Re: [PEDA] a lib. for everyone

2001-07-29 Thread Don Ingram

One step better  more in keeping with the Fashion...   oops I mean Software
Industry would be to use XML.  A textural database structure ( for want of a
more accurate definition ).

Cheers

Don Ingram
[EMAIL PROTECTED]

- Original Message -
From: Bagotronix Tech Support [EMAIL PROTECTED]
To: Protel EDA Forum [EMAIL PROTECTED]
Sent: Friday, July 27, 2001 7:14 AM
Subject: Re: [PEDA] a lib. for everyone


  It should be noted that the Protel ASCII format is not only text based,
 but
  it is self-documenting. All the records and fields are named wherever
they
  occur. The library format is binary only, but one could import parts to
a
  PCB file and then create a library from the PCB file.

 True, but I think the reason for the gripe is that Protel files are not
 native ASCII.  To get ASCII, you have to export files, an extra step that
 has to be done after each change to the original binary file.

 I agree with Chris that Protel's native file format should be ASCII.  Back
 in the days when we were using 286 PCs, the faster and smaller binary file
 formats made the difference between snappy and sluggish.  Anyone remember
 OrCad SDT III where you had to compose and decompose the libraries?  OrCad
 was fast, fast, fast, but used native binary formats.  Nowadays with
 stratospheric CPU clocks and RAM sizes, it is less of an issue.  The EXEs
 are 10 times the size of the board data!

 Board data should be native ASCII because years later when someone wants
you
 to resurrect or modify an old design, you have data that you can read
 without the program that created it.  You probably won't have that old
 program years from now!  Even if you do, will it run on the computer you
 have then?  I recently read a science fiction book set in the present time
 where one the characters in the story printed on paper anything he wanted
to
 save for the future.  I thought this was hilarious until the next
paragraph,
 where he explained that paper records lasted longer than disk data (true)
 and would always be compatible with future eyeballs (also true).  We have
 some floppies from late 80's/early 90's that are not readable anymore due
to
 magnetic decay.  But we have plenty of paper office records from that time
 that are still quite readable!

 Best regards,
 Ivan Baggett
 Bagotronix Inc.
 website:  www.bagotronix.com


 - Original Message -
 From: Abd ul-Rahman Lomax [EMAIL PROTECTED]
 To: Protel EDA Forum [EMAIL PROTECTED]
 Sent: Thursday, July 26, 2001 5:02 PM
 Subject: Re: [PEDA] a lib. for everyone


  At 03:06 PM 7/26/01 -0400, chris mackensen wrote:
  because
  cadence's files are all OPEN and TEXT based and NOT ENCAPSULATED in a
  DATABASE (and NOT BINARY), it was a matter of my perl script just
parsing
  Excel sheets of pin data (automatically saved as text files over the
web)
  and generating the concept/cadence schematic part as the appropriate
TEXT
  files.
 
  It should be noted that the Protel ASCII format is not only text based,
 but
  it is self-documenting. All the records and fields are named wherever
they
  occur. The library format is binary only, but one could import parts to
a
  PCB file and then create a library from the PCB file.
 
 
  [EMAIL PROTECTED]
  Abdulrahman Lomax
  P.O. Box 690
  El Verano, CA 95433
 


* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-27 Thread Steve Wiseman



On Fri, 27 Jul 2001 [EMAIL PROTECTED] wrote:
 The pin data probably does exist somewhere as a text file or bonding details of
 die pad location ( in x,y ) to pin number on the package. But for manufacturers
 to put these into Protel specific format, its not going to happen soon.

I don't need it in Protel format, just plain text, CSV, Excel, _anything_
more machine readable than pdf, which is people-readable. 
I'm prepared to write (or have written) perl / python / whatever proglets,
or even editor macros, to turn the manufacturers info into an ASCII PCB
part - there's no magic to it at all. Typing these things in, over and
over again, is futile, error-prone and dull. 

 There is
 possibly a legal implication in providing partial schematic details and if the
 format was a text file of pin numbers and names ready to be imported into CAD
 SCH editors it may not be where the manufacturers want to be involved in.

They want to be involved in selling their products, and having them
work. Xilinx came to terms with this, and it's likely to just be a matter
of time before it's commonplace. 

 For high pin count devices the pin data is always in manufacturers data sheets
 as tables spanning  multiple pages and giving pin number, name and details. Im
 speculating here but this could be scanned/ ocr'd and then maybe put in Excel
 and massaged into some sort of format.

But this is a load of reverse engineering that's error prone and futile. 

 But the missing link is Protel providing an easy and workable schematic editor
 able to import some form of this data. Another one for Mr Protel.

To be honest, I'm happy with what we've got. ASCII libraries are fine. 

Steve


* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-27 Thread Steve Wiseman



On Thu, 26 Jul 2001, chris mackensen wrote:

 for example, to get an Excel spreadsheet of pin data, one only has to copy
 and paste from PDF to Excel (usually through MS-Word, to get the tables
 correct)

Hmm, the massaging this needs normally makes it very painful. I've not
tried going through Word (I don't have Word to try it on...). Sometimes, a
combination of txt copy from pdf, plus save pdf as rtf, gets a usable
result, but neither copes gracefully with things like manufavturers
leaving table entries blank for power pins, etc. Still, better than
nothing!

 I used to have an automatic Excel sheet to schematic part generator

At a previous employment, I had a VHDL to Altera Part generator, which was
sweet - consistent pin names from source to PCB, correct pin
directions. It was driven from the post-layout Altera tools' output.

Steve


* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-26 Thread Brian Guralnick

For validation, on the PCB side, I used to place a 'scanned version' in
a translucent state, of the part over the footprint.  I've been able to find
and correct many footprint problems in advance using this technique.  If
such an ability was built into Protel  a copy of the scan was kept with the
footprint in the library this would help ease proofing of the footprint.

As for part naming, I think that only exact part numbers, down to the
full suffix will do in such a database.  Though you can search for just the
prefixes in the database, we don't need the 'Napster' equivalent happening
here, but it will happen anyways.

I also think that each component should also come with it's own
footprint, with the attached component name to it.  Otherwise, matching
component footprint name could create a cross mess.

I don't think that pin 1 should always be center.  For example, SMD LED
which require placement based on their center of the footprint, or even a
slight offset which may be by design.

We would need a database sharing tool, like sort a like 'Napster'.  When
searching for a particular part, a spec column would list:
1) user's source of part
2) user who entered part, date of entry
3) component part number / type / included /
4) footprint part number / type / included / metric / imperial / pin1
center /

Though, to properly set this up, I have no clue there...


Brian Guralnick



* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-26 Thread Bagotronix Tech Support

 As for part naming, I think that only exact part numbers, down to the
 full suffix will do in such a database.  Though you can search for just
the
 prefixes in the database, we don't need the 'Napster' equivalent happening
 here, but it will happen anyways.

The problem with full suffix is that many parts use suffix modifiers for
non-physical specification.  For example, temperature and speed grades.
Perhaps some standard for indicating wildcards in those suffix character
positions is needed.  For example, MAX186??AP is a MAX186 of any precision
grade and temp grade, with an SSOP20 package.  Note:  this is not an
endorsement of Maxim parts!  Does anyone know if the wildcard character '?'
is acceptable in footprint names?  I haven't tried it.  If not, some other
non-alphanumeric character must be used (maybe '_').  Of course, this
example is a chip.  Many special footprints will be connectors and passives,
not chips.  The situation still applies, however.  AMP (or Tyco, whatever
they are calling themselves this week) connecters come in many variants with
similar footprints.

About Napster-like file sharing, the problem with it is that the person who
has what you want has to be online when you want it.  I don't like the idea
of exposing my workstation PC to the outside world with an open tunnel
through the firewall, and having to leave it on all the time.  That's why we
need a website to host this thingy.

Best regards,
Ivan Baggett
Bagotronix Inc.
website:  www.bagotronix.com


- Original Message -
From: Brian Guralnick [EMAIL PROTECTED]
To: Protel EDA Forum [EMAIL PROTECTED]
Sent: Tuesday, July 24, 2001 11:29 AM
Subject: Re: [PEDA] a lib. for everyone


 For validation, on the PCB side, I used to place a 'scanned version'
in
 a translucent state, of the part over the footprint.  I've been able to
find
 and correct many footprint problems in advance using this technique.  If
 such an ability was built into Protel  a copy of the scan was kept with
the
 footprint in the library this would help ease proofing of the footprint.

 As for part naming, I think that only exact part numbers, down to the
 full suffix will do in such a database.  Though you can search for just
the
 prefixes in the database, we don't need the 'Napster' equivalent happening
 here, but it will happen anyways.

 I also think that each component should also come with it's own
 footprint, with the attached component name to it.  Otherwise, matching
 component footprint name could create a cross mess.

 I don't think that pin 1 should always be center.  For example, SMD
LED
 which require placement based on their center of the footprint, or even a
 slight offset which may be by design.

 We would need a database sharing tool, like sort a like 'Napster'.
When
 searching for a particular part, a spec column would list:
 1) user's source of part
 2) user who entered part, date of entry
 3) component part number / type / included /
 4) footprint part number / type / included / metric / imperial / pin1
 center /

 Though, to properly set this up, I have no clue there...

 
 Brian Guralnick




* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-26 Thread Andrew J Jenkins
 


Re: [PEDA] a lib. for everyone

2001-07-26 Thread Dennis Saputelli

i would agree with steve's comments here
sometime i have spent longer looking for a footprint than it would take
to make one (simple ones only that is)
also it is true that they can and sometimes should be slanted to the
project at hand
sch parts, however, don't strike me as being hard at all

Dennis Saputelli

Steve Wiseman wrote:
 
 On Thu, 26 Jul 2001, Ian Middleton wrote:
 
  It could also ask about paste, solder, plating and tenting overrides as
  well. It could also allow mixing on units when entering (mills and mm).
 
 you can type xxxmm and yyymil - works nicely. The rest of the wishlist I
 also wish for...
 
 (The library discussion is interesting, but I'm an untrusting person, and
 do my own library parts. The effort taken to validate a footprint is
 similar to the effort to make it, and I often slant footprints to a
 particular project / assembly tactic - for instance, if I know a board's
 going to involve a lot of debug with a scope, the QFP pads all grow a
 bit...) I also don't find making footprints that much of a chore,
 certainly nothing compared to the hell that is building schematic
 parts. Oh for a pdf - library converter that'd get the pin types, names
 and numbers right!
 
