PCB Libraries Forum Homepage
Forum Home Forum Home > General > General Discussion
  New Posts New Posts RSS Feed - Creating a Centroid /CPL file- Bottom side data
  FAQ FAQ  Forum Search   Events   Register Register  Login Login

Creating a Centroid /CPL file- Bottom side data

 Post Reply Post Reply
Author
Message
Eugenia View Drop Down
New User
New User


Joined: 05 Mar 2020
Status: Offline
Points: 2
Post Options Post Options   Thanks (0) Thanks(0)   Quote Eugenia Quote  Post ReplyReply Direct Link To This Post Topic: Creating a Centroid /CPL file- Bottom side data
    Posted: 05 Mar 2020 at 3:01pm
I want to write a program to generate a pick&place file, and the characteristics it must have, seem straightforward to implement, except that I cannot find information about how the coordinates and rotations must be defined for components placed on the bottom layer. 

Is the Origin always the same, namely, the bottom left corner of the PCB seen from the top side or does it change for each side? 

And similarly, about the zero orientation. Defined as seen from the top side (mirrored image), or from the bottom side? 

I'd appreciate any help provided.

Back to Top
Back to Top
feynman View Drop Down
Active User
Active User


Joined: 06 Feb 2020
Status: Offline
Points: 12
Post Options Post Options   Thanks (1) Thanks(1)   Quote feynman Quote  Post ReplyReply Direct Link To This Post Posted: 22 Mar 2020 at 3:12pm
I'm afraid there is no standard for component rotations/coordinates on the top side, let alone for the bottom side :)

Regarding the coordinate origin: What I see the most is the origin at the bottom left of the PCB (when seen from the top side), while the origin does not change for each side. However, the x coordinate changes sign when a component is mirrored to the bottom side.
For example, a component at 5/10 (x/y) on the top side is at -5/10 when mirrored to the bottom side. This corresponds to the direction the pick&place machine has to move relative to the origin.
But I've also seen files where the x coordinate did not change sign for bottom components.

Regarding the zero orientation: Some CAD packages leave the rotation at 0° when a component is mirrored to the bottom side, some CAD packages change it to 180°.

The bottom line is, that there seem to be two conventions:
1) A "see though" convention
2) A "bottom view" convention

The first convention is the perspective of the PCB layouter, who sees everything in a "see through" view in his CAD tool. So two components on top of each other (one on the top, one mirrored the bottom side) have identical coordinates and identical rotation.

The second convention is the perspective of someone who actually holds the PCB in his hands.
Let's assume there is a diode mounted on top and the same diode mounted on bottom as well. And both diodes have 0° rotation. The visual orientation (e.g. "cathode on the left") of the diodes should be the same, whether looking at the bottom diode from the bottom side or looking at the top diode from the top side.

Since your assembler is the one holding the PCB in his hands, one might argue that the "bottom view" convention is the more suitable one.

Long story short: I don't think these conventions matter too much :) Since there is no standard format, your assembly house will most likely look through your pick and place files and edit/translate it for their needs.
Back to Top
Tom H View Drop Down
Admin Group
Admin Group
Avatar

Joined: 05 Jan 2012
Location: San Diego, CA
Status: Offline
Points: 5718
Post Options Post Options   Thanks (1) Thanks(1)   Quote Tom H Quote  Post ReplyReply Direct Link To This Post Posted: 22 Mar 2020 at 4:55pm
The original goal was to create a Zero Component Orientation to eliminate or reduce the up front costs in the assembly process for all packages that must be inserted with a specific rotation. 

Resistors are not in this class of components, but most other parts are. 

IPC broke their own rule by switching the Pin 1 location at the last minute at the 2004 IPC APEX meeting in Anaheim, CA. The IPC-SM-782 was around for 18 years with Pin 1 Lower Left. One person (Jamie Smith) who never attended any Land Pattern meetings or any IPC development meetings attended the 2004 IPC APEX meeting and he questioned the Pin 1 Zero Component Orientation should be changed to the Upper Left because JEDEC and EIA published Pin 1 in Upper Left. There was a vote on this decision and everyone voted with Jamie and JEDEC, reversing an 18 year standard. 

Then 2 years later, IEC (the world standard) published their Zero Component Orientation with Pin 1 Lower Left. And most the component manufacturer's display their package dimensions with Pin 1 Lower Left. 

Lots of confusion over this issue. IPC-7351C was going to introduce Rotation A (Pin 1 up) and Rotation B (Pin 1 down), but I'm not sure anymore what IPC is doing with 7351C. It's been in limbo for so long that I don't know if it will ever be released. 

But the goal was to reduce setup time in the assembly process, but we're not there yet.

Stay connected - follow us! X - LinkedIn
Back to Top
 Post Reply Post Reply

Forum Jump Forum Permissions View Drop Down



This page was generated in 0.156 seconds.