X-Authentication-Warning:	delorie.com: mail set sender to geda-user-bounces using -f
X-Recipient:	geda-user AT delorie DOT com
X-Envelope-From:	paubert AT iram DOT es
Date:	Thu, 25 Jun 2015 23:24:25 +0200
From:	"Gabriel Paubert (paubert AT iram DOT es)" <geda-user AT delorie DOT com>
To:	geda-user AT delorie DOT com
Subject:	Re: [geda-user] [RFC][PATCH] PCB: Allow non rounded clearances for
	rectangular/square pins and pads
Message-ID:	<20150625212425.GB9528@visitor2.iram.es>
References:	<20150625163731 DOT GA18117 AT visitor2 DOT iram DOT es>
	<201506251651 DOT t5PGpqmO020374 AT envy DOT delorie DOT com>
MIME-Version:	1.0
In-Reply-To:	<201506251651.t5PGpqmO020374@envy.delorie.com>
User-Agent:	Mutt/1.5.21 (2010-09-15)
X-Spamina-Bogosity:	Unsure
X-Spamina-Spam-Score:	-0.2 (/)
X-Spamina-Spam-Report:	Content analysis details: (-0.2 points)
	pts rule name description
	---- ---------------------- --------------------------------------------------
	0.0 URIBL_BLOCKED ADMINISTRATOR NOTICE: The query to URIBL was blocked.
	See
	http://wiki.apache.org/spamassassin/DnsBlocklists#dnsbl-block
	for more information.
	[URIs: delorie.com]
	-1.0 ALL_TRUSTED Passed through trusted hosts only via SMTP
	0.8 BAYES_50 BODY: Bayes spam probability is 40 to 60%
	[score: 0.4460]
Reply-To:	geda-user AT delorie DOT com
Errors-To:	nobody AT delorie DOT com
X-Mailing-List:	geda-user AT delorie DOT com
X-Unsubscribes-To:	listserv AT delorie DOT com

On Thu, Jun 25, 2015 at 12:51:52PM -0400, DJ Delorie wrote:
>
> Not that I'm discouraging the patch (options are always good) but this
> is why we don't do square clearance by default:
>
> http://www.delorie.com/pcb/notches.html
>
> Just something to keep in mind :-)
>

Yes, I was aware of this. We already had this discussion a long time
ago and I completely disagree with you, since on my boards the case of
rounded corners looking better is the exception rather than the rule
(fewer than 10 pads out of 2000 for the board I mentioned).

The important difference is that you forced rounded rectangles upon
all the users, while this patch gives you the choice.

> IIRC I talked with a pcb fab "insider" at the time and he agreed that
> square clearances made sense for the mask (where X and Y alignments
> were independent) but round clearances were OK for copper (where the
> problem is etching, not X and Y alignment).

My favorite PCB manufacturer disagrees with you, and a trivial geometry
calculation shows that it's not that simple: right now I typically use
75µm soldermask clearance and 150µ copper clearance, and there is a
significant risk of getting the copper plane exposed with solder mask
misalignment, especially if there is some under-etching in the area.
In any case the copper edges are slightly rounded after etching but it's
not a problem.

However, there were at least 2 other reasons for which I wanted this:
- coplanar waveguide: it's quite easy to find DC blocking capacitors
  which match the width of the centerline, but the small slivers that
  appear on the edge of the ground lines are extremely ugly. Until
  now I had to resort to holes, which are a mess to edit when you move
  the line (infrequent I admit since on RF boards, the first thing you
  do is to route the high frequency signals manually, there are typically
  very few but they are critical and the reason of the existence of the
  board in the first place, and they are intimately connected with the
  mechanical design).

- for fine pitch components, when the distance between pads is less
  than twice the clearance, I also find the "wavy" shape of the plane
  edge is really disgusting.


    Regards,
    Gabriel

- Raw text -