 Steve
 

-- 
___
www.integratedcontrolsinc.comIntegrated Controls, Inc.
   tel: 415-647-04802851 21st Street  
  fax: 415-647-3003San Francisco, CA 94110

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-26 Thread Ted Tontis

I would like to try and stay with Wind Rivers standards. They have
already done most of the leg work on this subject. I know most have a lot
going on and I wanted everyone to have time to look over some of the
documentation on Wind Rivers setup, and comment on the possible problems or
conflicts they may see in there setup.
As for the one who should be leading the group on this matter, my
intentions where to make this a group project and have everyone openly voice
there opinion and ideas. Everyone's concerns should be addressed no matter
what they are, is that not why this group was set up in the first place?
***(I vote no leader)***
(IMO) I do not believe one person should be the deciding factor for
new ideas or possible revisions brought on by other users or them selves (it
makes it look like one is trying to take credit for someone else's
idea/work). Mind you this was not my idea. I got the idea from Tom Hausherr
from Wind River, after I sat in one of his white paper sessions at the IPC
designers council symposium. What he said struck a cord and I felt it would
be a very helpful tool for everyone in the group, someone new to Protel, or
another designer using another CAD pkg to shorten design time. Lets face it
we all work with father time on or backs to get a project out in time.

(step off the soap box)

Thank you,

Ted 


-Original Message-
From: Andrew J Jenkins [mailto:[EMAIL PROTECTED]]
Sent: Thursday, July 26, 2001 11:04 AM
To: Protel EDA Forum
Subject: Re: [PEDA] a lib. for everyone


On 11:10 AM 7/26/2001 -0400, Bagotronix Tech Support wrote:
Does anyone know if the wildcard character '?'
is acceptable in footprint names? 

Good question. According to a Windriver document (Getting20%Started.doc in
the autocad directory), under the sub-heading Pads software decal list
defaults, item 3 states that the ? is illegal, along with other
standard illegal characters  *   ; , 

Decals are footprints, yes? (i'm kinda ignorant of what PADS folk call
footprints)

I believe that Ted said he basically wants to follow the standards
previously set by the folks at Windriver, who have already begun the process
for the PADS community specifically, but also for the EDA community at large
by virtue of their having already solved many of the administrative
questions.

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-26 Thread chris mackensen


 Oh for a pdf - library converter that'd get the pin types, names
 and numbers right!

One solution may be to look at ClientBasic as Sch:PlacePin has some
parameters that govern Name,Number,X,Y, etc. etc.  but ClientBasic (VB for
all intents and purposes) is a super major pain in the ass to do any string
manipulation (compared to perl ;-))

I am also sure a server could be written using the Protel API, but (as in
all cases herein) a very specific (say for example) data file (or
spreadsheet) format would have to be defined  but a database file might
be better (in that the format does not matter, rather the field names and
types do).

for example, to get an Excel spreadsheet of pin data, one only has to copy
and paste from PDF to Excel (usually through MS-Word, to get the tables
correct)  This will work if the PDF datasheet is not pages of embedded
images and actually contains real text.

some PDFs are locked which is where ghostview came in (a free'er PDF type
viewer which can do a text extract of pages specified in a range).
passwords were usually ignored in ghostview

As another solution attempt, I did try to do a database import in a schlib,
but it does not seem to work (even though a lot of the server processes in
schlib are really sch processes, hence Sch:ImportSchematicFromDatabase
doesn't work even though Sch:PlacePin does (in theory))!  if they could
at least add that feature like how they have it on regular sch's then this
could work very well from the excel/access/export_to-DB4 then import pin
database lists into protel approach...

I used to have an automatic Excel sheet to schematic part generator
it was a web automated and COM/OLE Automating perl script (perl is free and
open to everyone) but it was tailored to Cadence unfortunately because
cadence's files are all OPEN and TEXT based and NOT ENCAPSULATED in a
DATABASE (and NOT BINARY), it was a matter of my perl script just parsing
Excel sheets of pin data (automatically saved as text files over the web)
and generating the concept/cadence schematic part as the appropriate TEXT
files.  It was then a matter of pushing pins around in the cadence
graphical editing environment as the perl script instantiated the pins all
in separate columns based upon pin type (power and ground pins were in two
rows on the top and bottom of the schematic/library graphics page
respectively).  This was a very easy solution to typing in 1200 pins (and
pin names in three different (3 * 1200 = 3600??yikes!) text files
w/different formats as required by some of the processes of the day in
cadence land) Oh the savings in part-debug time when human fat fingering
of the keyboard is eliminated.

I wish that I could get into protel without any specific requirements on
which language to use (like, for example, by using COM/OLE Automation... I
use it in Perl, C++, C++ Builder and any other language du jour) instead of
*HAVING* to use extremely painful Delphi (C++ Builder would be a good
compromise, because then I would write an OLE Automation wrapper (language
independent) using RAD around protel's EXE/DLLs once the NDA is lifted heh
heh heh)... sorry, my personal bias.  I digress...

Cheers,
-chris




* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-26 Thread Abd ul-Rahman Lomax

At 12:28 PM 7/26/01 -0500, Ted Tontis wrote:
 I would like to try and stay with Wind Rivers standards. They have
already done most of the leg work on this subject.

An obvious step to take would be for a user to move that the Wind Rivers 
standards be adopted, and then we could go over them and discuss any 
possible changes. I have not seen those standards yet, so I am not 
specifically recommending this, but some of us *have* seen them.

And I'd suggest that the motion be made on the library list (join by 
sending mail to [EMAIL PROTECTED]), where some kind of 
formal process can be followed and the archive will serve as a focused 
record of discussions and decisions. On the association list, we have 
regular threads which are for free general discussion and then formal 
threads, with the subject line beginning with MOTION:, followed by a brief 
title. The formal threads are strictly monitored by the chair, who can 
delete irrelevant material from the formal threads in the list archive. 
Normally this is not necessary The MOTION threads become the formal 
record of the committee.

I strongly suggest that any *formal* discussion be moved to the library 
list, for the same reason that projects involving detail are moved to 
committee in any large organization. As I conceive of it, the committee 
will only make recommendations and further general user input would be 
requested before finalizing anything.


[EMAIL PROTECTED]
Abdulrahman Lomax
P.O. Box 690
El Verano, CA 95433

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-26 Thread Abd ul-Rahman Lomax

At 03:06 PM 7/26/01 -0400, chris mackensen wrote:
because
cadence's files are all OPEN and TEXT based and NOT ENCAPSULATED in a
DATABASE (and NOT BINARY), it was a matter of my perl script just parsing
Excel sheets of pin data (automatically saved as text files over the web)
and generating the concept/cadence schematic part as the appropriate TEXT
files.

It should be noted that the Protel ASCII format is not only text based, but 
it is self-documenting. All the records and fields are named wherever they 
occur. The library format is binary only, but one could import parts to a 
PCB file and then create a library from the PCB file.


[EMAIL PROTECTED]
Abdulrahman Lomax
P.O. Box 690
El Verano, CA 95433

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-26 Thread Bagotronix Tech Support

 It should be noted that the Protel ASCII format is not only text based,
but
 it is self-documenting. All the records and fields are named wherever they
 occur. The library format is binary only, but one could import parts to a
 PCB file and then create a library from the PCB file.

True, but I think the reason for the gripe is that Protel files are not
native ASCII.  To get ASCII, you have to export files, an extra step that
has to be done after each change to the original binary file.

I agree with Chris that Protel's native file format should be ASCII.  Back
in the days when we were using 286 PCs, the faster and smaller binary file
formats made the difference between snappy and sluggish.  Anyone remember
OrCad SDT III where you had to compose and decompose the libraries?  OrCad
was fast, fast, fast, but used native binary formats.  Nowadays with
stratospheric CPU clocks and RAM sizes, it is less of an issue.  The EXEs
are 10 times the size of the board data!

Board data should be native ASCII because years later when someone wants you
to resurrect or modify an old design, you have data that you can read
without the program that created it.  You probably won't have that old
program years from now!  Even if you do, will it run on the computer you
have then?  I recently read a science fiction book set in the present time
where one the characters in the story printed on paper anything he wanted to
save for the future.  I thought this was hilarious until the next paragraph,
where he explained that paper records lasted longer than disk data (true)
and would always be compatible with future eyeballs (also true).  We have
some floppies from late 80's/early 90's that are not readable anymore due to
magnetic decay.  But we have plenty of paper office records from that time
that are still quite readable!

Best regards,
Ivan Baggett
Bagotronix Inc.
website:  www.bagotronix.com


- Original Message -
From: Abd ul-Rahman Lomax [EMAIL PROTECTED]
To: Protel EDA Forum [EMAIL PROTECTED]
Sent: Thursday, July 26, 2001 5:02 PM
Subject: Re: [PEDA] a lib. for everyone


 At 03:06 PM 7/26/01 -0400, chris mackensen wrote:
 because
 cadence's files are all OPEN and TEXT based and NOT ENCAPSULATED in a
 DATABASE (and NOT BINARY), it was a matter of my perl script just parsing
 Excel sheets of pin data (automatically saved as text files over the web)
 and generating the concept/cadence schematic part as the appropriate TEXT
 files.

 It should be noted that the Protel ASCII format is not only text based,
but
 it is self-documenting. All the records and fields are named wherever they
 occur. The library format is binary only, but one could import parts to a
 PCB file and then create a library from the PCB file.


 [EMAIL PROTECTED]
 Abdulrahman Lomax
 P.O. Box 690
 El Verano, CA 95433


* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-26 Thread Clive . Broome



The pin data probably does exist somewhere as a text file or bonding details of
die pad location ( in x,y ) to pin number on the package. But for manufacturers
to put these into Protel specific format, its not going to happen soon. There is
possibly a legal implication in providing partial schematic details and if the
format was a text file of pin numbers and names ready to be imported into CAD
SCH editors it may not be where the manufacturers want to be involved in.

For high pin count devices the pin data is always in manufacturers data sheets
as tables spanning  multiple pages and giving pin number, name and details. Im
speculating here but this could be scanned/ ocr'd and then maybe put in Excel
and massaged into some sort of format.

But the missing link is Protel providing an easy and workable schematic editor
able to import some form of this data. Another one for Mr Protel.







