Design Rules For PCB

Overview

For this week we have to Characterize the design rules for your PCB production process: document feeds, speeds, plunge rate, depth of cut (traces and outline) and tooling.
In New Cairo lab we have two milling machines which are (MDX-40A & SRM-20). We tested both machines and got the best results from the SRM-20.


Know your Machine

MDX-40A

SRM 20

Machine

MDX-40A
SRM 20

Work Area

12" x 12" x 4.13"

8" x 6" x 2.38"

Table Size

12" x 12"

9.14" x 6.17"

Max. Tool Size

1/4" or 6mm

1/4" or 6mm

Spindle Speed

4500-15000 rpm

3000-7000 rpm

Operating Speed

1.97 in/sec or 50mm/sec

1.18 in/sec or 30mm/sec

Power

110V / 2.1A

110V / 2.5A

Materials

MDX-40A
References

For MDX-40A.
For SRM-20.


The Process

A) Fab Modules

A CAM program that will help us translate our design from Picture To G-code to Work with.

Responsive image

open fabmodules then choose the input to be an image.png.

Responsive image

choose the output to be RML.


Responsive image

For the process choose either traces or outline.

Responsive image

choose the machine, zero XYZ, choose no.of offset and the cut depth.


Responsive image

calculate then save.


B) Roland MDX-40A

For the machining Roland MDX-40A was our first machine to use for that test.

Responsive image
  • Our instructor subdevided our mission and set a primary time table for our work.
  • Responsive image
  • Unscrewing the old sacrificial layer off of the machine work bed.



  • Responsive image
  • Creating a new sacrificial layer from a cork sheet.
  • Responsive image
  • Applying double tape on the cork so we can fix it to the machine. Avoiding air bubbles and tape overlapping to get a perfectly level flooring to our FR01.


  • After fixing the sacrificial layer to the machine, fixing the FR 01 to it by double tape with the exact same processand precautions, We tried the test a couple of times but results didn't get any better.

    Responsive image
  • Removing the brokenend mill using an allen key.
  • Responsive image
  • Installing the new end mill carefully about two fingers out.



  • Responsive image
  • The test failing for the second time as the second time due to unlevelled sacrificial layer.
  • Responsive image
  • We removed the cork layer as it proved its inefficiency. And started searching for a more suitable material.



  • Responsive image
  • We chose to laser cut an new sacrificial layer out of MDF.
  • Responsive image
  • We fixed the new layer with double tape just like before.



  • Responsive image
  • To assure perfect leveling, we milled / scraped a whole layer off of our piece of MDF.
  • Responsive image
  • Then we fixed the FR01 again and started testing one more time.



  • Responsive image
  • Resutls from MDX-40A didn't get any better.

  • C) Roland SRM-20

    Having had hard time with the MDX-40A, we decided to switch to the SRM-20.
    We went through the same initial steps preparing the machine, the work bed, fixing the FR01 with double tape and setting the zeros.
    The best test resutls

    Responsive image
  • Performing the test on the SRM-20.
  • Responsive image
  • Some of the results.



  • Responsive image
  • The best result.



  • Conclusion:

    1- High quality results was from SRM-20. 2- For milling, the best end mills are: [1/64", V bit 30/0.3 and V bit 30/0.2]. 3- For cutting, we tested the 0.1, it worked fine but we needed to adjust the borders.

    Responsive image
  • End mills test results summary.