[emvali]

REGARDING ALLOWABLE COOLANT PRESSURE, TEST THE ORIGINAL PUMP FOR PRESSURE AND FLOW RATE AND TEMPERATURE RISE. IS THERE A RECOMMENDED INLET TEMPERATURE TO BE CONCERNED WITH, TO AVOID CHILLING ONE END OF THE TUBE OR CAUSING EXCESSIVE TEMPERATURE DIFFERENTIAL ALONG THE LENGTH OF THE TUBE? IN A VERTICAL APPLICATION, COUNTER FLOW, INLET AT TOP, OUTLET AT BOTTOM ORDINARILY SHOULD YIELD MORE EVEN COOLING.

BTW, ALL CAPS IS EASIER ON ARTHRITIC FINGERS.

Thanks Ken,

Check the drawing somewhere below, the flow is indeed from top to bottom.
The idea is to use CNC coolant fluid, where the container's capacity is almost 70L and the pump is powerful enough to raise the fluid high enough.

[emvali]

I have a Camtech L20 55w with a veritical mounted laser. System works fine but I sure the laser assembly has been build for the veritical mount. Has an adjustable slide tube to focus the lens and is air cooled using 6 fans. The Z axis height contol is manual. I had intended to add a stepper and slides for auto control but now I'm not sure if the conveniance is worth the additional weight.

Be forwarned that there is a great deal of side to side force in this arrange and a budjet laser head may not hold up.

danr50,

I don't understand what do you mean by side to side force.
The attachement I think using is mounted on the machine's Z housing, which is static, not moving.
Regarding up/down movement I have 2 options:

1- Using a sliding tube as in your setup, which is moved by Z
2- Manual table height adjustment - this CNC mill has this option too.

As I am not constantly using Z movement except for initial setup, the only thing that moves is the table on X/Y - so there are no side to side forces, unless I don't understand you.

[danr50]

No, my bad quick scan of your post. Reread and understand. You definately won't have a problem on a stationary mount.

[Jim Moore]

I am sorry you are discovering the pain of bad equipment. You have already given up on the table, why do we suppose the laser is better? Have you done a mode burn or other mode evaluation with the raw beam to qualify the laser on its own?

Supposing the laser is good and can be mounted vertically (with manufacturer approval) I see nothing wrong with your splicing plan. Another alternative to consider is a fiber delivery system. I was just at a laser processing conference where passing several kilowatts down a 300 um fiber was commonplace. I'm sure there is an even broader offering at 40 watts.

If you had more power, could you use a longer final lens and get more depth of field? Can you use a surface follower, perhaps a capacitive non-contact one? Is there an option for focus control? Can the control learn the waviness of the table?

Good Luck,

Jim Moore

[emvali]

I am sorry you are discovering the pain of bad equipment. You have already given up on the table, why do we suppose the laser is better? Have you done a mode burn or other mode evaluation with the raw beam to qualify the laser on its own?

Supposing the laser is good and can be mounted vertically (with manufacturer approval) I see nothing wrong with your splicing plan. Another alternative to consider is a fiber delivery system. I was just at a laser processing conference where passing several kilowatts down a 300 um fiber was commonplace. I'm sure there is an even broader offering at 40 watts.

If you had more power, could you use a longer final lens and get more depth of field? Can you use a surface follower, perhaps a capacitive non-contact one? Is there an option for focus control? Can the control learn the waviness of the table?

Good Luck,

Jim Moore

Hi Jim,

I had no big expectations so the pain is not big too.

My only request was to be able to cut plastic materials, acrylics, in a simple fast and convenient way and occasional engraving of serial numbers, labels and etc.

One reading this thread might ask why do I need the laser when the CNC can do most jobs too - here the answer is with some of the advantages of laser cutting vs. milling, starting with the width of the cut and ending with the clamping of the material on a CNC table.

I know one possibility is to use a fiber, regarding sophisticated controls to compensate for table's waviness the answer is no, the machine's control is very basic. Let's remember that the acrylic sheet thickness itself is not exactly even, 10% tollerance is very common. The machine has no Z control (it is a initial setup made manually), also the machines with electric table height control are very basic so the follower practically can't be used.