Steve Wiseman [EMAIL PROTECTED] on 07/27/2001 03:19:34 AM

Please respond to Protel EDA Forum [EMAIL PROTECTED]

To:   Protel EDA Forum [EMAIL PROTECTED]
cc:(bcc: Clive Broome/sdc)

Subject:  Re: [PEDA] a lib. for everyone





On Thu, 26 Jul 2001, Dennis Saputelli wrote:

 sch parts, however, don't strike me as being hard at all

My current design, with 3 devices over 250 pins, has reminded me how dull
generating sch parts can be. It's just galling that the information _must_
be available within the IC manufacturers, and if it was available in a
format more machine-readable than pdf, library generation would be a case
of dragging pins round rather than dull-yet-critical typing, and I'm idle.

Steve







* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-25 Thread Clive . Broome



Agreed the Protel libraries (along with any others) need to be properly
verified. The point is trying to cram all these footprints into one super
library. If you just take the Protel semiconductor libraries as a basis for
working out the number of parts, plus variations as you describe, its rather
unwieldy. Add connectors (1K?) switches (1K?) general components (1K?) thats my
estimate. Did the original poster say he had 8500 footprints? In one library?
Thats serious library management problems.
The usual compliants about the footprint wizard and 'pathing' the footprint back
to a source library might solve a lot of the requirements for having 'verified'
footprints. I wouldnt use any part I didnt make myself but trying to make it in
the footprint wizard, it involves an estimation and a calculation and crossed
fingers Its a joke. Mr Protel this needs attention

BR Clive









John Haddy [EMAIL PROTECTED] on 07/25/2001 02:04:02 PM

Please respond to Protel EDA Forum [EMAIL PROTECTED]

To:   Protel EDA Forum [EMAIL PROTECTED]
cc:(bcc: Clive Broome/sdc)

Subject:  Re: [PEDA] a lib. for everyone



Under no circumstances should the Protel libraries be copied! They
are riddled with errors (I'd list them all as unverified). For
instance, any metric pitch component that's been created with the
footprint wizard is likely to be wrong.

The biggest hurdle I forsee is the widely differing requirements
of any single footprint. For example: a footprint for wave solder
is different from that for reflow; the IPC worst-case design
methodology generates overly conservative footprints; a designer
working on high density layouts will want bleeding edge land
pattern designs, not general purpose ones with excessive silkscreen
clearances (for example). So just with these few variations, we're
up to at least four variants of every footprint!

Much as I applaud the sentiment, I can't see a single library ever
providing a single source set of footprints.

I regularly use components that have land pattern designs tailored
to the manufacturing process; e.g. I might create an SC70-6 footprint
which is fine for the current project because I'll be using a
0.004 paste stencil thickness, however the same design may well cause
manufacturing issues if it's used with a 0.006 paste stencil!

We're going to need much more sophisticated design rules before I'd
consider using anybody else's designs (so that I can check, for
example, paste aperture ratios given a specified screen thickness)

Just my $0.02

John Haddy

 -Original Message-
 From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
 Sent: Wednesday, 25 July 2001 12:03 PM
 To: Protel EDA Forum
 Subject: Re: [PEDA] a lib. for everyone




 Im assuming that the existing PCB footprint libraries on the
 Protel website
 would be a subset of this super library.
 The libraries are:

 BGA531 footprints in 9 libraries based on pin pitch
 Through Hole  107 footprints in 5 libraries
 QFP's 226 footprints in 12 libraries based on pin
 pitch and
 index position (with duplicated footprints)
 Chip carrier74 footprints  in 2 libraries
 SOP   129 footprints  in 2 libraries

 Plus a number of manufacturer specific and general libraries.
 Taking just the
 semiconductor footprints, thats
 1067 footprints that will be bundled into the super library that
 someone will be
 remaking. Or will these footprints
  just be copied from the existing Protel libraries? If these
 seperate libraries
 get jumbled into one library, will
  it make using and finding easier? Who will be doing the work on this?



 ___

 Clive Broome
 IDT Sydney Design CentrePh: +61 2 9763 3513
 8 Bayswater Dr, HomebushFax:+61 2 9763 3409
 Sydney,  NSW, 2127  Email:[EMAIL PROTECTED]
 Australia

  Australia's Leading Semiconductor Designers
 ---








 Ted Tontis [EMAIL PROTECTED] on 07/24/2001 04:50:18 AM

 Please respond to Protel EDA Forum [EMAIL PROTECTED]

 To:   Protel Forum (E-mail) [EMAIL PROTECTED]
 cc:(bcc: Clive Broome/sdc)

 Subject:  [PEDA] a lib. for everyone



 Would there be any interest in a PCB footprint lib. with all the parts you
 would ever need for free. I ask this because I am working on
 trying to get a
 large lib. in Protel. It would have the silk screen, a fence that would be
 on the last electrical layer to avoid component placement conflictions,
 assembly art work, pin 1 id. All parts would be in mm I welcome any input
 towards this idea weather it be good or bad.

 Ted Tontis C.I.D.
 Engage Networks
 316 N. Milwaukee Street
 Suite 214
 Milwaukee WI, 53202
 PH 414-273-7600 ext. 7607
 FX 414-273-7601












* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http

Re: [PEDA] a lib. for everyone

2001-07-25 Thread Ian Wilson

On 08:55 PM 24/07/2001 -0700, Abd ul-Rahman Lomax said:

As I stated, I do prefer that we move this discussion to the new list, 
[EMAIL PROTECTED]

You may prefer but the vast majority if people responding do not want 
another mail list it seems...

Ian Wilson

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-25 Thread Ted Tontis

I started the thread on this subject. I knew the size of this
project was going to be large, that is why I posted the thread in the first
place. I wanted to see if anyone else would like to help or even if anyone
wanted to have a large library system that was free. My idea was to get
everyone who wanted to try and shorten there design time and get there input
and ideas.
I do not have 8500 parts in on library please give me some credit.
I said that I am working on importing 8500 parts into Protel from another
cad system.  There are other designers doing this and they would most
likely trade footprints for footprints. The scope of this goes further then
here, and the possibility to increase the library could come from other
designers or as Mr. Lomax stated help from Protel. I am sure Protel is
following this thread closely, if we where to take up this project Protel
would be able to say we have the largest certified free library. That would
boost there sales to unbelievable numbers.
As designers we sometimes have to make the impossible happen, and we
do make it happen. So why then all the negative remarks?

Ted


* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-25 Thread Brad Velander

Abd ul-Rahman,
a footprint without associated paste mask detailing is only an
incomplete solution. It's like having the PCB without components or a
soldering iron to attach them. So everybody goes off into their corner and
duplicates their time and efforts to prescribe and describe the paste mask
for each and every footprint design? This would probably lead to footprint
designs which had not considered paste mask or varying levels of paste
deposition during the design.
What was the point of this thread? I think you may have overlooked
the obvious, that is the footprints should be a complete solution for all
and eliminate duplicated individual effort.
To this end the solder paste mask deposition should be specified
from the start and the whole library designed to work with that deposition
of solder paste. Otherwise you do nobody any favor if one footprint works
with 4 mil solder but the one you choose to locate directly beside it was
meant for 7 mil solder deposition.

Brad Velander,
Lead PCB Designer,
Norsat International Inc.,
#300 - 4401 Still Creek Dr.,
Burnaby, B.C., V5C 6G9.
Tel. (604) 292-9089 direct
Fax (604) 292-9010
website www.norsat.com


 -Original Message-
 From: Abd ul-Rahman Lomax [mailto:[EMAIL PROTECTED]]
 Sent: Tuesday, July 24, 2001 9:56 PM
 To: Protel EDA Forum
 Subject: Re: [PEDA] a lib. for everyone
 
 
SNIP
 
 I'd assume that the libraries would not normally contain paste mask 
 information.
 
SNIP
 
 [EMAIL PROTECTED]
 Abdulrahman Lomax
 P.O. Box 690
 El Verano, CA 95433
 

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-25 Thread Brooks,Bill

Hi Ted,
Wow... this really got some very interesting attention... Good points all.
Ted, It sounds like you are working with those guys over at WindRiver in San
Diego. They have undertaken a project to 'standardize' all footprints in
millimeters and provide them with development tools and accessories to the
world for free (with WindRivers name all over it of course). Tom Hausherr, a
very experienced and driven PCB designer and leader in the PADS users group
in San Diego,  is coordinating the effort there along with consultation with
the leaders of the IPC Designers Council and volunteer designers and has
made considerable progress. The only drawback, that I can see, is that the
libraries that are being created are in PADS, somewhat useless to other
cad systems. Is this where you got the inspiration to try to get this going
for Protel? It is really great to see the good will of so many technical
people being exercised for the common good of the industry as a whole. I
applaud your effort to make a difference! Keep it up! You 'nay' sayers...
its already happening on a pretty big scale, I have seen the progress Tom
Hausherr has made... its impressive. If the work is divided up among a large
group its very possible to do amazing things... Protel people are some of
the best technical people around... think about it. You can make a
difference, even if its only a small difference, it adds up. You are a great
bunch keep up the ideas... :)

- Bill Brooks

Bill Brooks 
PCB Design Engineer 
DATRON WORLD COMMUNICATIONS INC.
3030 Enterprise Court 
Vista, CA 92083 
Tel: (760)597-1500 Ext 3772 Fax: (760)597-1510 
mailto:[EMAIL PROTECTED] 
IPC Designers Council, San Diego Chapter 
http://www.ipc.org/SanDiego/
http://home.fda.net/bbrooks/pca/pca.htm



-Original Message-
From: Ted Tontis [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, July 25, 2001 7:30 AM
To: 'Protel EDA Forum'
Subject: Re: [PEDA] a lib. for everyone


I started the thread on this subject. I knew the size of this
project was going to be large, that is why I posted the thread in the first
place. I wanted to see if anyone else would like to help or even if anyone
wanted to have a large library system that was free. My idea was to get
everyone who wanted to try and shorten there design time and get there input
and ideas.
I do not have 8500 parts in on library please give me some credit.
I said that I am working on importing 8500 parts into Protel from another
cad system.  There are other designers doing this and they would most
likely trade footprints for footprints. The scope of this goes further then
here, and the possibility to increase the library could come from other
designers or as Mr. Lomax stated help from Protel. I am sure Protel is
following this thread closely, if we where to take up this project Protel
would be able to say we have the largest certified free library. That would
boost there sales to unbelievable numbers.
As designers we sometimes have to make the impossible happen, and we
do make it happen. So why then all the negative remarks?

