Week 8

GROUP ASSIGNMENT

R O L A N D modela pro M D X - 5 4 0

Setting the Modela pro MDX-540

The machine we have in the fablab Ulima is the Roland Modela pro MDX-540 First we have to calibrate the workspace and material. Keeping in mind that the cut will start in the first quadrant and cut from bottom to top and to the right.

SETTINGS

So, to start using the machine, first we have to change the mill, set the origin for the mill and calibrate the Z axis before running the software to start.

First, to turn on the machine, press the button that is on the lower front part of the machine.

Then, be aware of the emergency button in the front that will stop and re start the machine and the settings.

Set the mill in the origin to start, according to the first quadrant.

The control let you set the origin while moving the X and Z point. For the Z set select the Z0 SENSE option and with the support set the height of the mill.

Go up and down from left to right the mill with the jostick above, right for up and left for down.

There will be two different settings for the mill to raster and the mill to cut.

For this work area, it has a support of acrilic to cushion the cut. To ensure the material is properly positioned, we tape it down with double-sided tape. Sometimes the material may be slightly warped and not straight.

Finally open the VPanel for Modela Pro II app in the computer and select the file of the g code to cut.

SETTINGS

g r o u p T E S T S

First to generate the gcode, we import the PNG file to MODS. And set the option for raster andthen for cut mill.

SETTINGS for MODS

I made 3 tests in ceramic. The first from above the picture was with a mill of 0.8 and I realise I did bad the Z setting in the machine. Because the road was too deep.

The second test was with a mill that was plain in onw side and still with V angle with 30 degrees. The thing was that the second mill was a little broken, which made a bad finish, however the depth was better.

The last test was the best finish. With a new mill and 0.09 for cut depht It made the best raster and almost all the template.

solding PCB final test with a 3d printed case.

d e sign the B O A R D

First I redesign the board I made in week 6 in KiCad because I made it in Eagle in Fusion and the fab library didn't had the 3d components. Also I found Kicad easier to work.

KiCad redesign and 3d board

Here I had to update the footprint of some components exactly for the ones I had. I made the width roads of 0.8mm.

Update a footprint

width roads

For the contour I wanted to make and organic shape following my drawings of nature. However I could not foind a way to make curves in Kicad. The best I could make was to upload a reference picture and make the contour as a polygon. I tried the bezier tool but did not make the contour just some lines.

width roads

F L A T C A M

I exported the .gerbr file from Kicad and generated the G code from setting the mill options. First It will generate the routing and then the CNC objet so after this settings finally generate de cnc code for the machine.

Left png from kicad and right some drawings in Illustrator to have an organic edge.

Before sending the code to the machine, I simulate the g code in https://ncviewer.com/ where I could see the layers of cut and if there is no extra dots from the board to delete.

Left png from kicad and right some drawings in Illustrator to have an organic edge.

I cut the first try in fiberglass because I wanted to try another material. However this is more toxic as it left a strange smell and cutting chip. Also with a mill of 0.8 that cut the tests in ceramic did not cut this material. Also i found the surface too rough. In this test I could manage to see if the text was in the correct layer and size to raster and it was. Also here is where my instructor and me realized that the space i left for the free pins to connect with another board will not work because it wasn't in the edge. So the other board will have to be upside down to fit. Also this free pins could not work because it also needed free GND and VCC pins to connect.

First try in fiberglass.

M O D S

As I wanted to try a png for better edges with some drawings, I also redesign the Kicad file by correcting the errors from the week6 design and added extra free pins for GND, VCC, and Analog pins. So I have now 7 extra free pins to connect next week another board for a rgb led that will fit exactly in this straight edge.

FINAL schematic and pcb design

I didn't like the contour from kicad, and as MODS make the gcode from a PNG I exported the kicad gerber to Gerber2png and make a contour in Illustrator with the png in real size so it doesn't scale, and added some other drawings to personalize the board and test other figures.

Left png from kicad and right some drawings in Illustrator to have an organic edge.

With the same settings from the group tests I sent the file in mods. However the first try of contour I forgot to update the pcb design and it cut the wrong file. But I realize I was sending the invert file of contour in mods and make the contour another road. Then, I correct this inverting the contour as the image above, and make the correct edge BUT it somehow offset the cut and almost cut one road in the left. I tried to cut another one but the machine is in used and I will be cutting another one this weekend.

Final tests of contour.

s o l d e r i n g my P C B

Soldering tests with hot air and soldering iron

In the lab there was two options to sold. The soldering iron that I had used before and the hot air that works with a paste. I tried both and first makde some tests in the boards that were wrong. Once I felt confident I try with the paste first. You clearly can difference which are from the paste and which from the soldering iron. I found the pste very practical for long soldering but as it is liquidd it expands way too much with the air pressure and it is difficult to manage. I think this will be more easy if I clear all the pcb board and only left the roads.

Soldering tests with hot air and soldering iron

From also week 6 I design a case for the pcb board and printedit in PLA. I was very nervous it didnt fit as I had to make an offset because as I mention before the machine cut inside the edge contour and not in the axis. Making the board a little smaller.

3D printed case for the board

I tested the board with the multimeter to see if ther was continuity in the roads and If i made the soldering right before testing a promgramming code from week 4.

Testing continuity

I used the codes programmed in week 4 to turn on 3 leds by themselves within 1 second between each and then controlled by a button.

New try: Control the intensity of the led with a button.

This week I tried something new: to control the intensity of the leds by a button. I hd to search a lot between tutorials and some help of copilot to understand certain parts of the codes I found and finally made this wich I understood reads first the state of the button and counts the press and has some levels of color brightness if the button state changes.

const int LED1 = D4;
const int LED2 = D5;
const int LED3 = D6;
const int BOTON = D3;

int buttonState = 0;      // Estado actual del botón
int lastButtonState = 0;  // Estado anterior del botón
int pressCount = 0;       // Contador de presiones
unsigned long lastDebounceTime = 0;
unsigned long debounceDelay = 50; // Tiempo de antirrebote
int brightnessLevels[] = {0, 64, 128, 192, 255}; // Niveles de brillo
int currentBrightness = 0; // Brillo actual de los LEDs

void setup() {
  pinMode(LED1, OUTPUT);
  pinMode(LED2, OUTPUT);
  pinMode(LED3, OUTPUT);
  pinMode(BOTON, INPUT);
}

void loop() {
  int reading = digitalRead(BOTON);

  // Si el estado ha cambiado
  if (reading != lastButtonState) {
    lastDebounceTime = millis();
  }

  if ((millis() - lastDebounceTime) > debounceDelay) {
    // Si el estado ha cambiado y ha pasado el tiempo de antirrebote
    if (reading != buttonState) {
      buttonState = reading;

      // Incrementar el contador en la transición de LOW a HIGH
      if (buttonState == HIGH) {
        pressCount++;
        if (pressCount >= sizeof(brightnessLevels) / sizeof(brightnessLevels[0])) {
          pressCount = 0; // Volver al nivel más bajo
        }
        currentBrightness = brightnessLevels[pressCount];
        analogWrite(LED1, currentBrightness);
        analogWrite(LED2, currentBrightness);
        analogWrite(LED3, currentBrightness);
      }
    }
  }

  lastButtonState = reading;
}

NEXT: In the next weeks I need to try with an rgb led to learn how to programm the chenge of color and the mix of color I want for my final project.

---

F I L E S

1. pcb kicad
2. pcb gerber
3. pngs
4. 3d printed case