measure the diameter with a caliper and choose a ream that is .003" larger, that should be a slip fit. Mc Master Carr will have one. You will regret a true press fit at .001" over.

Tom B.

Laser Diode Question

Post by Tom B. » Sat Sep 04, 2010 7:45 am

I bought a few diodes and mounting kits from these guys a few years ago:

http://www.optima-optics.com/ld_kit.htm#LDM

Thorlabs has similar (more expensive) kits, some with cooling etc.

Joe Farina

Laser Diode Question

Post by Joe Farina » Sat Sep 04, 2010 12:36 pm

Colin Kaminski wrote:Reams are very accurate. So, measure the diameter with a caliper and choose a ream that is .003" larger, that should be a slip fit. Mc Master Carr will have one. You will regret a true press fit at .001" over.
Thanks Colin. I checked http://www.mcmaster.com and saw two reamers, part numbers 8803A424 and 8803A426.

I also measured my 445nm diode with a digital caliper (resolution .0005 inch). This particular diode has a base diameter of 5.57mm.

5.57mm = .2193 inch. So if .001 is added to this, you would get .2203, if .003 is added you would get .2223.

The reamers from McMaster are .2205 and .2220. Which would you suggest?

Thieu

Laser Diode Question

Post by Thieu » Sat Sep 04, 2010 3:16 pm

Hi Joe, if something goes wrong and the hole becomes too large, you can put a piece of aluminum foil over the hole and press the diode through that. It will fill up the gap. Don't forget to make a hole in the foil afterwards :-) If the hole is too tight, you can always sand down a tiny little bit of the diode's edge, sometimes there's a burr. Or ream the hole with a tiny litte bit of sidewards pressure. So there's pretty much room for error if the fit's not correct right away.

Locked