*******************************************************
Todays forums are sponsored by Ian Martin Limited
Engineering/Technical Placement Specialists
www.ianmartin.com
*******************************************************

I've talked my colleague about what can and can't be done with RS-274X - he 
seems to be a minor expert on the subject. You are quite correct in saying 
that the primitive for the thermal has four spokes and apparaently this 
can't be easily changed, but as you say macros can be devised. 
Unfortunately, the arc method produces quirky looking results when there is 
a large relief and I'm told that to overcome this would require a moderately 
complex "super-macro" for each pad with multiple (overlapping) arcs drawn 
for each segment. Multiple groups of such arcs could then be extended as 
necessary (not just 90 degrees ie 4 arcs) to provide any number of "thermal 
arms" so to speak. Quite a lot for a single ground plane connection!

He seems to believe that he needs to control the thermal reliefs carefully 
for several applications - high frequency RF and power supplies to name but 
two. I spent an hour with him and I think that the following would be a 
decent spec:

The software that he uses gives exactly the control that Mr Lomax is talking 
about and then some. He can manually override any thermal relief with any 
number of thermal arms, at any angle, of any width. On an individual pad 
basis, he can extend the width of the thermal gap. The software seems to use 
the same method of plane generation that Protel uses for regular power 
planes - that is to say a negative image as opposed to polygon fill but most 
impressively, he can put a (polygonal) plane onto _any_ layer not just a 
designated plane layer ie. there seems to be no distinction between regular 
and plane layers. This makes split planes trivial and he seems to use these 
as if they are going out of fashion. Planes are visible on screen and when 
(regular) tracks pass through the plane, they dynamically cut a gap (what 
PADS used to call trace ploughing). He can blowout sections of plane using 
various "cutting" tools and it detects when pads are isolated by the 
blowouts (although not dynamically) - in fact all DRC seems to be batch 
based.

It also fully supports padstacks so he's got octagonal pads on the top and 
bottom and round in the middle. [Somewhat off subject] he can control via 
tenting top and bottom separately on a via by via basis - something that was 
discussed here recently. But the shocker seems to be the price tag (sub US 
$400 per year).

I'm going to open that large bottle of scotch and try not dilute it with too 
many tears.


Doug


>At 11:39 AM 10/4/01 +1000, Geoff Harland wrote:
>>What I was saying previously was that the RS274X standard does not support
>>"flashing" thermal relief patterns having just two openings. Given that
>>situation, the associated arcs (of these patterns) should (normally) be
>>"drawn" instead.
>
>RS-274X has a primitive called a "thermal," yes, and it has four spokes by
>nature. But RS-274X was designed to be *very* flexible, so basically *any*
>shape can be defined as an aperture macro.
>
>When Protel allowed differing numbers of spokes (many programs don't), it
>is rather obvious that no one ever told the people writing the photoplot
>routines. This has been broken, I suspect, for a long time.
>
>Ultimately, I think we should go to defined padstacks, the existing Protel
>pad shapes are one of the major limitations of the program. I would
>definitely use a chamfered rectangle (kind of like an "octagon" should be,
>but not regular and not broken) if Protel supported it. I used the Tango
>"rounded rectangle" all the time. (that's a rectangle with rounded 
>corners).
>
>Yes, padstacks are complicated but if the program automatically generated
>padstacks to match the existing status quo, but then we could lift the hood
>and edit and modify or add to those padstacks, we'd have our cake and full
>stomachs as well, to mix a few metaphors.
>
>So the photoplot routines could simply define a macro where the existing
>basic RS274X shapes won't work, and then flash the macro code. Easy peasy,
>best of both worlds: a drawn shape but compact plot files.
>
>(As an aside, such creatures would make back-conversion from Gerber easier,
>because they would define a connection to the plane just as a thermal
>presently defines such a connection. Drawn thermals are harder to recognise
>automatically.)
>
>[EMAIL PROTECTED]
>Abdulrahman Lomax
>Easthampton, Massachusetts USA
>


_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp

* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* To post a message: mailto:[EMAIL PROTECTED]
*
* To leave this list visit:
* http://www.techservinc.com/protelusers/leave.html
*
* Contact the list manager:
* mailto:[EMAIL PROTECTED]
*
* Forum Guidelines Rules:
* http://www.techservinc.com/protelusers/forumrules.html
*
* Browse or Search previous postings:
* http://www.mail-archive.com/proteledaforum@techservinc.com
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * *

Reply via email to