Looking for the lost laser

[Bruce Boone]

I have been looking for years for a cutting laser that's powerful enough to get through around 2mm of titanium, yet can still be used in raster mode for engraving of 3D texture and for marking and deep engraving on metal. In the past, it seemed to always be a high powered CO2 laser, the size of a room, fed only by G-code for cutting, or a small CO2 or YAG for marking with no capability to cut metal. There seems to be no laser that I've found that can bridge that gap. I have a YAG welding laser, which has power to blast right through a nickel, but it maxes out at 20 pulses per second (or about 4 PPS at that power) so won't work well for cutting or especially engraving. I also have a Universal sealed CO2 that does a great job on wood, but not enough umph to get through metal. It would seem I need around 400 watts in the conventional CO2 route to do what I need.

I have recently seen a fiber laser at a jewelry show. They have a crazy small spot size, (.0003") so we were able to deeply engrave a titanium ring (about .030" deep) with a 20 watt laser. They run at something like 200kHz and are meant for marking metal or cutting fine stuff like stents. I am having that laser company do their homework to see if they could produce the laser I want. It would take a galvo head and being able to tie it to a rotary axis for cutting. For some reason, nobody else has seemed to do it. It seems like such a thing would be in a lot of demand. Fiber lasers are unbelieveably efficient (like 35%) and plug into the wall and are air cooled. It's a whole different ballgame when those become more popular.

I used to work on a 1500 watt CO2 laser, so I've seen what it takes to cut bike tubes. With a fiber laser, it might only be something like 80 watts or thereabouts. It doesn't seem to be such a tough thing to do, but I guess there are technical reasons having to do with frequencies that work best for certain applications.

Has anyone had experience with a laser that can bridge the gap?

[paul mott]

Bruce,

Unless I have misunderstood your post - it may be that the gap you are trying to bridge is too wide (a car that can compete in Indy or F1 yet still be able to pull a trailer).

Why not consider two machines ?

Paul.

[Viktor Voroncov]

Paul, excellent post :) and I absolutely agree with you. Practically impossible have high quality CUTTING and ENGRAVING in one laser.

[Lee DeRaud]

Why not consider two machines ?What he said.

Seems like for the kind of money it will take to get the cutting capability you want, the salesman might throw in a decent "engraving laser" (low/mid-power CO2) for free.

[Viktor Voroncov]

Lee, why to me sales managers offer only rubber mats for my car as addition :) I also want laser, even low powered :)

[Bruce Boone]

The cutting lasers I've seen are generally in the area of $150K + They also need lots of power like 3 phase 460 volts, take consumables in the form of CO2, Nitrogen, and Helium, and are generally huge machines. I don't have the room or the power for one of those. I have my shop in my home. I don't know if a sealed CO2 can get the power I need. I haven't seen such an animal.

The fiber lasers in 40 watts are around $60K, and it will do a lot of what I am looking to do. They are weak on cutting through the 2mm titanium though. I just hate to spend that kind of money and still not be the final answer. I would rather spend $80 or so and be done with it if there's a machine that can do it. My Universal does a good job with Cermark, but it's not that durable and gets worn off rings over time. I'd like a better marking solution that does deep etching and do away with the ceramic. I just have never found a cutting solution at anything near an affordable price and one that's made for small stuff and that I can power with 220V. It seems redundant to own 4 lasers, all with slightly different characteristics.

[rich shepard]

Don't know about lasering titanium as it burns just like magnesium. When we machined it at GE we had to have special fire extinguishers by the machine at all times. They had 4 in the building and if they were in use you did not even turn your machine on. you could be fired on the spot if the extinguisher was not there and you were cutting.
rich

[Bruce Boone]

It cuts pretty well with a laser. I used to cut bike tubes of it. It can use an argon shield gas but it's not necessary. Only thin chips will burn.

