gEDA-user: Din connector footprints and symbols
Hi all. Over a year ago I used gSchem/PCB to produce a board design for which footprints/symbols were not part of the supplied sets. After much head scratching and re-reading the documentation I produced the files attached here. Unfortunately I chose to produce m4 versions which I then built into my version of gSchem/PCB. They seem to work as far as I can tell and I offer them as-is under the GPL. Feel free to modify them in any way you deem fit. I am not too sure how to contribute these but hope this is an acceptable method! I am not a skilled circuit designer but I like to follow the discussions on the list. Many thanks to all those who contributed code to this project over the years. Colin Ager Garboldisham Norfolk UK PS. I originally sent this last June but I didn't see it in my received mail so it may have gone astray # -*-m4-*- # # For 32-Pin Type D Type Connectors # Colin Ager 10/01/05 # Derived from work by Volker Bosch # # Note that all odd no pins are omitted in this connector # # $1: canonical name # $2: name on PCB # $3: value # $4: requested rows a,c # define(`PKG_DIN41_612FEMALE_D_TYPE', `define(`MAXX', 420) define(`MAXX1', `eval(MAXX -170)') define(`MAXX2', `eval(MAXX -40)') Element(0x00 $1 $2 $3 50 100 3 200 0x00) ( # Row a ifelse(index(`$4', `a'), `-1', , forloop(`i', 1, 16, `PIN(200, eval(300 + 200 *i), 60, 30, eval(i*2)a) ')) # Row c ifelse(index(`$4', `c'), `-1', , `define(`MAXX', 520)' forloop(`i', 1, 16, `PIN(400, eval(300 + 200 *i), 60, 30, eval(i*2)c) ')) # Fixing Holes Pin(290 180 120 80 M1 0x01) Pin(290 3720 120 80 M2 0x01) # Outline of outer part of moulding ElementLine( 80 80 MAXX 80 20) ElementLine(MAXX 80 MAXX 3820 20) ElementLine(MAXX 3820 80 3820 20) ElementLine( 80 3820 80 80 20) # Outline of female part of moulding ElementLine(120 320 MAXX1 320 10) ElementLine(MAXX1 320 MAXX1 360 10) ElementLine(MAXX1 360 MAXX2 360 10) ElementLine(MAXX2 360 MAXX2 3540 10) ElementLine(MAXX2 3540 MAXX1 3540 10) ElementLine(MAXX1 3540 MAXX1 3580 10) ElementLine(MAXX1 3580 120 3580 10) ElementLine(120 3580 120 320 10) # Mark: Pin 2a Mark(200 500) )') # -*- m4 -*- # # For 32-Pin Type D Type Connectors # Colin Ager 10/01/05 # Derived from work by Volker Bosch # # Note that all odd no pins are omitted in this connector # # $1: canonical name# Derived for 32-Pin Type D Connectors from the same series # Colin Ager 10/01/05 # $2: name on PCB # $3: value # $4: requested rows a,c # define(`PKG_DIN41_612MALE_D_TYPE', `define(`XPOS', `eval(300 + 200 * i)') define(`MINY', 300) Element(0x00 $1 $2 $3 520 550 0 200 0x00) ( # Row a ifelse(index(`$4', `a'), `-1', , forloop(`i', 1, 16, `PIN(eval(3600 - 200 *i), 300, 60, 30,eval(i*2)a) ')) # Row c ifelse(index(`$4', `c'), `-1', , `define(`MINY', 100)' forloop(`i', 1, 16, `PIN(eval(3600 - 200 *i), 100, 60, 30, eval(i*2)c) ')) # Reverse side contact pin forloop(`i', 1, 16, `ElementLine(eval(XPOS-100) MINY eval(XPOS-100) 375 40) ') # Fixing Holes Pin( 200 400 120 80 M1 0x01) Pin(3700 400 120 80 M2 0x01) # Boundary of M1 ElementLine( 100 300 320 300 20) ElementLine( 320 300 320 395 20) ElementLine( 320 395 320 620 10) ElementLine( 320 620 200 620 10) ElementLine( 200 