Ted


* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-25 Thread Abd ul-Rahman Lomax

At 07:10 PM 7/25/01 +1000, Ian Wilson wrote:
On 08:55 PM 24/07/2001 -0700, Abd ul-Rahman Lomax said:

As I stated, I do prefer that we move this discussion to the new list, 
[EMAIL PROTECTED]

You may prefer but the vast majority if people responding do not want 
another mail list it seems...

Mr. Wilson, we know, is allergic to new lists and only accepts them with 
difficulty. However, there will need to be a *lot* of detailed discussion, 
I expect, and we will need polling tools (if large numbers want to get 
involved, otherwise it could be done more informally).

There are already more subscribers to the new list than there are people 
responding in this thread, so Mr. Wilson's comment about vast majority 
may be a bit biased. I know that I'm now suppressing my own responses here. 
Obviously, I can't stop users from discussing the issue here, but, if past 
discussions are a guide, there will be a lot of talk and little or no 
action coming out of it. We will need *ratified* standards or we will end 
up with a library that many designers won't want to use; I want to maximize 
consensus on the standards so that they will truly represent the collective 
wisdom of our fellow users.


[EMAIL PROTECTED]
Abdulrahman Lomax
P.O. Box 690
El Verano, CA 95433

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-25 Thread Abd ul-Rahman Lomax

At 09:29 AM 7/25/01 -0500, Ted Tontis wrote:

 As designers we sometimes have to make the impossible happen, and we
do make it happen. So why then all the negative remarks?

All I know is that if you want to accomplish something valuable, get used 
to it. It is *much* easier to sit back and find fault with a proposal than 
to work out how to make it successful, and most people, or at least a 
sizable number, take the easy way.


[EMAIL PROTECTED]
Abdulrahman Lomax
P.O. Box 690
El Verano, CA 95433

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-25 Thread Bagotronix Tech Support

 You may prefer but the vast majority if people responding do not want
 another mail list it seems...

 Mr. Wilson, we know, is allergic to new lists and only accepts them with
 difficulty. However, there will need to be a *lot* of detailed discussion,

I don't think a new list is needed to handle PCB footprint message traffic.
After all, aren't we all using Protel?  The footprints being distributed
would be in Protel data format.  Therefore, it's a Protel EDA Forum issue,
right?  The Protel EDA Forum is not just about workarounds and software
bugs, right?

I think a lib for everyone is a great concept, but it will always be an
unfinished work.  I would prefer if there was a web page we could download
PCB footprints from.  Not a huge library that may or may not have what we
are actually looking for, but individual footprints in their own little
files.  For example, if I am looking for an Amp RJ-45 footprint, I should
not have to download the entire AMP library before I find out the footprint
I want isn't in it.  If possible, the footprints should be in Protel ASCII
format so they could be used with older (pre-99SE) versions of Protel.

If we don't get a web page, we can still ask others in this list for
footprints.  I did so 2 weeks ago and got several responses that were
exactly what I needed - thanks to all who responded.  But if we do get a web
page for footprints, we could proactively contribute footprints.

As far as verification of parts goes, I oppose any bureaucratization that
would hinder or slow down the footprint distribution process.  This sounds
like more total quality management business school bullsh*t.  The buyer
beware - free stuff may only be worth what you pay for it.  Check the
footprint yourself before you use it.  I do.

Best regards,
Ivan Baggett
Bagotronix Inc.
website:  www.bagotronix.com


- Original Message -
From: Abd ul-Rahman Lomax [EMAIL PROTECTED]
To: Protel EDA Forum [EMAIL PROTECTED]
Sent: Wednesday, July 25, 2001 12:56 PM
Subject: Re: [PEDA] a lib. for everyone


 At 07:10 PM 7/25/01 +1000, Ian Wilson wrote:
 On 08:55 PM 24/07/2001 -0700, Abd ul-Rahman Lomax said:
 
 As I stated, I do prefer that we move this discussion to the new list,
 [EMAIL PROTECTED]
 
 You may prefer but the vast majority if people responding do not want
 another mail list it seems...

 Mr. Wilson, we know, is allergic to new lists and only accepts them with
 difficulty. However, there will need to be a *lot* of detailed discussion,
 I expect, and we will need polling tools (if large numbers want to get
 involved, otherwise it could be done more informally).

 There are already more subscribers to the new list than there are people
 responding in this thread, so Mr. Wilson's comment about vast majority
 may be a bit biased. I know that I'm now suppressing my own responses
here.
 Obviously, I can't stop users from discussing the issue here, but, if past
 discussions are a guide, there will be a lot of talk and little or no
 action coming out of it. We will need *ratified* standards or we will end
 up with a library that many designers won't want to use; I want to
maximize
 consensus on the standards so that they will truly represent the
collective
 wisdom of our fellow users.


 [EMAIL PROTECTED]
 Abdulrahman Lomax
 P.O. Box 690
 El Verano, CA 95433


* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-25 Thread Max Henzi

From my standpoint, it is counterproductive to start a new list to deal with
this issue.  Having multiple lists merely serves to fragment the user
community.  If the topic is germane to Protel users, then it should appear
on this list.  I for one, do not have the time or the inclination to monitor
numerous lists.  I guess I need to be included as one who is allergic to
new lists and only accepts them with difficulty.

Best regards,

Max P. Henzi, CEO
Lavenir Technology
2440 Estand Way
Pleasant Hill, CA 94523
voice: (925) 680-7400  fax: (925) 686-5131 
email: [EMAIL PROTECTED]
Web: www.lavenir.com 

 -Original Message-
From:   Bagotronix Tech Support [mailto:[EMAIL PROTECTED]] 
Sent:   Wednesday, July 25, 2001 10:11 AM
To: Protel EDA Forum
Subject:Re: [PEDA] a lib. for everyone

 You may prefer but the vast majority if people responding do not want
 another mail list it seems...

 Mr. Wilson, we know, is allergic to new lists and only accepts them with
 difficulty. However, there will need to be a *lot* of detailed discussion,

I don't think a new list is needed to handle PCB footprint message traffic.
After all, aren't we all using Protel?  The footprints being distributed
would be in Protel data format.  Therefore, it's a Protel EDA Forum issue,
right?  The Protel EDA Forum is not just about workarounds and software
bugs, right?

I think a lib for everyone is a great concept, but it will always be an
unfinished work.  I would prefer if there was a web page we could download
PCB footprints from.  Not a huge library that may or may not have what we
are actually looking for, but individual footprints in their own little
files.  For example, if I am looking for an Amp RJ-45 footprint, I should
not have to download the entire AMP library before I find out the footprint
I want isn't in it.  If possible, the footprints should be in Protel ASCII
format so they could be used with older (pre-99SE) versions of Protel.

If we don't get a web page, we can still ask others in this list for
footprints.  I did so 2 weeks ago and got several responses that were
exactly what I needed - thanks to all who responded.  But if we do get a web
page for footprints, we could proactively contribute footprints.

As far as verification of parts goes, I oppose any bureaucratization that
would hinder or slow down the footprint distribution process.  This sounds
like more total quality management business school bullsh*t.  The buyer
beware - free stuff may only be worth what you pay for it.  Check the
footprint yourself before you use it.  I do.

Best regards,
Ivan Baggett
Bagotronix Inc.
website:  www.bagotronix.com


- Original Message -
From: Abd ul-Rahman Lomax [EMAIL PROTECTED]
To: Protel EDA Forum [EMAIL PROTECTED]
Sent: Wednesday, July 25, 2001 12:56 PM
Subject: Re: [PEDA] a lib. for everyone


 At 07:10 PM 7/25/01 +1000, Ian Wilson wrote:
 On 08:55 PM 24/07/2001 -0700, Abd ul-Rahman Lomax said:
 
 As I stated, I do prefer that we move this discussion to the new list,
 [EMAIL PROTECTED]
 
 You may prefer but the vast majority if people responding do not want
 another mail list it seems...

 Mr. Wilson, we know, is allergic to new lists and only accepts them with
 difficulty. However, there will need to be a *lot* of detailed discussion,
 I expect, and we will need polling tools (if large numbers want to get
 involved, otherwise it could be done more informally).

 There are already more subscribers to the new list than there are people
 responding in this thread, so Mr. Wilson's comment about vast majority
 may be a bit biased. I know that I'm now suppressing my own responses
here.
 Obviously, I can't stop users from discussing the issue here, but, if past
 discussions are a guide, there will be a lot of talk and little or no
 action coming out of it. We will need *ratified* standards or we will end
 up with a library that many designers won't want to use; I want to
maximize
 consensus on the standards so that they will truly represent the
collective
 wisdom of our fellow users.


 [EMAIL PROTECTED]
 Abdulrahman Lomax
 P.O. Box 690
 El Verano, CA 95433


* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-25 Thread Ted Tontis

AJ
If you get a chance look over the Defining a CAD library at
www.pcbstandards.com and see what wind river has set up for there standards.
I thought it would be a good idea to try to stay with there standards as we
could use there libraries and they could use ours. It would cut development
time down considerably.

Ted

-Original Message-
From: Andrew J Jenkins [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, July 25, 2001 12:35 PM
To: Protel EDA Forum
Subject: Re: [PEDA] a lib. for everyone


On 10:03 AM 7/25/2001 -0700, Abd ul-Rahman Lomax wrote:
At 09:29 AM 7/25/01 -0500, Ted Tontis wrote:

 As designers we sometimes have to make the impossible happen, and
we
do make it happen. So why then all the negative remarks?

All I know is that if you want to accomplish something valuable, get used
to it. It is *much* easier to sit back and find fault with a proposal than
to work out how to make it successful, and most people, or at least a
sizable number, take the easy way.

Actually, based on the thread so far, your observation is false. The
majority of respondants in this group have had positive comments. Only two
or three found outright fault with Ted's Tontis' proposal and denied any
possibility of its fruition or their eventual use of such a library.

Obviously, I can't stop users from discussing the issue here

No, you can't. Thanks Mark.

I would urge other readers/writers to avoid the splintering that Mr Lomax is
engineering within this group.

Remember the old Roman trick: Divide and conquer.

aj


* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-25 Thread Andrew J Jenkins

On 01:10 PM 7/24/2001 -0700, Brad Velander wrote:
Andrew,
 please don't throw me into the simple nay-sayers


Nope. I didn't. Just doubting Thomas, to whom I originally replied.

  I have my doubts
about whether it could be accomplished amongst a group as diverse as ours,
but it can be done. My concerns would be more for the diverse opinions and
desired features within the library and coming to some consensus on a
standard approach.