[Dan Hintz]

Bruce,

Good to see you on the boards (again)... still loving my ring :)

I think what you're running into when you say "in demand" falls short when you add in the money factor. As you've already noted, fiber lasers are still quite pricey, particularly for the garage mechanic. I would love nothing more than to cut 1/8" metal with an all-in-one machine, but adding another $80k machine to do so means I would need to be aiming for that specific market. Your business model fits that nicely, but most would not want to limit themselves to a few mm of cutting capability... if they spend that kind of dough on a machine, they don't want to limit their capabilities (i.e., jobs).

[Bruce Boone]

I've only recently learned about fiber lasers, but from what I've seen, they should make CO2 obsolete very shortly. Solid state, no cooling, 100,000 hour life expectancy, 35% or so efficient, tighter beam, etc. They are just high tech enough that I'm not comfortable making it myself. I've built 4 lasers and have the mirrored ruby rod and flashlamp for a ruby laser yet to be built. I've been into lasers since middle school. The fiber lasers are such a perfect idea; a long thin lasing medium with plenty of chances to have the light reflect and excite the right atoms; long and thin means it expels heat effectively. They can also use more than one diode to fire them, so power can easily be multiplied. It seems like this will be the answer I'm looking for, but the applications lab needs to catch up a bit.

[Dan Hintz]

I've only recently learned about fiber lasers, but from what I've seen, they should make CO2 obsolete very shortly.
Not likely, at least not anytime soon. The issue is with the wavelength... the 10.6um of CO2 works great with organics (wood, plastics, etc.), but is mostly reflected by metals. The wavelength of fiber at around 1um, however, is readily absorbed by metals but hardly touches most organics (a few plastics are readily marked by fiber).

Now, if they figure out how to get a fiber coating to lase at 10um (give or take), then we'll talk...

[Bruce Larson]

It cuts very well with a plasma torch, and a torchmate or similar can be had for reasonable costs.

[Dan Hintz]

A plasma torch doesn't have nearly the same quality nor resolution of cut as a laser.

[Bruce Boone]

The cut of a fiber laser is awesome. I believe Epson has one now. I still think they will take over the CO2 market soon enough when costs come down. They are bound to as there's not much to them; just an LED and a fiber for the smaller ones. No perfectly mixed gasses, high voltage, coolant lines, and perfectly aligned mirrors for the resonating chamber. They don't wear out or have to be refilled. The wavelength isn't that much of an issue when the power density is that good. Still plenty of power for stuff like wood and plastics, even in a 10 watt machine. The keyboards of computers are done with that wavelength. They are so fast that the whole thing is done in like a quarter second or so. That's with galvo mirrors though.

You would suppose that it would be as simple as hooking up a more powerful fiber laser to an existing XY table. Maybe it's not that simple.

[Dan Hintz]

Here's one way to prove it to yourself. Use a 30W CO2 system and cut some 1/8" plywood... now do the same with a 30W fiber system (assuming the fiber will even cut through that wood with any precision without significant charring). Neglecting the fact that the fiber system costs 5 times as much as the equivalent power CO2 system, I think you'll see why current fiber systems cannot even begin to compare to CO2 systems for a wide range of materials.

Wavelength plays a huge role, and it simply cannot be ignored. I've toyed with purchasing a fiber system (or at least modifying my current CO2 carriage to hold a fiber when I need the capability, which would be significantly less expensive since the only purchase would be the fiber/power system and some home-machined holders for the fiber at the carriage), but it's so low on the priority list I can't be much bothered with it other than occasionally Googling systems for daydreaming material.

EDIT: I also wanted to add that fiber systems do wear out. As with all semiconductors, the drive diodes have a finite lifetime and lose power over time. It is the same quantum mechanism used as in LEDs, which do not have the promised 100,000 hour lifetime the marketing geeks would like for us to believe. I cannot say I've looked into the lifetime of the frequency-shifting phosphors in the fiber itself, but the drive diodes definitely have a measurable lifetime.

[Bruce Boone]

Still, dozens or hundreds of time more life than CO2. I'll stick to my prediction. We'll see in a few years.

