gEDA-user: Unresolved rat lines, zero-ohm resistor, wire bridge
Hello, I am trying to design a single-sided board with SMD components only (no drilling). The toporouter (which is absolutely awesome, btw.) routes almost all rat lines with only several left unresolved. But, what now? I can do several things: 1) Make the PCB and connect suitable places with wire. disadvantage: The PCB cannot be published without further explanation. And, it is not beautiful. 2) Insert dummy components like zero-ohm resistors or jumper wire in schematics with the hope, that some traces can be routed below the components. disadvantage: The schematics looks crazy. Moreover, surprisingly, it does not help. The autorouter sees different circuit and magically designs different traces. Often, number of unresolved rat lines increases. And, it is totally unpredictable where exactly to insert the dummy components and how many of them. 3) Make double sided PCB and the other side realize with wire bridges only. Disadvantage: This is a bad idea as number of vias is much higher that number of unresolved rat lines. 4) Use #1 but do some manual post-processing. disadvantage: At any change in the schematics the manual work must be done again. The best solution (for me) would be #3 if: - the number of vias would be as small as possible - vias should be in pairs so that the wire connects exactly two. Or this: - if some rat lines cannot be solved, make a pair of pads (or pins) for them. Does anyone have an idea? Thank you very much, Jan Martinek ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Unresolved rat lines, zero-ohm resistor, wire bridge
I'm not sure I understand the problem with #1. Can't you take the mostly routed design and back annotate the jumpers so that they are parts in the original schematic? Then you get what you are looking for in #3 which you think is the best approach. Rick At 12:17 PM 10/16/2010, you wrote: Hello, I am trying to design a single-sided board with SMD components only (no drilling). The toporouter (which is absolutely awesome, btw.) routes almost all rat lines with only several left unresolved. But, what now? I can do several things: 1) Make the PCB and connect suitable places with wire. disadvantage: The PCB cannot be published without further explanation. And, it is not beautiful. 2) Insert dummy components like zero-ohm resistors or jumper wire in schematics with the hope, that some traces can be routed below the components. disadvantage: The schematics looks crazy. Moreover, surprisingly, it does not help. The autorouter sees different circuit and magically designs different traces. Often, number of unresolved rat lines increases. And, it is totally unpredictable where exactly to insert the dummy components and how many of them. 3) Make double sided PCB and the other side realize with wire bridges only. Disadvantage: This is a bad idea as number of vias is much higher that number of unresolved rat lines. 4) Use #1 but do some manual post-processing. disadvantage: At any change in the schematics the manual work must be done again. The best solution (for me) would be #3 if: - the number of vias would be as small as possible - vias should be in pairs so that the wire connects exactly two. Or this: - if some rat lines cannot be solved, make a pair of pads (or pins) for them. Does anyone have an idea? Thank you very much, Jan Martinek ___ 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: Unresolved rat lines, zero-ohm resistor, wire bridge
Hi, no, it really does not work. I think you are suggesting #2. The autorouter is unpredictable. If I change anything in the schematics, the autorouter comes with different design, often worse than before with more unresolved rat lines. Adding a jumper in schematics does not result into reducing rat lines. One example: I had six unresolved rat lines. I added six resistors into appropriate places in schematics. And, voila, I ended up with _nine_ unresolved rat lines and almost no traces went underneath the resistors. The autorouter did not find the solution. Jan Martinek On 10/16/2010 06:53 PM, Rick Collins wrote: I'm not sure I understand the problem with #1. Can't you take the mostly routed design and back annotate the jumpers so that they are parts in the original schematic? Then you get what you are looking for in #3 which you think is the best approach. Rick At 12:17 PM 10/16/2010, you wrote: Hello, I am trying to design a single-sided board with SMD components only (no drilling). The toporouter (which is absolutely awesome, btw.) routes almost all rat lines with only several left unresolved. But, what now? I can do several things: 1) Make the PCB and connect suitable places with wire. disadvantage: The PCB cannot be published without further explanation. And, it is not beautiful. 2) Insert dummy components like zero-ohm resistors or jumper wire in schematics with the hope, that some traces can be routed below the components. disadvantage: The schematics looks crazy. Moreover, surprisingly, it does not help. The autorouter sees different circuit and magically designs different traces. Often, number of unresolved rat lines increases. And, it is totally unpredictable where exactly to insert the dummy components and how many of them. 3) Make double sided PCB and the other side realize with wire bridges only. Disadvantage: This is a bad idea as number of vias is much higher that number of unresolved rat lines. 4) Use #1 but do some manual post-processing. disadvantage: At any change in the schematics the manual work must be done again. The best solution (for me) would be #3 if: - the number of vias would be as small as possible - vias should be in pairs so that the wire connects exactly two. Or this: - if some rat lines cannot be solved, make a pair of pads (or pins) for them. Does anyone have an idea? Thank you very much, Jan Martinek ___ 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 ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
