lab11 / eagle

Custom scripts and libraries for Eagle PCB

Lab11 Eagle Support Files

Part libraries, packaging scripts, CAM jobs, and other files we use with EAGLE.

Required package installation

Ubuntu (assuming default python 3.8)

sudo apt install pdftk unoconv texlive-core
pip install -r requirements.txt

Process

We highly recommend reading over some of the Lab11 Eagle Documentation.

Generally:

• Generate bom with ulp. Sort by values and output to csv.
• Generate gerber files with eagle
• Open csv, add notes, delete rows for DNP footprints, and save as xlsx
• Run the main eagle.py script from the board file directory

eagle's Issues

Renumber script errors with a float input

For instance if I have a resistor valued 0.12 and other resistors valued 10k it causes this error:

File "/home/adkins/Research/eagle/scripts/eagle_renumber.py", line 148, in
parts = collections.OrderedDict(sorted(parts.items(), key=lambda t: resolve_unit(*t[0])))
TypeError: unorderable types: str() < float()

HTS221 device not correct

Dear
For the HTS221 in this eagle lib the pin assignment is not correct.
As it is in this lib the pin assignment is accoring figure 2 of the HTS221. BUT this is the bottom view!.
(So my small PCB is now wrong )

Eagle Renumber Script fails with error unorderable types

When I run the renumber script on debug backplane with prefix "L" it fails with the error:

File "/home/adkins/Research/eagle/scripts/eagle_renumber.py", line 165, in
parts = collections.OrderedDict(sorted(parts.items(), key=lambda t: resolve_unit(*t[0])))
TypeError: unorderable types: str() < float()

bom to digikey script doesn't include all part numbers

Only lists a single part in the part list, when many are identical:

Eagle script doesn't package multiple drd files generated by the use of blind vias

In eagle, if a blind via is used, the excellon.cam generates multiple drd files named "<project_name>.drd.<first_layer><last_layer>".

For example, if I have a lot of through vias and some vias that just go from layer 01 to layer 02, it would generate two files:

<project_name>.drd.0102
<project_name>.drd.0116

The eagle script does not seem to package these files when it zips them together, presumably because it is only looking for .drd not *.drd'

The renumber script only works for single letter prefixes

For instance it won't renumber LED as a prefix. This might be related to the pull request for prefixes that were too long. I'll look into it, putting it here so we don't forget.

renumber script doesn't handle non-contiguous values

For example if I have parts C3 and C1, it will try to renumber to C1 and C2, but on the renumber phase there is a key error (probably because C2 never existed).

Eagle Renumber Script changed device name

When running the eagle renumber script it changed

device="0603_RES" -> device=0Ω603_RES" in the eagle source file. It clearly should not change the device name.

It seems to have happened on a part with a value of zero, and the suffix was not added to the value field in that line like it should have been. Perhaps the search and replace used to add the proper suffix to the value found the first zero in the line instead of finding value="0"?

eagle renumber suggested prefix suggestion incorrect

well this is just wrong:

The following parts have different suffixes:
C22 : 3.3 nF, C3 : 3.3 nF, C7 : 3.3 nF, C20 : 3.3 nF, C27 : 3.3 nF, C46 : 474 nF, C47 : 800 nF
And would look like this with suffix correction:
C22 : 3.3 uF, C3 : 3.3 uF, C7 : 3.3 uF, C20 : 3.3 uF, C27 : 3.3 uF, C46 : 474 uF, C47 : 800 uF

Manufacture / distributors code

Hi, I am one of the developer of KiCost, a tool in Python to get the prices from the distributors web site originally did to KiCad, but I expand to CSV BoMs.
I saw that you library already have some web site distributors codes and want to check the compatibility, we use the format distributor+'#' (e.g. mouser#, digikey#, ...) and generic manf# for manufacture number.
Cold you attach here some CSV BoM lists to my tests?