The task is to keep everything at a reasonable cost, suitable for small and occasional use - believe me that when I need to cut large quantities of something I take it to the guys with the Mazak.

The laser tube with it's basic power supply at the given price of $130 good for at least 1000 hours of use at full current needed for cutting worths replacement like a lamp bulb.

I think this is the advantage here, the heart of the system at low cost and yes, relatively low quality too - I could not give up with the table issue, I could change several parts and make it work perfectly, but then the carriage assembly waits to be enhanced and a bunch of other parts. Sometimes is better to start something in a simple way from the beginning rather than patching a bad design.

I think that by keeping everything at the basics and thinking of quick tube change on the design can be a nice idea for many that are looking for a cheap method or equipment.

[BrianT]

Vertical mounting is not recommended due to any tiny manufacturing leftovers ending up on the exit mirror. The energy density is extremely high at that point and any contamination will be burnt toast in no time, discolouring the exit mirror and maybe even cracking it.

Any decent laser tube will have been manufactured in a clean room so SHOULD not have any physical impurities but you never know.

HTH
BrianT

[emvali]

Brian,

This is an interesting point, specially with the Chinese tubes.

Hmmm.... this is bad news, I had a feeling there is a catch somewhere.

[Jkirk3279]

This is just the sort of thing I've been thinking about.

I have a Precix 9000 CNC machine. I realized when I got it there was the possibility of mounting a laser on the Z axis gantry opposite from the router motor.


The questions I came up with regard to the control software, etc.

Once you take your laser tube out of the chinese box and remount it on the CNC machine, you will need a way to turn the laser on and off.

You can't use the chinese laser controller to run your CNC table (unless by some miracle that controller could run the far larger CNC machine servo motors).

You could put in a terrain-following switch that toggles the laser on and off and ensures the laser beam only fires when it's in focusing range.

Use your CNC software (what are you running with now?) to raise and lower the Z axis and treat the laser as if it's a router bit.

When the Z axis goes up, the laser shuts off, when it drops down, the laser engages.

Any other solution, like software on/off, would likely require writing your own CNC software's post processor as well as hacking the serial port or USB connection.

Regarding the optical fiber approach: this is the ideal solution for me.

Exposed mirrors, blah.

Lots of possibility for failure. And always the risk that a misaligned mirror might mean catching a laser beam on your retina.

The Optical fiber though eliminates that danger.

Set your CNC software to ... a straight sided end-cutting bit the same diameter as the laser beam.

The hardware on/off terrain switch isn't ideal of course. You'll have just a moment of dwell time as the laser is triggered before the X-Y movement starts.

If your software allows for it, set it to start cutting AWAY from the workpiece, that should solve the problem.

[emvali]

This is just the sort of thing I've been thinking about.

I have a Precix 9000 CNC machine. I realized when I got it there was the possibility of mounting a laser on the Z axis gantry opposite from the router motor.


The questions I came up with regard to the control software, etc.

Once you take your laser tube out of the chinese box and remount it on the CNC machine, you will need a way to turn the laser on and off.

You can't use the chinese laser controller to run your CNC table (unless by some miracle that controller could run the far larger CNC machine servo motors).

Absolutely not, I have no intention to use the laser's controller board. The power supply is a separate box with inputs for on/off and intensity (current) control.

You could put in a terrain-following switch that toggles the laser on and off and ensures the laser beam only fires when it's in focusing range.

Use your CNC software (what are you running with now?) to raise and lower the Z axis and treat the laser as if it's a router bit.

When the Z axis goes up, the laser shuts off, when it drops down, the laser engages.

Any other solution, like software on/off, would likely require writing your own CNC software's post processor as well as hacking the serial port or USB connection.

This depends on the machine and it's structure.

Almost every CNC machine will have discrete outputs controllable by M code or a special group of G codes.

If not, a mechanical or optical switch can be used as you mentioned.
For example, my Hartford open mill has fixed head assembly from which the spindle only moves up/down, while on the VMC the whole spindle and Z assembly moves.

Regarding the optical fiber approach: this is the ideal solution for me.