Re: gEDA-user: Unresolved rat lines, zero-ohm resistor, wire bridge
No, I am not suggesting #2. You don't want to reroute the design after you add the jumpers. Once you have a routed design, add the jumper pads to the layout so that wires can be added to the bottom of the board to complete the unrouted connections. Then use back annotation to update the schematic and you are done! DO NOT try to auto-route the layout again from the schematic. As you say, this does not work well. You have to accept the fact that if the auto-router does not complete the routing, you have to manually route the remainder of the board. Once you do manual touch-up of any kind, that is no longer a part of the automatic process and will need to be redone if you want to change the design later. In your case, if you want to auto-route the board again, you need to remove the jumpers from the schematic, make the changes to the schematic, rerun the auto-route, do your touch-up again, back-annotate the schematic and be happy. :^) Rick At 01:29 PM 10/16/2010, you wrote: Hi, no, it really does not work. I think you are suggesting #2. The autorouter is unpredictable. If I change anything in the schematics, the autorouter comes with different design, often worse than before with more unresolved rat lines. Adding a jumper in schematics does not result into reducing rat lines. One example: I had six unresolved rat lines. I added six resistors into appropriate places in schematics. And, voila, I ended up with _nine_ unresolved rat lines and almost no traces went underneath the resistors. The autorouter did not find the solution. Jan Martinek On 10/16/2010 06:53 PM, Rick Collins wrote: I'm not sure I understand the problem with #1. Can't you take the mostly routed design and back annotate the jumpers so that they are parts in the original schematic? Then you get what you are looking for in #3 which you think is the best approach. Rick At 12:17 PM 10/16/2010, you wrote: Hello, I am trying to design a single-sided board with SMD components only (no drilling). The toporouter (which is absolutely awesome, btw.) routes almost all rat lines with only several left unresolved. But, what now? I can do several things: 1) Make the PCB and connect suitable places with wire. disadvantage: The PCB cannot be published without further explanation. And, it is not beautiful. 2) Insert dummy components like zero-ohm resistors or jumper wire in schematics with the hope, that some traces can be routed below the components. disadvantage: The schematics looks crazy. Moreover, surprisingly, it does not help. The autorouter sees different circuit and magically designs different traces. Often, number of unresolved rat lines increases. And, it is totally unpredictable where exactly to insert the dummy components and how many of them. 3) Make double sided PCB and the other side realize with wire bridges only. Disadvantage: This is a bad idea as number of vias is much higher that number of unresolved rat lines. 4) Use #1 but do some manual post-processing. disadvantage: At any change in the schematics the manual work must be done again. The best solution (for me) would be #3 if: - the number of vias would be as small as possible - vias should be in pairs so that the wire connects exactly two. Or this: - if some rat lines cannot be solved, make a pair of pads (or pins) for them. Does anyone have an idea? Thank you very much, Jan Martinek ___ 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 ___ 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: Unresolved rat lines, zero-ohm resistor, wire bridge
On Sat, Oct 16, 2010 at 01:49:57PM -0400, Rick Collins wrote: No, I am not suggesting #2. You don't want to reroute the design after you add the jumpers. Once you have a routed design, add the jumper pads to the layout so that wires can be added to the bottom of the board to complete the unrouted connections. Then use back annotation to update the schematic and you are done! DO NOT try to auto-route the layout again from the schematic. As you say, this does not work well. You have to accept the fact that if the auto-router does not complete the routing, you have to manually route the remainder of the board. Once you do manual touch-up of any kind, that is no longer a part of the automatic process and will need to be redone if you want to change the design later. In your case, if you want to auto-route the board again, you need to remove the jumpers from the schematic, make the changes to the schematic, rerun the auto-route, do your touch-up again, back-annotate the schematic and be happy. :^) If you do a diff on the schematic between these steps, it will be easy to script the undo-change-redo sequence if you expect to make a lot of changes. -- Andrew Poelstra Email: asp11 at sfu.ca OR apoelstra at wpsoftware.net Web: http://www.wpsoftware.net/andrew/ ___ geda-user mailing list geda-user@moria.seul.org http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
