2W 445nm Diode laser cutter/engraver in a suitcase

Over Christmas, I was mentioning to my brother that I purchased a CO2 laser cutter and was in the process of upgrading it. He mentioned that he would love to use it to cut airbrush stencils (he does auto detailing.) The stencil material is about 3 mil to 10 mil thick in various materials including mylar.

Anyway, I said the CO2 laser cutter was probably overkill for something like this and claimed that I could make a laser cutter in an aluminum briefcase for less than $300 using a laser diode. So, he took me up on the challenge and here we are now.

Looking around at options, I couldn't find any good sources for laser power vs. cutting capability. So, I looked for what what the most powerful I could get for a reasonable price. I settled on a 2W 445nm laser diode with a GaAs lens, copper host, and a built in driver (as I understand this diode can go up to 3W if an appropriate driver is made and the laser is pulsed) for simplicity. $125 shipped. I also picked up an aluminum heatsink to fit the laser.

I received the items today, so the build has officially started.

Before I get too far into it, I wanted to mention this: I set up the laser with a momentary pushbutton just to make sure it was working. Knowing how powerful these are (and not having the proper safety glasses yet) I set up a black target, looked it the opposite direction of the beam and pressed the button. It freaked me out how bright this diode is. So, now I am quite paranoid about the safety of this laser and know I need to do a lot more research before calling the build finished. These things are no joke! It may have been misinformation, but I had been reading that at about 3W, these 445nm laser diodes are as powerful (cutting ability) as my 40W laser. I am skeptical of that, but that was what I stumbled on in several forums.

On to the build...

So we start with a typical aluminum equipment case. Dimensions are 17.5x12.5x6 inches. The stencil material is found in either 8.5x11 or 8x10 sizes. I decided on a 12x12 cutting area since the electronics will take up very little room. I am basing the XY stage on the CoreXY design which is very compact and takes up only a little more room than the cutting area itself.

I decided that I wanted to use parts that I could find in a typical hardware store as much as possible. I also have a 3D printer which will assist me in a great deal to get just the right design.

The corexy platform consists of two steppers with gears/pulleys and 8 idlers. The motors are mounted parallel to each other and both are driven at the same time to produce the motion. Since the carriage will only be carrying the diode with heatsink, I am able to get away with a very lightweight system. Instead of timing belts and gears, I will be using Spectra 100lb braided fishing line and custom 3D printed pulleys. The Y axis rods are standard 1/4" steel rods you can find in just about any hardware store and 1/4" brass bushings. The X axis consists 1/2" aluminum C channel.

Attachment 263112 All the parts with the C channel and laser heatsink for clarity
Attachment 263114
The laser carriage up close with details shown

Attachment 263116
And the guides for the Y axis. The other parts shown are the motor pulley, idler pulley, spacer, and the stepper mount

Attachment 263118The laser mounted in the carriage. That is actually a prior iteration. But the new iteration is just beefier and a better fit.

Attachment 263120 The gist of the mechanism. I need to cut the metal rods and C channel drill and rivet the guides to the C Channel, and polish the rods to ensure uniform thickness.

The aluminum case will serve as the majority of the structure. The rods will be mounted to the walls of the case. The rear pulleys will be mounted on a bracket that will also attach to the back wall. Same with the steppers on the front wall. I tore out the felt lining from the case mainly for static reasons. I haven't decided what I will re-line it with yet.

Attachment 263122Cheap 5V geared steppers common on Ebay (2 for $6, I think) with the gearing, they achieve a step angle just under 0.9 degrees. The gearing is very strong, I can barely turn the pulleys with my fingers. I will be using these same motors, but I ordered another set with a more friendly mounting arrangement.

For the electronics, I will be using an Arduino Uno with CNC shield and DRV8825 stepper drivers capable of 1/32 microstepping. A customized GRBL firmware will be loaded and many g-code sender programs are capable of supporting GRBL. This may get a customized open-source treatment, though.

I started to cut a window in the aluminum case, but realized after seeing how bright that laser is, that standard anti-glare plexi isn't going to cut it. I am trying to research what I can use that would be relatively safe. Transparent red acrylic might work with a UV protection film applied.

The case has the combination locks on it, and I will be interfacing interlock switches to these that will only allow the laser to operate when the case is latched shut. The case will be decorated with brass hardware and the build area lined (and stiffened) with anodized sheet metal.

cool!

with laser machines its alwasy wise to protect your eyes. There are glasses you can buy specifically made to block out wavelengths from certain lasers. Most machines come with a pair, but may pay to invest in a pair?

Cant wait to see the finished product.

Side note... be wary of what materials he wants to cut (i am not familiar with mylars chemical properties), some vinyls etc are bad news for lasers as they give off hydrocholoic acid :/

The bought stencils made of mylar are lasercut, so no issues there. There are several posts about using it for solder stenciling as an example. My only real concern is that the uv diode may not be able to cut it since it is transparent/translucent. I keep reading that a visible beam laser cannot cut transparent or white materials, yet I keep seeing examples of people doing it. So, *shrug*

I see posts like this where people do it: DIY laser cutter built to make stencils | Hackaday

Notice he is using a 1W diode... lol. I went way overkill here. Thankfully this first one is actually mine :) My brothers will probably get the 1W diode to knock the cost down quite a bit. The article above also mentions that the builder planned to add a slide out drawer to put the stencil blank in and no windows. Hmm, not a bad idea and easy to do with the corexy platform. The entire build area is un-restricted.

You can read more about it here: CoreXY | Cartesian Motion Platform

And here's a picture of the concept:

Attachment 263150
image credit: DUBL-CROSS corexy gantry by BV3D - Thingiverse

As for safety goggles... absolutely. I only have ones designed for the CO2 laser. But I ordered up a pair for the blue diode. I won't be playing with the laser without them again, trust me. The entire room lit up from that laser.

One application I had planned for this was to expose photosensitive PCBs by drawing the traces right into it. Though I am sure 2W is again way overkill. With the carriage design, it should be easy to swap out different lasers.

you would wont to add a small fan to blow the fumes away from the laser smoke can wreck the lens

Quote:

---

Originally Posted by daniellyall

you would wont to add a small fan to blow the fumes away from the laser smoke can wreck the lens

---

There will be a 12V air pump for air assist and 120MM fans in the case. I just haven't gotten that far in the build details yet. Definitely requirements. :)

Have you been able to actually cut anything with this laser? That doesn't look that heavy and I do a lot of cardboard cutting and I always am thinking it's a waste of the laser time so I was wondering if it would adapt to the CO2 laser head for cardboard runs? Are these life limited like the CO2?

Quote:

---

Originally Posted by buddydog

Have you been able to actually cut anything with this laser? That doesn't look that heavy and I do a lot of cardboard cutting and I always am thinking it's a waste of the laser time so I was wondering if it would adapt to the CO2 laser head for cardboard runs? Are these life limited like the CO2?

---

Not until I am absolutely 100% positive and educated about the safety.

Here is an example of what a 2W laser can do!:

https://www.youtube.com/watch?v=PbWGjPByXKk

https://www.youtube.com/watch?v=a0AD5kJ6Wm0

This is just at 1W (warning annoying music):
https://www.youtube.com/watch?v=__5CpKlr4-4

This thing is no freaking joke. Far more powerful than I had recknoned and I have much more respect for it than I did at the outset. The module I bought has the 2W driver which is not directly adjustable. I cannot turn it down for focusing and alignment tests or cutting/engraving lighter materials. I am looking at other drivers or the possibility of rolling my own PWM controlled supply and getting a diode with GS lens and with the diode pre-pressed into the housing (the hardest part.) With a proper driver, you can drive these down into the milliwatts and up to about 3W with proper cooling. At 2W, you are driving 1.8A into the diode. One watt is about 750mA. Any constant current supply will work. Many people use LM317 in current regulation mode to build simple drivers. I just ordered an adjustable 3A constant current regulator board for $5 from China meant for LEDs. Point being that the drivers are the relatively easy part.

I have been also reading that this 2W laser should only be run for a a couple minutes max without active cooling. If you care for them properly, they should last several times longer than a CO2 tube. Heat is really the only thing that kills them.

I think a 2W diode like this is also way overkill for cutting cardboard and would likely catch it on fire very quickly. Even at 1W, it is probably overkill.

