Mail*Link(r) SMTP               ASS: X S Solder on SOT-23

Kenny,

Your response of "no response" I find very interesting.

IMO, I would not allow the filling of the strain reliefs in the component
leads for any "serious" Class 3 product.  The following are my
concerns/thoughts:

1)  With solder against the component body, significant heat has been
transfered through the component lead(s) through the lead and into component
body seal.  This could have degraded the adhesive bond between the lead and
plastic body molding compound.

2)  Differences of thermal expansion coefficients between the component and
the printed board interconnection substrate during thermal/power cycling/shock
will strain the (assumed weakened) bond between the component lead and
component body.  Re-occuring cyclic reverse-stress could lead to component
failure.

3)  Mechanical shock/vibration will likewise subject the component lead and
component body to the same type of possible failure mode as in concern/thought
2).

However it may be acceptable, depending on the life-cycle environment (as you
will read, this pun was not intended) for the poroduct.  If the component were
mounted on an assembly that was a medical implant, such as a heart pacemaker,
it would probabibly be acceptable, because if the inside of the ol' human body
is subjected to thermal shock/cycling or mechanical shock/vibration to damage
the assembly/component, I think some other part of the 'ol body will
deteriorate functionality before the SOT-23's soldered connection.

Disclaimer - this is my opinion, and would most probably not endorsed by my
employer.

Ralph Hersey
e-mail:  [log in to unmask]

--------------------------------------
Date: 4/30/96 7:17 AM
From: Kenny Bloomquist

On Friday, April 26th I wrote:

We are discussing excess solder on a SOT-23 for a critical DOD application
per MIL-STD-2000A.  The condition is solder contacting the body of the
component between the PWB and the body, also solder contacts the lead seal.
Per 2000A this "shall not" be done but it is not in Table 1 as requiring
disposition.  My real question is, what is the risk for this condition and
does anyone have any supporting data (good, bad or otherwise)?

I was suprised that no one sent a response.  I know this is a touchy subject
but any information would be greatly appreciated.

Thank you in advance for all responses.

[log in to unmask]


------------------ RFC822 Header Follows ------------------
Received: by quickmail.llnl.gov with SMTP;30 Apr 1996 07:15:30 -0700
Received: from ipc.org by simon.ipc.org via SMTP (940816.SGI.8.6.9/940406.SGI)
	 id IAA11945; Tue, 30 Apr 1996 08:54:36 -0700
Resent-Date: Tue, 30 Apr 1996 08:54:36 -0700
Received: by ipc.org (Smail3.1.28.1 #2)
	id m0uEF6w-0000CSC; Tue, 30 Apr 96 08:04 CDT
Resent-Sender: [log in to unmask]
Old-Return-Path: <[log in to unmask]>
Message-Id: <[log in to unmask]>
X-Sender: [log in to unmask]
X-Mailer: Windows Eudora Version 1.4.4
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Tue, 30 Apr 1996 06:09:41 -0700
To: [log in to unmask]
From: [log in to unmask] (Kenny Bloomquist)
Subject: ASS: X S Solder on SOT-23
Resent-Message-ID: <"VhA0C1.0.lO7.o_WXn"@ipc>
Resent-From: [log in to unmask]
X-Mailing-List: <[log in to unmask]> archive/latest/3726
X-Loop: [log in to unmask]
Precedence: list
Resent-Sender: [log in to unmask]