I was looking at the lasersaur project to see how they’re doing.

I’ve expressed concern about how long it’s taking them given the funding and the number of sponsors waiting for them.

The trigger for my concern is that the project isn’t really open source so it does not have the benefit of the crowd. That aspect has concerned me because without the crowd the project may not mature as fast. Also the design may not be suitable for as many people.

I think it’s a huge disservice for the project to close the doors to the public.

Myself, I’m pretty new to all this CNC stuff (I took interest in it this passing summer).  So I’m just an reasonably enthusiastic newbie.

But I was a little concerned when I saw the statement in the picture I’m showing to the right.

Finally, the first lines of firmware have been written and the the barebone gantry is moving sweetly. We are more and more leaning towards GCode (or a subset thereof). Many people have commented on this and tried to push the community in that direction — maybe rightly so.

Our initial hesitation was that GCode seemed super messy and bloated for the lasersaur. After taking a look how reprap-related project use a subset of it we think much higher of it.

Reading through the source code of grbl and fiveD is also super helpful. Still if somebody knows of a technical paper on stepper motion, acceleration, deceleration strategies we would use it as our night time text for the weeks to come”.

The statement “leaning towards GCode” got my goat a little for some reason. Like I said, I’m no expert but the whole motion control thing seems really well solved. For instance this CNC system costs about 150 bucks. It includes everything you need to run 3 steppers motors (that can control three axis) from a computer using very well proven software such as Mach 3 or EMC2. A laser really only needs two axis so it’s actually overkill. This still leaves the laser uncontrolled. Well this CO2 laser power control was less than 50 bucks shipped. You connect it to the standard high voltage supply that is used with the CO2 lasers.

My intuition is telling me they are trying to improve or reinvent something that is already working very well.

This is a double edged sword. On one hand it’s important to take new paths. Especially the untraveled paths. So that’s good argument to get off the beaten path. But is the control of steppers a show stopper for people who want to get a laser cutter?

So the other hand is that they are trying  to fix something that isn’t really broken.

There must be something else about a laser cutter that makes it difficult to obtain. I guess if the motive of the project is for the participants to learn about motion control from an engineering standpoint, then that’s cool. But the project has a bunch of sponsors so that’s different.

This whole thing is good food for thought for me.

As many of you may know I am working on an open source project called Marauder which will be an accessible CNC mill that can cut plastics, wood and such (and a plastic extruder).

I guess I need to find improvements to make it easier (and less expensive) for people.

One area that I’m working on is different bearing systems. It’s a little like the Lasersaur project because I’m challenging existing solutions that work well.

It really takes a lot of time to figure out what the real barriers to ownership are.

CNC Laser Cutter construction

I decided to make a CNC laser cutter with some material handling capabilities for my model plane project.

Water Cooling

I have some surgical tubing for the water jacket (the CO2 laser tube is water cooled). It seems to brown the water and I don’t like that since it might accumulate some material in the water jacket. I realize I can clean the jacket with hydrochloric acid but I figure it’s a better idea to not have any material going through the system.

The laser itself has some nice white colored tubing on it to connect the various water jackets. I’m curious if anybody knows what it is so I can replace my surgical tubing with it.

Some of my friends are saying it’s silicon surgical tubing or fuel line from a motorcycle shop. I will check that out the next time I’m at the bike shop.

CNC Software

I went with EMC2 which runs under Linux. It is very easy to use. You need to download an ISO file from then burn a CD from the ISO. Boot from the CD and presto! You have a CNC controller running off the CD without affecting the host computer. I elected to install the Ubuntu Linux software with the EMC2 right onto the computer so that I could boot from the hard disk. This allowed me to run updates and such. It upgraded everything to the latest. This is a Ubuntu 8.04 LTS. I think they are working on getting the next one going which is Ubuntu 10.04 LTS. I will upgrade when I become aware that it is ready and it is stable enough.

My lab does not have any network cables running from my server room. So I installed a Wireless Bridge. I made if from an WRT54G v8 (Linksys router) by installing DD-WRT open source software onto it. I then configured it as a wireless bridge and had it join my network which has a higher end router (also running DD-WRT). Now I have 4 eithernet jacks to plug stuff into to have internet access in my lap. I can remotely access the lab computer across the network (this includes remote desktop and ssh).

EMC2 is pretty easy to configure and run. It didn’t take too much fiddling to get it programmed with all the parameters of my gantry system.

CNC Steppers, Controllers and Power

I initially started to build circuits to control stepper motors. I discovered that you can buy components that are not too expensive. I got a xylotec 4-Axis kit. It comes with 4 large steppers with double shafts, the 4 channel controller and a 24V power supply. The EMC2 software had this outfit listed so it was easy to get up and running. All I really had to do was tin all the wire ends and screw all the wires together. Apparently with this outfit, if you wire something wrong it fries. So I was extra careful. I’m very pleased with this system. It is very powerful and is quiet.


I made a gantry out of 1″ aluminum angle. I used the contraptor methodology of drilling holes and using nuts and bolts. I designed my own sliding elements to use ball bearings. I used yoyo bearings. I used XL timing belts and pulleys which don’t seem to have any back lash.

Also, I went through all the belts sizes and prices to find that 88″ belts were relatively inexpensive. I made a spreadsheet to compare every belt size to it’s price so I knew if I should buy belt by the foot or to a specific size. I eventually went with 88″ belts and decided to just pull the slack out. You can see how I did this in the pictures. You can also see the gantry detail and the Y-Axis truck detail. It runs pretty fast and smooth.