[Dan Hintz]

Hardly dozens or hundreds... diode packs have a (roughly) 10k-hour lifetime. In a production environment of 8/hr day, 5 days/wk, that's about 4-5 years. CO2 cartridge users usually see this kind of lifetime (disregarding bum cartridges), and some are still running strong long beyond that. Diode packs can't be regassed or their RF drive electronics replaced, which brings tubes back to "like new" condition, they must be tossed and replaced wholesale... that's usually at a cost of $1k-1.5k. A quality diode pack will cost significantly more.

I don't disagree that there will be some great advances in the coming years and prices will drop, but I think you're trying to make comparisons of apples and oranges. The two technologies are aimed at different markets for a reason; the differing output wavelengths cannot be interchanged (yet). CO2 is used to cut thick metals because you can essentially throw tons of power at it, reflectance inefficiencies be darned because it's easier/cheaper to get 2%-efficient CO2 at 2kW than a 30%-efficient fiber at 300W. You're not going to see fiber cutting wood because the efficiency scale is tilted all too heavily in CO2's favor. That's not about to change unless they can create kW fiber systems for pennies/Watt.

[Robert Walters]

It says it can cut thin metal sheets...
http://www.pryormarking.com/marking-products/laser-marking/yf50.html

[Dan Hintz]

Robert,

Man, that's a tiny video of it cutting metal... you'd think they could afford to blow it up some. 50W+ is really beefy for a fiber setup. The sheet they're cutting looks relatively thin (but the small window only allows me to make an educated guess), like 30 mils... I'm guessing that thickness because the sheet appears slightly warped (by heat) by the end of the cut.

That other video for marking is definitely using a galvo system.

EDIT: I also see them listing diode life at 50k hours... I highly question such a lifetime. If you always ran the system at less than full power, then you could increase the life, but that's a marketing ploy, not realistic.

[Bruce Boone]

I'm just sayin' This was done with a 20 watt laser. My 135 watt CO2 laser can't come close to that even at 100% power and speed at 1%

That's almost what I need. Maybe the 40 watt can do it, or maybe it will take 80 or so. The beam is so concentrated that it should be able to burn through about anything. Maybe the laser you have seen is different, but this one has a beam width of .0003" so it's just sick what it can do with very little power. They make stents with these things. It's funny to me that they can blast through silver sheet fairly easily, whereas that's pretty hard to do with CO2.

I think that with most things electronic, China will see there's profit to be made, tool up and flood the market with cheap lasing engines sometime.

[Dan Hintz]

I'm just sayin' This was done with a 20 watt laser. My 135 watt CO2 laser can't come close to that even at 100% power and speed at 1%. That's almost what I need. Maybe the 40 watt can do it, or maybe it will take 80 or so. The beam is so concentrated that it should be able to burn through about anything. Maybe the laser you have seen is different, but this one has a beam width of .0003" so it's just sick what it can do with very little power. They make stents with these things. It's funny to me that they can blast through silver sheet fairly easily, whereas that's pretty hard to do with CO2.
The small dot size is due to the shorter wavelength... fiber is roughly 1/10th the wavelength of CO2, so you get a focal size of roughly 1/10th that possible with a CO2. A typical CO2 beam with a 2"FL lens will give you a 3-5mil spot... with proper collimation optics, you can get that down to 1-2mils. At that point, you're beginning to bump up against physical and quantum/wavefront limitations. But again, that wavelength difference makes cutting metal with CO2 the realm of multi-hundred W (and kW) systems.

At what speed were those two cuts made? If it was relatively fast (a handful of seconds), the 20W may be adequate for what you need. If it was really slow (30-60 seconds or more), bumping up to a higher wattage would be recommended. Though you have to ask yourself, how often would you make such a cut, and would it be worth really bumping up the price just to save a few seconds of processing time.

I think that with most things electronic, China will see there's profit to be made, tool up and flood the market with cheap lasing engines sometime.
China is good about copying mechanical systems and some basic software, but fiber systems are beyond the ability of the typical Chinese garage copier. They may be able to cobble together a less expensive system using someone else's diode pack and fiber, but those will remain out of their reach for some time. The diode packs require serious clean-room capabilities (even the Chinese knock-off LEDs can't compete with a quality version, and they've been trying for years), and the doped fibers also require specialized equipment with high-purity materials.

