New to Arduino but Familiar With Electronics...


I'll get right to it. I recently ordered an Arduino UNO board. I plan to build a machine that can cut paper with high precision. I am planning on achieving this by using either a linear encoder or a stepper motor. I am leaning more towards the stepper motor design as it would be difficult to find a custom length printed encoder track (if anyone knows where to find these, please do let me know!).

The designs on the paper are all the same size. We could print them out at precise distances from each other and tell the machine where to cut. However, the machine needs to be able to grab any sized paper and detect marks on the paper that tells it where it is on the paper and where the cutting job ends. The paper is 24" wide by varying lengths (our printer is an EPSON FN6400 and utilizes this paper). I don't think I will need to make the paper longer than around 50" and can be made shorter than 50".

For now, I am only concerned with building a machine that can plot X, Y coordinates accurately using either a motor or encoder as I feel this is the most important part of this project. How difficult will something like this be? Will the code be extremely complex or rather simple? Any help, suggestions, and ideas would be greatly appreciated, as this is the first time I've ever messed around with an Arduino (these things are pretty amazing!).

Sincerely, ~Sat

Honestly, it seems more like a mechanical design problem. Almost any microprocessor can handle the control.

Are you after building something like this: Arduino Controlled T-Slot XY Table ?

“high precision” - what you mean by that ?
within 1mm? within 0.1mm? within 0.01mm ?

“linear encoder or a stepper motor” ?
With just a stepper motor you’d rely on the motor not ‘missing steps’, this is how most 3D printers work and give resolution to 0.1mm or so.
With a “linear encoder” you still need some sort of motor, and you’d be better off with a rotary encoder for a servo or stepper.

You can buy ‘linear scales’ (used with DRO’s for milling machines) which will read down to 0.002mm, but they ain’t cheap.

“detecting marks”
So you want to XY scan the paper with some sort of sensor, then do some cutting ? with a motorized cutter ?

I think you really need to sit down and fully define exactly what you want.

In any case, if you end up needing to drive 3 steppers and looking at a few sensors then a UNO can do that… but I expect your full list of requirements is going to end up more than that.


Hi, While you are figuring out a complex mechanical design, just get started with Arduino. Some examples and learning materials HERE. and LOTS more on the Net...

DISCLAIMER: Mentioned stuff from my own shop...

The limiting factor may be in the detection of the marks, specifically the fuzziness of the print process.

The "marks" are called fiducials. Spell checker doesn't like that, but I don't care. The phonetic is correct. We require them on circuit boards and panels of circuit boards so the pick-and-place machine will know if the board is skewed from the original programming and make allowances. You also see them on paper from high speed printers.

You haven't described how your machine will react to the fiducials. I finally had to use a dictionary for the spelling. Will the paper be moved or will the cutting program be smart enough to rotate the directions to fit the paper?


Salut, tout le monde,

Thank you for all the input. I was not expecting this many replies :3

To answer some questions, I want the machine to be accurate down to 1mm or less if possible. The machine will scan a paper that is 24" wide and vary in length up to 5' (comes off a roll). The machine will only be able to work up to 5' paper length and nothing more.

I need the machine to scan the side of the paper looking for a set of marks that were made by our printers. One mark tells the machine where the start of the paper (work area) is, then scan up to a determined length looking for another mark to tell it where it ends. The images will be printed in rows going from one side of the paper to the other in a fixed matrix so that the machine will simply need to cut the area where an image is supposed to be, regardless of whether there is an image printed or not. All images will be of the same shape and size.

Additionally, each row will have a start and end mark (essentially the same mark used to determine the start of the work area/paper will also be used as the start of the first row). The machine will have to look for the end mark, then scan a predetermined distance looking for another mark in order for it to know if another row of items is present. If none are found, the machine simply counts one row then returns to work on that row which was detected and cut it. Once it's all done, it returns itself to the zero position. These marks will be used for the machine to align itself with the paper.

I hope this is making any sense. I am trying my best to describe it. If not, I happen to be an artist and I would be willing to illustrate it in order to show you what it is I am trying to describe.

For now, I'd like to build a small prototype that simply cuts one or two shapes where I tell it to using marks on a small sheet of paper, possibly size A4 letter. I also plan to use centimeters as the preferred unit of measurement for the matrix on the paper. However, I need it to be precise to 1mm, as I need it to cut inside of the image to about 1mm or 2mm (I'll decide on which after). This will get rid of any possible stroke lines that the original image may possess that are unwanted.

Our printing program (Wasatch) allows us to print images in a matrix that also uses the cm. So, we could place the images in predetermined areas. The image will always be printed in that very same spot each time.

I believe this is possible and should not prove to be too difficult to make, even if I am not too familiar with Arduino. It'll be a learning experience for sure, which is where the fun of this project lies :)


  1. What type of scanner could I use? I hear Pixy could work. Would a Pixy camera be overkilled when I could get away with a cheaper and more reliable IR scanner of sorts?

  2. Where could I purchase parts such as chrome plated rail bars that are 5' in length? (I am willing to make the machine shorter, perhaps to about 4' but no less wide than 24".)

  3. Can all this be accomplished?

Any questions, please ask. I will answer any questions to try and explain what it is I'm planning on building. It is essentially an X, Y plotter but with a few extra bells and whistles, and a very sharp blade :3 Thank you all very much and I hope I managed to give you guys a better idea.

Merci beaucoup, ~Sat

Can all this be accomplished?

Yes but it is a very ambitious and advanced project and I don’t know if you have the skills. It is going to be expensive. Mechanically it can be very difficult to get the stability.

A Pixy will probably not be good enough and it is not by any means over kill, in fact I don’t think it is up to the job. For that bit you might need a Raspberry Pi with a high resolution camera.