620 100 620 20) ElementLine( 100 620 100 300 20) # Boundary of M2 ElementLine(3580 300 3800 300 20) ElementLine(3800 300 3800 620 20) ElementLine(3800 620 3700 620 20) ElementLine(3700 620 3580 620 10) ElementLine(3580 620 3580 395 10) ElementLine(3580 395 3580 300 20) # Pin edge of body ElementLine( 320 395 3580 395 20) # Pin area Boundary ElementLine( 200 620 200 800 20) ElementLine( 200 800 3700 800 20) ElementLine(3700 800 3700 620 20) #Mark Pin 2a Mark(3400 300) )') v 20040111 1 P 400 12500 100 12500 1 0 1 { T 400 12500 5 10 0 1 0 0 1 pinnumber=1 T 400 12500 5 10 0 1 0 0 1 pinseq=1 T 500 12500 5 10 1 1 0 0 1 pinlabel=2a } P 400 11700 100 11700 1 0 1 { T 400 11700 5 10 0 1 0 0 1 pinnumber=2 T 400 11700 5 10 0 1 0 0 1 pinseq=2 T 500 11700 5 10 1 1 0 0 1 pinlabel=4a } P 400 10900 100 10900 1 0 1 { T 400 10900 5 10 0 1 0 0 1 pinnumber=3 T 400 10900 5 10 0 1 0 0 1 pinseq=3 T 500 10900 5 10 1 1 0 0 1 pinlabel=6a } P 400 10100 100 10100 1 0 1 { T 400 10100 5 10 0 1 0 0 1 pinnumber=4 T 400 10100 5 10 0 1 0 0 1 pinseq=4 T 500 10100 5 10 1 1 0 0 1 pinlabel=8a } P 400 9300 100 9300 1 0 1 { T 400 9300 5 10 0 1 0 0 1 pinnumber=5 T 100 9400 5 10 0 1 0 0 1 pinseq=5
Re: gEDA-user: licensing (GPL or otherwise) for hardware?
Karel == Karel Kulhavy [EMAIL PROTECTED] writes: On Thu, Dec 14, 2006 at 09:17:33AM +0100, David Kuehling wrote: For PCB that might be the same: if you distribute gerber files, you distribute the *output* of PCB, which obviously doesn't contain literal code from the footprints. If you distribute a .pcb-file, that is What about derived work? Isn't a Gerber derived work from the PCB? If someone paints a painting on canvas and I make a digital shot, it will be a different object (painting on canvas vs. data in a memory card), but it represents the same work and is covered by the copyright too. I must admit that you seem to be right. Including footprints and creating a gerber file is quite similar to statically linking a program with a library. Maybe I got too lost in the PostSript analogy. The problem with M4-symbols is, that the output directly contains input code. It's a preprocessor after all. For Postscript typesetting prologues that's not the case. Postscript-snipsets can be regarded as programs, that when run create some unrelated output (such as a rasterized image or whatever). But for highly parametrized M4-symbols I will still claim, that the output after preprocessing is _output_ and not a derived work. Think of the M4-Macro as a program with some printf(... %s...) in it. You wouldn't normally consider program output generated this way a derived work? Maybe it's the amount of printf's compared to the amount of non-printing control code that makes the difference here ? Just take tragesym, Latex, dvi2ps, Lout, psnup, a2ps etc. All those programs create output via lots of printf() and thesse format strings get partially into the output. Still nobody would consider the output a derived work of the program. regards, David -- GnuPG public key: http://user.cs.tu-berlin.de/~dvdkhlng/dk.gpg Fingerprint: B17A DC95 D293 657B 4205 D016 7DEF 5323 C174 7D40 ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Re: licensing (GPL or otherwise) for hardware?
