More KiCad modules

kicad-appicon-128

I’ve added a bunch of modules to my KiCad library. These have not yet been vetted, so approach with caution. Here’s what’s new or changed:

  • New library mfk-ipc-7351-diode.mod with:
    • mfk-SOD3716X135N (SOD-123 diode)
  • New library mfk-smd-not-ipc.mod with:
    • mfk-fairchild-SOT23 (Fairchild version of SOT-23)
    • mfk-fairchild-SOT23_213 (as above but with “BEC” pinout)
    • mfk-fairchild-SOT233 (Fairchild version of SOT-233)
    • mfk-fairchild-SOT23_2313 (as above but with “BCEC” pinout)
  • Changes to mfk-ipc-7351-transistor.mod:
    • Deleted mfk-IPC7351-trans-SOT65P210X110-6N_10mil
    • Changed silk width in mfk-SOT65P210X110-6N (SOT363) to 10 mil.
  • New part in mfk-vishay.mod:
    • mfk-Vishay-TSOP341xx-alt (alt version of mfk-Vishay-TSOP341xx)
  • New part in mfk-keystone.mod:
    • Added mfk-KEYSTONE-621

KiCad workaround for mounting hole too close to pad

KiCad-mounting-pin-violation-workaround-crop

Here’s a workaround I used in a KiCad layout that involved a DRC error with a module’s (non-conducting) mounting pin. The mounting pin is physically close enough to an electrical pin that it makes the DRC clearance test fail. The proximity isn’t actually a problem because the offending pin is just so much mounting foo for the part, but KiCad doesn’t know the difference.

The workaround is to edit the pad for the mounting pin and assign it the same pad number as the pad the DRC thinks it’s too close to. (See the two pads numbered 3 in the above image). Kludgey, but it silences the (not really an) error. I made this change in the PCB layout rather than in the library module as it won’t be a problem on boards with smaller copper clearance.

BTW, the module is for a Keystone 970 RCA jack and is available in my Bitbucket repo.

PDF output from KiCad in Linux from SVG

kicad-appicon-128

The ps2pdf-d script I developed earlier worked fine for getting PDF from KiCad‘s Eeschema using Postscript as an intermediate format, but it falls short with Pcbnew because there is no way in Pcbnew to make a Postscript file that merges layers. Using File > Plot, you can pretty easily get Postscript files of individual layers, but you can’t get a merge of, say, the top copper, top silkscreen, and drawing layers.

But you can do this with File > Print SVG. You’ll need to play around with the dialog to get the SVG you want, but it’s pretty simple. Thus, svg2pdf-d was born:

#!/bin/bash

# DESTRUCTIVELY convert all svg files in working directory to pdf.
# Source file extension is case sensitive!

# Requires inkscape.
# Mithat Konar 2013 <http://mithatkonar.com>

OPTIONS=""

FILES=$(ls -1 *.svg)

# There are two alternatives below for doing the conversion.

#~ ## BEGIN ALT_1:
#~ ##     This approach isn't optimal because it restarts inkscape per
#~ ##     file. However the simplicity lends itself to porting (e.g., to
#~ ##     windows BAT.
#~ for file in $FILES
#~ do
    #~ base=$(basename $file .svg)
    #~ inkscape --without-gui ${OPTIONS} ${file} --export-pdf=${base}.pdf
#~ done
#~ ## END ALT_1

## BEGIN ALT_2:
##     This approach seems to be better because it starts one inkscape
##     instance for all files, but it also involves a temp file.
# Make a temp file to store batch commands.
CMDFILE=$(mktemp) || { echo "Failed to create temp file"; exit 1; }

# build up the list of commands
for file in $FILES
do
    base=$(basename $file .svg)
    echo "${file} --export-pdf=${base}.pdf" >> $CMDFILE
done

# Process commands in a batch.
DISPLAY= inkscape ${OPTIONS} --shell < $CMDFILE rm -f $CMDFILE echo ## END ALT_2 # Delete old files. # Since inkscape exits with 0 even with errors, we need to explicitly check # for conversion before deleting originals. rv=0 for file in $FILES do     base=$(basename $file .svg)     if [[ -f ${base}.pdf ]]; then         rm $file     else         echo "$file not converted." 1>&2
        rv=1
    fi
done
exit $rv

svg2pdf-d requires Inkscape, which is not really that light. But it’s arguably something that should be in your FOSS toolkit anyway.

The Inkscape conversion used in svg2pdf-d seems to work fine with the SVG files generated by Eeschema as well. I’m seeing only minor differences between the SVG→PDF versus PS→PDF conversions, and the SVG→PDF files appear to have truer color. So the SVG approach may be the better one in general.