Features are the stumbling block. As has been mentioned, one person's preferred 
footprint may not be another's. (my efforts, for instance, have been primarily limited 
to hand-soldered PCBs, resulting in a need for much larger pads than might be expected 
with other soldering processes) 

This leads me to the conclusion that either multiple working footprints for a given 
base footprint will have to be individually generated, or another process which 
allows for capturing the variations within a single, generalized part, like folding an 
agreed-upon group of mfg methods requirements into each footprint, by use of the 
available layers, or by use of some other trick(s). Unless some trickery is 
practiced, I'm concerned that the library size might grow beyond the expected 10K 
count value that has been offered.

 Many of us have that tendency to ramble, as long as it is a flow of
thoughts that others can discuss and build on what is the problem with it?

None that I can think of. I was just practicing my apologetics ;) 

aj

P.S. In answer to your question/comment a few days ago, I've been here all along, 
never left, just on Hiatus for the last couple months, and so fell into the 
nether-state of Lurker...Too much work on the decrepit 100+ yr old house to 
accomplish, and too little time to accomplish it in...(right now, it's just too 
danged hot here in Cleveburg to work outside in the sun all day long, (ex: 88F  
70%RH @ 5pm yesterday) so I have un poco tiempo with which to play before dragging 
my butt back out into the lobster pot every so often.

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-25 Thread Andrew J Jenkins
 


Re: [PEDA] a lib. for everyone

2001-07-25 Thread Andrew J Jenkins
 


Re: [PEDA] a lib. for everyone

2001-07-25 Thread Andrew J Jenkins
 


Re: [PEDA] a lib. for everyone

2001-07-25 Thread Abd ul-Rahman Lomax

At 01:11 PM 7/25/01 -0400, Bagotronix Tech Support wrote:
As far as verification of parts goes, I oppose any bureaucratization that
would hinder or slow down the footprint distribution process.  This sounds
like more total quality management business school bullsh*t.  The buyer
beware - free stuff may only be worth what you pay for it.  Check the
footprint yourself before you use it.  I do.

As I would see it, a part that is submitted would immediately be available 
to all, but it would be flagged as unverified.
Yes, caveat emptor.

[EMAIL PROTECTED]
Abdulrahman Lomax
P.O. Box 690
El Verano, CA 95433

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-25 Thread Phil So

The suggestions for locating the library on a web site has a lot of merit,
especially in view of the concerns summarised here by Andrew.  Software that
is used for on-line catalogue shopping could possibly be adapted, or used as
a model, to allow users to browse through the various permutations possible
for any particular footprint.  This would probably produce the results that
the users would want sooner, less expensively and more reliably than trying
to have the same facility incorporated within Protel.  It would then be
possible to show all of the specifications and parameters related to the
footprint, such as paste screen thickness, reflow vs solder wave vs hand
soldering, margin from actual part to the part outline etc.  Some guidelines
should be set regarding the information to be included in the spec. sheet
for each footprint.  It would then be left up to the user to decide how
applicable a particular footprint would be in his situation.  This approach
would allow Protel's programmers to stick with improving the basic product.
The web site could be supported by one or more sponsors, donations or
subscriptions.  There are probably enough participants in this forum who
have the skills to develop such a web site.  People are already putting time
into developing Protel servers and sharing them with other forum members.

Just a thought or two.

Phil So


 -Original Message-
 From: Andrew J Jenkins [mailto:[EMAIL PROTECTED]]
 Sent: Thursday, 26 July 2001 05:56
 To: Protel EDA Forum
 Subject: Re: [PEDA] a lib. for everyone
 
 
 On 01:10 PM 7/24/2001 -0700, Brad Velander wrote:
 Andrew,
  please don't throw me into the simple nay-sayers
 
 
 Nope. I didn't. Just doubting Thomas, to whom I originally replied.
 
   I have my doubts
 about whether it could be accomplished amongst a group as 
 diverse as ours,
 but it can be done. My concerns would be more for the 
 diverse opinions and
 desired features within the library and coming to some consensus on a
 standard approach.
 
 Features are the stumbling block. As has been mentioned, one 
 person's preferred footprint may not be another's. (my 
 efforts, for instance, have been primarily limited to 
 hand-soldered PCBs, resulting in a need for much larger pads 
 than might be expected with other soldering processes) 
 
 This leads me to the conclusion that either multiple working 
 footprints for a given base footprint will have to be 
 individually generated, or another process which allows for 
 capturing the variations within a single, generalized part, 
 like folding an agreed-upon group of mfg methods requirements 
 into each footprint, by use of the available layers, or by 
 use of some other trick(s). Unless some trickery is 
 practiced, I'm concerned that the library size might grow 
 beyond the expected 10K count value that has been offered.
 
  Many of us have that tendency to ramble, as long as 
 it is a flow of
 thoughts that others can discuss and build on what is the 
 problem with it?
 
 None that I can think of. I was just practicing my apologetics ;) 
 
 aj
 
 P.S. In answer to your question/comment a few days ago, I've 
 been here all along, never left, just on Hiatus for the last 
 couple months, and so fell into the nether-state of 
 Lurker...Too much work on the decrepit 100+ yr old house to 
 accomplish, and too little time to accomplish it in...(right 
 now, it's just too danged hot here in Cleveburg to work 
 outside in the sun all day long, (ex: 88F  70%RH @ 5pm 
 yesterday) so I have un poco tiempo with which to play 
 before dragging my butt back out into the lobster pot every so often.
 


The contents of this E-mail may contain information that is legally
privileged and/or confidential to the named recipient. This information is
not to be used by any other person and/or organisation. The views expressed
in this document do not necessarily reflect those of the company. 


* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-25 Thread Abd ul-Rahman Lomax

At 02:07 PM 7/25/01 -0400, Andrew J Jenkins wrote:
The problem, as I see it, with traffic in multiple locations is that it's 
end result is fragmentation of the group and

Meeting in committee is hardly fragmentation. The records of the 
committee are open to the public and will remain so, and anyone who wishes 
to join the discussion may do so. Further, the committee will report back 
to this list and anyone who does not like its recommendations will be free 
to express it.

I don't know that Mr. Jenkins is aware that I was elected chair of the 
Protel Users Association. I am the owner of record for the Protel-users 
yahoogroups lists, but I only hold those lists as trustee for the 
association; my decisions regarding them are subject to association review, 
and not only can I be overruled by the association, I *have* been overruled 
in one case.

Every organization which has accomplished something significant has learned 
to divide and delegate responsibility. Not everyone wants to participate in 
every activity, and we are already bleeding subscribers to the Techserv 
list, people who unsubscribe simply because the volume of mail becomes too 
much. I spoke to a number of these at the last PC Design Conference West.

(There is a list, protel-users-announcements, which is a *very* low traffic 
list, and I recommend that anyone who would like to receive association 
announcements without maintaining active list subscription(s) join that 
list. A piece of mail to [EMAIL PROTECTED] will 
accomplish it. It's a moderated list specifically to keep it from receiving 
traffic which is not association-sponsored.)

I know that some of us, even some of those who have been active in 
association affairs (such as Mr. Wilson, who has done a great deal and who 
deserves our gratitude), dislike the proliferation of lists. Some 
organizations similar to ours have a single list, but have required subject 
headers, and people can set their subscription preferences to include or 
exclude particular subjects. That's another way of accomplishing a similar 
purpose.

However, requiring that all users receive all mail is the same kind of 
error as was made by many cooperative organizations back in the seventies: 
everything was decided by the entire organization. Problem was, it was very 
difficult to get everyone to agree, and a single argumentative person could 
effectively prevent anything from being done. In the end, the organizations 
went according to the decisions of those who could afford the time to sit 
through long-winded plenary sessions. And, usually, after a few years, the 
organization was gone.

I'd prefer to have an opt-out system, where everyone was subscribed to all 
the lists, and then could opt out of the ones not of interest to them. If 
Mr Jenkins had his way, it would seem, *we couldn't opt out.* Is this 
really what we want?

Techserv has periodically popped in here to stop discussions it deemed 
off-topic. Originally, the yahoogroups lists were started for two reasons:

(1) to provide a backup list for when the Techserv list was down, which 
happened too many times.
([EMAIL PROTECTED])

(2) to provide a place or places to discuss or announce matters which were 
against Techserv policy, such as announcements regarding protel license 
resales (protel-users-resale) or miscellaneous discussion (protel-users-misc).

If you want to talk fragmentation, it would be in the recent founding of 
the Open Forum by Techserv, which was a direct duplicate of 
protel-users-misc. *That's* fragmentation.

Response to protel-users-library has been substantially greater than was 
the response to any other of the accessory lists. In less than one day, 17 
users have joined. There has been, as yet, no discussion on that list; I 
assume that it will start up when the energy being spent here quiets down.

Yahoogroups lists have archives. The only reason that we have an archive 
for the Techserv list is that one was started on yahoogroups. 
(Protel-users-PEDA-archive). On the same day that this archive was opened, 
there was apparently activity attempting to start up an archive on the 
Techserv site. It was not completed.

An archive (minutes) is essential for a working committee. If the minutes 
of a committee are filled with material irrelevant to the purpose of the 
committee, they will be next to useless. That's why we need a separate 
list. It is for order, not for fragmentation. Discussion lists are great 
for asking questions of a broad audience and for discussion. They are *not* 
so great for making decisions.


[EMAIL PROTECTED]
Abdulrahman Lomax
P.O. Box 690
El Verano, CA 95433

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* 

Re: [PEDA] a lib. for everyone

2001-07-24 Thread Coleman, Tim

I think this is a cool idea. I would think Protel would most certainly be
intrested in helping, it adds value to the product. They already have
standards in place and would want us to work to the same standards. That may
be the only drawback to Protel involvement, unless we agree with thier
standards. I would suggest that that might be a good starting point, to look
at the Protel standrad for footprint design and ask the question 'is it good
enough?' 'Can it be improved?'