The Optical fiber though eliminates that danger.

The advantage with the optical fiber will be that the tube can be installed in a convenient place. Also, it will allow me to use a higher power type, such as 100-150W which are almost 2000mm long.
Another advantage will be the possibility to leave the tube in the original horizontal position, solving the problem raised by BrianT.

So... where can these fibers be found?

[zax15uk]

Most fiber coupled systems are laser diode based but there are some CO2 options:

http://www.jtingram.com/id10.html


Cost is likely to be prohibitive.

Zax.

[KEN BANGERTER]

I SUSPECT YOU ARE SWAPPING A SIMPLE KNOWN ENGINEERING PROBLEM FOR A COMPLEX UNKNOWN ONES AND WILL HAVE TO DEAL WITH MANY HARDWARE AND SOFTWARE DEVELOPMENT PROBLEMS BEFORE YOU ARE UP AND RUNNING RELIABLY.

HAVE YOU HAD CONVERSATIONS WITH SEVERAL INDIVIDUALS WHILE ACTUALLY HANDLING THE MACHINE AND THROWING AROUND IDEAS?

CONSIDER THAT BOLTED JOINTS THAT LOOSEN IN OPERATION MAY BE FIXED WITH A LOCK WASHER, SELF-LOCKING OR PREVAILING TORQUE NUTS OR LOCKTITE. FLIMSY STRUCTURE CAN BE BEEFED UP, SLOPPILY FITTED MOVING JOINTS CAN BE BUSHED OR IMPROVED BY USING NEEDLE BEARINGS IN PLACE OF BUSHINGS, AND PROPERLY FITTED SHAFTS.

AS A RETIRED PRODUCT DEVELOPMENT CONSULTANT WHO IN THE 1960s AND 70s DESIGNED MANY PNEUMATIC ASSEMBLY TOOLS, INCLUDING TOOLS FOR MANUFACTURING OF THE MANNED LUNAR LANDERS, ULTRASOUND IMAGING SYSTEMS THAT YIELDED A POLAROID PRINT OF THE CROSS SECTION OF A SUBJECT, MECHANICAL POWER TRANSMISSION SYSTEMS, LASER ALIGNMENT SYSTEMS AND SHADED POLE MOTORS, I FOUND THAT TALKING WITH THE CLIENT WHILE STUDYING THE PROBLEM, ALWAYS FOUND SIMPLE SOLUTIONS TO COMPLEX PROBLEMS.

IN MY EXPERIENCE, ALMOST ANYONE CAN IMPROVE AN EXISTING DEVICE, AFTER RECOGNIZING ITS SHORTCOMINGS. AS YOU MENTIONED, THIS IS AN “OCCASIONAL USE” MACHINE.

STAY WITH THE KISS PRINCIPAL, IT MAKES LIFE EASIER.

[emvali]

Ken,

As an engineer I believe in 2 things:

1 - As Einstein once said.... immagination is more important then knowledge.
2 - Think unconventional, get out from the CUBE.

Nobody said that if the current structure of a laser engraver contains 4 mirrors and the installation looks as it does everybody should follow like a herd of sheep.

I am a beleiever in the power of such forums for so called brain storming discussions - I find not only enthusiasm but lots of practical experience.

Yes, I am throwing an idea, already found issues I could never imagine together with a lot of interest which as I assumed from the beginning are a result of the same bad experience I had.

I see nothing wrong about it.

I think that it will take less time to design this CNC attachement rather than solving problems of a bad manufactured machine. It is like with software, sometimes it will take less time writing a new program instead of "entering in another programmer's head and logics" in order to solve a problem.

[emvali]

By the way,

When I say "think unconventional - get out from the cube"

I don't mean the movie - there is an example that instead of initiating all they should have done is just wait and this is not our situation.

What I mean by this sentence is not to be square, not thinking as everybody does - this is why I used to like Apple more than IBM and was a fan of Steve Jobs.

