Welcome to Laser Pointer Forums - discuss green laser pointers, blue laser pointers, and all types of lasers

Buy Site Supporter Role (remove some ads) | LPF Donations

Links below open in new window

FrozenGate by Avery

Just got these IR dump diodes

LightSourceBlue

New member
Joined
Sep 4, 2024
Messages
21
Points
3
I managed to get a deal I could not pass on these 4 x 400W (claimed) IR pump diodes.

I intially tested one prior to any cleaving of the coupled fiber optic cables, and as I started to increase the current (I started low) something started to burn in the fiber optic cable so I quickly stopped and waited for the cleaver I ordered. Today the cleaver arrived and I stripped & cleaved the cable (did not polish though), and again first go something burnt in the cable so I stopped again but this time cleaned the cable as it looked like it was just some left over cladding burning up. Once I cleaned that up, the cable isn't burning up anymore and can burn paper when close to the end of the cable - however not much else even at full power (33v 20amps). I'm hoping nothing has reflected back into the diode and damaged anything when there was some burning in the cable on the first two attempts.

Is it possible to a clean cleave and still have output safely without a plug or polishing?

I'm still waiting to receive the right size SMA 905 ends (the first ones I think are too small - 125um, the core is apparently 220um and estimated the cable is 360um), and also epoxy to do a proper cleave & connection so I can get a lens on.
 

Attachments

  • WhatsApp Image 2024-10-13 at 02.24.10.jpeg
    WhatsApp Image 2024-10-13 at 02.24.10.jpeg
    521.4 KB · Views: 14





Those are pump arrays made up from many diodes knife edged inside then fiber coupled and you MUST cleave and polish the end of the fiber, they simply cut the fiber when they remove them from service, likely from banks used in large industrial fiber laser systems.

The end of the fiber must be cleaved and polished otherwise it will burn backwards when you power it up.
The end of the fiber becomes an optical instrument, like a lens, if not polished well you will get poor results, like running a laser beam through an opaque diffuser.

You also need to use thermal compound and a heatsink if running for more than a few seconds.

Make sure to wear proper laser safety goggles any time you are working with lasers over 5mw, especially NIR lasers.

Note as those diodes inside the arrays are used they may suffer from lower efficiency which means heat buildup and dimming, so try running the array at less than full rated current if you notice it starts out strong then gets weak, also use a heatsink.
 
Last edited:
I would probably recommend getting a Fiber optic Power meter kit, possibly a splice connection check tool?

Is it possible to just buy some new cable and replace the entire run/loop?

Which cables are you referring too? The output cable or the cables within the box?

Just because you’re not seeing any burns after cleaving, doesn’t mean the process was done correctly. Are you using a manual cleaver or higher end ARC cleaver?

Remember, the cable is glass, any small imperfections will result in a major loss of output power and as you’ve seen, could result in the light escaping the cable path and burning the conduit. The process is like surgery in a clean room, every single step needs to be done perfectly.

I’ve been cutting glass for over 26 years, my father owns a glass company in Southern California, so I have extensive knowledge and experience with the material. Regardless of the size or application, glass is a tricky material.

Are all 4 units resulting in the same issue? And are you using any type of focusing/BE lens?

Those units should definitely be able to burn thru steel in seconds at 400W output..

Here’s a YouTube video of a professional performing some splicing, even he has to redo the process a few times, checking the results with his tools/meters along the way.


I would recommend looking up more videos on the subject, idk if it’s going to be more affordable to buy more splicing tools, or just buy premade/specific length cables for you’re project.

If all 4 units are the same, I would use a power meter/cable checker to find the best condition cables and then install them into 1 of the units to perform a test.
 
Another thing I’m noticing, the cables coming from your bench supply look like they’re way too small of AWG wire to run 20A thru, if those are the generic wires with alligator clips, I would recommend making a new set of cables to supply power to them. For 20Amps, I would recommend running nothing smaller than 10AWG silicone/copper wires. Don’t use the alligator clips, make a proper soldered connection, i would recommend using some Amass XT60 connectors so you can solder leads onto each box, and have 1 power cable from your bench supply. For this current level, I also recommend not using the standard male probes going into the bench supply, get some 10AWG sized, insulated Fork spade terminal connectors. Tin the tips of the wire, slide your fork spade terminal connector on, crimp properly, then finish it off by soldering it. Throw 2 layers of some shrink tube with glue on and you’re good to go.

I actually use 14AWG silicone/copper wires coming off my 60V/5.2Abench supply. Here’s pics of my bench supply leads, since I’m only running 5A, I still use alligator clips when connecting them to a driver for quick tests, if I’m going to be running any type of extended time tests, I’ll add some leads to the driver I’m going to run power too, then I can remove the alligator clips and use the probes.

IMG_5644.jpeg
IMG_5645.jpeg

You’ll also need to mount them on a proper sized heatsink w/ thermal grease and some cooling fans. I see you have them on an aluminum work block, so they NEED to be bolted down with paste between the unit and the block. If it were me, I would buy some big 40mmx80mm TEC plates to mount between the fiber box and the block for the proper cooling. These units were most likely TEC AND water cooled in their previous homes/.
 
Last edited:
Once you get your fiber end cleaved and polished you will likely want a fiber laser collimator lens..... it's best to put a termination on so you can attach your collimator, or you can rig something up but it needs to be aligned.


 
Lot's of good advice here!

