Ralph, Ken
The other thing to consider is that as the base material stress relieves /
subsides, the fastener is no longer at the spec'd torque.
The process I prefer to use for this sort of application is to put the
fastener on hand tight, wave solder, and then torque after wave.  One can
avoid the haloing and have the fastener at the required torque this way.

regards,

Graham Collins
Process Engineer
Litton Systems Canada, Halifax Facility

-----Original Message-----
From: Ralph Hersey [mailto:[log in to unmask]]
Sent: Wednesday, June 24, 1998 8:49 PM
To: [log in to unmask]
Subject: Re: [TN] Measling/Crazing


Hi Ken--

Chafin, Ken G. wrote:

> We have a FR-4 board (11" by 8" by 0.093") on which is mounted a large
> sink beneath two power transistors.  When the screws which secure the
> transistors to the heat sink and to the board are torqued to 5
> inch-pounds, significant measling beneath the heat sink (visible from
> solder side) occurs when the board is flow soldered.  When the torque is
> reduced to 3 inch-pounds the measling disappears.
>
> Does the above information suggest there is a problem here beyond a
> simple process problem which is addressed by reducing the torque from 5
> inch-pounds (which the drawing specifies) to 3 inch-pounds?  (For the
> record, I don't know if this should be called measling or crazing--both
> the mechanical and thermal stress are required to produce the
> phenomenon.
>
> It seems odd that the relatively small amount of torque change could
> have such a dramatic effect--if the material is within specification
> limits and the flow solder process is close to optimal.
>

Per the current IPC-T-50, and IPC-A-600 terms and definitions, the condition
you have would most
probably be "Haloing" and not "Crazing" or "Measling".

"Measles" only occur at the weave intersections of the glass-fabric
reinforcement material and
are limited to the size of the intersection.

"Crazing" most frequently occurs due to mechanically induced stress and is
somewhat "round" in
shape, crazing can also be thermally induced and is generally located
between component lands
(see IPC-A-600 D (not the E) page 6).

"Haloing" by definition is "Mechanically-induced fracturing of delamination,
on or below the
surface of a base material, that is usually exhibited by a light area around
holes or other
machined features." (per IPC-T-50).

By you description, most probably, the product has a "whitish-like" or other
light colored (with
respect to the color of the base material) in a radial pattern around the
fastener.

In my previous (work) life, we experienced the same effect, either after
flow soldering or after
product was used.  This was due to heat transfer from the cooling sink or
power semiconductor
through the fastener into the base material.  The heat increased the
stress/strains in the base
material around the fastener as well as changes the mechanical and other
properties of the resin
system.

At the greater fastener torque, the insulative base material is under
significant compressive
stress,  when the heat from the flow soldering process is conducted into the
base material, the
base material will stress relieve.  As the resin (epoxy) softens due to
temperature, the glass
fiber bundles will redistribute (tend to flatten out) due to the fastener's
compressive stress,
and coupled with the reduced modulus of the resin.  The resin will attempt
to expand
isometrically, it can't due to the fastener, therefore there will be some
radial displacement of
the resin away from the fastener.  All this material movement results in
shear stresses/strains
at the glass fiber/resin interface.  The result is a change in the
refractive index of light in
the base material due to the separation -- classic haloing.

At the lower fastener torque, the base material compression won't be quite
as much, and is below
the threshold for causing haloing.

Reliability concern -- in general there should be no problems, use the
IPC-A-600E's acceptance
criteria if there are no other requirements.

CAUTION -- EXCEPT if the fastener torque is being used to maintain either
electrical continuity
or heat transfer between the members being mechanically joined.  (In the
application this might
be true -- the fasteners may be used to mount a power semiconductor to a
heat sink, and the
printed board; and the fastener is also being used to maintain electrical
conductivity between
the semiconductors case and conductive patterns.

It is very poor design practice to use any plastic materials for any
mechanical connections that
are required for either electrical or heat transfer applications, unless a
spring washer is used
to ensure there is sufficient compliance in the joint members.  A
"Bellville" or other "spring
washer" will ensure the desired fastener tension is maintained due to
thermal
expansion/contractions, and plastic flow of base materials under the
operating conditions and for
the life expectancy of the product.

If the design is using the mechanical fasteners for electrical and/or
thermal design
consideration, there will be a latent failure problem whether you have
haloing or not.

Ralph Hersey

Ralph Hersey & Associates
3885 Mills Way
Livermore, CA 94550-3319
PHN/FAX: 925.454.9805
e-mail: [log in to unmask]

################################################################
TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
################################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following
text in the body:
To subscribe:   SUBSCRIBE TechNet <your full name>
To unsubscribe:   SIGNOFF TechNet
################################################################
Please visit IPC web site (http://jefry.ipc.org/forum.htm) for additional
information.
For technical support contact Hugo Scaramuzza at [log in to unmask] or
847-509-9700 ext.312
################################################################

################################################################
TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c
################################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body:
To subscribe:   SUBSCRIBE TechNet <your full name>
To unsubscribe:   SIGNOFF TechNet 
################################################################
Please visit IPC web site (http://jefry.ipc.org/forum.htm) for additional information.
For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312
################################################################