Re: gEDA-user: [pcb] drill helper
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi Good idea, what would be the much smaller drill hole? If it's too small, it can disappear with toner transfer, if it's too big it doesn't center the drill. I would suggest ~0.4 mm. I suggest using a formula like this: When using the drill helper, make drills holes 1/3 or 1/4 of their size, but no smaller than 30mil. I can easily drill 30mil holes without any special helpers and 30mil is large enough for any etching technique. On the other hand I do have problems centering drills for larger mounting holes. A 30mil hole in the copper is too small to have any mechanical effect on a large drill bit, so it would need to be proportionally larger for larger holes. Best regards Tomaz -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.6 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFGSZ1qsAlAlRhL9q8RAmN7AKC0TAo+uqybwmvMW+6TbeEMpnZe8QCgqflF kDWXbGw+rqriEhSvDdISLZQ= =4JMf -END PGP SIGNATURE- ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: [pcb] drill helper
Nice! I also proposed this some time ago. There should be three options for PS output: 1) full size drills to see how the board will look like 2) reduced holes as template for manual drilling 3) no holes at all for automatic drilling (usable if there are problems with gerber output, there may some problem with exact scaling but may be the only way if no gerber - postscript converter is available). The best size for 2 should be about 15 mils, but can be made configurable (just as default line width or via size). Current dril helpers are not good since they are dots and will push a drill off the proper place. Wojciech Kazubski Having made a number of boards at home now, I'm thinking the drill helper could be better. My local copy does this instead: If you check drill helper, all pins are drawn with a much smaller drill hole. I.e. it doesn't draw the normal-sized hole at all, so if the drill is off-center, you won't have a gap between the hole and the copper. Instead, it only etches a small amount of copper at the center, to help you center the drill. Comments? ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: [pcb] drill helper
On 5/15/07, DJ Delorie [EMAIL PROTECTED] wrote: Having made a number of boards at home now, I'm thinking the drill helper could be better. My local copy does this instead: If you check drill helper, all pins are drawn with a much smaller drill hole. I.e. it doesn't draw the normal-sized hole at all, so if the drill is off-center, you won't have a gap between the hole and the copper. Instead, it only etches a small amount of copper at the center, to help you center the drill. Following is my experience from many PCBs: With a Proxxon mini-bench drill and standard 0.8 mm drill, guiding holes are a must as the drill is so soft that it bends. The guiding hole should be smaller than the drill to work as a funnel for the drill. I use 0.5 mm holes. With a Proxxon you can litteraly see how the drill bends to enter the guide, and you can feel how the PCB is moved by the tension for the drill to straighten out. (We are talking about very small measures of movement and tension here) For a LABO drill or a CNC no guiding holes should be used as these drills are so hard that they will break if they hit the edge of the hole. They normally don't need guides anyway. With some experience, drilling with a Proxxon mini-bench is faster than using a specialized LABO for holes around 0.8 mm as a normal drill is more forgiving than the LABO one. And you also don't need to hit the hole straight as the drill pulls the board in. You can also re-drill a hole which is impossible with a LABO. -- Svenn ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: trace length?
