Largest Manufacturing Technology Community on the Web
Stay Connected:
Finishing building a homemade fiber laser machine...
- Thread starter arkcnc
- Start date
- Replies 61
- Views 43,596
You have some pump diodes for a fiber laser. No precision metal cutting capability with just a pump diode.
This may be OK as your motion system is not capable of precision cutting with a fiber laser as mentioned before. Others who have fitted their plasma or router table with a fiber laser have found the cut looking more like what a rabid rat would accomplish.
This may be OK as your motion system is not capable of precision cutting with a fiber laser as mentioned before. Others who have fitted their plasma or router table with a fiber laser have found the cut looking more like what a rabid rat would accomplish.
A company I used to work for built a CNC ultrasonic knife. Well, bought the knife, built everything around it. The system was cutting multiple layers of unidirectional prepreg and worked quite well.
Strostkovy
Titanium
- Joined
- Oct 29, 2017
I'm a fairly novice laser operater, but wouldn't the pump diodes have to have equal or better beam quality than the fiber that just amplifies it? Am I missing something?
The servos that I am using are high precision ones (40,000 cpr)
You really think that I going to spend that kind of $$ without investigate or buy just low quality components to save some pennies???
I been building motion NC systems for loooong time...
I invite you to watch my videos. My youtube channel: Ark Channel
Mechanical I can hold easily .00025" tolerance with these machines.
Strostkovy
Titanium
- Joined
- Oct 29, 2017
700 watts of coupled diodes is just as bright as a single diode amplified to 700 watts. There may be some cutting difference due to them having slightly different wavelengths and not adding into a coherent beam, but for situations like this delivered power per unit area is pretty much the biggest factor.
There is no way any laser can produce a beam finer than its pump source, and I doubt the coupling is much worse than the distortion caused by the amplifier.
Milland
Diamond
- Joined
- Jul 6, 2006
- Location
- Hillsboro, New Hampshire
I'll tell you, that was one skilled rabid rat that cut that part. Give that rat a raise! But, uh, don't let it bite you...
Not just that, but its also well worth remembering most of us here posting home made machines have well above the average in house manufacturing capabilities compared to a typical zone member. Can't speak for ark, but in my case the base most of my machines get built around is not extruded alu section with roller bearings running on it but serious fractions of a ton of heavy steel plate with precision ground surfaces and industrial preloaded hiwin ball way systems, positioned by good preloaded twin nut ball screws, hung on pre loaded angular contacts, you know just like a "real commercial CNC"
What might look like a simple zone style gantry will probably have multiple times the cost in raw materials alone in it, but the results really can be in a different ball park accuracy wise. Equally for most of us here with say conventional machines accurate even to the thou range, its relatively easy to then make cnc's that are at least a magnitude more precise.
You are absolutely right!
I have both on my system; acceleration and precision!
Hahahaha. Yes it is! very skilled rabid rat!!!
I will post it, let me record one
So, I'm sorry for resurrecting this relatively old thread, but it happens to be the second Google result for "diy fiber laser" and it's irking me that it may give people the impression that this is the route they should take.
Firstly, everything @IkeHarris has said is factually accurate and pretty much everything @arkcnc has said is incorrect. @arkcnc, what you have are continuous wave multimode fiber coupled diode lasers - you have not constructed a fiber laser. Your lasers have a spectral width of about 5nm centered at 976nm (which happens to be the peak wavelength Ytterbium absorbs). This relatively wide spectral width is going to impact your ability to achieve perfect focus, and you are not going to be able to pulse the beam correctly to achieve the high energy density that fiber lasers are known for. With that said, you will be able to cut thin plywood.
A fiber laser is constructed using pumping diodes, just like what you have, but they are used only to provide an energy bias to doped fiber optic cable. A single mode pulsed laser diode is then used to add to that bias and elevate the overall energy in the doped fiber to initiate lasing. This single mode diode needs to be about 5% of the overall power (read: expensive), and its pulses are what dictate the temporal aspect of the fiber laser's output (very important to achieve ablation versus simply melting the material).