The diode and heatsink are very light. The copper host is probably the heaviest part. I will measure it on a kitchen scale, but the weight is no more than a small handful of pocket change. After seeing some videos on youtube (search M140 laser or 445nm laser on youtube) I am less skeptical about the power of this diode coming close to the power of the 40W CO2. The main difference is that CO2 being IR can burn a wider range of materials while a visible laser will just pass through transparent materials and reflect from white or shiny materials. It will heat them, but won't be able to cut them.

Quote:

---

Originally Posted by buddydog

I was wondering if it would adapt to the CO2 laser head for cardboard runs?

---

Missed this question. The heatsink is 19mm at the widest part and 15.5mm in the thinnest part. The space between the fins is 1.9mm. The head mount on the K40 is 12mm dia. So, you would need a modified holder. I redesigned the holder in sketchup, so you could easily change the diameter and make a holder that fits. For that matter, it shouldn't be much trouble to make up an adjustable holder. I can throw something together for you

I don't know the focal length of the laser lens. It should be adjustable. It is a GaAs lens, so the focal length is typically longer than ZnSE by design.

Hmm, I just went into my sketchup model to mash something up and post it and the thickness is 2mm, and the holder diameter is 15.7mm. So, it would work as is (by opening the front of the holder to slide the heatsink in.

EDIT: Here you go:
K40 Laser Cutter New Lens mount by Retroplayer - Thingiverse

Attachment 263174

You can use zipties, wire, or whatever around the prongs to secure the lenses, but should be a tight fit already for both.

Metal has been cut, rods polished (though I need to make a jig for this to be more accurate... I do have an old cordless drill with dead batteries...)

I began some assembly and realized I needed to tweak a few things.

So, the carriage was redone: Bigger front face to mount a fan or peltier/fan. Added tabs to the edges to mount the filament line instead of screw holes on the top. Tweaked the contact points with the heatsink to lessen the amount of contact (in case it gets hot enough to melt the ABS.)

Attachment 263176Attachment 263178

I designed the rod ends that will get installed in the walls of the case.

The Y Guides were redone to fit two brass bushings (one on each end) and I thickened the tab that goes into the C channel stock. It was far too thin and cracked when I installed the rivets. Also with the rivets, you cannot remove the carriage. I will be switching this up to 4mm screws and the tabs are now the full depth of the C channel. I also adjusted it to be printable in the orientation to give it the most strength. It still needs supports, but on a superficial face than can be sanded.
Attachment 263180

I eventually found some acrylic window material with laser safety ratings, but it is $200 per sq. ft!!!

So, instead... I will be installing a cheap board camera and LCD monitor for a simulated window. I can replace alot of board cameras for $200!! I will need to get one with auto-iris control or modify the lens so it doesn't get saturated when the laser comes on. But if mounted properly, it shouldn't suffer any direct hits and if it does... these things are like $20. Much cheaper than blindness.

EDIT: Two cameras and a 7" LCD - $55 Cameras are probably crap, but I expect them to give up their sight for me so no complaints at $15 a pop.

Build is slowing down a little. I am waiting on some parts right now and also doing a lot of research to understand this laser better. I am also tweaking the mechanical stage right now since I have the metal cut and I can dry-test things a bit. Identifying several things I need to tweak and also trying to make the printability bullet-proof as much as possible (3D printing is an art form sometimes!) You want to ensure you have a good first layer surface, that the layers are orientated for strength, and that you reduce the amount of support material especially in areas where tolerance is very important. The carriage is the hardest part.

For the carriage, I was originally going to use round standoffs and bearings to create rollers, but the ABS on aluminum moves really smoothly as it is. I shaved off .2mm and will try applying teflon tape to the interface between the plastic and metal to make things glide really really well. But even as-is, it moves with very little resistance once I sanded and buffed the aluminum C-Channel.

The 1/4" steel rods I picked up from the hardware store are a different animal. They are not precision rods, so I am experiencing slight sticking in a few places on the bushing. I do have some 6mm precision rods and linear bearings coming, but I really wanted to design this with hardware store bought parts. One way to get the rods smooth is to attach the rods in a drill chuck or lathe and sand with ever increasing grit paper. Stopping to keep an eye on the diameter in several places (you can mark areas where it is sticking.) You need to move the sander up and down the spinning rod at a relatively even pace. It doesn't need to be absolutely perfect, but you don't want sticking from too much thickness and you don't want a bunch of slop from too thin, either.

I have also needed to return to work starting today, so that will interfere a bit. Anyway, I should be at a point to show off the mechanical stage under GRBL control in the next few days I think.

Just wondering if your Lazer would cut into 2 materials,

1) Corrugated Cardboard with a thickness of around 35mm ( 1 1/2 inches )

2) An outline Drawing onto MDF ( Craftwood ) with a cutter depth of around 1 to 2mm ( 1/32 inch )

Thanks

Ken

Quote:

---

Originally Posted by KenR60

Just wondering if your Lazer would cut into 2 materials,

1) Corrugated Cardboard with a thickness of around 35mm ( 1 1/2 inches )

2) An outline Drawing onto MDF ( Craftwood ) with a cutter depth of around 1 to 2mm ( 1/32 inch )

Thanks

Ken

---

If you have read up to this point, you can see that I haven't finished the build yet. I also haven't tested much with this laser yet because I am waiting on my safety goggles. I could only take a few guesses to answer your questions, but at the laserpointer forum, I am sure there are many with more experience that could answer.

For my simple answer: with enough time and energy, you could likely cut through just about anything as long as it will absorb the energy of the laser, the point remains focused, and it does not reflect the beam. However, that doesn't mean it will be pretty and certainly not safe if you have no way of keeping flames down like filling the cutting area with an inert gas.

I don't believe this laser is what you would want to cut those materials. The cardboard would likely catch on fire after only a mm or 2.

I suspect this diode can do what you are asking with the MDF (which is essential high-density cardboard really) at 1/32 inch (which is .79mm) but likely not much more without more time and a way to keep the workpiece cool during the cut.

With any visible light laser, you are limited in the materials you can cut by how much of the light they absorb vs. the light they reflect. And reflecting is extremely dangerous! This is what makes the CO2 laser so effective as a cutter. MDF may be too reflective for this laser. I am not sure.

I think a CO2 laser is more suited for your task. But not so much for the 35MM cardboard. The limitation would mostly be that the depth of field of the optics in the CO2 cutter is only about 4mm before the beam becomes unfocused which will burn instead of cut. If you moved the workpiece closer as you cut, you could go deeper, but then the edges of the cone shaped beam which is now unfocused will hit the edges of the cut and start a fire. I personally would just using a jigsaw to cut that. Much faster and more safe.

I suspect this is the #1 cause of the fires people report (an unfocused beam either by not setting the correct distance of material to the lens, trying to cut too deeply, or trying to cut material that diffuses the beam and makes it unfocused.)



Once I get both of my cutters up and running, I will be experimenting in the area of maximizing the depth of cuts. I have some ideas for techniques to try, but I wouldn't throw them out there without trying them first. :)

I use cardboard for fixturing aircraft parts and the only way to make that work is cutting multiple sheets and gluing them together over a known good form. A jig saw wouldn't be accurate enough for my purposes. The 40w cheapie is dead on so works good. Cardboard is crappie to cut, smelly and flames up a lot.

I am wondering just how valuable the auto z set will be for hobby use? After using mine for a while I find I prefer the spacer tool. The one advantage the auto z would have maybe is safety possibly. Any time I am fiddling inside the case I remember when I first got my machine and unknown to me (at first) it was intermittently firing on it's own from time to time. I am just lucky I always respected the beam path with my fingers. By the way after using this thing I am not so afraid of it and the one interlock I hooked up is now disconnected. You do have to open it from time to time and it's a bit like a propeller on an airplane (except not near as dangerous) and you just have to always respect it. Just like my large metal lathe, I keep bulk quantities of chop sticks at the machine and any time I open the lid the first thing I do is grab one and that's what I move things around with if the power is on to the machine. Helps me to remember what I am doing. During alignment I always have a paper precede any reaching in to the beam area or I stuff a piece of aluminum up in the corner to block the beam entrance although I usually just turn the power off to make any adjustments. I am used to stopping at critical points, taking a breath and thinking before doing something I know could hurt me. Just like pausing at the end of the runway, looking for traffic,scanning controls and instruments and checking traffic again and THEN committing to going. With the CO2 I only have my regular glasses which will block the beam and I wear them all the time.

By the way if you use the K40case there is a hole in the bottom, I think the air entrance, and I have had it scorch a paper I left under it once. I fashioned a piece of sheet aluminum to be under there to catch parts and stop the beam. All aluminum should be dull in finish so it absorbs, not reflects.

The K40 25 watts is I think a lot different reflection and safety wise than say 200 watts.Still a good machine to develope good habits on.

Quote:

---

Originally Posted by buddydog

I use cardboard for fixturing aircraft parts and the only way to make that work is cutting multiple sheets and gluing them together over a known good form. A jig saw wouldn't be accurate enough for my purposes. The 40w cheapie is dead on so works good. Cardboard is crappie to cut, smelly and flames up a lot.

I am wondering just how valuable the auto z set will be for hobby use? After using mine for a while I find I prefer the spacer tool. The one advantage the auto z would have maybe is safety possibly. Any time I am fiddling inside the case I remember when I first got my machine and unknown to me (at first) it was intermittently firing on it's own from time to time. I am just lucky I always respected the beam path with my fingers. By the way after using this thing I am not so afraid of it and the one interlock I hooked up is now disconnected. You do have to open it from time to time and it's a bit like a propeller on an airplane (except not near as dangerous) and you just have to always respect it. Just like my large metal lathe, I keep bulk quantities of chop sticks at the machine and any time I open the lid the first thing I do is grab one and that's what I move things around with if the power is on to the machine. Helps me to remember what I am doing. During alignment I always have a paper precede any reaching in to the beam area or I stuff a piece of aluminum up in the corner to block the beam entrance although I usually just turn the power off to make any adjustments. I am used to stopping at critical points, taking a breath and thinking before doing something I know could hurt me. Just like pausing at the end of the runway, looking for traffic,scanning controls and instruments and checking traffic again and THEN committing to going. With the CO2 I only have my regular glasses which will block the beam and I wear them all the time.

By the way if you use the K40case there is a hole in the bottom, I think the air entrance, and I have had it scorch a paper I left under it once. I fashioned a piece of sheet aluminum to be under there to catch parts and stop the beam. All aluminum should be dull in finish so it absorbs, not reflects.

The K40 25 watts is I think a lot different reflection and safety wise than say 200 watts.Still a good machine to develop good habits on.

---

I can imagine. Not sure cardboard is something I expect to cut. It usually cuts so easily with a knife. But then I don't do anything thick or high precision with cardboard. I imagine that the bulk of what I will use the laser for is making control panels, cutting thin plywood for structural things, and the occasional engraving.

As for the auto Z thing. It is more just to see if it can be done and knowledge to store away for a bigger DIY build. Whenever I get a new toy, I always try to learn as much as I possibly can about it. Been disassembling, modifying, and breaking things since I was 8 years old. :) Hey, sometimes they even worked again after I put it back together! I mentioned in another thread how I had made a plasma cutter out of an old 5in B&W TV. I did that when I was about 13 years old. Looking back, I am lucky I didn't seriously injure myself with some of the experiments I tried.

Look at this way - I didn't "need" a laser cutter either, but I bought one. ;) I am not sure if I can do it, but one project that lead up to buying a laser cutter is that a friend came to me asking about a project to honor her late husband who was a detective. She has a high resolution digital copy of his fingerprint and we came up with an idea to etch the fingerprint into a looking glass and edge-light it with UV LEDs. Then put the hole thing into a pretty stand. If I can't etch into the looking glass, I will etch into the black stand material dry-brush UV ink into the etch and just light it with UV so the fingerprint glows. It would be especially awesome if the etch was so shallow or so small you couldn't see it without looking through the glass.

(I get wrapped into some crazy projects for sure.)

Yes.. Don't talk to me about projects, although I did manage with a friends help to digitize a recently found acetate recording from 1938 of a grandmother I never knew. We did it this morning. Pretty weird to hear her singing. Almost all that's left of her and a strange connection to the actress Cornelia Otis Skinner? Hers was a very tragic story with trickle down historical problems to today. Still nice to check one off today as finished. Made good use of my 1940's RCA recording studio!

By the way Retro i like "reading you ". Good fun here!

I *think* I have a the parts finalized. Here are some screen shots and descriptions. Once I assemble and verify, I will post them up on thingiverse if anyone would like them.

