That's correct, I was talking about a -very- simple one which has enough capability of cutting a coke can. For positioning/cooling and such complexity=3Dmore cost should be introduced. For the pcb, drilling of thin copper (0.5 mm) can be done with 1W power but the substrate is the problem due to it's internal structure. Soldering and drilling copper can be done with a 1W YAG. If it's the only option you have, you will find ways to make the whole PCB using 1W YAG if you put enough effort like finding the approximate values which will is enough to seperate copper/substrate molecules from each other. Maybe you'll need non-standard PCB's to do so, but it's doable assuming the specially crafted conditions. Frequency of the source is another aspect, you will need to pump more power as you higher the frequency. There is no such thing as UV laser, there are lasers down to 400nm (SSD) and wavelengths below this limit is achieved using nonlinear optical crystals. If you higher the frequency using such technique you will need to pump more power at input of the diode. 1W entering a KDP will not come out as 1W, increasing the frequency has a cost in decreased power. So it's approximately constant relation between intensity/frequency assuming a constant source. It all yields to energy which is just enough force to seperate the molecules of the target object which is fundamentally a Van Der Waals force. The theory makes it clear that it is doable, it's a waste of energy to use 500W laser to cut copper and such and is all originating from the imperfection of the practice. I saw 1W 808nm laser diode cutting aliminium very easly. I have no idea about the values right but 3000W for a 100W YAG sounded a little exagurated to me. It will not be a continuous drive for sure, power of PWM should be exactly calculated. And yes TE cooling suck too much power if it's done right. On Thu, Jun 23, 2011 at 12:56 PM, Tony Smith wrote: >> Information in this thread is misleading unfortunately. You don't need >> a gas laser to do this. They are big and heavy. There are up to 100W >> laser diodes at 808nm wavelength (IR). Interaction of wave with matter >> is divided into 3 parts which are ; reflection / absorbsition / >> >> >> As Bob stated earlier in this thread, fiber lasers are cool and they >> can have tunable frequency outputs. This tunability =A0will be between >> approximately 1200nm-1700nm in communication wavelength range.Even if >> you change the source of your fiber system (put a lower wavelength >> diode) this system will not work accurately. When you have smaller >> wavelength penetration depth will be lower thus reducing the quality >> of cutting process. Also fiber cable will melt if you put a really >> high power laser diode at source side. Tunable lasers are not for >> cutting and such, they are for communications. >> >> Go and buy a 1W 808nm laser diode, they will usually operate between >> 1.8V to 2.2V. Design a simple PWM driver circuit and interface to PC >> using USB or RS232 then you will have a simple laser cutting system ; >> BOM; 50-60$ > > > Misleading is right, your $50-60 leaves out steppers, drivers and other > hardware. =A0I suppose you could always wave it about. > > I'd also love to see that 1W laser cut anything but paper rather slowly. > Burning isn't cutting, edge quality matters. > > While YAG (fibre, diode etc) is better than CO2 for various reasons, it > isn't practical, hence almost all industrial machines are gas CO2. =A0YAG= is > very inefficient and uses an enormous amount of power, a 100W YAG needs > about 3000W input, and that's before you add cooling, as you may imagine = to > need a lot of cooling. =A0Even then it'll only cut less than 0.5mm of > aluminium. > > A 100W CO2 laser is pretty useless for aluminium (maybe 500W), but it'll = be > much cheaper to buy & run. =A0It needs about 500W input (An RF CO2 would = be > about 1000W). > > CO2 for acrylic & organics, high power CO2 for metal, YAG for precision > metal or stuff CO2 can't do. > > One nice thing about laser CNC is building a 200W machine is about the sa= me > as a 20W one, you can just about swap tubes and carry on. =A0I had a 35W = laser > for a while and I'm contemplating building a 150-200W one. =A0I think I c= an do > it for not much more than $3000, and that's mostly tube. =A0Why, of cours= e, is > why not. > > Tony > > -- > http://www.piclist.com PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > --=20 http://www.piclist.com PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .