View Full Version : New user - Laser head

Adrian Hill
05-30-2007, 7:06 AM

Let me first introduce myself. My Name is Adrian Hill and I live in Cape Town, South Africa. My business does a variety of weird and wonderful thigs ranging from electronic product development through to CNC retrofits.

A customer of mine wants us to add laser head to an existing CNC machine. We can mill all the required attachment from Aluminium and we can also do the electronics, but the issue is the laser.

There has been a lot of talk about CO2 and YAG lasers but I haven't been able to find a laser supplier. Many companies will supply the entire machine, but where can one obtain only the laser?

Then of course - which is the appropriate laser to use and at what power. The customer wants to engrave wood, plastics and light metals.


PS - Hi Rodney

Mike Null
05-30-2007, 7:31 AM
I think you may be swimming upstream on this one but give these folks a try.


Rodne Gold
05-30-2007, 8:03 AM
Hiya Adrian
Don't go down that route , its not just a simple retrofit of a laser , there are a huge amount of other issues.
The system will be open and thus dangerous
The laser itself needs a controller and software to fire at the right time
The laser will need an optical path added as well as a good power supply.
There will be issues with ramping and the inertia of the system , it will be exceedingly slow
You will pay a fortune for something like a 50-100w sealed beam.
There are about another 100 reasons I can think of as well!!!
This is a job you may never get right.
Tell your customer to bite the bullet and get a commercial small system , can be had as low as R120k (prolly less than a retrofit would ultimately cost.) Maizeys (gcc), Melco (gravograph), Rubber stamp and engraving (trotec) AMS (jhb - zenith yag and epilog ) can all help you out with one.

Richard Rumancik
05-30-2007, 11:28 AM
Hello Adrian

If by an "existing CNC machine" you mean a milling machine, then that would probably be the wrong tool to add laser engraving capabilities. Cutting capability perhaps, but not engraving. You would not be able to achieve the necessary speed for raster engraving on a mill.

You specified engraving needs only, but soon they would say "why can't we cut with this?" But engraving and cutting requirements are quite different in terms of software/firmware.

If you mean retrofitting a CNC router, then it may be possible, but it would be an expensive path to take. If you are designing a product for only one customer you have to put a lot of development work and research into a single unit. It sounds like your company has not done any laser-related work so there would be some learning involved. The software and firmware development alone would be a significant task. I think you would soon regret signing up for this project.

Whether router or mill, you would need to dedicate the machine to laser, in my opinion. You don't want to be routinely installing and uninstalling the "laser attachment."

I am assuming your customer feels they will save money going down this path, but it will be expensive and time consuming, and you might not be able to reach the ultimate goal. If after a year of development, it "sorta" works . . . well, maybe there are other options.

If their budget is limited then they could consider a used machine. Someone really needs to define the goal clearly first. What is the max. area they want to engrave? If small logos and part numbers then a galvo laser may be suitable. What features of laser do they need/find attractive? Small CO2 lasers don't "engrave" metal. (You can mark some metals with Cermark.) Yags are better for metal marking (no chemicals needed) but generally used in the "blackening" mode, not for material removal. CO2s are better for cutting plastics and wood. You won't get everything in a single laser system.

Maybe they should consider adding an engraving spindle to their mill and buy suitable software for line engraving (like MillWrite or V-Carve etc).

If you can elaborate on their needs, then it would be easier to make suggestions.

Adrian Hill
05-30-2007, 1:29 PM
Hi all,

I don't think that a milling machine is inappropriate at all. The machine is fairly fast (I don't know what the expected feed rate is on a laser), but I am sure that just as feed rate and tool RPM on the mill can compensate for one another, the same should be for feed rate versus laser power.

From a software point of view I also don't see the difference between milling and raster engraving. I run parallel toolpaths using 3mm ball nose tools at a 3% stepover which is close to raster engraving. I suspect that the laser power can be artificially controlled by Z height and laser on and off by other M codes.

The CNC side is simply not an issue - we currently run a number of retrofitted machines including mills and routers. Stepper motors can be used but servos are preferred. One could of course build the entire machine from scratch, and again, that is not an issue - all the parts are locally available off the shelf.

The only issue I can see is the laser.

I agree that the machine should be dedicated though.

I agree with Rodne that using a standard laser will be a problem. I do think that there must be complete lasers available with fibre optic delivery systems..

The need is to do engraving and cutting fast and cost effectively.

If one consideres the actual cost of an off the shelf machine versus a DIY build I must say I'm for the DIY build. I can safely say that I can handle all the software, PC interfacing and CNC hardware issues.

Rodne Gold
05-30-2007, 2:28 PM
For raster type laser engraving (IE ALL laser engraving) speeds of up to 1.5m/sec are required
We have a Tekcel CnC router we investigated having a laser retrofitted to , it's just not viable.
You could fit a laser to one of these if all you wanted to do was cut , but you would have to have a flying optic setup. IE the laser mounted where the spindle is will be very very heavy.
a 25 watt synrad tube will cost you about R40 000 without a controller or power supply.
You wont find a system that works on an xy basis that delivers significant power via optic fibre.
25 watts is not worth doing the mod for , so to get like a 50-100w tube with controllers will cost like 100k or so.
Get the number for Stan Woolford in durban , he might aid you as he is the local repair guy for laser tubes. Engraving supplies in pretoria will have it (part of maizeys)

Adrian Hill
05-31-2007, 5:54 AM
Hmm, there is a fair difference between laser cutting and laser engraving. The feed rate for laser engraving is really high.

I do't think that it would be viable to build a laser engraver - a Laser cutter maybe..

I'll speak with Stan.

Rob Bosworth
05-31-2007, 12:00 PM
Adrian, I've been in the laser business since 1981. Most of my experience comes from the design and manufacturing side of this business. Several times a year we get calls from people who have decided that they can save all kinds of money by building their own laser system or they are going to retrofit a laser to another piece of equipment. For years, I have tried to explain that it is possible to do something if a person is extremely handy and a tad bit lucky, but usually they end up with an expensive piece of equipment that has been compromised for both the original operation and for the laser processing. So I do not try to disuade anyone any longer. I just say, that's a great idea. Hook that plow up to your family sedan and use it to plow a field. Both a tractor and a sedan have forward and reverse, and have fuel consuming engines.

Major CO2 laser suppliers:
Universal Laser Systems
Synrad Laser
Coherent Laser
Deos Laser (a Div. of Coherent)
Rofin Sinar
PRC Laser

Good luck.

Richard Rumancik
05-31-2007, 11:02 PM
Adrian, you sound as though you are comfortable with the integration aspects of this project, so if the laser is the only holdup here are some ideas. I hope you are not offended with some of the comments that members have made. As we don't have a complete picture of what you want to be able to do, your customer's budget and their expectations it can be difficult to know the best approach.

I know it would be great if there were some gadget you could chuck in the spindle and make the toolholder a laser. I don't think this currently exists. CO2 lasers don't work well with fiber optics. There may be some special cases but generally CO2 lasers use mirrors. Most of the medical CO2 lasers are constructed with mirrors and articulated connections between rigid tubes. Lots of work is being done on fiber lasers and diode lasers these days so maybe there will be some innovations in the future that would make it simpler to do a conversion like this.

In the meantime, if you want to convert a mill into doing CO2 laser work then there may be some options. I don't know what kind of mill your customer has but I'm visualizing a Bridgeport type CNC mill. Start off with a knee mill if possible as it's easier (since you can manually jog the table height for focus). Buy a Synrad or ULS tube and mount it off the mill parallel to the long axis of the mill table, aiming it at the spindle. (In actual fact it could come in from behind the mill table at a any angle to save floor space as long as it points to the spindle). It would be best if the laser tube support structure could be attached rigidly to the milling machine base so there is no relative motion. Otherwise, mount it on a heavy steel table weighed down with concrete or iron. Lock the spindle so it does not turn, and provide an interlock to the spindle motor so it can't be turned on. Install a mirror and lens assembly on the tool holder so the incident laser beam is reflected downwards toward the table. In a milling type setup, you will be moving your material as opposed to moving the beam, so you only need one mirror with this concept. You can protect eyes from the beam path with an appropriate metal tube extending from the laser tube to the mirror assembly. (If you don't have a knee mill to enable focus adjustment, you could build a separate table attached to the bed that elevates.)

For cutting you may want to build a table that goes on top of the milling surface to contain the cutting surface/honeycomb etc. As cutting will occur at one point below the spindle, it should be fairly easy to set up fume extraction.

Build high walls around the whole thing so only the operator can get near the machine while in operation. Laser goggles would be mandatory in the area.

Order a copy of Mach3, or something similar, get your favorite CAD/CAM software that is capable of creating g-code, (and g-code suitable for raster) and you're all set to go. Set up the laser as an auxiliary axis. I don't know if Mach3 is suitable for raster or not. You could also consider something like the Testra box. Or look at Solustan. I haven't used any of these products myself.


In the absence of tool loads, you could operate it at relatively high speeds, maybe a few hundred inches per minute. That would still be slow compared to a raster laser system but may be adequate for what you are attempting to do. My personal feeling is that the setup described might be good for machine-shop cutting, marking, and prototyping, but doing raster photos etc probably isn't going to be an appropriate task for this machine.

If you could still use it as a CNC mill without major tear-down and set-up, that would be an interesting bonus. I suggested earlier that it should be dedicated to laser. But there may be a way to still use it as a mill. If the 45 degree mirror assembly was mounted as though it were a standard CNC tool, it would go into the holder repeatably each time. As long as the laser is on a stable heavy mount, it won't move. Then in order to get up and running all you need to do is lock out the spindle motor, install the mirror assemby, and manually rotate the spindle to the exact angle so the mirror is aligned with the laser. Then the spindle would be locked. This alignment process would not be difficult to achieve and could be made to be repeatable quite easily.

I did not mean to suggest earlier that a milling machine could not be converted to CNC laser. There are probably many ways and it might even be an enjoyable project. If you can get the financials to work, and are satisfied with the limitations and the customer is realistic in expectations, go ahead. I almost want to build one myself. The main cost would be in the laser, optics, and machine controller/software. If you decide to do a conversion, I'd be very interested in knowing the results.

Sam Gardner
06-01-2007, 1:34 AM
This is proving a very interesting thread and may I post a pic of a CNC mill fitted with a 200W laser used only for cutting. I saw the mill at work at an expo I attended in Singapore and I gather the machine was built in Asia most probably Singapore itself. So it seems it can be done but looking at the size of the mill head I do not see it capable of high speed and high quality raster engraving .

Richard Rumancik
06-01-2007, 3:57 PM
Physics is working against you when you try to move a hundred kg of steel back and forth really fast. That is why I originally said that a CNC mill would not be the best tool for raster. But if you were cutting out a part and needed a part number or mark on it, a small amount of raster might be economical. The 200 watt size is an interesting size for a small shop. The reason is that the 1-2 kilowatt metal cutting lasers don't work well for cutting thin stock. They are just overpowered for light work. So if one can find a market cutting thin materials like metal stencils or shims or metal gaskets etc then a retrofit might be possible. It would also be suitable for thicker plastics that could not be done on a laser engraver.

I would not use a retrofitted machine to compete with the lasers used by most of the members of this forum. I would find jobs that can't be done by these lasers, and which are unsuitable for high-power metal working lasers. I think there may be a niche in there somewhere.

Synrad has some interesting applications for 100-200 watt lasers on their website.

Neal Schlee
06-01-2007, 4:59 PM
Oh so that's why you see all those old Sedan's in farm fields in Wisconsin! :-)