Testing an old 20watt laser diode (fiber coupled) / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

Testing an old 20watt laser diode (fiber coupled)

first previous 6 7 8 9 next

Move Thread

LAN_403

Carbon_Rod

Sat Oct 18 2008, 05:25AM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

Steve is right about this issue...
However, the internal primary optics and bar are usually bonded to the conductor plates (not the case its bolted into).

What is the Vdrop across the bar during operation?

Indium Foil on a large water cooled copper slab is likely the best option for a twisted flange.

EEYORE

Sat Oct 18 2008, 05:28AM

Registered Member #99 Joined: Thu Feb 09 2006, 06:10PM
Location: florida, usa
Posts: 637

Carbon_Rod wrote ...

Steve is right about this issue...
However, the internal primary optics and bar are usually bonded to the conductor plates (not the case its bolted into).

What is the Vdrop across the bar during operation?

Indium Foil on a large water cooled copper slab is likely the best option for a twisted flange.

Hello,
The voltage drop across the laser at about 28amps was about 1.8volts. I didnt really tighten the laser down all that well, just used regular flathead screws and gave it a good twist. (Screws seemed kinda soft and tore without that much effort).
When running it, it seems to be just fine. Nice round red spot on the box before it begins to smolder...
Matt

...

Sat Oct 18 2008, 07:07PM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

The laser is fine, not sure if you guys have seen what is inside of one of these lasers, but there is a 'standard' 1" square by 3/8" thick copper block that has the laser diode soldered to it, that has a thin layer of normal white thermal compound and is bolted down to the main plate. You will have a hard time putting enough force on the lower plate to put enough of a bow in it to then put enough bow into the secondary copper block bolted on top to bow enough to crack the laser bar. You might be able to put enough of a bow in the laser to throw off the alignment of the fiber coupling, but the fibers are so big that even moving them a few mills would have a negligible impact on output power.

here is a dead 30w one that have (you can see the bar has been vaporised, poor thing)

At this point indium foil might not be a bad idea, although I am not even sure where to get a small piece of indium foil. I don't think places like mcmaster stock that stuff, you might try a thermal company like indium corp and beg for a sample. If you are feeling adventurous I could send some raw indium metal with your fiber/googles, you would jsut need to find a way to roll it into a sheet (it it malleable enough for that to work)

The laser is fine, its just that making a good thermal contact isn't an easy thing to do.

Carbon_Rod

Sun Oct 19 2008, 05:15AM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

Mr ... ,
that unit appears to be missing it's collimating rod lens along the bar.
Are these manufacturer's units fabricated in this manner (factory calibration for pitch)?

mattrg2,
The 1.8vDrop is usually minimal at threshold currents. The 30-40watt bars I know usually stabilize at about 2.1vDrop at about 20A. Vsupply was adjustable from 1.6 to 2.3 for characterization, but 2.2vDrop is normally considered the Max. Note the limits may vary depending on the laser product.

Some Fiber Coupled Collimators:

EEYORE

Sun Oct 19 2008, 05:38AM

Registered Member #99 Joined: Thu Feb 09 2006, 06:10PM
Location: florida, usa
Posts: 637

Hmmm...I wonder why my voltage drop is so low? It ought to be well above threshold.
For indium foil :

Matt

...

Mon Oct 20 2008, 08:31PM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

The coherent fap packages don't use any lenses between the bar and the fiber, they just use fibers that are about the same diameter as the emitters on the bar. A 19 emitter 50% fill bar will have emitters that are about 250uM wide, so you put a 300uM fiber in front of it and you can actually get very good coupling. There might be a lens ground into the fiber itself, but there definitely wasn't supposed to be one on the bar.

I am not sure if the the fibers are aligned individually, or if the fiber assembly is made on a jig with the correct spacing and then aligned as one piece. It appears that the whole fiber assembly was soldered down, so it seems like they probably mount the bar down, put the fiber assembly on an aligner jig, add some solder, maximize power, and let the solder cool.

The sro-opitics indium wouldn't work for mounting a FAP since they only have 1/2" wide foil, and cap you at 10" You might just be able to use a few pieces, but I think you would probably come up a little short (and would only have one chance at attaching your laser).

The voltage drop for the laser sounds about right to me, I find that you will see most of the voltage drop above 2v is in your leads to the laser, not so much in the laser die itself. For a fap package there is a thick copper plate, so you only see the diode drop, which shouldn't be above 2v at 30a.

EEYORE

Wed Oct 22 2008, 03:14AM

Registered Member #99 Joined: Thu Feb 09 2006, 06:10PM
Location: florida, usa
Posts: 637

Well I decided to order a new heatsink. Found a guy selling 5 pounders on ebay that claims they are extremely flat...I also am bidding on another laser. Hopefully I win, as I dont feel like bothering with this one anymore for now. (Likely needs better sanding after I miss-sanded it). It will be something for a rainy day...Lets see if I can get lucky and win one for 100$ too!

Matt

...

Sat Nov 22 2008, 05:27AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

I found these guys on ebay, they might help you with you laser troubles a bit

Especially if you could find copper heatsink, you could solder that mounting plate down if you wanted

EEYORE

Sat Nov 22 2008, 05:55AM

Registered Member #99 Joined: Thu Feb 09 2006, 06:10PM
Location: florida, usa
Posts: 637

Hello,
I have one

))
I ordered a FAP800 40watt module from outback6. It had the TEC and all inside, but I took it out. The tec needs 48volts 7amps, and that just adds to the problem. (Just moves heat around, at the expense of more trouble finding a supply). Looks like I am going liquid cooling for this thing. (From advice on alt.lasers).
I will use a kit for water cooling CPU's. (Copper waterblock, radiator, pump, antifreeze). Copper heatsinks are costly and very hard to find. (Large enough for one of these lasers). There are some 60watt coherent laser diodes with water cooling I may go for too.

BTW_Picked up some laser goggles on ebay! 45$ for some used Thorlabs LG1's. The OD curve shows 808nm being about OD4-5 or so. They block UV and IR. I just got them today. Excelent condition. I turned on a 2.4watt 808nm fiber coupled laser diode I have and the red glow instantly disappeared. So neat how that happens! Now I feel safe. Any idea if these will work at 40-60watt powers? Some guy claims to have blinded himself with a 60watt 808nm laser despite using OD12 goggles. Isnt power reduction= 10^-OD times the power of the laser? I dont see how hes not just some idiot BSing me? Seems OD4 would mean 4mW from a 40watt laser?
Matt

...

Sat Nov 22 2008, 06:53AM

Registered Member #56 Joined: Thu Feb 09 2006, 05:02AM
Location: Southern Califorina, USA
Posts: 2445

He was probably using OD12 googles for 1064nm or the likes. Yes, od4 googles will turn a 40w beam into 4mw, although be careful to not burn strait through them!

Also, keep in mind the diodes with direct water cooling aren't fiber coupled, rather they give a flat line like 1-2" long by a few hundred uM thick.

Water cooling is great for these things, although make sure to get a good waterblock. I have had good luck with the swiftech apogee blocks, and since they have the new apogee GT out now (it is the same thing as the apogee but it has a bow in it to it get better contact to a processor, completely useless for our application) and even some other block that is completely new out you can get the old apogee blocks for pretty cheep. that would also let you run the tec a lot easier, since you can dissipate a lot of heat with a radiator. And they are cheep :) If you get fed up with the prices for pc watercooling parts there are tables out there that give you part numbers for heater cores out of cars (they are used to heat the air for the heater) that can be had for cheep online or almost free at a junkyard.

first previous 6 7 8 9 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob