Cheapo Laser Cutter

The_Master

Wow. I'm interested in laser cutting 1/16" plywood for R/C airplane parts. Until yesterday I had no idea such a thing could be had for $400:

https://www.amazon.com/CNCShop-Engraving-Machine-Engraver-Woodworking/dp/B012F0LKHE

I was vaguely aware of homebrew diode-based CNC systems that could burn an image, but thought any usefully powered cutting system would cost $10k++ or whatever.

Supposedly the cuts are .001" resolution, burn marks can be minimal, and the cuts are fast. (video below)

According to reviews the software is junk, but I'm just interested in the hardware. Worst case scenario I would spend a couple days interfacing a micro to the stepper drivers, doing a canned production routine.

I'm posting here because some of you might find this interesting, and for getting feedback. For example, if I am envisioning just doing a custom controller, maybe I should be looking at a different (and maybe even cheaper) product in the first place?

xanadu

I've been watching reviews on them. The box is pretty small 300x200mm work area.

It seems easier to get this X-Y kit https://www.monoprice.com/Product?p_id=13960 and mount a laser on it. You still need an enclosure, but your work area would be much larger. It would end up costing less and be much easier to fix.

idbruce

xanadu

It seems easier to get this X-Y kit https://www.monoprice.com/Product?p_id=13960 and mount a laser on it.

Nice little kit

macrobeak

I have found that the Shapeoko was a great introduction to CNC learning and hobby work.
It is mature, well documented and has a good support community.
Many users have adopted it for laser cutting.
It is also easy to interface the Shapeoko controllers to Fusion360 for an efficient CAM workflow. Once again, there is good forum support.

W9GFO

The_Master wrote: »

I was vaguely aware of homebrew diode-based CNC systems that could burn an image, but thought any usefully powered cutting system would cost $10k++ or whatever.

Supposedly the cuts are .001" resolution, burn marks can be minimal, and the cuts are fast. (video below)

According to reviews the software is junk, but I'm just interested in the hardware. Worst case scenario I would spend a couple days interfacing a micro to the stepper drivers, doing a canned production routine.

Those $400 Chinese lasers are junk. Junk hardware, junk laser tube, junk electronics, junk support. You can get a decent Chinese laser for a couple thousand, or a domestic one for five times (or more) as much.

I would definitely stay away from the $400 lasers. Better to get something like a Shapeoko and add a diode laser to it if you really want to laser something.

JasonDorie

I bought a 700mm x 500mm, 100W Chinese import for $2500, delivered, and have been quite happy with it. I've heard from a variety of places that anything less than about $1200 is going to be *very* questionable and have a lot of corners cut. Just the laser controller for mine (the thing that controls the laser and stepper controllers) is about $350 by itself. A 40W laser tube is about $240. Now add steppers, motor controllers, optics, wiring, cable chain, enclosure, air pump, water pump, tubing...

Then think about the fact that this one, with a solid state 2.5W diode, is *also* $400: https://www.amazon.com/BIQU-Evgraving-Machine-Engraving-Engraver/dp/B01HCX191G/ref=pd_sbs_201_6

The_Master

Bought my $350 laser cutter. Laser tube works, power supply works, steppers work. So for the hardware, it payed off. I figure I might be able to get something running by today. There are probably a lot of people here with stepper motor experience....

The steppers are 17HM3448N-15AD, run at 24V. I was kinda surprised this voltage was used. (I'm new to steppers) I am unsure if any type of current limiting technique is used when the steppers are halted or are running slowly.

Here is what I would like to do today to keep it simple. Use the 24V because it is there. Or I can lower it. Simply run steppers at one quite modest speed. The chosen speed will be slow enough so that the jerk during the acceleration to this discrete speed won't cause any problems with the motor skipping any steps, thus negating the need accel/decel ramps.

The reason I suspect no complex current limiting technique is used is because when I look at sample stepper circuit diagrams online (using L298N), there is no provision for current limiting. (not in the chip either I believe) Although running at 24V in the case of the laser's steppers might be a totally different animal.

Keep in mind this is not a mechanical milling machine. The only load the 'gantry system' has to work against is a beam of light.

I was hoping anyone with stepper motor experience might let me know if I am way off the tracks here.

Phil Pilgrim (PhiPi)

Current limiting in stepper circuits is done by "chopping" the supply voltage going to the motor. Because the motor is inductive, the current builds slowly, so that when the desired current threshold is reached, the supply is cut off. When this happens, current continues to flow through the motor but begins to decay. When the current falls below a lower threshold, the supply voltage is applied to the motor again.

-Phil

JasonDorie

At 24v the steppers are almost certainly current limited. Higher voltages are used because it builds up the field quicker, and allows the steppers to be moved at a higher top speed, but it generally requires managing current or you'll fry the coils.

I'm assuming the laser has them already, but if not, there are tons of inexpensive off-the-shelf stepper drivers available for relatively cheap that will handle this, and microstepping, for you. I've used these and like them: http://www.studica.com/us/en/Makeblock/me-2h-microstep-driver/12008.html

I think you will also find that running at constant speed isn't terribly useful. It'll work, but you'll need to limit the beam strength so you don't burn the heck out of everything you're cutting. Proper laser drivers handle acceleration and deceleration, and vary the beam strength with the speed of the laser head so the target sees a consistent power level.

Bean

Another method I have seen is to put a series resistor in line with each coil. For example if the coil resistance is 10 ohms, but a 10 ohm resistor in series with it and apply twice the voltage the coil is rated for (24V for a 12V coil). When the power is first applied, the impedance of the coil is much higher than 10 ohms, so more voltage is seen across the coil (perhaps 18V) and less across the resistor (perhaps 6V). As the field builds, the coil impedance lowers to the nominal 10 ohms and the coil sees the normal 12V.

I don't know how effective this is, but it certainly is simple and cheap.

Of course you need to use high power resistors...

Bean