I’ll add I have added this to the wiki after a bit more testing making significant revisions.

Getting good PDF output from KiCad in Linux

kicad-appicon-128
Update: Turns out SVG might be a better base than PostScript for this. See PDF output from KiCad in Linux from SVG.

I’ve tried to get decent PDF output from KiCad‘s Eeschema a few different ways, but what I’ve found to work most reliably is to first export to PostScript (File > Plot > Plot PostScript) and then use the ps2pdf command from ghostscript to convert to PDF. (The Arch wiki has a good writeup on ps2pdf.)

The biggest problems with this are:

  1. It’s a lot of typing to get the conversion to happen (the minutia of which you won’t have memorized).
  2. It leaves you with both a PostScript and a PDF of the document(s), one of which is likely to get out of sync with the other, which may or may not be in sync with the actual schematic, etc.

To help with this, I use the script below. I drop a copy of it into the root of my KiCad projects and edit the OPTIONS as needed for the project. Then whenever I want PDFs of my schematics, I export PostScript from Eeschema and then click on this script in my file manager. Note that running this script will destroy any *.ps files in the directory—that’s by design.

#!/bin/bash

# DESTRUCTIVELY convert all postscript files
# in working directory to PDF.
# Requires ghostscript.

# Mithat Konar 2013 [http://mithatkonar.com]

OPTIONS="-dOptimize=true -sPAPERSIZE=11x17"

FILES=$(ls -1 *.ps)

for file in $FILES
do
  ps2pdf $OPTIONS $file && rm $file
done

I’ve not tried the script on any PostScript files other than those produced by Eeschema, but I’ve got no reason to think it won’t work on other PS files as well.

Stitching layers between copper zones in KiCad

old-sewing-machine-11284647472AxtX

Note: This is a 2011 repost from the now mostly defunct Biro Technology blog. I’m reposting it because it’s good information that belongs here. (P.S. Added it to the wiki.)

I’ve been working a bit with KiCad lately and have run into a problem in PCBnew with “stitching” (i.e., adding vias between) filled zones on top and bottom layers. This is something you typically do if you have flooded the unused spaces on both top and bottom of your board with copper and have connected the floods to ground or some other reference.

The KiCad FAQ outlines a process for doing this, and it works fine until you refill (i.e., re-pour) the zones–or the DRC refills them for you. When the zones are refilled, the vias you added for stitching become isolated from the zones and end up as little pads floating in space.

The problem and a workaround was discussed in a recent thread on the kicad-users mailing list. I wanted to summarize here the workaround in a slightly less terse way:

  1. Route the board and define your zones as you always have.
  2. Fill the zones as you always have.
  3. Select “Add tracks and vias” from the toolbar on the right.
  4. Click on an existing pad that’s connected to the zone’s net, drag the pointer a little bit to create a short track, then either (a) right-click and select “Place Via” or (b) type the ‘V’ shortcut.
  5. To add more stitching vias, continue to drag the pointer and type ‘V’ where you want to drop vias (or right-click and select “Place Via”).
  6. When you are done placing vias, hit the ‘End’ key on your keyboard (or right click and select “End Track”).

You can repeat this as many times as you want to create different clusters of stitches. When you refill zones, the vias will retain the connectivity information and work as expected.

OSH Park board order

PCB

Since Olimex’s PCB prototype service is currently offline, I decided to try another provider for a project I’m working on. The one I settled on is OSH Park, and I have to say the online ordering process worked far more smoothly than I expected. I uploaded a *.zip file full of Gerbers exported directly from KiCad, and the system recognized everything perfectly. Here’s hoping the boards turn out equally fine.

One bit of advice: As near as I can tell, KiCad’s layer alignment targets only add extra square inches to your design for which you will be charged, so you may as well leave them off.

First commercial PCB layout using libre tools

I recently delivered my first printed circuit board layout project using libre software. It’s an actively regulated, high current power supply for an audio equipment manufacturer, and it should be on the shelves in a month or so. The software used was KiCad (GPL2), though I did use FreeRouting (gratisware) to help route the board. I am quite happy with the results and the process. The work was done completely in Linux (Debian Wheezy, if you’re curious), proving that libre EDA–including the OS–is entirely possible.

Except for autorouting, the overall experience was comparable to working with my previous go-to package: the now-defunct WinQCad. While FreeRouting’s autorouter seems comparable to the best that other high-value EDA tools can presently offer, WinQCad’s was in a class of its own. FreeRouting still gets the job done, but it needs more hand-holding and prodding.

In addition to using it for PC layout, I am also using KiCad as a front-end schematic capture tool for SPICE simulations. Now that I’m over the worst of the learning curve, I’m really looking forward to doing more work in KiCad.