2'x4' HOBBY LASER TABLE (redux)

First off I made a huge mistake and accentually deleted my instructables account. Long story short remember that Autodesk and instructables are the same. The old archive of the article is |here| luckily the internet is forever, well at least the text is.

What are you looking at?

This is a CNC diode laser table. It is not fast but it can cut big projects. Its made of steel since its way cheaper than extruded aluminum and it is an exercise in bumbling through the unknown. When I started building this laser I had no idea what I was doing but figured heck its possible, people have built them before it is time for me to give it a try.

Getting the bug to build.

Early 2018 I was talking to a co-worker about how cool would it be to have a laser table, I have a 3d printer and it is great but has its limits with size print speed. The cheap Chinese lasers are nice but most of them don't even cut the size of a sheet of notebook paper. Building one would be too expensive.

At that time he said "You know the price has dropped a lot to build one."

I checked Amazon and saw a 3 axis Arduino CNC with motors, CNC shield and limit switches for ~70 dollars. I was in hook line and sinker.

What am I going to build?

• Why 2'x4'
• The expensive stuff is the hardware and laser. The frame is the cheap part. 500 dollars for a small machine but 10k for a big one that can do 2'x4'. Mainly because its in the size that can make production money. I'm not in the making money business I'm in the hobby business but if I figure out how to make a small machine I have everything to make a big one. I'm going to go big! I can get full sheets at Home Depot and have them cut them down to 2'x4' sheets easily.
• Aren't Diodes slower than CO2 lasers?
• The quick answer is yes they are. I am not making a production machine. The cost of the Diode itself is more expensive than the CO2 laser but I am not buying mirrors, water pumps and cooling for a more powerful laser and I don't have to make room for a 1m long laser tube.
• More on mirrors one of the big things with larger systems is keeping it square enough the laser doesn't lose focus a diode sits on top of the gantry and it isn't affected if it moves 3" or 3 miles.
• Where are your plans?
• Yeah besides the electrical I have zero. I winged most of this project. I wanted to tackle one problem at a time and proceed from there. The basic idea is a x/y laser with a large gantry to hold the laser, enclose it and start having fun with the build and building with it.
• Why use steel opposed to extruded aluminum?
• Yes, most DIY machines like this are made of extruded aluminum, from what I can tell this is mostly due to the fact that many people who DIY this type of machine cannot weld. Welding steel gives me more flexibility in how and where I attach items. I can drill and tap anywhere and weld at any point, with extruded aluminum you have minimal, dims you have to follow and can only mount realistically at the interval of the rails.
• Cost, welding has a higher upfront cost in tools but a much lower cost in material. Most of the square tube I have might be 2.20 a foot while extruded aluminum will be double that and I have to buy the T-slot fasteners.

I used a lot of tools but reality basic toolset is required. The main thing you need to buy is a laser module and an Arduino CNC kit.

To have a CNC machine you need to have a computer. Computer's don't interface well with machines that don't use USB.

The Arduino and CNC shield.

The kit I bought off of Amazon had everything short of the tool and linear bearings to make a CNC. If you are familiar with Instructables I bet you have seen an Arduino before, they make a CNC shield that most people use to make 3d printers using a program called GRBL.

Need to stay organized.

From the start, I wanted to treat the circuity portion like a proper control panel. I mounted the 12v power supply at the far end, then wired 2 terminal strips, +/- to land equipment then added a series of 120v and 12v relays to eventually act as safety switches finally mounting the Arduino at the far end. Once I got this system working I would eventually mount it.

Baby steps

I didn't go from I have no idea what I'm doing to build a laser table in 2 days. This journey started in Feb 2018 and is still going Feb 2019 as I am writing this. At this point the laser works just haven't enclosed it, hopefully by the end on this instructable you will be seeing a complete machine.