Stephen == Stephen Williams [EMAIL PROTECTED] writes: I believe deceiving for a material profit is a criminal act, but deceiving for a public benefit is not illegal. Well, where I live (USA) theft is theft, no matter what the motive. If I rob a bank and give all the money to a church or a charity, I'm still going to jail. Copyright infringement is not related to theft in any way as far as I understand US law. Sorry for the trolling, I just couldn't resist :) ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
gEDA-user: Footprint with Unusual Pads
Hi, Looking at the PCB docs it seems like SMD pads must either be round or rectangular, never both? I'm just looking at the recommended land pattern in this data sheet (PDF, page before last page): http://www.ti.com/lit/gpn/bq24070 They'r drawing longish pads which are round on one end and rectangular at the other end. Also the die-attach pad has a very unusual complex polygonal form. What's the right way to define such a footprint (preferably oldlib-style)? Add some additional copper polygons around the pad?? Or overlap multiple pads with the same name/number? Or just make all pads rectangular and hope that reflow soldering still works? Being very new to PCB layouting I have no clue :) Thanks for any help, best regards, David -- GnuPG public key: http://user.cs.tu-berlin.de/~dvdkhlng/dk.gpg Fingerprint: B17A DC95 D293 657B 4205 D016 7DEF 5323 C174 7D40 ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Footprint with Unusual Pads
On 15 Dec 2006 13:39:04 +0100, David Kuehling [EMAIL PROTECTED] wrote: Hi, Looking at the PCB docs it seems like SMD pads must either be round or rectangular, never both? I'm just looking at the recommended land pattern in this data sheet (PDF, page before last page): http://www.ti.com/lit/gpn/bq24070 They'r drawing longish pads which are round on one end and rectangular at the other end. Also the die-attach pad has a very unusual complex polygonal form. What's the right way to define such a footprint (preferably oldlib-style)? Add some additional copper polygons around the pad?? Or overlap multiple pads with the same name/number? Or just make all pads rectangular and hope that reflow soldering still works? Multiple pads (with the same the pin number) should work. Each pin would take two pads, the die-attach would take four pads. (* jcl *) -- http://www.luciani.org ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Symbol submission
On 12/14/06, Bob Paddock [EMAIL PROTECTED] wrote: If you use IPC-style names the alphanumeric sort works fairly well. If you add a mfg/mfg_pn suffix you can get a better sort. What happens when the company is bought by an other company? Happens all to often. You write a Perl script to update the names ;-) (* jcl *) -- http://www.luciani.org ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Footprint with Unusual Pads
DJ == DJ Delorie [EMAIL PROTECTED] writes: They'r drawing longish pads which are round on one end and rectangular at the other end. Also the die-attach pad has a very unusual complex polygonal form. Both can be made from multiple pads. See (also attached): http://www.gedasymbols.org/user/dj_delorie/footprints/test/bq24070.fp Hi, thanks alot. What's the right way to define such a footprint (preferably oldlib-style)? The m4 libraries are deprecated. Please create new symbols in newlib style. Also, you can create footprints inside pcb using lines and rectangles, which is easier than figuring out the m4 stuff. According to my copy of the PCB documentation oldlib is not concidered deprecated, despite the naming. I actually already started writing all my symbols in M4. I've worked with M4 in the past and being a programmer I just don't know how to use a mouse :) regards, David -- GnuPG public key: http://user.cs.tu-berlin.de/~dvdkhlng/dk.gpg Fingerprint: B17A DC95 D293 657B 4205 D016 7DEF 5323 C174 7D40 ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Footprint with Unusual Pads
According to my copy of the PCB documentation oldlib is not concidered deprecated, despite the naming. I actually already started writing all my symbols in M4. I've worked with M4 in the past and being a programmer I just don't know how to use a mouse :) Well, up to you then ;-) What we're planning on doing is running the M4 processor at build time, rather than at run time. Thus, pcb itself only need worry about newlib style footprints. The explicit use of m4 in pcb is a hack; different people prefer different languages (I use perl myself) but the common format is newlib; the Makefiles can run any program they need to to generate them. Plus, m4 isn't readily available on Windows platforms, so moving to newlib makes Windows support easier. ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Footprint with Unusual Pads
What we're planning on doing is running the M4 processor at build time, rather than at run time. U, why not have the develoeprs run it in maintiner-mode, and just distribute the newlib stuff to regular users? I tremble in fear about running the M4 thing as part of a normal user's build since it seems like another thing which can go wrong. Plus, m4 isn't readily available on Windows platforms, so moving to newlib makes Windows support easier. Another reason to have the developers run the M4 processor, and not the users. Stuart ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Footprint with Unusual Pads
U, why not have the develoeprs run it in maintiner-mode, and just distribute the newlib stuff to regular users? We could do that too. ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Symbol submission