I would also like to see more done to improve and increase the simulation
libraries. Perhaps the same approach could be made to generate simulation
ready models for a comprehensive simulation library and a method of updating
with new parts as time goes on.
But thats another topic and I guess that less folk would be intrested in it.
(although they might be if it was easier to get simulation ready models for
the parts they want to use)

TC

 Original Message-
 From: Abd ul-Rahman Lomax [EMAIL PROTECTED]  
 Sent: Tuesday, July 24, 2001 7:03 AM
 To:   Protel EDA Forum
 Subject:  Re: [PEDA] a lib. for everyone
 
 At 01:50 PM 7/23/01 -0500, Ted Tontis wrote:
 Would there be any interest in a PCB footprint lib. with all the parts
 you
 would ever need for free. I ask this because I am working on trying to
 get a
 large lib. in Protel. It would have the silk screen, a fence that would
 be
 on the last electrical layer to avoid component placement conflictions,
 assembly art work, pin 1 id. All parts would be in mm I welcome any input
 towards this idea weather it be good or bad.
 
 Of course I would be interested. I would be *more* interested if we were
 to
 have a project to develop library standards *before* we build a bunch of
 parts. We could set up Y.A.M.L. (Yet Another Mailing List) to function as
 a
 committee for that purpose.
 
 Another writer has suggested that it cannot be done.
 
 If it cannot be done, then surely Protel is wasting a lot of time and
 money
 trying to do it
 
 Now, who has more resources, Protel or the Protel user base?
 
 If anyone has any doubts about the answer to this question, I suggest a
 little reflection. All the resources Protel has come from the users (or
 their employers). Necessarily, those users provide only a small fraction
 of
 their resources to Protel. Further, the users are constantly gleaning
 actual experience with footprints from which Protel employees may be
 isolated.
 
 Of course, *organizing* the users is the difficult part. But it would be
 quite worthwhile. If a small percentage of the Protel designers were to
 devote a small percentage of their time to providing parts built to a
 defined standard, we could build a truly comprehensive library.
 
 *And we are already doing the hard part, building the parts!*
 
 Essentially, instead of dozens or hundreds or thousands of us building the
 same part, only one or a few parts need be built. (Variations on parts are
 necessary for different conditions.)
 
 While I would certainly like every part built exactly how I want it, I
 would give up that in order to have *validated* parts built according to a
 standard which I consider reasonable.
 
 What I forsee is a system whereby a user wishes to use a part which is not
 in the library we have created. He builds it, and he submits it. The part
 is posted as unvalidated. Another user, certified by the user group for
 this purpose, might check the part. The status of that part becomes
 checked, not validated. And then users who use the part will report
 their
 experience with it, particularly with actual fab and assembly. Reports
 regarding the part are tallied and ultimately the footprint is
 automatically given validated status because of a multiplicity of
 validations and an absence of complaints. (If there are validations *and*
 complaints, the matter becomes more complicated, I won't go into that
 now.)
 
 The key to this will be making the submission and retrieval and validation
 of parts *easy*, so that no user is unduly burdened. Protel might
 definitely help with that part. Ultimately the reporting mechanisms might
 be built into client, with a tool that sends a footprint on command to a
 configurable address, together with comment text.
 
 
 [EMAIL PROTECTED]
 Abdulrahman Lomax
 P.O. Box 690
 El Verano, CA 95433

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-24 Thread Andrew J Jenkins

On 09:52 AM 7/24/2001 +1000, Thomas wrote:

It cannot be done. 

While I understand the sentiment, boy do I just hate when people say such things...

After all, there are not all THAT many footprints. Components, yes, footprints...I 
don't think so...

aj

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-24 Thread David Cuthbert

Ted Tontis wrote:
 Would there be any interest in a PCB footprint lib. with all 
 the parts you would ever need for free. I ask this because I am
 working on trying to get a large lib. in Protel. It would have the
 silk screen, a fence that would be on the last electrical layer to
 avoid component placement conflictions, assembly art work, pin 1 id.
 All parts would be in mm.

Art Zirger started such a task many years ago (back in the Protel 1.x
days); he's not there anymore, but it looks like the workstation hosting
the server is still up at Caltech. See
http://leonardo.caltech.edu/protel/protel.html

This isn't really suitable for the current versions of Protel, but it
might give you an idea of what's involved.

Some of the issues, off the top of my head:
- Naming conventions.
- Differing process rules (I may want a 10 mil pad in one process, 5 mil
in another).
- Slightly different layouts amongst manufacturers.
- Revision control (not specific to this, but an issue with design
databases in general).
- All the random people-related issues involved in coordinating such a
task.

While I agree that getting a library with *all* of the parts I'll *ever*
need together is impossible, certainly it is possible to get a library
with *most* of the parts I need *today*.  Heck, if I had a footprint
library for the parts carried by DigiKey/Newark/Allied/etc., I'd be
extremely grateful.

For contributors to such a task, there's an issue with intellectual
property.  In my case, for example, I could use such a library at work,
but would have to run all submissions by our legal department.  insert
heavy sigh

Nonetheless, if you're up to such a task, I'll give you whatever support
I can in my spare time.


Dave Cuthbert
[EMAIL PROTECTED]

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-24 Thread Ted Tontis



-Original Message-
From: Abd ul-Rahman Lomax [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, July 24, 2001 1:03 AM
To: Protel EDA Forum
Subject: Re: [PEDA] a lib. for everyone

Mr. Lomax
snip
What I forsee is a system whereby a user wishes to use a part which is not 
in the library we have created. He builds it, and he submits it. The part 
is posted as unvalidated. Another user, certified by the user group for 
this purpose, might check the part. The status of that part becomes 
checked, not validated. And then users who use the part will report their

experience with it, particularly with actual fab and assembly. Reports 
regarding the part are tallied and ultimately the footprint is 
automatically given validated status because of a multiplicity of 
validations and an absence of complaints. (If there are validations *and* 
complaints, the matter becomes more complicated, I won't go into that now.)

The key to this will be making the submission and retrieval and validation 
of parts *easy*, so that no user is unduly burdened. Protel might 
definitely help with that part. Ultimately the reporting mechanisms might 
be built into client, with a tool that sends a footprint on command to a 
configurable address, together with comment text.

Mr. Lomax
You are right on the money this is how the validation works. when a
part is designed it goes in a special file on the web sight as a not
validated part. When a designer or Protel (if where lucky) can validate the
part the name of the person who validated it is put on the part and it is
then moved to the validated folder. I believe there is a lock on the
folders, We as designers can not change parts we post them and someone else
moves them to the validated and not validated folders.

Ted


* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-24 Thread Brad Velander

Andrew et al,
I get the feeling that many people think that there would not be too
many footprints. How many is not too many? My SMT+ footprint reference has
2474 footprints and adds at least 50 - 100 per year. Now a large number of
them would not typically be used by any one user but against a larger number
of users you would start to increase the total number required to rather
large numbers. (i.e. I don't use any BGA, QFP, not many connectors, large
VLSI, large SOIC, DIPS or leaded electrolytics, resistors, capacitors. Some
other users may use all or significant numbers of these.) Sometimes the
number of part variations elude us because we deal with our own little
universe of parts and part requirements.
Additionally there would be the work of documenting such footprints
and deriving a naming structure. Whew, a lot of work people. Not to mention
there will be several dozen opinions about how to construct everything based
upon biases, opinions, experiences and guidelines.

Brad Velander,
Lead PCB Designer,
Norsat International Inc.,
#300 - 4401 Still Creek Dr.,
Burnaby, B.C., V5C 6G9.
Tel. (604) 292-9089 direct
Fax (604) 292-9010
website www.norsat.com


 -Original Message-
 From: Andrew J Jenkins [mailto:[EMAIL PROTECTED]]
 Sent: Monday, July 23, 2001 9:48 PM
 To: Protel EDA Forum
 Subject: Re: [PEDA] a lib. for everyone
 
 
 On 09:52 AM 7/24/2001 +1000, Thomas wrote:
 
 It cannot be done. 
 
 While I understand the sentiment, boy do I just hate when 
 people say such things...
 
 After all, there are not all THAT many footprints. 
 Components, yes, footprints...I don't think so...
 
 aj

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-24 Thread Dennis Saputelli

wow, what ambition!
can you pls explain:
 2 There is a conductive fence on the lowest layer as to avoid part placement
 conflicts

a little more?
lowest layer = bottom layer?
and how does conductive help?

Dennis Saputelli

Ted Tontis wrote:
 
 The standards are all ready set. I do not have control of them, this task
 was put into motion sometime ago. If you want to look at the standards you
 can at www.pcbstandards.com . again I have no control over the set
 standards.
 anyone who thinks this can not be done is sorry to say wrong. We as
 designers make libraries every day. If we where to stick to the standards,
 follow them as they have been written, and everyone puts some time in the
 project the end result will be one large library that everyone will be able
 to uses. Just think of the amount of time you will save if you did not have
 to make a library. How many of us have the same foot prints in different
 designers .ddb, or have no real standard in there company and your library
 is a free for all.
 
 Here is a list of standards.
 
 1 All parts must be in metric.
 2 There is a conductive fence on the lowest layer as to avoid part placement
 conflicts
 3 for through hole parts the center is located on pin 1
 4 for SMT parts center is located in the center of the part
 5 each part has an assembly drawing
 6 parts naming i.e.. Small Outline Packages, 7.8mm Lead Span (Pitch 0.65mm)
 SOP78- Pin Qty
 
 There are more but you get the idea. The end result will benefit everyone in
 the group, I am doing this in my free time and I wanted to make sure that
 people would be interested before I dig in and get my hands dirty. I am
 currently working on translating 8500 parts into this format. Look over the
 standards and give me as much input weather it be positive or negative. I am
 also working on getting a forum started or get on the existing one as the
 project extends it self.
 
 Ted
 
-- 
___
www.integratedcontrolsinc.comIntegrated Controls, Inc.
   tel: 415-647-04802851 21st Street  
  fax: 415-647-3003San Francisco, CA 94110

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-24 Thread Ted Tontis

Dennis,
What I meant to say is that Protel offers up to 32 signal layers,
therefore the fence would be located on the 30th signal layer. since this is
a conductive layer you can set design rules or clearance constraints for
your components. It would solve the silk screen pad violations, help in
design spacing for test points, and rules for rework or placement. When you
are done laying out the board turn that layer off and your fence is gone.
Thus avoiding generating gerber data for that layer. Sorry I should have
explained my self in more detail.

Ted

-Original Message-
From: Dennis Saputelli [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, July 24, 2001 12:01 PM
To: Protel EDA Forum
Subject: Re: [PEDA] a lib. for everyone


wow, what ambition!
can you pls explain:
 2 There is a conductive fence on the lowest layer as to avoid part
placement
 conflicts

a little more?
lowest layer = bottom layer?
and how does conductive help?