So baby steps, GRBL has to be downloaded to a computer then you upload the firmware to your Arduino. They do a good job of explaining the process on their website. After I wired up the panel to give 12v to my board I upload the firmware and turned on the GRBL program to see if everything works. x/y motors spin when I tell it to go a direction. THE FIRST STEP ACHIEVED!

The previous step I got the motors to spin pieces of paper doesn't sound all that thrilling but it made my week.

What motors did I get

The kit came with Nema17 stepper motors, I had no idea what that meant at the time but googled it up. Steppers precisely control pitch and have high torque these motors have 200 steps per revolution so they can natively control 1.8 degrees per step. I wanted to gear it down so I had a 3/1 reduction, this would give me .6 degrees per step and up my torque without having to use half steps. I will explain half steps later when we talk more about the board.

To get the gear reduction I used GT2 belt and gears, the drive gear is a 20 tooth gear that is belted to a 60 tooth gear giving me a 3 to 1 reduction.

What do the motors move?

I, of course, wanted the motors to move the x and y-axes. The X will be a single motor on a bridge and they will have 2 motors that move the bridge over the work. This will give me decent control and the heavy moving would be shared between to motors.

Making the Y-axis mounts.

Bearing blocks are expensive, just throwing that out there. I found I could buy a stack of 5mm bearings for cheap and print my own blocks out. These got mounted to a larger frame to hold the gear reduction shaft above the motor then I bolted the motor down with the motor mounts the kit came with. The shaft is held into place with some set screw rings.

The base is 10 gauge steel, roughly 1/8" thk. It is way overkill but from my limited experience with machinery, a heavy base helps keep everything accurate.

How big to make it?

You need to have enough room to have your belts clear the sides of the work surface and allow your laser to to reach edge to edge without hitting anything. For my project I found about 4" on either side gave me enough room to mount my motors and I figured 3" on the bottom would give me enough for the head of the laser.

To lay this step I placed a 2'x4' piece of wood to simulate my part then I cut out I cut 2 48" long 1.25" square tubes to act as my left and right side. Clamped one square tube to the far side and move my parts around until I figured I had enough room. As I said earlier this turned out to the 4" and 3". Once I was happy with this I marked my sheet metal and cut it with a circular saw. A standard circular saw can do amazing work with a metal cutting blade.

After the base plate was cut I cut the 2 remaining square tubes to frame in my base plate and stich welded 1 on 12 skip weld. I did this as the weld is plenty strong for what I am doing and I didn't want the material to warp with excess heat. I figured I could seal the project with caulking at the end.

What do the axises move on?

For smaller projects, I have seen rods and that that is what my 3d printer runs on, they work well for short runs but long distances they tend to move due to not being supported. I chose a linear bearing, a cheap one from Amazon, not the multiple hundred dollar options you can get for real CNC machines. I made a decision a while ago that I was not going to make this thing accurate down to 1/128th of an inch. If I get 1/32" accuracy I would be happy. Making this decision drops the cost to around 1k opposed to several thousand dollars. Keeping precision over a long distance gets exponentially more expensive.

Y-axis bearings

I took the same 1.25" square tube and used it as a mounting rail for the bearing. Layed a line down the square tube marked and tapped the holes. To make the Y-axis square to the base I used 2 small pieces of 1.25" square drop at a spacer and clamped it to the side of the frame of the base. Then tack welded it into place. I repeated the same step for the other Y-axis bearing. I wasn't too worried about the two bearings being 100% parallel. I will explain that in a minute.

X-axis bearing

The X-axis bearing bridge is only hard mounted to the left side Y-axis, the right Y-axis bearing it sat in a U shaped pocket that fits very tightly. This is done for a few reasons, the first is with only being bolted to the left Y bearing makes the assembly easier to disassemble. The other reason is if the 2 Y-axis bearings are slightly out of parallel I have a slip joint that will not cause a bind.

Won't this cause crooked cuts

Technically yes but it will be so small I have yet to see anything I can measure. This is because I have 2 motors moving the Y-axis on either side of the bridge.