Ryan - On Tue, May 15, 2007 at 09:22:51AM -0400, Ryan Seal wrote: Is there a method in PCB that allows one to measure the trace length so that signals can be phase matched if needed? If not, this would be a nice feature. No, but I would be interested to help out, either coding or testing. I haven't yet gotten around to it by myself. - Larry ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: [pcb] drill helper
DJ Delorie wrote: My local copy does this instead: If you check drill helper, all pins are drawn with a much smaller drill hole. Comments? Yes. Great idea. I've been effectively doing that by my hole specs for prototypes not matching actual desired finished sizes. When you make such unreal entries, the data is less useful for simulations though -- you could not make an accurate model of pads and vias for analyzing capacitance or transmission lines. Having this optional and selectable without changing the correct design data will be a benefit. After my hand drilling experience, I feel guide holes of 1/4 the drill diameter are beneficial down to .4mm (.016 inch). 1/4 is better than 1/3 dia since the shape of a sharpened drill has the flat tip that is maybe 1/6 the diameter on the high speed steel twist drills I have seen. We have some suggestions that one ratio to drill diameter may not satisfy all, so I would like to specify everything about a certain hole size in my design as a line in the resource (Pcb) file for defaults, and as a similar overriding line at the top of the .pcb output file so PCB's output is as general as possible -- useful for manufacturing and simulating. This entry would give corresponding drill size and drill helper size for desired finished size, and override any default or formulaic sizes. Wouldn't it be nice if one of the output drawings showed standard holes and vias in a view that is just lines of vias/holes connected by traces. Right when you start a design before laying out any thing you could set your process etch undercut amount, then output the standard holes and vias drawing to gerber, view it with gerbv to check that all your desired standard holes were enough to get the design done with particular connectors and mechanical details. John Griessen ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: OT Transformer shop
Anybody know of a transformer shop in NA? Richard Sumner sumr.com He does production and custom transformers, coils and chokes. Reputed to be a good person to deal with. Phil Taylor ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
gEDA-user: CVS ebuilds for gentoo
Hi, I want to switch to the cvs version(s) of gschem, gattrib and pcb. Has anybody here got ready to use ebuilds for the named tools (or for the whole suite)? Greetings, Stefan signature.asc Description: OpenPGP digital signature ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: [pcb] drill helper
Good idea, what would be the much smaller drill hole? Currently, I use twice the minimum drill size (i.e. 8 mil diameter) unless that's bigger than the actual drill. If it's too small, it can disappear with toner transfer, 8 mil is about 5 dots wide on a 600dpi laser printer, and you don't need a perfect etch - just enough hole to get something etched. Thus, 8 mil is plenty for toner transfer. ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: trace length?
Is there a method in PCB that allows one to measure the trace length so that signals can be phase matched if needed? If not, this would be a nice feature. Put the cursor over the trace and type :Report(NetLength) It tells you which net it is and the length in the message log. Note that if the trace is not just a single path (i.e. it branches), you get the sum of all segments on all branches. Also, polygons aren't supported. It's not bound to any hotkeys, but of course you can bind it to something with the lesstif hid's pcb-menu.res. ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
gEDA-user: et - symbol translate not working?
On a very recent (May 7th*) CVS version of gschem, doing et 0 results in most of the symbol going off the navigable workspace (negative coordinates?). Does anyone else have this problem? Thanks, Cory Cross *about. How do I find out what my current checked-out version is? ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: et - symbol translate not working?
Does anyone else have this problem? I've seen it too. ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: et - symbol translate not working?
On Tue, 2007-05-15 at 13:58 -0400, DJ Delorie wrote: Does anyone else have this problem? I've seen it too. I'm about to commit a fix (one liner). Peter ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
gEDA-user: howto : strap use in pcb
