X-Authentication-Warning:	delorie.com: mail set sender to geda-help-bounces using -f
X-Recipient:	geda-help AT delorie DOT com
X-Authority-Analysis:	v=2.0 cv=W/m6pGqk c=1 sm=0 a=iT9oT/qeaKT7VIyqxLI7Lg==:17 a=vxbrVm3e934A:10 a=qvxnG7FKX0EA:10 a=AHZmWlJqWhoA:10 a=05ChyHeVI94A:10 a=N659UExz7-8A:10 a=ayC55rCoAAAA:8 a=o8dMl00YQu0A:10 a=lcUJnDsQAAAA:8 a=sHMjRCDmr7d97aIwFLQA:9 a=pILNOxqGKmIA:10 a=M3k38b3RpH0A:10 a=iT9oT/qeaKT7VIyqxLI7Lg==:117
X-Cloudmark-Score:	0
X-Authenticated-User:
X-Originating-IP:	67.249.82.108
Message-ID:	<50FDF351.6040303@twcny.rr.com>
Date:	Mon, 21 Jan 2013 21:02:57 -0500
From:	"Tom, KC2ZAT" <kc2zat AT twcny DOT rr DOT com>
User-Agent:	Mozilla/5.0 (Windows NT 6.1; rv:17.0) Gecko/20130107 Thunderbird/17.0.2
MIME-Version:	1.0
To:	geda-help AT delorie DOT com
Subject:	[geda-help] Need help with footprint I've been working on
X-Enigmail-Version:	1.5
Reply-To:	geda-help AT delorie DOT com

Hi list.
I have some footprints that I've been working on that I cannot get
right.  I'm hopeful some more experienced eyes can help.

What I want to do is use DB-9/15/25/37 (male and female) on the edge of
a PC board.  I'm trying to build 2-way radio programming equipment that
will fit in the shell of, for example, a DB-9 connector.

I got the idea from a web site I stumbled across.  They were doing
something similar but with perf board and boards from ExpressPCB (I think.)

I've tried many iterations with PCB (GUI) to create the footprints.
None were very good.  I recently read the file format information for
the footprint files then looked at my last attempt.  I discovered that I
could recycle parts of the design and fix quite a number of issues by
directly hacking the footprint with a text editor.  That left me with a
(probably) useful symbol but with one problem.

The mark was way off the footprint.  It was actually sitting "outside"
the symbol when I looked at it with PCB.  Much playing around with the X
and Y mark numbers in the ELEMENT header and messing with the MARK
statement would not fix the problem.  In fact, I could not even get the
mark to move.

Searching around gedasymbols.org got me a few things I thought I could
use as a model to figure out what I was doing wrong.  That's when I
discovered there were a few footprints that seemed to have what I
wanted.  The catch was no 15 and 37 pin footprints.  At this point, my
pride is at stake and I really want to get this to work.  (I had hoped
to contribute the footprints but they may not be needed.)

Here is the last "working" footprint.  It is a DB-9F but the mark is in
the wrong place.  I know the X and Y mark values are zero.  This is the
way I saved this version because it fixed all of my other problems.

Element["" "" "" "" 0 0 0 0 0 100 ""]
(
	Pad[5500 6600 12500 6600 9000 2000 10000 "" "5" ""]
	Pad[5500 12050 12500 12050 9000 2000 10000 "" "9" "onsolder"]
	Pad[5500 17500 12500 17500 9000 2000 10000 "" "4" ""]
	Pad[5500 22950 12500 22950 9000 2000 10000 "" "8" "onsolder"]
	Pad[5500 28400 12500 28400 9000 2000 10000 "" "3" ""]
	Pad[5500 33850 12500 33850 9000 2000 10000 "" "7" "onsolder"]
	Pad[5500 39300 12500 39300 9000 2000 10000 "" "2" ""]
	Pad[5500 44750 12500 44750 9000 2000 10000 "" "6" "onsolder"]
	Pad[5500 50200 12500 50200 9000 2000 10000 "" "1" "square"]

	)

This is my last attempt to get the mark where it belongs.
I tried switching to relative offsets.  I created the following from
scratch using a spreadsheet to help me get the pin spacing math right.
Please excuse the formatting here.  I'm doing the PCB work on Ubuntu
12.04LTS but Wordpad is scrambling some of the quoted text for some
reason.  I planned to worry about the text positioning after I fix the
mark.  Actually, I was not planning on any text on the footprint.

ELEMENT[0x00000000 “DB9F edge connector pads” “” “” 9000 27500 -22000
-6000 0 100 0x00000000]
(
PAD[5500 -21800 12500 -21800 9000 2000 10000 “1” “1” “square”]
PAD[5500 -16350 12500 -16350 9000 2000 10000 “6” “6” “onsolder”]
PAD[5500 -10900 12500 -10900 9000 2000 10000 “2” “2” ]
PAD[5500 -5450 12500 -5450 9000 2000 10000 “7” “7” “onsolder”]
PAD[5500 0 12500 0 9000 2000 10000 “3” “3” ]
PAD[5500 5450 12500 5450 9000 2000 10000 “8” “8” “onsolder”]
PAD[5500 10900 12500 10900 9000 2000 10000 “4” “4” ]
PAD[5500 16350 12500 16350 9000 2000 10000 “9” “9” “onsolder”]
PAD[5500 21800 12500 21800 9000 2000 10000 “5” “5” ]
)

This attempt results in a blank PCB on the screen and the following in
the message log.
ERROR parsing file '/home/kc2zat/Desktop/Untitled Document'
    line:        1
    description: 'syntax error, unexpected $undefined'
Note:  home directory is "/home/kc2zat"
Loading menus from /usr/share/pcb/gpcb-menu.res

It appears to be pointing at my ELEMENT statement on line 1.  I've read
and reread the DOC on the ELEMENT statement and I can't find anything
wrong.  Digging through Google only turns up stuff for RUBY and
esoteric, unrelated, and useless items.

A footnote here.  Looking at the ELEMENT write up in the PCB DOC, it
seems to conflict with the MARK write up in the same DOC.  It could be
just the way I'm reading the manual but both of them seem to point to
each other as the "old format" that is no longer recommended.

I would really appreciate any help and insights with this.
Thanks.
-- 
Tom, KC2ZAT
"Imagination is more important than knowledge." -- Albert Einstein

- Raw text -