Time to mount the laser.

I purchased an Endurance 10watt laser. This laser is designed to be able to retrofit on 3D printers so it has a simple bolting pattern to the back of the laser, the diode and fans come pre-assembled I just have to bring power to both the diode and the fans separately. The complicated part comes with the fact that using different lenses you have to adjust the height of the laser to focus it on the product you plan on cutting. Also the thickness of the product you have to adjust for this too.

I chose a manual Z adjustment

I know I could have gone with a cool stepper actuated adjustment I chose a simple manual method, I figure I once I set the height I won't be changing it for a while and a manual set up will be lighter and simpler. For the adjustment, I chose a 6" linear bearing with a 1/4" ACME thread screw to screw 6" long for my up and down adjustment. Once the height is set I would just screw down the free nut to lock it in place. Later I decided to weld another nut to the top of ACME thread so I can use a ratchet to quickly move focus the laser.

X-axis control is made with the same 20 to 60 GT2 pulley. Instead of having the bearing mounts to the front and behind the motor I opted to put both on one side of the motor. This is also where welding the assembly helped. I was able to weld it to the top of the angle web, something I couldn't do with extruded aluminum.

At this point we have bearings for the x/y-axises, have the laser mounted and have the motors positioned to do the work. Its now time to add tension pulleys to make a working belt system.

Tension pulleys

I added 3 tension pulleys to the system all of them are made the same way. I took the 1.25" square tube and cut out one wall completely so I had a U profile. From here I mounted 2 of the bearing blocks I printed before with a 20 tooth pulley. Measured out from the base to make sure they are square to the drive pulley and tacked it into place.

The Belts

You can purchase GT2 belt in long strips, I purchased mine in I believe 5-meter rolls. This makes making drive belts very easy I looped the belt between the 2 pulleys and cut it about 2" longer than I thought I needed as its easier to cut more off then cut a new belt.

Clamps

Since this isn't a system that will continuously revolve like the down gearing belt that is attached to the motor mounts I can just clamp the belts to the X-axis bearing bridge and to the Laser mount itself. You can see how I did this in the first 3 pictures of this step. I found the best tension would be to have the belt 2-3 teeth too short to easily loop the belt around the pulley. Clamp it the belt down then use a spare 5mm rod to pop the belt over a bearing.

Wireway

There will be wires everywhere along with an air assist line. If the system was small aka sub 12 inches you might be able to get away with a wire wrap and have it suspended from above. This will be moving 2' in one direction and 4' in another. For this, I chose to use a flexible wire way too keep all of this from getting tangled. In the 4th picture, you can see the first wire way, it loops on itself and lays on the base plate. The second wire way on the X-axis is supported by 3 1" wide pieces of flat bar.

Laser Bed

I talked to some friends and the suggested using a ceiling tile light louver as a laser grid. It comes in 2'x4' sections and is a uniform and really cheap so if it gets burned up over time it is an easy replaceable consumable.

Oh, I also mounted the CNC board.

Nothing complicated here just mounted it at the edge of the board out of the way and mounted it on some stands so I can put ductwork under it.

Wire the laser

The laser I purchased has a control box for it, it takes 120v in for power and uses a 0-5v signal to power the laser. The hardest part for powering up the laser was finding out where to land the control on the Arduino board. If your trying to find it this on the internet I will save you the time. Wire the 0-5v signal to the Z+ end stop plugs if your using a CNC shield. At this time the laser turns on and I was happy the main part of the machine works. Still not the time to start burning.

Calibration of the X/Y-axis

At this point in time, the machine doesn't know how to send a 1" move signal to the machine, 1" signal might be a 12" move or it might be a half inch move. In GRBL there is a multiplier for calibration.