Hello, I would know how to put straps on my pcb in order to reduce vias. Is it possible and how ? -- Ludovic SMADJA Le hasard, c'est Dieu qui se promène incognito - Albert Einstein HALTE AUX SPAMS : Cet email est signé. Pourquoi signer un email ?? http://www.cacert.org/help.php?id=2lang=fr_FR#whyEmails Mes coordonnées : Jabber ID : [EMAIL PROTECTED] MSN : [EMAIL PROTECTED] Clé PGP : DD76063A sur pgp.mit.edu begin:vcard fn:Ludovic SMADJA n:SMADJA;Ludovic adr:Elancourt;;France email;internet:[EMAIL PROTECTED] note;quoted-printable:JID (jabber ID) : [EMAIL PROTECTED] MSN : [EMAIL PROTECTED] x-mozilla-html:TRUE url:http://Castor-et-herlie.homelinux.org version:2.1 end:vcard signature.asc Description: OpenPGP digital signature ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: howto : strap use in pcb
On 5/15/07, Ludovic SMADJA [EMAIL PROTECTED] wrote: Hello, I would know how to put straps on my pcb in order to reduce vias. Is it possible and how ? If by a strap you mean a busbar then you could make a footprint that matches the mechanical specifications. (* 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: howto : strap use in pcb
Hi, It has been suggested before that you can designate one copper layer for this purpose, and just use vias instead of a footprint. Wherever you want a strap, just draw a copper line. Print the layer seperately and you have a reference for your wires. Doing it this way will allow the DRC to pass and the rats nest will work too. -- Mike Jarabek FPGA/ASIC Designer, DSP Firmware Designer http://www.sentex.ca/~mjarabek -- -Original Message- From: Ludovic SMADJA [EMAIL PROTECTED] Date: Tue, 15 May 2007 21:25:46 To:gEDA user mailing list geda-user@moria.seul.org Subject: gEDA-user: howto : strap use in pcb Hello, I would know how to put straps on my pcb in order to reduce vias. Is it possible and how ? -- Ludovic SMADJA Le hasard, c'est Dieu qui se prom�ne incognito - Albert Einstein HALTE AUX SPAMS : Cet email est sign�. Pourquoi signer un email ?? http://www.cacert.org/help.php?id=2lang=fr_FR#whyEmails Mes coordonn�es : Jabber ID : [EMAIL PROTECTED] MSN : [EMAIL PROTECTED] Cl� PGP : DD76063A sur pgp.mit.edu begin:vcard fn:Ludovic SMADJA n:SMADJA;Ludovic adr:Elancourt;;France email;internet:[EMAIL PROTECTED] note;quoted-printable:JID (jabber ID) : [EMAIL PROTECTED] MSN : [EMAIL PROTECTED] x-mozilla-html:TRUE url:http://Castor-et-herlie.homelinux.org version:2.1 end:vcard ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: howto : strap use in pcb
Thanks for your answer. It's for me the good solution. regards, Ludovic Mike Jarabek a écrit : Hi, It has been suggested before that you can designate one copper layer for this purpose, and just use vias instead of a footprint. Wherever you want a strap, just draw a copper line. Print the layer seperately and you have a reference for your wires. Doing it this way will allow the DRC to pass and the rats nest will work too. -- Mike Jarabek FPGA/ASIC Designer, DSP Firmware Designer http://www.sentex.ca/~mjarabek -- -Original Message- From: Ludovic SMADJA [EMAIL PROTECTED] Date: Tue, 15 May 2007 21:25:46 To:gEDA user mailing list geda-user@moria.seul.org Subject: gEDA-user: howto : strap use in pcb Hello, I would know how to put straps on my pcb in order to reduce vias. Is it possible and how ? ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user -- Ludovic SMADJA Le hasard, c'est Dieu qui se promène incognito - Albert Einstein HALTE AUX SPAMS : Cet email est signé. Pourquoi signer un email ?? http://www.cacert.org/help.php?id=2lang=fr_FR#whyEmails Mes coordonnées : Jabber ID : [EMAIL PROTECTED] MSN : [EMAIL PROTECTED] Clé PGP : DD76063A sur pgp.mit.edu begin:vcard fn:Ludovic SMADJA n:SMADJA;Ludovic adr:Elancourt;;France email;internet:[EMAIL PROTECTED] note;quoted-printable:JID (jabber ID) : [EMAIL PROTECTED] MSN : [EMAIL PROTECTED] x-mozilla-html:TRUE url:http://Castor-et-herlie.homelinux.org version:2.1 end:vcard signature.asc Description: OpenPGP digital signature ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
gEDA-user: Icarus bug?
I believe I found a bug in the Icarus VPI implementation, but I wanted
to check before filing a bug report.
I have code that works with ModelSim under Linux, but I want to
develop on my primary computer, which is a PowerBook, so I'm trying to
get it working with Icarus under OS X. Here's the troublesome bit:
void register() {
s_cb_data init;
vpiHandle cbHandle;
init.reason = cbStartOfSimulation;
init.cb_rtn = initializeSim;
init.obj = NULL;
init.time = NULL;
init.value = NULL;
init.user_data = NULL;
cbHandle = vpi_register_cb(init);
/* vpi_free_object(cbHandle);*/
}
void (*vlog_startup_routines[ ] ) () = {
register,
0
};
When I run it this way, it seems to work okay, but when I uncomment
the call to vpi_free_object, it causes Icarus to crash with a Bus
Error. The Verilog PLI Handbook says that it's good style to free
the callback handle, unless you need to use it somewhere else.
Does this look like a bug to anyone else?
