The influx of cheap laser cutters from China has been a boon to the maker movement, if at the cost of a lot of tinkering to just get the thing to work. So some people just prefer to roll their own, figuring that starting from scratch means you get exactly what you want. And apparently what [Mike Rankin] wanted was a really, really small laser cutter.
The ESP32 Burninator, as [Mike] lovingly calls his creation, is small enough to be in danger of being misplaced accidentally. The stage relies on tiny stepper-actuated linear drives, available on the cheap from AliExpress. The entire mechanical structure is two PCBs — a vertical piece that holds the ESP32, an OLED display, the X-axis motor, and the driver for the laser, which comes from an old DVD burner; a smaller bottom board holds the Y-axis and the stage. “Stage” is actually a rather grand term for the postage-stamp-sized working area of this cutter, but the video below shows that it does indeed cut black paper.
The cuts are a bit wonky, but this is surely to be expected given the running gear, and we like it regardless. It sort of reminds us of that resin 3D-printer small enough to fit in a Christmas ornament that [Sean Hodgins] did a while back. We’d suggest not trying to hang this on a tree, though.
33 thoughts on “Tiny Laser Cutter Puts Micro Steppers To Work”
Yay, resolution: ~10μm, backlash: ~500μm. I wonder where that backlash came from, is it from mechanism or from stage mounting?
I know, the backlash on the video is visible from miles away…
Since the laser does not put any cutting force on the machine (cough…mOmEnTum oF phOToN…cough…), i think the backlash problem can be easily solved by preloading the carriages. In such tiny machine, the rubberband would probably be enough.
Do you maybe know actual resolution of that drives (steps per mm)? I couldn’t find it anywhere. Also: nice build.
I’ll bet the gear train is highly suspect..
Yes there is quite a bit of slop in the whole gear mechanism.
The wobbly movement you’re seeing is not backllash- it’s slop in the “bearings” that ride on the “guide rails”. In that mechanism, the nut contacts the frame which serves as the guide. Sloppy fit between the nut and the frame allows the laser and stage to wander around as the screw turns. I think the linear motion assemblies from CD/DVD drives are a little better than this type.
There’s probably also some backlash in the mechanism due to the nut fitting loosely on the screw.
Can’t it just be solved with a spring pushing the carriage to one direction only?
My thoughts exactly. I have worked with similar linear actuators before, they are cheap and poorly made. Unusable for precision displacement.
What”s the laser output power? (Both specified and actual, which can be significantly different)
What’s the safe power limit for avoiding instant permanent eye damage?
Where is the shielding to prevent the beam entering eyes, either from reflections or directly if the laser becomes detached?
Remember the standard wall sign in laser labs: “do not look at laser beam with remaining eye”.
I guess the laser power is not much higher than fancy laser pointer. You can tell by lack of heatsinks or any other cooling. It probably only burns black stuff when properly focused and will get significantly less lethal with increasing distance, as it gets out of the focus…
Where instant and permanent blindness is concerned, guessing is negligent.
You might say it’s the blind leading the blind.
Aren’t most ‘clear’ plastics relatively opaque to light from these little lasers, or just CO2? A clear plastic soda bottle or other food container from the recycling bin would easily cover this tiny device while in use.
Right now i’m powering the laser with a regular +3.3V LDO and a resistor set limining the current to 150mA. I’d like to up that but only have the one diode and didn’t want to risk damaging it.
I would worry more about your eyes than the diode!
Can I suggest you get a large area photodiode (e.g. BPW34), reverse bias it with (say) 5V, focus the spot on it, and measure the photocurrent.
The *peak* sensitivity will be 0.5A/W and that will be reduced according to the wavelength as shown in the datasheet. From that you can estimate the power, e.g. at 450nm the sensitivity would be 0.1*0.5A/W, so a 10mA current means 200mW.
Then compare that with the laser safety regulations.
Do not look at open class 3B laser with remaining eye.
Yes indeed. From https://www.lasersafetyfacts.com/3B/ we have
“Class 3B visible-light lasers are hazardous for eye exposure. They can cause burns to the retina. A person cannot turn away or blink fast enough to prevent retinal eye injury from a nearby Class 3B laser. ”
“At the low end, around 5 to 50 milliwatts, a Class 3B laser poses a moderate risk of eye injury. … At the high end, around 250 to 500 milliwatts, even a brief exposure could cause retinal damage.”
“The Nominal Ocular Hazard Distance (NOHD) for a lower-powered 50 mW Class 3B visible-beam laser with a tight beam (0.5 milliradian divergence) is 330 ft (100 m). The Nominal Ocular Hazard Distance (NOHD) for the most powerful 499 mW Class 3B visible-beam laser with a tight beam is 1,050 ft (320 m).”
It it were and XY mechanism you could use it to etch designs in fingernails.
inkjet is easier to remove: https://www.youtube.com/watch?v=d6fvw838T04
What is the laser dot size?
To be really useful for a laser cutter/ etcher of this size is for it to work in really small dimensions.
Can it, for example, put a smiley face on Lincoln in a penny (the back side where he is sitting in the Lincoln memorial)?
Guesses are dangerous where blinding someone (else) is a possibility.
The only thing that matters are the hard numbers
Some small and easily available lasers put out 100* “the” eyesafe limit, so even 99% reduction would barely be enough!
Sanity check: what’s the laser’s wavelength? If visible and the bottle appears clear, then the laser will penetrate it!
 there isn’t a single limit; it depends on duration, wavelength, divergence, etc
It is manually focussed now by looks like a 30mil hole or so.
I used to work on a production laser system with about a 1 square meter work area, that had a clever mechanism where the rough positioning was done with x,y positioning, but fine positioning was done with a pair of x,y galvos. In a case like this where your working area may only be 2cm on a side, a pair of galvos might be all you need, and they have the advantage of vastly less mechanical backlash and extremely high speeds. There’s a big drawback of a more complicated optics.
Wouldn’t a “tiny stage” be a “soapbox”?
Not to be confused with 18000 laser pointers on a 2-axis mount.
Introducing the Melt-inator 6-5000
“No useful purpose” – at least he’s honest about that. What if someone were to take this work and put it to productive use? I think a more powerful laser would be required, and then there’s the mass of that, plus power, and heat dissapation.
I had a few linear motors and was just wondering what I could make with them. Positively nothing useful and i’m surprised the thing even worked at all.
The entire structure is the PCB — awesome! The world can always use more tiny motorized stages.
A good solution to mating boards on a right angle is tricky. These PEM standoffs with connectors is surprisingly rigid.
Put a Cherry MX stem on the build platform so you can mount a blank keycap on the stage and you have a custom mechanical keycap legend engraver!
Was just about to suggest the same thing.
I think he could reduce his backlash by mounting just a mirror on the carriage and have the laser stationary pointed at the mirror.. saves a lot of weight..
Never thought of that idea. Brilliant!
Please be kind and respectful to help make the comments section excellent. (Comment Policy)