StepsPerMM = (StepsPerRotation/mmPerRotation)/MicrosteppingFraction

Since I geared down opposed to using micro stepping my Microstepping Fraction is 1 so its was steps per rotation/mm per rotation to get my steps per mm. This will get you very close due to measuring and rounding errors. To correct this I taped a 1/32" welding rod to the laser, placed a ruler below the rod pointing at the 0mm and gave a command ton jog the laser 100mm in the X-axis. I would get a slightly different measurement and then I would use that measurement to correct my StepsPerMM

StepsPerMMnew = 100mm/measuredMM * StepsPerMM

I did this a few times until I could get out to 300mm and stay accurate as I could see. I chose 300 mm as its the full 12" ruler to a solid number.

Testing Testing Testing

Several days of testing followed, I found I developed a problem where the Y-axis it would not return correctly. I found that the pulleys would rotate on the bearing bar. This was fixed by filing small flats where the set screws would mount.

Also using the cutting lens I had to get very close to the wood to cut, while it worked it was so close I easily got debris building up on the lens. I searched and found a G-7 lense that would allow cutting at a farther distance from the wood ~6cm away. Since then I have had no issues.

My wife has been acid etching cutting boards for friends as gifts for a while now and they seem to love them. Our friends put them out in the center of their kitchens. At this time my wife has seen me make a box that can barely cut circles and squares. I quickly got her attention by burning our last name on a cutting board.

At this time a few safety notes, don't look at the laser, don't even look at the reflection of the laser. Make sure you have proper eye protection. Make sure you in a well-ventilated area, since this isn't inclosed right now I have my garage door open/cracked with a box fan blowing everything out, I am also sitting upwind so I don't blow the smoke into me.

If you were always leaving the machine in a well-ventilated area where you don't have to worry about people accidentally looking at the laser or reflections of the laser, you could be done. Its a fully functional laser table.

My machine was involved in a side project.

One of my friends loves his Tacoma so for his birthday some of our friends asked I could burn an image of his Tacoma for as a gift. I was feeling confident, I mean I could burn a circle how hard could it be. It was extremely hard to do. I learned a lot about my machine and how to burn an image. Due to the clandestine nature of this project, I have minimal pictures as I didn't want to any chance of sharing an image on Facebook of my progress and having a half image of a Tacoma in the background. This project was worked on for about 3 months until I got a good image.

Things I got out of this:

1. The stepper controllers do overheat if you don't have a fan on the heatsinks and they loose steps.
2. There is a small zone of not burning, burning an image and a black blob.
3. Images are really slow with my laser due to somewhat slow travel speed. Burn throughs are iffy, seems like the low-end movement is too high.

Fixes

1. I printed a duct for a 30mm fan to blow on the steppers.
2. Burn black as you can but use stippling, works like how a newspaper prints an image.
3. The higher speed I just had to drastically up the max movement speed of the machine. The low end was a bit harder to fix. Found out that upping the micro steps from 1 to 1/16 allowed my low-end movement speed to drop very low. Went from ~4.9 steps/mm to ~49.9 steps/mm.

The biggest issue is getting the reliability up for a continuous 6-hour project. Due to the nature of the test, I would only get 2-3 attempts a week. One thing that significantly helped reliability was using a program called Lightscribe, a very user-friendly UI that made the work easier set up. Also while the computer doesn't need to be exceptionally powerful you want to have a dedicated computer for the project. If you are doing something as simple as using chrome and you cause a small communication error to the laser due to any issue you might as well toss out the project. I even go as far as turning off wifi so there are no chances of background updates.

Making an exhaust system.

If you are familiar with making computers its a common practice to have more air trying to enter the computer than exit so the case is positively pressurized so you don't suck in dirt into sensitive parts. With this laser I wan the opposite. I know I won't be able to 100% seal the system, I want to pull as much air as possible out. For this I want an exhaust manifold in the back of the box to pull the smoke out. For this I let the laser do the work.