Benjamin
___
geda-user mailing list
geda-user@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
gEDA-user: Help
I start using gschem part of geda 20050313 version on Xandros os. It constantly when you list expect close the gschem leaving me with the unsaved portion to be reentered again, I know that this is an older version, but this version is the only version available from Xandros network. Is there any body that successfully installed and run a newer version on Xandros professional ver 4? And if yes can you supply me with the direction on how to install? I try to install the latest iso file but it went thru several error that I eventually stop it all the errors were basically the same The message were: ERROR WHILE INSTALLING with the choice of abort the installation or continue the log file was pointing out which program was installing at the time first was make utils_install, second make gsymcheck_install, third make gschem_install at this point I abort the installation. I also try to install the debian version the one marked stable using the help of xandros network it did not install. Thank you in advanced [EMAIL PROTECTED] ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Help
On Tue, 2007-05-15 at 21:09 -0400, Vincent Onelli wrote: I start using gschem part of geda 20050313 version on Xandros os. It constantly when you list expect close the gschem leaving me with the unsaved portion to be reentered again, I know that this is an older version, but this version is the only version available from Xandros network. There have been many bugs fixed since this version. I can't advise about Xandros, or installing a newer version, but you might like to try starting gschem from a terminal with: G_SLICE=always-malloc gschem This is known to work around an incompatibility between old versions of gschem and newer versions of glib (One of the program libraries gschem needs). If at all possible, try and get a newer version of gEDA, as a few other crashing bugs have since been fixed which won't be fixed by the above command line. [snip] I also try to install the debian version the one marked stable using the help of xandros network it did not install. Thank you in advanced As Wikipedia tells me Xandros is Debian based, it might be possible to use pre-build .debs from Debian. (I don't know). It will probably depend on which version of Debian Xandros is based on, and how far it deviated from Debian. What were the error messages when it failed to install, that might be of use to anyone who might assist you. Regards, -- Peter Clifton Electrical Engineering Division, Engineering Department, University of Cambridge, 9, JJ Thomson Avenue, Cambridge CB3 0FA Tel: +44 (0)7729 980173 - (No signal in the lab!) ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Icarus bug?
On Tue, 15 May 2007 21:01:40 -0400
Mike Jarabek [EMAIL PROTECTED] wrote:
Hi,
On Tue, 2007-05-15 at 15:47 -0700, Benjamin Ylvisaker wrote:
I believe I found a bug in the Icarus VPI implementation, but I wanted
to check before filing a bug report.
I have code that works with ModelSim under Linux, but I want to
develop on my primary computer, which is a PowerBook, so I'm trying to
get it working with Icarus under OS X. Here's the troublesome bit:
void register() {
s_cb_data init;
vpiHandle cbHandle;
init.reason = cbStartOfSimulation;
init.cb_rtn = initializeSim;
init.obj = NULL;
init.time = NULL;
init.value = NULL;
init.user_data = NULL;
cbHandle = vpi_register_cb(init);
/* vpi_free_object(cbHandle);*/
}
void (*vlog_startup_routines[ ] ) () = {
register,
0
};
Be careful here, you are passing a pointer to an auto variable into the
vpi_register() function. The lifetime of the variable may need to be
longer than the end of your function. The Verilog standard itself is
pretty loosy-goosy on the subject of who allocates and frees these
things.
I'd recommend declaring the 'init' as:
static s_cb_data init;
instead. Also, anything that the structure refers to should also be
declared static, or malloc()'ed. In fact, all of the examples in the
Verilog standard follow this 'static' convention.
This sounds like good advice in general, though my very limited
testing suggests that it works either way (static or stack allocated).
When I run it this way, it seems to work okay, but when I uncomment
the call to vpi_free_object, it causes Icarus to crash with a Bus
Error. The Verilog PLI Handbook says that it's good style to free
the callback handle, unless you need to use it somewhere else.
It might not be a good idea to free the callback object yet, since the
callback has not happened. The example in the Verilog standard does not
free the object either.
Normally you use the vpi_free*() functions to free things that are
returned to you from the simulator kernel.
Section 6.5 of the 2nd edition of the Verilog PLI Handbook states in
no uncertain terms that the callback handle should be freed after
registering a callback to avoid space leaks. However(!), I downloaded
the PLI Book code from the website and found that all the free calls
after callback registrations were removed. It seems that the
confusion is about whether the free call should free the callback data
structure itself or just the handle to it. I'm not worried about
leaking one or two handles in the whole application, so I'll err on
the safe but possibly leaky side for now.
Does this look like a bug to anyone else?
In some ways, yes. But it looks like a classic auto variable that gets
a pointer dereference that leads to a bus error, or worse, random stack
trampling. Those bugs are the most fun to find.
The new PLI Book code still doesn't use static allocation for the
callback struct, and it seems to work in both ModelSim and Icarus. I
don't want to get into too much of a habit of declaring things static
when they don't need to be. Does anyone know if Icarus and/or any
other simulators make a copy of the callback data structure, so that
it doesn't need to hang around after the registration call?
Benjamin
___
geda-user mailing list
geda-user@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
