Hi all. I have just purchased a Vevor 3 axis laser and cnc machine. Can this be used for making PCBs?
Thanks and regards
Randy
Thanks and regards
Randy
Hi Randy, I don't know this particular CNC , but in general, you should be able to use it.
What is the wavelength and power of your laser?
The most common way seems to be to clean the pcb, spray a suitable paint and use the cnc / laser
to burn off the paint on the parts which need to be ethched off and then etch the board in your favorite etchant.
There are a bunch of videos on youtube - see the comments too for useful additional info -
If you have a 385nm to 405nm , another option is to use Dry film photoresist and use the laser to expose the photoresist -
What is the wavelength and power of your laser?
The most common way seems to be to clean the pcb, spray a suitable paint and use the cnc / laser
to burn off the paint on the parts which need to be ethched off and then etch the board in your favorite etchant.
There are a bunch of videos on youtube - see the comments too for useful additional info -
If you have a 385nm to 405nm , another option is to use Dry film photoresist and use the laser to expose the photoresist -
The other option is to use an engraving bit - like for example here: https://www.open-electronics.org/engraving-pcbs-by-means-of-a-cnc/
The little CNC is not very rigid, and you are likely to get a lot of chattering if using the spindle to machine the tracks. If you could use the laser as part of an etching process it will be much more accurate (and quieter!)
I have the same CNC machine at home. It got me started but have since upgraded to a much larger machine.
Cheers,
SS
I have the same CNC machine at home. It got me started but have since upgraded to a much larger machine.
Cheers,
SS
Having done some milling experiments recently, It's definitely practical and doable. At a certain width it starts to get much more challenging; for example I've not been able to do traces thin enough for say a 3255 yet without instantly breaking the tiny end mills. The most important thing so far as I can tell is you need to find ways of fixing the board to be as flat as possible with no ability to flex. There is a fancy pcb mill available with software designed to track the board height variabilities and compensate. But the much more simple and cost effective solution is to just make damn sure the piece is good and dead flat. So routing traces on a two inch by four inch board is going to be considerably more predictable than a full size six by twelve with flexing down the middle. But for the types of projects you mentioned it should be pretty easy to work out a decent strategy. This only applies if you go the milling route, obviously.
I did this as an ESP32 breakout experiment and it was pretty easy once I figured out the flatness issues and made the necessary adjustments to my workholding setup.
I did this as an ESP32 breakout experiment and it was pretty easy once I figured out the flatness issues and made the necessary adjustments to my workholding setup.
It's definitely possible.
You can use the laser to etch a layer of paint or vinyl attached to the copper side and then etch the board in ferric chloride.
Then use the CNC to run the holes through the board.
I built a 2 axis pen plotter and use fine/ultrafine sharpies to draw the traces onto PCBs for etching. Then into my CNC for the holes and good to go!
This one was for a Lidar scanner driven by an arduino:
You can use the laser to etch a layer of paint or vinyl attached to the copper side and then etch the board in ferric chloride.
Then use the CNC to run the holes through the board.
I built a 2 axis pen plotter and use fine/ultrafine sharpies to draw the traces onto PCBs for etching. Then into my CNC for the holes and good to go!
This one was for a Lidar scanner driven by an arduino:
Thanks for showing the possibilities guys, I am very excited about the things I can make with this
the first one video is interesting, I guess the copper was coated with some black paint or so. It seems laser burns the paint, hence, the next step is putting the PCB into the acid to resolve an opened copper.
PS: actually, the second video is exactly about that. The expected resolution is VERY high, due to a marking-laser resolution being around 1/3 of hair.
Last edited:
I am very excited about the PCB possibilities with this unit!
In a few weeks time I will be in a position to think "bigger". I am trying to work out if this unit can be expanded for larger format cutting. I am interested in being able to cut 1.2m x 1.2m with the laser or bit. Has anyone tried expanding their machines?
In a few weeks time I will be in a position to think "bigger". I am trying to work out if this unit can be expanded for larger format cutting. I am interested in being able to cut 1.2m x 1.2m with the laser or bit. Has anyone tried expanding their machines?
Ballscrew length is a hard limitation to how far you can make the machine travel, so no, expansion is not an option. The improvements you can make are more along the lines of stiffening/rigidity, surface parallelism and flatness, and spindle improvements like power, bearings, etc. Sub Sonic basically confirmed that the machine is not rigid. Because the machine is minimal it will likely even require some amount of massaging to get it running reasonably well. This is why I mentioned earlier that if you minimize your work envelope your ability to maintain accuracy will increase. Machine tolerance is measured over a distance, meaning x amount of deviation over x distance; it's compounding the further you get from a given point.
There are too many details and tips n tricks to really lay out here and you are going to want to find the relevant youtube content dealing with the various pitfalls and how to mitigate them. But for starters you at least should have a couple precision surfaces, for example, a 25mm gauge block and one or more 123 blocks. Precision surfaces, meaning flat ground and parallel to within a very tight tolerance, is an absolute requirement for many reasons. But one key reason is you need some minimum standard to base any machine coordinate deviations off of. It helps if you can get the fancy gauges and measuring instruments but even they are not as essential as the basic flat/parallel surface.
Having done focusing of telescopes using a 100:1 planetary gearboxed focuser with a 32 microstep controller (over a million steps for the length or travel) running 12V chopper driver for a 2.3V stepper. You still have play between the teeth etc.
The 3018 are a lead screw driven direct drive unit with anti backlash nuts, so the resolution is really down to the stepper, which is not an issue.
The main limitation is the rigidity of the machine, perhaps the biggest challenge is stiffening up the X axis so the Z axis (& spindle) does not have flex. Unfortunately this unit has a lot of flex here, and it is something that cannot be easily fixed without overcapitalising on the unit. The Y and Z axes also have their issues.
I used mine to learn how to use a 3 axis CNC, and it was great for that. It is hard to damage if something goes wrong.
I now use it with a diode laser for labelling items such as kids’ pencils - the resolution of the CNC when used with a diode laser is surprisingly good, and it works very well for this purpose.
If using the CNC for routing, definitely give it a shot, just keep feed rates low and depth of cut shallow and see how you go. There’s only one way to find out if it will work for your project! 😊
Cheers,
Matt
The main limitation is the rigidity of the machine, perhaps the biggest challenge is stiffening up the X axis so the Z axis (& spindle) does not have flex. Unfortunately this unit has a lot of flex here, and it is something that cannot be easily fixed without overcapitalising on the unit. The Y and Z axes also have their issues.
I used mine to learn how to use a 3 axis CNC, and it was great for that. It is hard to damage if something goes wrong.
I now use it with a diode laser for labelling items such as kids’ pencils - the resolution of the CNC when used with a diode laser is surprisingly good, and it works very well for this purpose.
If using the CNC for routing, definitely give it a shot, just keep feed rates low and depth of cut shallow and see how you go. There’s only one way to find out if it will work for your project! 😊
Cheers,
Matt
Hello all
Thank you for the feedback on this. I am a bit confused, though. I have a friend who works in a shop that sells these kinds of things. He says that for the price range in retail shops, I would have to spend twice as much when compared to ali prices to get similar performance. Also, that these things are pretty modular and that it's possible to build a DIY frame from scratch and use parts from this machine. Anyway it's a learning process
As a next step for the workshop equipment, I do wish to end up with a laser cutter that can do 1.2m x 1.2m and cut through 18mm MDF. I have been recommended to go DIY here, build a frame and fit a one of those 3 lens 10w lasers
Thank you for the feedback on this. I am a bit confused, though. I have a friend who works in a shop that sells these kinds of things. He says that for the price range in retail shops, I would have to spend twice as much when compared to ali prices to get similar performance. Also, that these things are pretty modular and that it's possible to build a DIY frame from scratch and use parts from this machine. Anyway it's a learning process
As a next step for the workshop equipment, I do wish to end up with a laser cutter that can do 1.2m x 1.2m and cut through 18mm MDF. I have been recommended to go DIY here, build a frame and fit a one of those 3 lens 10w lasers
While its an interesting challenge are you sure its the best use of your resources? These guys https://jlcpcb.com/ are so cheap and fast for PCB's and plated through holes are much more reliable than wires soldered on both sides. They are now offering cheap 3d printing. At $2/board from Jlcpcb you would need to make a lot of PCB's to come out ahead financially doing it with CNC if you ever can. Cutting cabinet parts or machining chassis parts makes more sense for CNC efforts.
Hey man, the query about PCBs was for some one off things, I wouldn't want to wait around if I have tools on hand for prototyping something easy. Jlcpcb is a great find, I'll save that
I will be cutting a lot of foam, cabinet parts and things with this. I am making a lot of RC flight craft and watercraft, as well as some full scale things running on RC type hardware. I am also custom fitting out a 6m houseboat, a 4m water camper, a 3.5m sports fisher as well as building a flying watercraft to a new variant of the WIG using a new appendage that I think that I have invented for stabilising an aircraft. That's why I want to upgrade later to a larger format too
I will be cutting a lot of foam, cabinet parts and things with this. I am making a lot of RC flight craft and watercraft, as well as some full scale things running on RC type hardware. I am also custom fitting out a 6m houseboat, a 4m water camper, a 3.5m sports fisher as well as building a flying watercraft to a new variant of the WIG using a new appendage that I think that I have invented for stabilising an aircraft. That's why I want to upgrade later to a larger format too
I know somebody who built wire EDM machines from different bits, saved a lot of money.
PLCs and drive motors are cheap, $300 and up for basic PLCs.
If you look at it, the base, drive, guides, and controllers are basically made by different people.
Laser levels might work for checking, redoing the alignment is difficult.
Sell the machine , or use it for low resolution work.
Worn ball screws will also cause errors, the tool will never return to the same position for the next cut.
PLCs and drive motors are cheap, $300 and up for basic PLCs.
If you look at it, the base, drive, guides, and controllers are basically made by different people.
Laser levels might work for checking, redoing the alignment is difficult.
Sell the machine , or use it for low resolution work.
Worn ball screws will also cause errors, the tool will never return to the same position for the next cut.