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:
"Ralph Hersey" <[log in to unmask]>
Date:
21 Mar 1996 08:11:22 -0800
Content-Type:
text/plain
Parts/Attachments:
text/plain (130 lines)
                      Subject:                              Time:  8:20 AM
  OFFICE MEMO         RE>Fine Pitch in Motion               Date:  3/21/96

Leo <[log in to unmask]> 

Early last week, Leo made a post concerning fine pitch, I prepared the
following response and had trouble posting it on the technet, so I sent it
directly to him.  I haven't seen any other response to his post on the technet
postings I've received and I though the response would be of interest to some
of you other technetters.

I've got a feeling the following is more than what you wanted, but it will
provide you with some information that will help you to meet you and your
customers requirements.

In Leo's post, he had two main concerns:

1)  "We are going to be placing some 15 mil quad flat packs on a pcb assy to
be used in an automotive like enviornment (off road vehicle type of 
vibration is possible) and are wondering if there are any special processing
steps that we whould take to enhance field reliability."

2)  "Our customer complains that previously tested engineering samples that
they made had some problems with the fine pitch.  We've not encountered this
type of issue before.  We and our customer would appreciate any input you
could give us."

-------  Here goes  ----------------------

In order to help you and your customer it would be best to know what
your/their experience has been:   Do you/they know the failure modes?  Was it
due to shock/vibration, thermal, component lead or solder joint failure.  If
component lead failure, was it at the heel bend or the knee bend?  If the
solder joint, how were the leads bent/formed and the configuration of the
solder fillet.

The following are some of the "gereral" design and assembly requirements for
mounting "flat-pack", SO and similarily leaded components for "Serious, Class
3 High-Rel" applications".  "Automotive" is a severe operating environment and
therefore, you may want to consider "optimizing" the traditional IPC Class 3
design, manufacturing and acceptance requirements to go for additional
reliability.  I know that the following are not exactly easily achieved with
fine-pitch components, in particular with many (most) of the commercially
available components,  but when the product definition requirements for
quality includes reliability in adverse operating environments -- all
processes may not simple to achieve the objective.

Component Leads - inspect lead bends for lead and material cracks / fractures
and lead forming impressions.  Use "low stress" and modulus leads such as
copper -vs- Alloy 42 (nickel/iron)

Component Lead Angle - the optimum component lead angle for surface mounted
components is 60 degrees, the acceptable angle is in the range is 55-65
degrees for Class 3 Serious Hi-Rel; but the angle can be opened up to 45-65
degrees for existing designs where the component body has been reduced in
width.   Reason - the 55-65 degree lead angle optimizes the strain relief for
x-, y-, and z-axis displacements.  DO NOT USE ANY COMPONENT LEADS WITH THE
COMPONENT LEAD ANGLE GREATER THAN 65 DEGREES as the component lead (the leg
portion) will not compensate for  strains in the component lead (leg).  The
key to reliability is the heel fillet, and it is difficult to ensure a
suitable heel fillet for component lead leg angles greater than 65-70 degrees.
 At greater than 65-70 degrees, and a minimum height heel solder fillet, the
heel/land solder fillet is very subject to "peel" failure due to reverse
stress thermal and power cycling.

Land Pattern - the land pattern should be slightly wider than the max.
component lead width.  The minimum land length should be about 3X the width of
the lead.  The target land length condition should be about 6X the width of
the lead.  This allows for a slight toe fillet, a foot of up to 5X the lead
width, and the length of the land behind the heel of the component lead should
be ~0.3 the height of the leg. 

Solder Fillet - The land pattern needs to be designed to accomodate the toe,
side and heel fillets.  The heel fillet is the most critical part of the
solder joint and is responsible for about 70-80 percent of the "strength" of
the soldered connection.  The heel of the lead shall be located on the land
(no off-set or overhang).  The continuous side fillets extending from the heel
fillet and along both sides of the lead for about 2X the width of the lead are
the next most critical fillets.  The toe fillet is the least critical for the
longer foot lengths, For shorter foot lengths become more critical as the foot
approaches 2X the lead width.  Foot lengths less than 2X the lead width should
not be used for "Serious" Class 3 High-Rel.  If you are using solder paste,
care must be exercised to ensure liquidous soldering temperature is maintained
long enough to ensure there are no entrapped voids in the solder fillet.  The
optimum height of the heel fillet is ~0.3 (one-third) the height of the leg of
the component lead in order to provide the necessary stress relief due for
shock/vibration/thermal stress.  The optimum angle of the heel fillet between
the lead and the end of the land is ~45 degrees, in other words, the length of
the land behind the heel should be ~0.3 of the lead height in order to
optimize the formation of the heel fillet.  The minimum height of the solder
fillet (in
contact with the leg should not be less then the height of the top of the
foot, nor should the height of the solder fillet be greater than 0.5 the
height of the component leg.  For  Serious Class 3 High-Rel, the stress relief
bends shall not be filled with solder because they just concentrate the stress
some where else.

Component Mounting - the component body should have minimum standoff height of
the surface of the printed board.  Staking adhesives should be used with
caution, they should be low expansion and should be minimal thickness, less
than about 130 micrometers [5 mils (inch)].  The goal is not to have the
expansion of the staking material "pumping" the component in the z-axis
direction during thermal excersions (shock / cycling) and power cycling.

Component Lead Coplanarity - it's very critical, don't "force" leads into
coplanarity using a thermode or some other method of forcing the component
leads into the solder joint.  

Polymeric Coatings - Solder resist and conformal coating should be used with
extreme care, they can be reliability killers if they are under and "pump" the
component in the z-axis during thermal excersions and power cycling.

This information should give you some general help for you to improve you
product.  It will also give your processing/assembly personnel a challenge to
develop a viable productive process to meet you production/cost requirements. 
 But then, it all depends on how serious you (or your customer) are really
serious about achieving "Serious" Class 3 Hi-Rel product definition
requirements.

If you need any further information, you can be contacted as follows:

Ralph Hersey
Lawrence Livermore National Laboratory
Phn.:  510.422.7430
FAX:  510.424.6886
e-mail:  [log in to unmask]




ATOM RSS1 RSS2