TECHNET Archives

1996

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
"Bill Fabry" <[log in to unmask]>
Date:
8 Jan 1996 11:21:24 -0800
Content-Type:
text/plain
Parts/Attachments:
text/plain (79 lines)
        Reply to:   RE>Problems on PCMCIA Assembly
Bob:

PCMCIA edge connectors are double-sided, SMT straddle-mount type, typically
installed in-line after solder paste applications (prior to pick-place) or as a
2nd operation after assembly of PCA.  They are equipped with alignment "bosses"
that self-align the connector to notches in the PCB, making vision alignment
unnecessary.  Standard SMT design rules should apply for the pad design and
workmanship.  The major concern, from an assembly aspect, should be alignment
of the connector's lateral center-line and the edge of the PCB.

As with any straddle mount connector, its degree of "sag" (angular misalignment
of centerlines) is directly related to the contact shape from the connector
housing to the contact area as it grips the PCB.  If the solder "tail" of the
contact exits the plastic housing from the "outside" of the contact (away from
the PCB) and the connector is NOT mechanically supported during reflow, the
front of the connector can be allowed to droop downward during reflow, causing
detrimental misalignment.  The PCMCIA spec is not very forgiving !   However,
if the contact exits from the "inside" of the contact and the tail is long
enough, it CAN "self-align"  laterally as well.

There is a hot-bar installation machine that is used by a lot of the PCMCIA
memory card manufacturers (perhaps some of the other "Net Surfers" can fill in
the manufacturer's name for me).  The board is manually placed in the machine
VERTICALLY and the machine inserts the board into the fixtured connector and
uses hot-bar reflow to mass solder the contacts (30-45 sec cycle time).  The
same machine can be used to remove the connector just by heating the connector
first and then extracting the PCA.  Presumably, IBM and INTEL are users of
their add-on memory cards.

Hope this helps!

Bill Fabry
[log in to unmask]
--------------------------------------
Date: 1/8/96 6:55 AM
To: Bill Fabry
From: Robert Willis

I am currently evaluating connectors for PC card application as well as looking
at general problems during assembly of thin cards. I would appreciate any
information that engineers may be able to provide to assist my project.
My concerns are placement and vision alignment of connectors, any special
design
rules, connector packaging, reflow support, connector rework etc.

Bob Willis
SMART Group
2 Fourth Avenue, Chelmsford, Essex CM1 4HA. England.
Tel: (44) 01245 351502
Fax: (44) 01245 496123
Home Page: http:// Still Not Available


------------------ RFC822 Header Follows ------------------
Received: by rainbow.rasterops.com with SMTP;8 Jan 1996 06:52:15 -0800
Received: from ipc.org by simon.ipc.org via SMTP (940816.SGI.8.6.9/940406.SGI)
	 id BAA19912; Mon, 8 Jan 1996 01:00:50 -0800
Resent-Date: Mon, 8 Jan 1996 01:00:50 -0800
Received: by ipc.org (Smail3.1.28.1 #2)
	id m0tZCkS-0000OAC; Mon, 8 Jan 96 02:16 CST
Old-Return-Path: <[log in to unmask]>
Date: 08 Jan 96 03:18:35 EST
From: Robert Willis <[log in to unmask]>
To: IPC <[log in to unmask]>
Subject: Problems on PCMCIA Assembly
Message-ID: <[log in to unmask]>
Resent-Message-ID: <"A3ZpS1.0.bTA.0BDym"@ipc>
Resent-From: [log in to unmask]
X-Mailing-List: <[log in to unmask]> archive/latest/2168
X-Loop: [log in to unmask]
Precedence: list
Resent-Sender: [log in to unmask]






ATOM RSS1 RSS2