At 04:01 PM 2/7/2002 -0500, Jenkins, Charlie wrote:
>I have not tried this but select any track and then do a global to select if
>the track width matches 0.  Then only the 0 width stuff will be selected.

The track in this case was zero length, not zero width. Global edit will 
not catch this. Protel (incorrectly, in my opinion, does not display a zero 
length track. This is particularly anomalous because zero-length draws in 
gerber *are* drawn (and the result is as if the aperture were flashed). So 
if you import a gerber file with a zero-length track, that is, a draw with 
the same starting and ending point, it will become a zero-length track in 
Protel that does not display.

[Zero width track, incidentally, *does* display with a single pixel from 
whatever zoom, and it can readily be selected. And it plots, with a zero 
diameter aperture. I consider this correct behavior.]

There should be an option (under Preferences) to automatically delete 
zero-length track when it is created by an edit. Existing zero-length track 
should be respected, display, plotted, DRC'd, etc.

There are various methods to select zero-length tracks. It is not necessary 
to go to the ASCII database and, in fact, it would not be completely 
straightforward to find such track, (It would not be difficult in Excel or 
other spreadsheet, including the Protel spreadsheet, one could calculate 
track length and sort the records on that basis. The Protel spreadsheet, 
unfortunately, does not allow the setting of the selection field, as I 
recall. But you could give the track some length by changing one coordinate 
to a weird value, like 99.999, which could then be the object of a global 
edit, if one was still necessary)

As to the behavior of zero-length track: When moving the endpoint of a 
track, the track disappears as the cursor is moved to the endpoint of the 
track. I find that the track will edit normally, but because it is 
invisible, it may be difficult to place the cursor on the track to select 
and edit it.

The Query Manager cannot be used to select this track because the Query 
Manager does not allow formulas or variables in the selection fields, nor 
is track length one of the allowed variables.

I was able to make the track visible in the following way: I placed a large 
pad on the track, which took on the track's net. Immediately when the pad 
was released, the pad became visible, because primitives of the same net as 
a primitive with the focus display highlighted.

If the pad is "No Net," however, this does not work directly. A new pad of 
a different net, violating against the "No Net" pad, shows a DRC error, but 
the zero-length track does not display the error. The Violations Browse 
gives (in Details) the coordinates of the zero-length track, which could 
then be used by the Query Manager to select it.

There may well be a fast way to find this track and select it for deletion. 
If it were easy to use formulas in the spreadsheet, that might cut it. Any 
clues? (I know I could do it in Excel, but it would be much faster if the 
native spreadsheet could be used.) The problem is that the spreadsheet 
includes units in the coordinates, so a direct subtraction does not work. I 
could not find a *simple* way to convert the coordinate text to an actual 
number. Sure, I could parse the strings, but, hey, if you are going to feed 
us numbers with units, fix the spreadsheet so that it will do math on 
numbers with units!!!

(All is forgiven if there is a simple function I overlooked. Some forms of 
Basic will evaluate strings to numbers by ignoring any text characters 
beyond the numbers. That's all that would be needed. On the other hand, if 
there is a function, no kudos to the documentation people.... (does Protel 
have documentation people?)

Abdulrahman Lomax
Easthampton, Massachusetts USA

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
* To leave this list visit:
* Contact the list manager:
* Forum Guidelines Rules:
* Browse or Search previous postings:
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Reply via email to