Dennis Saputelli

Ted Tontis wrote:
 
 The standards are all ready set. I do not have control of them, this task
 was put into motion sometime ago. If you want to look at the standards you
 can at www.pcbstandards.com . again I have no control over the set
 standards.
 anyone who thinks this can not be done is sorry to say wrong. We as
 designers make libraries every day. If we where to stick to the standards,
 follow them as they have been written, and everyone puts some time in the
 project the end result will be one large library that everyone will be
able
 to uses. Just think of the amount of time you will save if you did not
have
 to make a library. How many of us have the same foot prints in different
 designers .ddb, or have no real standard in there company and your library
 is a free for all.
 
 Here is a list of standards.
 
 1 All parts must be in metric.
 2 There is a conductive fence on the lowest layer as to avoid part
placement
 conflicts
 3 for through hole parts the center is located on pin 1
 4 for SMT parts center is located in the center of the part
 5 each part has an assembly drawing
 6 parts naming i.e.. Small Outline Packages, 7.8mm Lead Span (Pitch
0.65mm)
 SOP78- Pin Qty
 
 There are more but you get the idea. The end result will benefit everyone
in
 the group, I am doing this in my free time and I wanted to make sure that
 people would be interested before I dig in and get my hands dirty. I am
 currently working on translating 8500 parts into this format. Look over
the
 standards and give me as much input weather it be positive or negative. I
am
 also working on getting a forum started or get on the existing one as the
 project extends it self.
 
 Ted
 
-- 
___
www.integratedcontrolsinc.comIntegrated Controls, Inc.
   tel: 415-647-04802851 21st Street  
  fax: 415-647-3003San Francisco, CA 94110

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-24 Thread Andrew J Jenkins

On 08:54 AM 7/24/2001 -0700, Brad Velander wrote:
Andrew et al,
 I get the feeling that many people think that there would not be too
many footprints.

IMO, too many is a relative term. 2500 is nothing compared to the discrete number of 
components available, footprint(s) aside.

  How many is not too many? My SMT+ footprint reference has
2474 footprints and adds at least 50 - 100 per year. ...
 Additionally there would be the work of documenting such footprints
and deriving a naming structure. Whew, a lot of work people. Not to mention
there will be several dozen opinions about how to construct everything based
upon biases, opinions, experiences and guidelines.

I agree, it would/will be a daunting task, to be sure, but not impossible, by any 
means. And, while your ref shows ~2.5K footprints, as you say, many are used by very 
few, but also, many are probably also obsolete. (granted, I can't say for sure, but I 
have my suspicions...) 

As an example, I once created over 100 unique multi-layer Autocad blocks in less than 
a day, most of which included much greater levels of detail than the average Protel 
block. 
Tedious, yes, impossible, no.

However, my main reply point was to the comment It can't be done. It's an irritating 
statement, that I find is most often used by those who are nay-sayers, or those who 
are simply trying to use 1st year psychology to goad a person into action by negative 
reinforcement. 

I simply have an annoying tendency to ramble. g

regards,

aj

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-24 Thread Brad Velander

Andrew,
please don't throw me into the simple nay-sayers, I have my doubts
about whether it could be accomplished amongst a group as diverse as ours,
but it can be done. My concerns would be more for the diverse opinions and
desired features within the library and coming to some consensus on a
standard approach. Would these be IPC standard? With what tolerances
applied? What difference in basic part dimensions constitutes the design of
a footprint variation? Etc..
While the AutoCAD blocks are a good example, I want some coordinate
control in Protel that would match AutoCAD, then boy could we do footprints.
Right now I find the greatest issue with making footprints is simply the
weak dimensional coordination supported in Protel. I would gladly do
footprints in ACAD any day.
Many of us have that tendency to ramble, as long as it is a flow of
thoughts that others can discuss and build on what is the problem with it?

Brad Velander,
Lead PCB Designer,
Norsat International Inc.,
#300 - 4401 Still Creek Dr.,
Burnaby, B.C., V5C 6G9.
Tel. (604) 292-9089 direct
Fax (604) 292-9010
website www.norsat.com


 -Original Message-
 From: Andrew J Jenkins [mailto:[EMAIL PROTECTED]]
 Sent: Tuesday, July 24, 2001 12:10 PM
 To: Protel EDA Forum
 Subject: Re: [PEDA] a lib. for everyone
 
 
 On 08:54 AM 7/24/2001 -0700, Brad Velander wrote:
 Andrew et al,
  I get the feeling that many people think that there 
 would not be too
 many footprints.
 
 IMO, too many is a relative term. 2500 is nothing compared 
 to the discrete number of components available, footprint(s) aside.
 
   How many is not too many? My SMT+ footprint reference has
 2474 footprints and adds at least 50 - 100 per year. ...
  Additionally there would be the work of documenting 
 such footprints
 and deriving a naming structure. Whew, a lot of work people. 
 Not to mention
 there will be several dozen opinions about how to construct 
 everything based
 upon biases, opinions, experiences and guidelines.
 
 I agree, it would/will be a daunting task, to be sure, but 
 not impossible, by any means. And, while your ref shows ~2.5K 
 footprints, as you say, many are used by very few, but also, 
 many are probably also obsolete. (granted, I can't say for 
 sure, but I have my suspicions...) 
 
 As an example, I once created over 100 unique multi-layer 
 Autocad blocks in less than a day, most of which included 
 much greater levels of detail than the average Protel block. 
 Tedious, yes, impossible, no.
 
 However, my main reply point was to the comment It can't be 
 done. It's an irritating statement, that I find is most 
 often used by those who are nay-sayers, or those who are 
 simply trying to use 1st year psychology to goad a person 
 into action by negative reinforcement. 
 
 I simply have an annoying tendency to ramble. g
 
 regards,
 
 aj

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-24 Thread Thomas

I was just trying to inject a little reality into the discussion.
It didn't seem to me that the original poster had any idea of the enormity
of the task he had set himself.

I too will be glad to help out, and will join the discussion group on yahoo.

Tom.

 -Original Message-
 From: Andrew J Jenkins [mailto:[EMAIL PROTECTED]]
 Sent: Wednesday, 25 July 2001 5:10 AM
 To: Protel EDA Forum
 Subject: Re: [PEDA] a lib. for everyone

-snip-

 However, my main reply point was to the comment It can't be 
 done. It's an irritating statement, that I find is most 
 often used by those who are nay-sayers, or those who are 
 simply trying to use 1st year psychology to goad a person 
 into action by negative reinforcement. 
 
 I simply have an annoying tendency to ramble. g
 
 regards,
 
 aj
 

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-24 Thread Clive . Broome



Im assuming that the existing PCB footprint libraries on the Protel website
would be a subset of this super library.
The libraries are:

BGA531 footprints in 9 libraries based on pin pitch
Through Hole  107 footprints in 5 libraries
QFP's 226 footprints in 12 libraries based on pin pitch and
index position (with duplicated footprints)
Chip carrier74 footprints  in 2 libraries
SOP   129 footprints  in 2 libraries

Plus a number of manufacturer specific and general libraries. Taking just the
semiconductor footprints, thats
1067 footprints that will be bundled into the super library that someone will be
remaking. Or will these footprints
 just be copied from the existing Protel libraries? If these seperate libraries
get jumbled into one library, will
 it make using and finding easier? Who will be doing the work on this?



___

Clive Broome
IDT Sydney Design CentrePh: +61 2 9763 3513
8 Bayswater Dr, HomebushFax:+61 2 9763 3409
Sydney,  NSW, 2127  Email:[EMAIL PROTECTED]
Australia

 Australia's Leading Semiconductor Designers
---








Ted Tontis [EMAIL PROTECTED] on 07/24/2001 04:50:18 AM

Please respond to Protel EDA Forum [EMAIL PROTECTED]

To:   Protel Forum (E-mail) [EMAIL PROTECTED]
cc:(bcc: Clive Broome/sdc)

Subject:  [PEDA] a lib. for everyone



Would there be any interest in a PCB footprint lib. with all the parts you
would ever need for free. I ask this because I am working on trying to get a
large lib. in Protel. It would have the silk screen, a fence that would be
on the last electrical layer to avoid component placement conflictions,
assembly art work, pin 1 id. All parts would be in mm I welcome any input
towards this idea weather it be good or bad.

Ted Tontis C.I.D.
Engage Networks
316 N. Milwaukee Street
Suite 214
Milwaukee WI, 53202
PH 414-273-7600 ext. 7607
FX 414-273-7601






* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-24 Thread Brian Guralnick

email glitch...

For validation, on the PCB side, I used to place a 'scanned version' in
a translucent state, of the part over the footprint.  I've been able to find
and correct many footprint problems in advance using this technique.  If
such an ability was built into Protel  a copy of the scan was kept with the
footprint in the library this would help ease proofing of the footprint.

As for part naming, I think that only exact part numbers, down to the
full suffix will do in such a database.  Though you can search for just the
prefixes in the database, we don't need the 'Napster' equivalent happening
here, but it will happen anyways.

I also think that each component should also come with it's own
footprint, with the attached component name to it.  Otherwise, matching
component footprint name could create a cross mess.

I don't think that pin 1 should always be center.  For example, SMD LED
which require placement based on their center of the footprint, or even a
slight offset which may be by design.

We would need a database sharing tool, like sort a like 'Napster'.  When
searching for a particular part, a spec column would list:
1) user's source of part
2) user who entered part, date of entry
3) component part number / type / included /
4) footprint part number / type / included / metric / imperial / pin1
center /

Though, to properly set this up, I have no clue there...


Brian Guralnick



* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-24 Thread Abd ul-Rahman Lomax

At 12:03 PM 7/25/01 +1000, [EMAIL PROTECTED] wrote:
Taking just the
semiconductor footprints, thats
1067 footprints that will be bundled into the super library that someone 
will be
remaking. Or will these footprints
  just be copied from the existing Protel libraries? If these seperate 
 libraries
get jumbled into one library, will
  it make using and finding easier? Who will be doing the work on this?

As I stated, I do prefer that we move this discussion to the new list, 
[EMAIL PROTECTED]

But here is what I would see. Most of us have had the experience of using a 
Protel part that was not, shall we say, what we would like. The holes are 
too small, for example, or something else about the part was not satisfactory.

The Protel parts would be dumped into the library, assuming Protel 
consents, of course, as Protel-supplied. There would still be a review 
process for all these parts. That process might have a low priority since 
there is so much other work to do.