On Dec 15, 2006, at 2:21 AM, Karel Kulhavy wrote: If I won't be able to submit through the old mechanism I won't probably submit anymore. gedasymbols.org are just too complicated for me. Subscription, password management, directories, CVS - takes too much time. gEDA should dump cvs and switch to subversion ... ;) -a ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Footprint with Unusual Pads
John Griessen wrote: Stuart Brorson wrote: Plus, m4 isn't readily available on Windows platforms, so moving to newlib makes Windows support easier. Another reason to have the developers run the M4 processor, and not the users. Since it's GPL, we have to delete the use of M4 entirely to delete it in a user's source build process. That is simply not true. The suggestion was to only run m4 when you configure with --enable-maintainer-mode. So clearly all the required sources are there and available. Here is another example, the user build process does not use autoconf but rather the output (configure) but we do provide the source (configure.ac). My thoughts were either the user build always runs the m4 stuff (not unreasonable, if you don't have m4, then chances are you don't have the other tools like /bin/sh and make and a c-compiler) or to include the m4 output so that the build rules don't get triggered unless the user touches the m4 input files. I like this much better than the maintainer mode stuff anyway. We get way too many complaints as it is from users who build from cvs but who don't read README.cvs to see you should disable building the docs or build in maintainer mode. I don't want to add to that mess. -Dan ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Footprint with Unusual Pads
Since it's GPL, we have to delete the use of M4 entirely to delete it in a user's source build process. IANAL but one could consider that M4 is a standard part of the operating system that the user just doesn't choose to install most of the time. The operating system in this case being cygwin, which has m4 (when installed) but most *users* of geda won't have installed it, just like they probably won't have installed gcc, which is also needed for the build, if then use pre-built geda binaries. This is completely different than something the users *can't* install, which would violate the GPL. The user always has the choice of not installing some tool if they don't want it and don't need it. This is the same issue as needing gcc, make, ld, etc. I asked RMS about this a long time ago, when I was developing DJGPP, and they modified the GPL to include the standard development tools in the os exceptions clause (DJGPP wasn't part of MS-DOS but you can't expect people using it to have to give DJGPP to every user of, say, Quake 1) ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Symbol submission
gEDA should dump cvs and switch to subversion ... Note that gEDA, PCB, and gedasymbols.org each have their own - independent - repositories. The PCB admins have at least talked about the difference between cvs and svn. There are ups and downs to either choice. Svn, for example, doesn't have the concept of modules - you either get the whole tree, or none of it. OTOH, cvs won't let you check out a branch as of a specific date. Svn also needs twice the local disk space. ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Symbol submission
On Fri, 15 Dec 2006 11:31:39 -0700 Andy Peters [EMAIL PROTECTED] wrote: -Cut- gEDA should dump cvs and switch to subversion ... ;) -a fan_flames(lares) svn++ -Lares pgp7tmHXF18Z8.pgp Description: PGP signature ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
gEDA-user: fun/easy BGA part
Ok, a s.e.design thread has got me thinking about pan-frying BGAs. Now I'm curious. What's a good BGA to start with? Good is defined as: 1. Inexpensive, cause I'll probably toast a few. 2. Useful, because it's good to see results. 3. Easy to design around, because I want to worry about the soldering process, and not burn through a number of chips getting the design 4. Require a minimum of support components. 5. Not so complex that lots of tiny vias would be needed. If I can get by with two rows pulled out on the top with 7/7 rules, and the remaining rows tied to gnd/vcc through one or two big vias, that's acceptable. Hmmm... BGA challenge anyone? ;-) ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: gschem symbol search order
Hi Patrick: El vie, 15-12-2006 a las 17:46 +0100, Patrick Bernaud escribió: Carlos Nieves Ónega writes: [...] Ack! This is changed behavior. It seems that the order is now different then what I expected. libgeda/gschem should always search the component libraries in the reverse order (that is your local libraries should be found first), however it is not doing that. This is a bug and will be fixed. Thanks for the bug report. Hopefully it's fixed now in HEAD. I am curious on the reason that made you clib_directories from a GSList to a GList. Correct me if I am wrong but this code is correct and does not require a GList. The bug is not in libgeda/s_clib.c but in all the places where s_clib_search_basename() is used: the function calling it must be careful enough to take the last item from the list. The bug is HERE (libgeda and gschem). So a better way to fix this problem would have been to use g_slist_last() on the returned list and not to modify s_clib.c. Here are the reasons: - All the existing code asserts that the first node returned by s_clib_search_basename is the first entry to be used. This makes sense. - It also makes sense to change the search function instead of changing every function calling it. - The above two points suggests to a change in the search function, searching backwards instead of forwards. - I like to use a double-linked list whenever I need to go forward and backwards in a list. - The penalty of using a GList instead of a GSList is only a pointer for each library node. - No user will be using thousands of libraries at the same time so the memory consumption won't be increased so much. - It makes programmer's life easier. Regards, Carlos ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Footprint with Unusual Pads