The most significant differences between an actual fiber laser and the fiber coupled diodes @arkcnc has are that a fiber laser lases at a very specific and narrow spectral width and - most importantly - fiber lasers can dump all that biased energy in a very short period of time. With 700 watts of pump diodes, if you modulate the diodes at a high frequency all you're going to get is a fraction of the output power based on the duty cycle - it's always going to be, at maximum, 700 watts. However, if those 700 watts are used as bias in a fiber laser and then lasing is controlled by a single mode pulsed diode, an amount of < 700 watts can be dumped over a very short period of time which results in a substantial increase in energy density.
You can imagine the "fiber" part of a fiber laser as an optical capacitor of sorts. In the same sense that a 9v battery may tingle your tongue when licked, using that 9v battery to charge a capacitor can result in an output capable of spot welding.
Firstly, everything @IkeHarris has said is factually accurate and pretty much everything @arkcnc has said is incorrect. @arkcnc, what you have are continuous wave multimode fiber coupled diode lasers - you have not constructed a fiber laser. Your lasers have a spectral width of about 5nm centered at 976nm (which happens to be the peak wavelength Ytterbium absorbs). This relatively wide spectral width is going to impact your ability to achieve perfect focus, and you are not going to be able to pulse the beam correctly to achieve the high energy density that fiber lasers are known for. With that said, you will be able to cut thin plywood.
A fiber laser is constructed using pumping diodes, just like what you have, but they are used only to provide an energy bias to doped fiber optic cable. A single mode pulsed laser diode is then used to add to that bias and elevate the overall energy in the doped fiber to initiate lasing. This single mode diode needs to be about 5% of the overall power (read: expensive), and its pulses are what dictate the temporal aspect of the fiber laser's output (very important to achieve ablation versus simply melting the material).
The most significant differences between an actual fiber laser and the fiber coupled diodes @arkcnc has are that a fiber laser lases at a very specific and narrow spectral width and - most importantly - fiber lasers can dump all that biased energy in a very short period of time. With 700 watts of pump diodes, if you modulate the diodes at a high frequency all you're going to get is a fraction of the output power based on the duty cycle - it's always going to be, at maximum, 700 watts. However, if those 700 watts are used as bias in a fiber laser and then lasing is controlled by a single mode pulsed diode, an amount of < 700 watts can be dumped over a very short period of time which results in a substantial increase in energy density.
You can imagine the "fiber" part of a fiber laser as an optical capacitor of sorts. In the same sense that a 9v battery may tingle your tongue when licked, using that 9v battery to charge a capacitor can result in an output capable of spot welding.
TravisMitchell
Plastic
- Joined
- Mar 8, 2023
I was very intrigued by this thread, but it seems the project was never finished or published. For anyone interested, I took it upon myself to build a fiber laser cutter. If you’re interested in building your own, you can see how I did it here
xg1745
Plastic
- Joined
- Jan 17, 2011
- Location
- AB, Canada
Since this thread seems to back already, curious if anyone building a fiber laser is still around. Considered doing something similar on zero budget: got 8 100W diode array driver units (diode with power supply, timing, cooling and fiber output) . These are 976nm range output.
Could I dump the fiber leads into a resonator cavity from an old lamp pump YAG? In theory it should be more efficient at absorbing the 976nm IR from the diodes than everything the flash bulb made. Not sure what final output would be but should be far better than the flash pump (total W in to lasing W out).
Also have a 1Kw flash pump YAG laser with couple spare rods sitting around.
If i could get 200W+ 1064nm output all the junk could be put to some serious use.
Could I dump the fiber leads into a resonator cavity from an old lamp pump YAG? In theory it should be more efficient at absorbing the 976nm IR from the diodes than everything the flash bulb made. Not sure what final output would be but should be far better than the flash pump (total W in to lasing W out).
Also have a 1Kw flash pump YAG laser with couple spare rods sitting around.
If i could get 200W+ 1064nm output all the junk could be put to some serious use.
Xg,
You would get very little output for your efforts. Nd:YAG wants 808nm for pumping. The diodes you have are for ytterbium doped fiber pumping.
Ike
xg1745
Plastic
- Joined
- Jan 17, 2011
- Location
- AB, Canada
Are Yb:YAG rods common or easily obtainable? Seems like it could be cheaper to just swap the rod in the pump chamber than to get a combiner and proper Yb doped resonator fiber. Also I think for experimental cobbling safer cost wise as the fiber/combiner can easily fail from contamination or poor lapping. Thank you for the info.