I finally got the right sized SMA905 connector (330um in this case, I tried 440, 220um, before realising 330um was the right fit), the right glue (the first time round I didn't realise I need to wait 4 hours for it to set), polishing pads, cleaver, etc - and had my first proper attempt at polishing the fiber. It looked fairly smooth when I finished it, however when I turned it only and slowly upped the current to around 2-3 amps (10-15% of rated) it showed light for a short bit and the stopped. Prior to turn on the power, I checked the end and it looked good -I tried to use the microscope thing I bought but it doesn't have an SMA connector! I didn't take a close up before, below is only after - but definitely didn't look like this before. I think the issue could have been the final cleave after the epoxy had set (I did one prior with the bigger cleave which I think was ok), I didn't have the pen type cleave so I had to get to cleave a bit of an angle with the bigger cleave which is not designed to do it with a fiber connector on the end - this didn't look straight, so maybe the fiber cracked when I was polishing it. Or maybe I polished too far.

fiber burnt sma905.png


This is the instructions I followed to cleave and polish

Those are pump arrays made up from many diodes knife edged inside then fiber coupled and you MUST cleave and polish the end of the fiber, they simply cut the fiber when they remove them from service, likely from banks used in large industrial fiber laser systems.

The end of the fiber must be cleaved and polished otherwise it will burn backwards when you power it up.
The end of the fiber becomes an optical instrument, like a lens, if not polished well you will get poor results, like running a laser beam through an opaque diffuser.

You also need to use thermal compound and a heatsink if running for more than a few seconds.

Make sure to wear proper laser safety goggles any time you are working with lasers over 5mw, especially NIR lasers.

Thanks for the polishing tips, this helped - I slowly making progress to getting these running! 100% agree, I need to make sure these don't fry. I've got some thermal compound and heatsinks ready when I am able to run them long enough, I'm checking the heat after each time powering them on, and never leave it on for more than a few seconds. They are warming up a little bit (a tiny bit more than room temp), but nothing quick enough to warrant a heatsink yet.

I have been using broader spectrum safety classes - the same ones for when I'm using coloured diodes, but invested in some specific IR googles which are a bit easier to see through. Much higher OD rating as well.
 
I would probably recommend getting a Fiber optic Power meter kit, possibly a splice connection check tool?

Is it possible to just buy some new cable and replace the entire run/loop?

Which cables are you referring too? The output cable or the cables within the box?

I've been looking for a fiber optic meter kit that can do 400w+, I only have a TEC + multimeter which I saw off someone else on here do - which works for low power diodes, but concerned about using it on ~400w laser source. I might have found some used options.

It doesn't look like it's possible to replace the cable, they do look like the kind that can be easily replaced. I am not comfortable yet with trying it, for risk of not being able to use a cable going forwards.

The fiber optic cables I'm referring to, are the ones that terminate out of the pump diode, and then come out of it - but does not have a SMA connector on it, just a cut fiber cable.

Just because you’re not seeing any burns after cleaving, doesn’t mean the process was done correctly. Are you using a manual cleaver or higher end ARC cleaver?

Remember, the cable is glass, any small imperfections will result in a major loss of output power and as you’ve seen, could result in the light escaping the cable path and burning the conduit. The process is like surgery in a clean room, every single step needs to be done perfectly.

I’ve been cutting glass for over 26 years, my father owns a glass company in Southern California, so I have extensive knowledge and experience with the material. Regardless of the size or application, glass is a tricky material.

Are all 4 units resulting in the same issue? And are you using any type of focusing/BE lens?

Those units should definitely be able to burn thru steel in seconds at 400W output..

I'm using a cleaver like the one below, but I just bought a pen style one and I should get it in a week or so. Do ARC cleavers typically do cleaving and splicing? And don't require any polishing once done?

I can definitely see how hard this is after the latest attempt, I'll have to keep trying - I'm making sure to keep the current at threshold and for a much shorter period of time until I get the cleave/polish & SMA connector setup correctly.

71yLTRvdTQL.__AC_SX300_SY300_QL70_FMwebp_.webp

If all 4 units are the same, I would use a power meter/cable checker to find the best condition cables and then install them into 1 of the units to perform a test.

It would be great if I could change the fiber cable easily, definitely would prefer that!
 
You don't need 400W power meter. just couple a bit of light out using microscope slide and measure that.

Don't even try replacing fiber on them, you'll trash the module.

Are this Md3+ or Yb3+ pumps? I guess the later
 
You don't need 400W power meter. just couple a bit of light out using microscope slide and measure that.

Don't even try replacing fiber on them, you'll trash the module.

Are this Md3+ or Yb3+ pumps? I guess the later

As in measure a quarter of the beam or something along those lines? Just to see that's it's within the scope of what it should be as opposed to an accurate measurement?

I'm not sure on Md3+ or Yb3+, I know they are 915nm IR.
 
Ah, ytterbium laser. Not nearly as interesting anymore :(

Not quarter. Fresnel reflection form glass at nearly 0AOI should be around 4%, so if you insert a single microscope slide into the beam, you should get 8% reflection, attenuating 400W to mere 32W. If your power meter can't handle that then attenuate further with second reflection
 
Ah, ytterbium laser. Not nearly as interesting anymore :(

Not quarter. Fresnel reflection form glass at nearly 0AOI should be around 4%, so if you insert a single microscope slide into the beam, you should get 8% reflection, attenuating 400W to mere 32W. If your power meter can't handle that then attenuate further with second reflection

That is good to know, that's definitely worth trying. Is it just a standard/any microscope slide or does it need to be a specific kind? What happens to the remaining 92% of the light? Is it reflected, or will the slide heat up? Assuming not the second as I would imagine the slide wouldn't last for long.
 
Still if you'll ever be throwing them out, I'll find a good home for them.
 





Back
Top