How about this: when we fab a board and we assemble it and everything is 
good with the footprints, we make a project library and send it off to the 
library center. The parts are compared with existing parts; if they match 
an existing part, that part gets a validation mark. This could be pretty 
much automated. When a part has several hundred validation marks, I'd think 
it we could consider it rather thoroughly checked!

The tools to bring this together don't exist, as far as I know. But unless 
we describe them, they are not likely to come into existence. Protel, as I 
mentioned, might well help with the logistics, but the project should be -- 
*must be* -- a user-sponsored and controlled project. The work for putting 
together the library, the *hard* work, we are already doing: we are 
building parts and we are getting feedback on them. Protel could not afford 
to buy this! But Protel *could* make it easier for us to bring the data 
together, perhaps some HTML code, perhaps the labor of a coordinator, it 
will be up to them what they are willing to do. I do know that they are 
friendly to this kind of idea, if the legalities are properly arranged.

I'll also note that Cadence has a full-fledged support site now which is a 
cooperative effort between the users and the corporation.

Because we are isolated and not organized, it looks to us like Protel is 
large and we are small. That is an illusion!


[EMAIL PROTECTED]
Abdulrahman Lomax
P.O. Box 690
El Verano, CA 95433

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-24 Thread John Haddy

Under no circumstances should the Protel libraries be copied! They
are riddled with errors (I'd list them all as unverified). For
instance, any metric pitch component that's been created with the
footprint wizard is likely to be wrong.

The biggest hurdle I forsee is the widely differing requirements
of any single footprint. For example: a footprint for wave solder
is different from that for reflow; the IPC worst-case design
methodology generates overly conservative footprints; a designer
working on high density layouts will want bleeding edge land
pattern designs, not general purpose ones with excessive silkscreen
clearances (for example). So just with these few variations, we're
up to at least four variants of every footprint!

Much as I applaud the sentiment, I can't see a single library ever
providing a single source set of footprints.

I regularly use components that have land pattern designs tailored
to the manufacturing process; e.g. I might create an SC70-6 footprint
which is fine for the current project because I'll be using a
0.004 paste stencil thickness, however the same design may well cause
manufacturing issues if it's used with a 0.006 paste stencil!

We're going to need much more sophisticated design rules before I'd
consider using anybody else's designs (so that I can check, for
example, paste aperture ratios given a specified screen thickness)

Just my $0.02

John Haddy

 -Original Message-
 From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]]
 Sent: Wednesday, 25 July 2001 12:03 PM
 To: Protel EDA Forum
 Subject: Re: [PEDA] a lib. for everyone




 Im assuming that the existing PCB footprint libraries on the
 Protel website
 would be a subset of this super library.
 The libraries are:

 BGA531 footprints in 9 libraries based on pin pitch
 Through Hole  107 footprints in 5 libraries
 QFP's 226 footprints in 12 libraries based on pin
 pitch and
 index position (with duplicated footprints)
 Chip carrier74 footprints  in 2 libraries
 SOP   129 footprints  in 2 libraries

 Plus a number of manufacturer specific and general libraries.
 Taking just the
 semiconductor footprints, thats
 1067 footprints that will be bundled into the super library that
 someone will be
 remaking. Or will these footprints
  just be copied from the existing Protel libraries? If these
 seperate libraries
 get jumbled into one library, will
  it make using and finding easier? Who will be doing the work on this?



 ___

 Clive Broome
 IDT Sydney Design CentrePh: +61 2 9763 3513
 8 Bayswater Dr, HomebushFax:+61 2 9763 3409
 Sydney,  NSW, 2127  Email:[EMAIL PROTECTED]
 Australia

  Australia's Leading Semiconductor Designers
 ---








 Ted Tontis [EMAIL PROTECTED] on 07/24/2001 04:50:18 AM

 Please respond to Protel EDA Forum [EMAIL PROTECTED]

 To:   Protel Forum (E-mail) [EMAIL PROTECTED]
 cc:(bcc: Clive Broome/sdc)

 Subject:  [PEDA] a lib. for everyone



 Would there be any interest in a PCB footprint lib. with all the parts you
 would ever need for free. I ask this because I am working on
 trying to get a
 large lib. in Protel. It would have the silk screen, a fence that would be
 on the last electrical layer to avoid component placement conflictions,
 assembly art work, pin 1 id. All parts would be in mm I welcome any input
 towards this idea weather it be good or bad.

 Ted Tontis C.I.D.
 Engage Networks
 316 N. Milwaukee Street
 Suite 214
 Milwaukee WI, 53202
 PH 414-273-7600 ext. 7607
 FX 414-273-7601







* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-24 Thread Abd ul-Rahman Lomax

At 02:04 PM 7/25/01 +1000, John Haddy wrote:
Under no circumstances should the Protel libraries be copied! They
are riddled with errors (I'd list them all as unverified). For
instance, any metric pitch component that's been created with the
footprint wizard is likely to be wrong.

First of all, I doubt that the Protel footprints were created with the 
footprint wizard! If Protel employees had to use the wizard, it would be, 
shall we say, ... improved.

I'd list the Protel footprints as Protel-supplied. To call them 
unverified would be rude, even if, technically, from the user 
perspective, they are unverified.

It is important that they be in the database so that users can verify them.

The biggest hurdle I forsee is the widely differing requirements
of any single footprint. For example: a footprint for wave solder
is different from that for reflow; the IPC worst-case design
methodology generates overly conservative footprints; a designer
working on high density layouts will want bleeding edge land
pattern designs, not general purpose ones with excessive silkscreen
clearances (for example). So just with these few variations, we're
up to at least four variants of every footprint!

Right. But it is a question of standards. The standard to which a footprint 
is designed would be coded into the name of the footprint.

The library would be indexed to manufacturer part numbers. (Schematic 
libraries likewise). I see no other way to cleanly organize the parts. 
Obviously, many different MFR part numbers will point to the same symbol.

Much as I applaud the sentiment, I can't see a single library ever
providing a single source set of footprints.

For some of us, the user library (actually it will be a library of 
libraries) will be sufficient for most uses. There will always be a need 
for users to make new footprints or to modify existing ones.

[...]

We're going to need much more sophisticated design rules before I'd
consider using anybody else's designs (so that I can check, for
example, paste aperture ratios given a specified screen thickness)

I'd assume that the libraries would not normally contain paste mask 
information.

The libraries will be as good as the rules used to design them, and the 
rules will be as good as we make them.

[EMAIL PROTECTED]
Abdulrahman Lomax
P.O. Box 690
El Verano, CA 95433

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-23 Thread Thomas

All the parts? I don't think you know the enormity of the task you have set
yourself.

Protel has a full time team doing this. All be it poorly.

For example, are you going to do all the footprints for MOVs, gas arrestors,
Tranzorbs, 10A inductors etc...

It cannot be done. 


 -Original Message-
 From: Ted Tontis [mailto:[EMAIL PROTECTED]]
 Sent: Tuesday, 24 July 2001 4:50 AM
 To: Protel Forum (E-mail)
 Subject: [PEDA] a lib. for everyone
 
 
 Would there be any interest in a PCB footprint lib. with all 
 the parts you
 would ever need for free. I ask this because I am working on 
 trying to get a
 large lib. in Protel. It would have the silk screen, a fence 
 that would be
 on the last electrical layer to avoid component placement 
 conflictions,
 assembly art work, pin 1 id. All parts would be in mm I 
 welcome any input
 towards this idea weather it be good or bad.
 
 Ted Tontis C.I.D.
 Engage Networks
 316 N. Milwaukee Street
 Suite 214
 Milwaukee WI, 53202
 PH 414-273-7600 ext. 7607
 FX 414-273-7601
 

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *



Re: [PEDA] a lib. for everyone

2001-07-23 Thread Abd ul-Rahman Lomax

At 01:50 PM 7/23/01 -0500, Ted Tontis wrote:
Would there be any interest in a PCB footprint lib. with all the parts you
would ever need for free. I ask this because I am working on trying to get a
large lib. in Protel. It would have the silk screen, a fence that would be
on the last electrical layer to avoid component placement conflictions,
assembly art work, pin 1 id. All parts would be in mm I welcome any input
towards this idea weather it be good or bad.

Of course I would be interested. I would be *more* interested if we were to 
have a project to develop library standards *before* we build a bunch of 
parts. We could set up Y.A.M.L. (Yet Another Mailing List) to function as a 
committee for that purpose.

Another writer has suggested that it cannot be done.

If it cannot be done, then surely Protel is wasting a lot of time and money 
trying to do it

Now, who has more resources, Protel or the Protel user base?

If anyone has any doubts about the answer to this question, I suggest a 
little reflection. All the resources Protel has come from the users (or 
their employers). Necessarily, those users provide only a small fraction of 
their resources to Protel. Further, the users are constantly gleaning 
actual experience with footprints from which Protel employees may be isolated.

Of course, *organizing* the users is the difficult part. But it would be 
quite worthwhile. If a small percentage of the Protel designers were to 
devote a small percentage of their time to providing parts built to a 
defined standard, we could build a truly comprehensive library.

*And we are already doing the hard part, building the parts!*

Essentially, instead of dozens or hundreds or thousands of us building the 
same part, only one or a few parts need be built. (Variations on parts are 
necessary for different conditions.)

While I would certainly like every part built exactly how I want it, I 
would give up that in order to have *validated* parts built according to a 
standard which I consider reasonable.

What I forsee is a system whereby a user wishes to use a part which is not 
in the library we have created. He builds it, and he submits it. The part 
is posted as unvalidated. Another user, certified by the user group for 
this purpose, might check the part. The status of that part becomes 
checked, not validated. And then users who use the part will report their 
experience with it, particularly with actual fab and assembly. Reports 
regarding the part are tallied and ultimately the footprint is 
automatically given validated status because of a multiplicity of 
validations and an absence of complaints. (If there are validations *and* 
complaints, the matter becomes more complicated, I won't go into that now.)

The key to this will be making the submission and retrieval and validation 
of parts *easy*, so that no user is unduly burdened. Protel might 
definitely help with that part. Ultimately the reporting mechanisms might 
be built into client, with a tool that sends a footprint on command to a 
configurable address, together with comment text.


[EMAIL PROTECTED]
Abdulrahman Lomax
P.O. Box 690
El Verano, CA 95433

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*  - or email -
* mailto:[EMAIL PROTECTED]?body=leave%20proteledaforum
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *