The Hell Ya Beller Fun with hot, pointy, sharp, and caustic stuff.

23Jan/121

Laser Rebuild – Part 7 (Finishing Up)

Now that the laser is firing all the way through the beam path and the path can be controlled by the computer, we're down to the detail:

Controlling the Laser

I don't have too much to say here because I got incredibly lucky and someone else did all the heavy lifting for me.

Users on Buildlog.net have done lots of work making it possible to control the laser from EMC2.  User BenJackson has published his configuration files and adapting them to my machine was a piece of cake.

His configuration file includes a custom HAL component that does PPI (pulse per inch) control.  With this configuration, I have three "knobs" I can turn to adjust how the laser cuts or engraves material.

  • Feed Rate controls how fast the laser is moving over the material
  • PPI controls how many 3ms pulses are delivered for each unit of travel (inch or mm)
  • Laser Power.  EMC2 can vary the PWM signal to the laser power supply to control the power delivered to the tube.

By playing around with these three, I've found I can go from drawing a light brown line on a piece of paper all the way up to burning clear through 5mm wood or 1/4" acrylic!  Very Nice!   I've got plenty of learning still to do but it's looking good.

Cooling

The basic buildlog setup is a 5 gallon bucket of water, and an aquarium pump.  I need to tidy up my setup but it's working for now.  The only part I'm missing is a flow switch.  The control card has an input for a switch that is tied into the safety interlock.  It should prevent the laser from firing if the water isn't flowing.  I don't have a switch yet so I've hotwired around it.  Not good.

 Shroud

The glass tube hanging off the back makes me nervous.  I'm not sure if there's any dangerous radiation emitted from the sides of the tube, but I'm not taking any chances.  Besides, it's just a matter of time before I drop the broom handle or something else on to the tube and bust it.  I had a local sheet metal shop bend and weld up a shroud that fits over the tube and screws to my mounting rails.  After painting it, I think it came out nice.

IMG_0591

 

Exhaust

When the laser is burning anything, and especially plastics, it puts off a lot of smoke and foul smelling crap.  I have a big dust collector for the CNC machine.  I'll add a drop near where the laser will end up so I don't have a dryer hose trailing across the floor.

And the results?  This is 1/4" plywood cut through.

IMG_0590

 

12Jan/120

Laser Rebuild – Part 6 (Alignment)

Since my tube mount is completely custom, I have to manually align the tube with the first mirror.   The C02 laser is invisible (and dangerous) so aligning it correctly can be a trick.  I saw a solution on buildlog that I thought was a very clever hack.  The guy used a reprap to print out two disks the same diameter as the tube.  One disk has a very small center hole, the other has a hole that exactly fits a laser pointer.  He mounts them in the brackets and adjusts the laser to shoot through the small hole, into the first mirror.  Once the alignment is correct, he loosens just one screw to remove the disks and insert the tube.  At that point, the alignment should be very close.

I have a reprap but I had some time to kill before Christmas while I was waiting for my power supply to arrive and needed some lathe practice.  I turned these two 'cookies' and bored one to fit a cheap laser pointer I have.

IMG_0518

31Dec/111

Laser Rebuild – Part 5 (Bench Testing)

Not much to say about this.  It's scary cool!   The tube glows pinkish purple and the wood block ignites instantly.  The only thing I've seen ignite faster was the imagination of my 12 and 10 year old sons.  Something about that glow inspires a guy to try stupid stuff.  Fortunately eye-protection was the order of the day and no injuries were reported.

IMG_0557

21Dec/110

Laser Rebuild – Part 4 (laser mounting)

Since I'm replacing the RF tube with a Chinese glass tube, I'll have to fabricate the mounting brackets.  The buildlog 2.0 laser uses plastic brackets cut out of HDPE on a CNC router.  I've cut HDPE before and it's great stuff to work with on the router and I even have enough left over from an earlier project.

I started by mounting two pieces of 1/8" aluminum square tube to the existing mounting holes on the frame of the laser.  This will give a nice solid base to attach the brackets to.  It also has clearance behind for routing wires and enough room to turn the bottom adjusting screw.

