I find that as long as the board is 0.060" or thicker, the automated
board routers in general use will hold to +/-0.003" or better when
measuring across a board (board length, width, etc.). The problem is
what the router was aligned to when they do the job. Aligning to
circuit board traces would require an optical alignment system--forcing
you to use high end vendors. Same for vias. The problem is that you
are probably mounting your board with special plated or un-plated
mounting holes. If they're plated, they are basically vias and done in
a separate process from the board route--so expect at least +/-0.005" of
error from those points to the board edge. But if they are unplated,
the board can be built so that they are routed with the same tool as
used to cut the board out and in the same process--so +/-0.003" or even
better (the largest element of the error is the error in diameter of the
route bit!) is achieveable (again, assuming you're not using thin
laminate). Note that I said the manufacturer "can" do it this way--you
need to work with someone who is smart enough to grasp this tolerancing
issue. Otherwise they could drill you tooling holes in a different
step. 0.050" thick laminate is still probably OK, but it will move a
little more when the route bit starts to wear. The real trouble is
0.030" and thinner, which readily buckles while you're trying to route
it.
Wayne Thayer
>>> [log in to unmask] >>>
Jim,
If your board is cost sensitive and you do not suffer from very tight
tolerance requirements on the housing, I would suggest specifying
+/-0.010".
I've seen local board shops here struggling a bit with the +/-0.005"
(especially when you mention the complex curved outline) and, assuming,
that
they are at least as good as in other parts of the world, that could
mess up
your yield.
And, finally, I can only join the voices that warned you against
punching.
In case someone does provide you with a good explanation why some
"special
punching process" would yield good results - there is reason to consider
(maybe there are things we haven't met yet), but not the standard
punching.
The shearing and tearing of glass fibers during punching leaves you with
very rough edges and damage to the glass/resin interface further into
the
board. Such damaged interface welcomes the capillary diffusion of water
along the glass bundles, resulting in nasty shorts.
Yehuda Weisz
-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of James Verrette
Sent: Wednesday, August 09, 2006 4:54 PM
To: [log in to unmask]
Subject: [TN] Routed edge tolerance
I am trying to figure out what the minimum clearance would need to be
between a PCB and a housing that it fits into. The piece I am not sure
on is the tolerance of the routed edge of the PCB. The PCB is about
.050 thick and the outline is quite complex with curved edges and
cutouts. What is the maximum variation in the routed edge that I can
expect from the typical mid-high volume board fabricator? I have been
told +/-.005 hole to edge and +/-.005 edge to edge. But what is the
yield at this tolerance? It is just for straight line routs? The board
is 4 layers and cost sensitive, so we don't want to over specify the
edge dimension tolerance and drive up cost or limit ourselves to select
few board shops.
I have also seen quotes from Asian sources recommend punching instead of
routing. Where is a good source of layout design guidelines for this
process? Does it yield good results? I have some knowledge of
singualtion punching, but not in the board fabrication process.
Jim Verrette
Electrical Engineer
---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e
To unsubscribe, send a message to [log in to unmask] with following text
in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt or (re-start) delivery of Technet send e-mail to
[log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to
[log in to unmask]: SET Technet Digest
Search the archives of previous posts at:
http://listserv.ipc.org/archives
Please visit IPC web site
http://www.ipc.org/contentpage.asp?Pageid=4.3.16
for additional information, or contact Keach Sasamori at [log in to unmask]
or
847-615-7100 ext.2815
-----------------------------------------------------
__________ NOD32 1.1699 (20060809) Information __________
This message was checked by NOD32 antivirus system.
http://www.eset.com
---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e
To unsubscribe, send a message to [log in to unmask] with following text
in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt or (re-start) delivery of Technet send e-mail to
[log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to
[log in to unmask]: SET Technet Digest
Search the archives of previous posts at:
http://listserv.ipc.org/archives
Please visit IPC web site
http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100
ext.2815
-----------------------------------------------------
---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives
Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100 ext.2815
-----------------------------------------------------
|