If I had to take a guess (and this is just a guess based upon what I know of the current level of Chinese technology and capabilities), I wouldn't expect to see a street-level price fiber system for 10-15 years, at best, and that is probably an optimistic view.

[Bruce Boone]

The cut was with several passes. We ran into an issue with the focal cone getting blocked by the side walls as the cut got deeper. It lost a lot of efficiency and the surrounding metal got hot and reduced efficiency further. It would spark greatly at first then dimish as more passes were added. It took a few minutes to get that far. I assume I can dial in settings to do better than we did at the show, but I don't think it's quite enough power. The application guy was headed to a show in Europe then he was going to experiment with 40 watts.

It seems that the 20 and 40 watt engines are made for marking, and the more powerful ones like 80 watts are designed for welding. It has to do with pulses per second and tying into the galvo mirror. I'd love to be able to do cutting and/or deep engraving on the same ring. It would open up a lot of possibilities that I can't really do now.

[Dan Hintz]

If you're interested in cutting, rather than marking (I think a fiber-marked ring around the entire circumference would look sharp), I would generally suggest carriage system, not a galvo. In your case, however, if rings are all you're using it on, the working area is small enough that a galvo would be perfectly acceptable (as well as opening up other possibilities outside of ring marking).

Drilling to any depth is always an issue for the reason you mentioned... power loss due to edge effects. Still, rings should not pose a major problem at 1/4" thick or less.

[Bruce Boone]

They are normally done by galvo, but I wouldn't be opposed to doing an XY with rotary. The galvo is fast, but it normally does one section of the rotary, completes it, then rotates to do another section. It indexes rather than using full 4th axis. That works fine for marking, but is not so good for cutting. It would mess with the focal plane as well as have overlap in cuts. I'm also having the applications guy check to see if the galvo could work with true rotary cuts. It might be a show stopper if they can't do that.

It doesn't seem that tough to couple an engine with an XY rotary table, but maybe for some reason it is.

[Zsolt Paul]

It was fun reading through this post. Very informative!

[Robert Walters]

Man, that's a tiny video of it cutting metal... you'd think they could afford to blow it up some.

Not to worry, they get SMALLER! Or should that be SMALLER
http://www.pryormarking.com/marking-products/laser-marking/laser-markers.html


Here's the direct video link:
http://www.pryormarking.com/images/stories/video/50watt.flv

You can use VLC to view it - http://www.videolan.org/vlc/

[paul mott]

Just putting in my 2d's worth.

As Dan mentioned earlier different types of laser operate at different wavelengths and are better at processing different materials.

Watts for watts YAG at 1064nm will make a better job with metals than CO2 at 10600nm.

Watts for watts CO2 at 10600nm will make a better job with acrylic than diode at 808nm.

You could go on making a list of comparisons but it is 'horses for courses' here - The 'best' type of laser used is related to the task it has to perform.

(try cutting acrylic at 808nm or try putting CO2 at 10600nm down a fiber optic and you will see what I mean - it is not impossible just difficult).

Paul.

[Dan Hintz]

Bruce,

Is there a particular manufacturer you're in talks with? A true rotary, even with a galvo, should not be an issue. If you are cutting on an object as small as a ring, an indexer will cause issues (as you found out) and a true rotary is the proper tool.

[Bruce Boone]

Laserstar is the company I saw at the show.

[Michael Hunter]

China is good about copying mechanical systems and some basic software, but fiber systems are beyond the ability of the typical Chinese garage copier. They may be able to cobble together a less expensive system using someone else's diode pack and fiber, but those will remain out of their reach for some time. The diode packs require serious clean-room capabilities (even the Chinese knock-off LEDs can't compete with a quality version, and they've been trying for years), and the doped fibers also require specialized equipment with high-purity materials.

If I had to take a guess (and this is just a guess based upon what I know of the current level of Chinese technology and capabilities), I wouldn't expect to see a street-level price fiber system for 10-15 years, at best, and that is probably an optimistic view.



Check out dreamslaser.net. Some serious machinery there including fiber and CO2 with RF excited tubes at looow prices. Also galvo with 1000x1000mm working area.