The stage will be H-bot/Core-XY based for its simplicity, accuracy, and compactness. There are many advantages to the CoreXY design, but its biggest disadvantage is that it doesn't do well with a lot of weight. The requirements for my build don't require any significant weight. Rigidity will be built in as part of the structure itself. The rods will mount end to end to the briefcase walls. They will be fixed with end mounts and then set-screw stops will be used to stiffen the rods. The biggest fail here would be not drilling the 2 holes perfectly at the same height. What, 2 holes? Yeah, only two because the back end of the rods will be mounted in the rear pulley holders.

Attachment 263380 Attachment 263382

I was going back and forth with the carriage design whether I wanted bearings, rollers, or as I mentioned above just teflon tape to reduce friction. I decided to place two holes in the carriage to which 3x5x3 bearings could be installed that would press against the inside of the aluminum C-channel. I also beefed the carriage up even more. I made it 30mm wide so that a 25mm fan or peltier could be installed flush against the laser heatsink. I also made sure there was room to run an aluminum tube down the side for an air assist and room on the top to mount the endstop switch. Though I am not settled on the placement of the switches yet. So, no holes. I can drill them to make them fit in a few different places. Always leave options!

Attachment 263384

I completely redid the guides (the part that rides on the rods for Y movement). I couldn't get them to print well before and realized I needed to go basic. They run the full depth of the aluminum channel, stick up 3mm above the channel for clearance of the pulleys, and I shortened the part that slides into the C-Channel so I didn't need to use as long of a piece of C-Channel. I will be mounting these with countersunk screws to the aluminum channel so the screws are flush and out of the way.

Attachment 263386

The GRBL board allows for endstops at both extremes of the axis. I will most likely be taking advantage of that. Once I finally get this all installed, I can start sorting out the steps/mm and stepper current.

I have ordered a dimmable LED driver capable of 2.1A. This will replace the driver in the laser diode. The dimmer works using either a 20K potentiometer between the two terminals or a 0-10V signal. I ordered an I2C DAC for the option of controlling the power with the GRBL shield, but in thinking about it, I will probably just install either a potentiometer or a rotary switch with a set of fixed resistors for manually setting it. There shouldn't be much need for varying the power of the laser during cutting. The diode current shouldn't exceed 1.9A, so there will need to be a resistor in series with the potentiometer to set the max current (when pot is 0ohm, the resistance will equal my resistor) This will be a multiturn trimmer pot for accuracy. I also picked up a Lasorb surge arrestor to protect against ESD. These diodes are extremely sensitive to that. I am crossing my fingers that I don't damage it removing the installed driver! But I need to be able to crank this thing down to safer levels when needed.

As mentioned above, a camera and 7" LCD are coming. This will be the virtual window. Not really necessary part of the build, so I can still claim I did it under $300 ;) Well, I have spent more than that already on this, but since I bought things I didn't end up using and had to buy some tools/supplies I didn't have prior, I still think I could claim under $300 even with the camera and monitor. The diode without the driver and with a three element lens instead of the G-2 can be gotten for about $60-80 instead of $125. And 3W is likely overkill, you could buy a 1W diode which is even cheaper still.

The aluminum case actually takes a huge chunk of the price at $30. I really need to find a cheaper source or make a case from wood instead. If only I had a laser cutter. Oh wait! Yeah, I need to get back to my K40 upgrade soon.

Tune in next time!

I am too lazy to draw up all the staple parts into 3D models, so I realize it may be hard to visualize what I am talking about. Patience, my friends. There will be photos of the actual assembly soon and it will make perfect sense.

I mentioned that the riskiest part was making sure that all the holes are lined up and at the same height for the rods. Well, I will be designing some jigs to help with that. Then I just place them inside the case on the floor and up against the wall and drill. This will ensure accurate hole placement. Another very useful purpose for a 3D printer or laser cutter!

Quote:

---

Originally Posted by buddydog

By the way Retro i like "reading you ". Good fun here!

---

You just love my 'scope creep!' I tend to get a bit unfocused in my personal projects and wander a bit as I explore new ideas. Sometimes it kills a build completely, but it does usually result in learning new things for next time. I am one of those that learn by doing. I can play with math and simulation all day, but it doesn't really sink in until I just take a hammer and duct tape it together and try it.

Or it could be that we apparently both work in the aerospace field. ;)