Perhaps I misunderstand. There's a clause in the GPL that says you have to provide ..., plus the scripts used to control compilation and installation of the executable except anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system. I put m4 in that exception category. ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Footprint with Unusual Pads
Dan McMahill wrote: John Griessen wrote: delete the use of M4 entirely to delete it in a user's source build process. only run m4 when you configure with --enable-maintainer-mode. Oh, I thought a wrong def. of user build. Sure that will work. JG ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: fun/easy BGA part
On 12/15/06, DJ Delorie [EMAIL PROTECTED] wrote: Ok, a s.e.design thread has got me thinking about pan-frying BGAs. Now I'm curious. What's a good BGA to start with? Good is defined as: 1. Inexpensive, cause I'll probably toast a few. 2. Useful, because it's good to see results. 3. Easy to design around, because I want to worry about the soldering process, and not burn through a number of chips getting the design 4. Require a minimum of support components. 5. Not so complex that lots of tiny vias would be needed. If I can get by with two rows pulled out on the top with 7/7 rules, and the remaining rows tied to gnd/vcc through one or two big vias, that's acceptable. You could try some DC-DC chips like the ON-Semi NCP1523, 8 pins and only five or six passives. TI or National may have others. (* jcl *) -- http://www.luciani.org ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Symbol submission
On Dec 15, 2006, at 12:31 PM, DJ Delorie wrote: gEDA should dump cvs and switch to subversion ... Note that gEDA, PCB, and gedasymbols.org each have their own - independent - repositories. The PCB admins have at least talked about the difference between cvs and svn. There are ups and downs to either choice. Svn, for example, doesn't have the concept of modules - you either get the whole tree, or none of it. Actually, that's not true ... but let's leave that for a different thread and a different list. OTOH, cvs won't let you check out a branch as of a specific date. Svn also needs twice the local disk space. svn keeps an untouched local copy of whatever you check out so you can do diffs, reverts, etc without hitting the network. Again, though, off topic, and I suppose I should apologize for bringing it up! -a ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Symbol submission
DJ Delorie wrote: Svn, for example, doesn't have the concept of modules - you either get the whole tree, or none of it. OTOH, cvs won't let you check out a branch as of a specific date. Svn also needs twice the local disk space. I'm wondering how git and svn compare... nirvana-wise I've been using svn some lately to keep my circuit design work in. There seems to be a way to deal with just sub directories of a tree, after you have the whole tree. If you move that subdir to a new place, it still functions as a svn working copy of that much of the tree, and you could then delete the rest with no impact on the repository. I've not noticed any quick way to break a repository into chunks after making it. I think that could be done by taking a section of updated working copy and stripping out all .svn dirs, then import it into a new subdir of the same repository. That would result in a repository with two top dirs, I think. After that tested out OK, the copied original dir could be svn deleted. The history of it and repository filesize would still be there. The new history would start off in the new repository dir. John G ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Symbol submission
Actually, that's not true ... but let's leave that for a different thread and a different list. Ok, but I do want to mention that the gcc and binutils/gdb/newlib/cygwin repositories are pondering this same question, and I have a lot of info from those discussions. It's not a cut-n-dried decision. We've discussed it internally in RH but the lack of a cvs modules equivalent was a blocking factor. ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Symbol submission
On Dec 15, 2006, at 2:20 PM, John Griessen wrote: DJ Delorie wrote: Svn, for example, doesn't have the concept of modules - you either get the whole tree, or none of it. OTOH, cvs won't let you check out a branch as of a specific date. Svn also needs twice the local disk space. I'm wondering how git and svn compare... nirvana-wise I've been using svn some lately to keep my circuit design work in. There seems to be a way to deal with just sub directories of a tree, after you have the whole tree. If you move that subdir to a new place, it still functions as a svn working copy of that much of the tree, and you could then delete the rest with no impact on the repository. You don't have to check out the whole tree! Just check out the part you need. I've not noticed any quick way to break a repository into chunks after making it. I think that could be done by taking a section of updated working copy and stripping out all .svn dirs, then import it into a new subdir of the same repository. That would result in a repository with two top dirs, I think. After that tested out OK, the copied original dir could be svn deleted. The history of it and repository filesize would still be there. The new history would start off in the new repository dir. Dunno exactly what you mean by break a repository into chunks after making it, but if you decide that you don't like where something lives in the repo, it's really easy to move things around (retaining history, too): $ svn move -m Comment about moving url://path/to/source url://path/ to/new/location where url is replaced by your access method (svn:, svn+ssh:, http:, https:) -a ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Symbol submission