[lasersafe1]

Vertical = NO. You can however keep your 3 mirror setup on the moving Y axis and maintain a proper focus as the X travels back and forth. The primary advantage of this setup is the ability to have a very large Y axis movement without worrying about the divergence issues. Divergence will only be a factor on the X axis and should be quite acceptable over the distance shown here. This setup keeps the focusing mirror from your Chinese laser and doesn't need a lens. All optics and laser move with the Y axis. The focus mirror is the only moving part on the X axis.

[Jkirk3279]

Most fiber coupled systems are laser diode based but there are some CO2 options:

http://www.jtingram.com/id10.html


Cost is likely to be prohibitive.

Zax.

It looks good. So the next question is, does the Chinese laser unit have any convenient mounting threads... or will it be possible to mount a socket on the surface to accept the fiber optic cable?

[lasersafe1]

It looks good. So the next question is, does the Chinese laser unit have any convenient mounting threads... or will it be possible to mount a socket on the surface to accept the fiber optic cable?

I can assure you. The cost of a single fiber launch system will be the equivalent of purchasing MANY of the Chinese CNC machines. Don't even go down that road. It is quite inefficient to launch free space tube laser into a fiber. It is much more common these days to actually produce the laser with a fiber. The fiber contains the gain medium and the cladding of the fiber is pumped with diode lasers. A 10 Watt fiber laser from IPG is $35,000. This is the bottom end of their offering.

[Jkirk3279]

Vertical = NO. You can however keep your 3 mirror setup on the moving Y axis and maintain a proper focus as the X travels back and forth. The primary advantage of this setup is the ability to have a very large Y axis movement without worrying about the divergence issues. Divergence will only be a factor on the X axis and should be quite acceptable over the distance shown here. This setup keeps the focusing mirror from your Chinese laser and doesn't need a lens. All optics and laser move with the Y axis. The focus mirror is the only moving part on the X axis.

Yes, this is a good sketch, thanks. It's sort of what I was thinking of, if the fiber optic cable option didn't work.

What concerns me is the following:

We're now considering sending a laser beam farther than the original specs on the chinese laser would have called for.



If there's any dirt or grit on the CNC table's bearings, you could get a shaky laser beam when the bearings go over the bump.

I know my CNC table is kind of dusty, most of the time. But I'm not shooting a laser beam around, so that doesn't bother anything.


In the original box, there wouldn't be any dust: and the shorter travel path helps too.

But the farther you send the beam the more even small variations could throw off the beam, correct?

I figure that the beam path bounces through three mirrors, as per your sketch.

Then assume there is a bit of sawdust or speck of something on the guide rods or screws, whichever the CNC machine uses.

Say the beam is diverged from it's true path 1/1000 of an inch. Now instead of perfectly hitting the second mirror, it's a fraction off; the second mirror also picks up the shock, and the beam gets diverged another 1/1000th of an inch, and then again at the third mirror.

Which is why the optical fiber is a better idea, IMHO. The beam will bend with the fiber: no mirrors to worry about.

You could compensate by meticulously cleaning the CNC machine and putting in larger mirrors.

Maybe some shock absorber system?

[emvali]

The glass laser tube has no thread, I think only the metal case lasers do have such outlet and they are far more expensive than the glass types.

Anyway, if the fiber cost is as mentioned, this will be the end of the idea.

The CNC machine I am talking about is a vertical mill, where the table only moves, the laser tube assembly is absolutely static.

I will attach a photo soon.

[lasersafe1]

We're now considering sending a laser beam farther than the original specs on the chinese laser would have called for.


In the original box, there wouldn't be any dust: and the shorter travel path helps too.

This layout actually has a shorter throw than the original Chinese system, so divergence and the alignment are less of an issue. The Y axis can be a mile long and it wouldn't change a thing as far as beam quality is concerned.

You are kidding yourself if you think there is less dust in the box. There is an exhaust fan that is constantly pulling out smoke during the cutting. You can only draw air out if you are letting new air in.

Yes, you are going to want good bearings and races. You won't be building an MDF system with drawer slides from Home Depot. You should shoot for the precision you would find in a typical printer or plotter. That being said, I was quite surprised to see the simplicity of the moving parts in the Chinese unit I purchased. They were quite innovative in their selection of low cost materials to serve the function.