IMG_0500

13Dec/110

Laser Rebuild – Part 2 (control)

I started looking around for a board to replace the original controller electronics.  My priorities:

  • Has to work with EMC2.  That's actually pretty easy to do since EMC2 is so flexibile, but I was looking for something a little more "off the shelf" compatible.  I'm not afraid to hack on an EMC2 configuration, but it's not my strong suite and I want easy.
  • Besides driving the steppers and laser, it needs to handle as much of the other stuff as possible.  Door switches, cooling pump, air assist, etc.
  • At least 1/8 microstepping.
  • Needs to fit in the cabinet along with a power supply and any fan, wiring, etc.
  • Not too expensive.
Also, it would be nice if it supported some direct control so I don't have to do everything from the PC.
I found a few solutions that were interesting but once again, the Buildlog guys seem to have anticipated my needs perfectly.  The laser interface driver uses pololu drivers which I'm familiar with from my Reprap RAMPS board.  The pololu stepper drivers are on a separate little board so they can be replaced easily if something "unexpected" happens.  Most of all, I'm happy to see an active community of guys using this hardware in conjunction with EMC2 to drive a laser.
The board is set up to control 3 Axis.  My Z axis is manual, so I have one driver free.  I'll either convert Z at a later date or maybe think about something like the rotational engraver on this page.
Here's what the stock cabinet looked like:
laser_guts
After gutting it I had a few leftover parts:
removed
After mounting the new controller, it looks like this:
NewGuts
The three black posts are stand-offs for the top panel.  The new controller supports some manual control, so I'll build a new panel to go there.
The DB-25 connector and the pololu microstepping controls fit almost perfectly into the existing holes in the chassis.  To make it fit, though, the mounting holes on the right edge of the board were under the lip of the right sidewall.  I couldn't drill and tap under there so I had a to make a small acrylic part.  The controller is mounted to the acrylic on that edge and the acrylic is screwed into the original holes.
installed
I already had a 24V power supply salvaged from a printer and it fit nicely in the back.  The filter, relay and chocolate-block terminal strip are the only things I saved.
After wiring it up and hacking up a configuration in EMC2, I have it moving the X and Y axis, detecting the limit switches, and detecting the safety switches on the doors.  It should also be tripping the relay, but I think the relay may be bad.
axis
I don't have any documentation on the number of steps per inch for the axis.  I emailed one of the guys from hacklab.to and he got me very close.  Theirs runs 500 half/steps per inch. I'm going to run everything in metric so I converted. Then I clamped a digital caliper and started fine tuning by trial and error. I found a value of X steps per mm got me within 9 one-hundredths of a mm over a 100 mm travel. Not bad.

calibrate

11Dec/111

Laser Rebuild – Part 1 (Research)

My first step in resurrecting this machine is research. Less than 5 minutes of hunting around on Google has uncovered a couple of incredibly useful sources:

I don't have any experience with lasers and I'm rather fond of my ability to see so I started by reading A LOT on the grandaddy of all internet laser resources.  Sam's Laser FAQ.

Buildlog.net  - This is the home of the fascinating Buildlog 2.0 laser cutter project.  An opensource design for a DIY laser that can be build for less than $2000.   The most interesting part is the community that has sprung up around the design and is collaborating on improvements.   There's a wealth of information in the forums including this thread about another ULS rebuild.  Also worth looking at is the Makerslide project that originated with Buildlog founder Bart Dring.

Hacklab Toronto did a ULS renovation a few years back.  It's well documented here.

After reading through these projects and studying my own machine, here's what I think I know:

Optics:  Seem to be in good shape but filthy.  I'm going to do a thorough cleaning following the procedure I found documented in a ULS manual.  While I have them apart, I'll take a closer look to see if they've suffered any permanent damage.

IMG_0497

Linear motion:  Steppers and guide rails seem to be in good shape.  Both axis work great.  The X axis belt is a bit loose but doesn't seem to be losing steps.
laser_internals

Laser:  No joy.  I'll be replacing this with a 40W Chinese tube and power supply.

laser

Cooing: The original RF laser was air cooled but the replacement will be water cooled.  Need a whole new system.

Ventilation:  The chassis is designed to conduct air across the build area.  Again, it's filthy but intact.

Motion control:  The original electronics are designed to be driven by a windows printer driver.  They're serious antiques at this point.  I plan on driving it from EMC2 and will need a new controller.

laser_guts

Safety interlock and E-stop.  All the switches and the by-pass key work on the original.  I should be able to wire them into a replacement controller.

by-pass

Build Platform:  A 1/4" aluminum plate.  Scored in places but serviceable for now.

IMG_0454

6Dec/1115

Sliptonic has a (frickin) laser!

Well, not quite....yet.

I found an old Universal Laser Systems model 25A for sale at a surplus auction.  It looked to be in good shape and went cheap.  After I got it home and cleaned up, I found that I couldn't get the laser head to fire.  I called in some expert assistance from Columbia Gadget Works, but no luck was to be had.  I've since learned that RF laser tubes have a life expectancy of about 10 years before they need to be re-gassed.  Re-gassing this one is prohibitively expensive and would *only* get it back to 25W.  Instead, I'm looking at replacing the RF laser with a chinese glass tube.  This should take it to 40W.

Either way, this looks to be a fun project.  Here's a picture.

 

 

2011-11-15 10.49.19