Since this is my first 3d built item it I cut was a bit of a learning curve, burn 1 (image 1 here) you can see I had straight edges, this did not want to go together. Burn 2 (image 2) I used tongue and groove to lock it together and it is set up for 6 120mm fans. I should exchange out the air in the box every 10-20 seconds.

If you are a wood worker... I'm sorry.

The fans are set up to run off of 12v DC power similar to computer case fans... exactly like computer case fans. Instead of setting up a system where I strip every wire and run it back to a terminal block for ~20 bucks you can get a computer case fan hub. In my mind worth it for the simplicity and the capability of easily adding more fans.

While wiring this up I added 2 more terminal strips for 12v DC power. For the terminal strips I did not want to use a jumper bar because I didn't like the idea of an exposed hot bar and if for some reason I need to use the strip to act as a junction not tied into a common +/- I can just unplug a jumper. I chose the top of the frame this will make more sense later as it will be easier to access once the case is made. Installing the fans I rand into a hurdle, computer case fans are designed to be screwed into some thin sheet metal. The screws that came with the fans didn't have enough length to grab the wood through the fan case. I have some rubber inserts I thought would work but the "case" side of the insert will not work with 3/16" thk wood. So I just used a dab of caulking in each of the 4 corners. Its generic white caulk, not a construction adhesive so if a fan dies I can pull the fan off later.

The goal is to have bolt on panels so I can access the components inside of the laser if/when its needed and add some more rigidity to the project. About midway through the project, I made this section of the frame to let me know of the envelope I had to deal with. I will need to add at least 2 doors, (1) the man part door (2) the electrical maintenance door.

The walls will be bolted to the frame to allow whole sides to come off if needed.

Air blowing on the laser impact location does a few things.

1. It helps blow away residue.
2. It acts like stoking a fire, hotter the burn the deeper the cut.

If you are working on a part that requires several passes you develop a carbon layer on top of the cutting surface that makes it much harder to burn through. So adding air will allow the laser to do more work.

I have seen many people use an aquarium aerator pump. I have this old airbrush pump I was given years ago. I am using what I have and it puts out an impressive volume. It is tubed to a 1/8" brass tube that is clamped to the laser mount. I slowly bent the tube in stages in a vice to keep it from kinking.

Completely sealing the case sounds like a good idea but I need air to leak in so I can have air leave exhausting the smoke from the laser. Leaving small slits along the bottom of the case will allow air to be sucked in along with 2 120 mm fans blowing air in. 1 over the electricals and 2 in the front of the case to set the flow of air.

I found camper seal foam tape is 1.25" meant to be for my 1.25" frame. The doors will get the window seal as its designed to move more than the camper seal.

As much as I hate to admit I did. The machine was on its way to be a monster 4'x5' machine and we didn't have room for it. After talking to the wife we decided its best to stop the project now and disassemble the parts so we can use them elsewhere.

We bought into the Snap Maker kickstarter. It will be the laser we want and I can retrofit the larger laser to it. They are also planing on making 2m beams for it so we can actually make a larger machine. It will do 3d printing, laser, and router table. With expansions for rotary, 1 hp router and the list goes on. The wife and I are both happy about it.

I have since cut apart the laser table, stored away all the electrical parts, use some on some car projects and the laser is stored in a safe location. The table it self was cut apart and made into a better welding cart the 1/8" thick base plate and frame was kept as a 4'x5' table top that I put on saw horses if I need additional work space.

Its a sad day for DIY but it was a very fun time. People I know asked why are you building a laser table when you can just buy. I told them it was for the fun of it and to see if I can. When I cut it apart they were shocked I would do it. I know know I can make a laser table, made a few things with it and a better option showed up. I am going to be able to much more with the new machine and it will be a much better tool than the one I was making. Mine was going to be very analog while this one will have some huge software advantages.

Anyway if you plan on making your own laser it is totally doable. I had almost no knowledge of how to do any of this when I started and even after throwing in the towel I thoroughly enjoyed the project and its not going to stop me from making relatively absurd tools in the future.