There seems to be a way to deal with just sub directories of a tree, after you have the whole tree. If you move that subdir to a new place, it still functions as a svn working copy of that much of the tree, and you could then delete the rest with no impact on the repository. But if you do an update to a specific version (or date) after moving the subdir, it re-creates the subdir. The thing that svn can't do is check out a tree WITHOUT getting specific subdirs. For example, getting pcb without the documentation. This is a big problem for binutils/gdb/newlib/cygwin because they share a repository and a lot of common top-level files. ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: fun/easy BGA part
You could try some DC-DC chips like the ON-Semi NCP1523, 8 pins and only five or six passives. TI or National may have others. Hey, I could actually use one of those in the furnace board to make 3.3v from 5v. Do they make one that can do 5v to 3.8v ? The gumstix has its own LDO regulators so you can't feed it clean 3.3v, and they don't have the dissipation to handle 5v input. OTOH I was thinking of something on the order of a 5x5 or 6x6 grid. A 6x6 can be broken out on one layer, no vias needed. ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: fun/easy BGA part
On 12/15/06, DJ Delorie [EMAIL PROTECTED] wrote: You could try some DC-DC chips like the ON-Semi NCP1523, 8 pins and only five or six passives. TI or National may have others. Hey, I could actually use one of those in the furnace board to make 3.3v from 5v. That's why I mentioned it. Do they make one that can do 5v to 3.8v ? I don't know. Since I do not have a successful ball-grid challenge under my belt I usually don't look at parts in ball-grid packages ;-) (or QFN's!) For buck converters I have been using the LT1616. The LT1933 is lower frequency but has the same pinout. Both are in SOT23-6 packages. (* jcl *) -- http://www.luciani.org ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Clarifying the License issues for gaf and PCB
On 12/15/06, Ostheller, Joel A. [EMAIL PROTECTED] wrote: I am currently doing my first gEDA PCB board and I have to say that I am worried that footprints will be error prone. It is disheartening to hear that someone with more gEDA experience then me, also shares that opinion. If only there was a flag which footprints were verified on an already created gEDA PCB Errors in footprint (and symbol) libraries are an EDA problem not just a gEDA problem. You need to verify that a footprint matches the manufacturer's specification and your process specification. Different manufacturer's sometimes have different specifications for the same package style. After I verify a footprint I move it to a released for production directory. I have gsch2pcb only use footprints from that directory. (* jcl *) -- http://www.luciani.org ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: fun/easy BGA part
If you want to build something you could attach to your gumstix, you could look at some various audio codec's. The AIC33 from Texas Instruments comes to mind -- I think that's available in a 5x5 or 6x6 array. 8x8, but they only use the outer two rows. I've discovered that many TTL latches are available in small BGA, like the venerable 74373. ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Clarifying the License issues for gaf and PCB
After I verify a footprint I move it to a released for production directory. I've started grouping my library into vetted and not vetted, with the vetted ones being the ones that have been fab'd and produce working boards. I suppose I could add more steps, like paper verified. There seem to be two mistakes I've made: wrong footprint, and wrong pinout. The first can be checked with a printout; the second usually waits until the board is made and tested. ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Clarifying the License issues for gaf and PCB
DJ Delorie wrote: After I verify a footprint I move it to a released for production directory. I've started grouping my library into vetted and not vetted, with the vetted ones being the ones that have been fab'd and produce working boards. I suppose I could add more steps, like paper verified. There seem to be two mistakes I've made: wrong footprint, and wrong pinout. The first can be checked with a printout; the second usually waits until the board is made and tested. Over at peeron.com (well known to Lego fans :) they have already solved this problem for Lego set inventories. In a nutshell, they track who submitted a kit inventory and how the inventory was obtained, and also track 2 (maybe more?) reviewers who independently verify the inventory. May I suggest that gedasymbols do something similar? Perhaps tracking: 1. submitter name 2. submitter has fabbed (y/n) 3. Reviewer #1 4. Reviewer #1 method: a) paper verified b) fabbed 5. Reviewer #2 6. Reviewer #2 method: a) paper verified b) fabbed -dave ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user