wafer-map 1.0.14

Semiconductor Wafer Mapping

Plots up a wafer map. Used in semiconductor processing and analysis.

Features

• Mouse and keyboard shortcuts!

• Knows SEMI M1-0302 wafer sizes!

• Something else!

Installation

Install from PyPI.

pip install wafer_map

Usage

<I still need to fill this out in detail.>

The easiest way to use this to to:

0. Import the wm_app module:

   >>> import wm_app
   

1. Set up your data as a list of (grid_x, grid_y, value) tuples:

   >>> data = [(grid_x_1, grid_y_1, data_1),       # 1st die
   ...         (grid_x_2, grid_y_2, data_2),       # 2nd die
   ...         (grid_x_3, grid_y_3, data_3)        # 3rd die and so on
   ...         ]
   

2. Call wm_app.WaferMapApp.

   >>> wm_app.WaferMapApp(data,
   ...                    die_size,
   ...                    center_xy,
   ...                    dia,
   ...                    edge_excl,
   ...                    flat_excl)
   

   The input parameters for WaferMapApp are:

   die_size:

   The die size in (x, y). Units are mm.

   center_xy:

   The grid (x, y) coordinate that represents the physical center of the wafer.

   dia:

   The wafer diameter. Units are in mm.

   edge_excl:

   The exclusion distance measured from the edge of the wafer. Units are in mm.

   flat_excl:

   The exclusion distance measured from the wafer flat. Units are in mm. Cannot be less than edge_excl.

3. An image should appear. Yay! Play around with it: middle-click+drag to pan, scroll wheel to zoom. See “Keyboard Shortcuts and Mouse Usage” section.

Example

There is an example file which somewhat demonstrates how to use this package. At the very least, you can run the example file and see how this wafer mapping software looks.

Navigate to the wafer_map directory in your python installtion (../Lib/site-packages/wafer_map) and run example.py in your cmd prompt or terminal:

python example.py

Example.py generates a fake data set and then displays it in 3 different ways:

1. As a standalone app. This requires only calling a single function in your code.

2. As a panel added to your own wx.Frame object. This allows you to add the wafer map to your own wxPython app.

3. As a standalone app, but this time plotting discrete (rather than continuous) data.

Nomenclature

For the entire project, the following nomenclature is used. This is to avoid confusion between a die’s coordinates on the wafer (floating-point values representing the absolute postion of a die) and a die’s grid location (integer row-column or x-y values that are sometimes printed on die).

coordinate:

Floating-point value representing the exact location of a die on the wafer. Also sometimes called ‘coord’

The coordinate origin is the center of the wafer and the center of the FloatCanvas panel.

grid:

Integer value representing the printed die. Can only be mapped to a coordinate if a grid_center is defined.

Each grid line falls on a die’s center.

grid_center:

The (float_x, float_y) tuple which is coincident with the wafer’s center coordinate (0, 0).

This is the only grid value that can be made up of floats.

row:

Alias for grid_y.

col:

Alias for grid_x.

Keyboard Shortcuts and Mouse Usage

No matter if you use the standalone app or add the panel to your own wx.Frame instance, keyboard shortcuts work. I’ve only added a few so far, but I plan on adding more.

The panel also supports mouse controls. Middle click will pan, mouse wheel will zoom in and out.

Home:

Zoom to full wafer

O:

Toggle display of wafer and exclusion outline

C:

Toggle crosshair display

L:

Toggle legend display

Notes

This package has been released to version 1.0.0. What this means is that it should be usable in an engineering-type environment. I’m starting to use it heavily myself. It’s not very customizable yet, but I don’t need that capability yet. You can see the roadmap at: https://github.com/dougthor42/wafer_map/milestones

There’s still a fair amount of code cleanup and refactoring to do, especially on the wm_legend.py module (as that was made last). So please do judge my coding style too harshly (though constructive criticism is much appreciated!)

Requires: wxPython

Current capabilities:

1. Draw wafer outline and flat or notch.

2. Draw edge exclusion outline.

3. Draw wafer center crosshairs.

4. Accept continuous or discrete data and color accordingly.

5. Provide zoom and pan capabilities.

6. Mouse-over to display die coordinate and value

7. Legend Display for both continuous and discrete data

Changelog

• 1.0.11 / 2015-10-28

  • Implemented #31 - added radius of die and mouse to status bar.

• 1.0.10 / 2015-05-04

  • Playing around with imports. Trying to make it so that if I’m running a file from my development directory, it imports all the development modules and if I’m running something else, it imports from the released (site-packages) directory.

  • Fixed Issue #30

• 1.0.9 / 2015-05-04

  • Fixed flicker issue.

• 1.0.8 / 2015-05-04

  • I messed up on the PyPI release, so I have to release under a new file name. Oh well…

• 1.0.7 / 2015-05-04

  • Minor import refactoring

  • Added some docstrings

• 1.0.6 / 2015-04-01

  • Refactored wm_core.draw_wafer_outline so that there are fewer branches

  • Added wm_core.calc_flat_coords to reduce code duplication. This function calculates the start and end coordinates of a horizontal chord below the circle origin whos length spans a given angle. See doctring on wm_core.calc_flat_coords for more info.

  • Fixed issue where a flat exclusion of 0 would not work

  • Fixed issue where an exclusion of 0 would prevent the flat exclusion from being drawn.

• 1.0.5 / 2015-03-30

  • Added optional “grid_center” input to gen_fake_data

  • Fixed Issue #28: Updated wm_utils.linear_gradient to act on HSL data rather than on RGB data.

  • Updated documentation for items in wm_utils.

  • Added option to plot the die centers as small red dots.

  • Added option to have the wafer map use a constant number of items for discrete legend.

• 1.0.4 / 2014-12-29

  • Changed import statements to not be weird.

• 1.0.3 / 2014-12-17

  • Fixed Issue #9: Users can now change the high and low colors for continuous data by passing in arguements or by using the app menu: Options --> Set High Color or Set Low Color

  • Fixed Issue #26

  • Fixed Issue #25: Continuous data now generates colors from a single algorithm.

  • Fixed Issue #14: Discrete Data now uses an acceptable algorithm for determining colors.

  • Fixed Issue #16: The plot now zooms to fit upon first draw. However, this created issue #21.

  • Started to add unit tests

  • Updated fake data generator to accept parameter inputs. Any missing parameter is randomly generated.

  • The legend for continuous data now fills the entire available vertical area.

  • Added a color for invalid data points such as NaN or Inf.

  • Plot range can now be set manually. If it’s not set, then it uses the 2nd and 98th percentiles.

  • Added yellow circle representing the wafer as if the flat did not exist.

  • Created wm_constants.py to contain various default values such as colors.

  • Some other changes that I can’t remember and foolishly didn’t write down.

• 1.0.0 / 2014-12-05

  • Official release.

  • The Legend should now work as intended.

• 0.6.0 / 2014-12-04

  • Closed issues #1, 2, 3, 4, and 6 in the tracker.

  • Updated gen_fake_data to use better algorithm and actually output correct data.

  • Updated wm_core.WaferMapPanel so that the status bar text displays the correct grid values. Verified working with all sorts of grid_center values.

  • Last Update before release, yay! All that’s left is to get the legend working.

• 0.5.0 / 2014-12-02

  • renamed wafer_map.py to wm_core.py.

  • Finally figured out the imports for running in development from my own dev directory vs running in “production” from the site-packages directory.

• 0.4.0 / 2014-12-02

  • Massive change to package hierarchy - separated app, frame, info, and fake data into individual modules.

• 0.3.0 / 2014-12-01

  • Added kb shortcuts and menu items for display toggle of wafer outline and crosshairs.

  • Added placeholder for legend and kb shortcut for display toggle.

  • Added option for plotting discrete data.

• 0.2.0 / 2014-11-26

  • Made it so a wafer map can be plotted with a single command.

  • Updated example.py to demo single-command usage.

• 0.1.0 / 2014-11-25

  • First working code. Added example file.

• 0.0.1 / 2014-11-25

  • Project Creation