Over the last ten years there has been a number of reasons why chip capacitors
have generally been found to fail during production and in the final products,
these will be outlined first.

The design and method of component manufacture can affect the component's
ability to stand up to modern manufacturing methods.  In cases where components
are selected at the top of the capacitance range for a given package size, due
to the number of plates the components can be more sensitive to fracture
failures. 

Often the component manufacture will suggest de-rating parts for a given
process; this is common in many designs. The internal construction and lay up of
the internal electrodes may also affect the robustness of the parts. This does
vary with different part suppliers and should be evaluated during initial
component assessment.

Components, when they are supplied, may be damaged prior to removal from the
packaging; this has certainly been seen in the past. Internal cracking may be
present and is then exaggerated by the assembly process. Examination of selected
components in their packaging may provide examples of defective parts. 

Machine centring and placement of components on to the board has been a problem
area over the years. Again this is exaggerated where the parts are less robust.
(In America SMT magazine in the early days called Nutshell News the editor
called the machine chip crackers, boy was he right). Recently it has been the
use of board support pins for double sided work on placement and screen print.

Poor positioning of pins under components does not allow the board to move when
the component is placed. The incorrect height of board support pins can also
exaggerate the problem. If the pins are adjusted too high when the board is
clamped for placement the board is actually bowed upwards during placement.
Minor cracking may occur and is then shown up during soldering and test.
 
Wave and reflow soldering temperature settings again have been shown to affect
component reliability. Minimising the temperature of solder baths to 240oC and
controlled board pre-heating has mostly eliminated the wave soldering problems
in the industry. 

Any good quality part suitable for use in board designed for manufacture should
be able to stand immersion in a solder bath at 260oC with no pre heat. Refer to
the IPC deign rules and process compatibility document for components. The IEC
documents have the same guide.

It is uncommon for the reflow soldering process to cause failures unless parts
have already been damaged. If the quality of the capacitor is poor with internal
voiding or stress cracks the reflow process is less likely to cause problems.

In the case of wave or reflow soldering, if the board is poorly supported and
the board is allowed to sag the board may be warped after soldering. When the
board cools the board is left in a warped state. If any subsequent process
straightens the board it may cause component cracking. A possible example is
in-circuit test where the board is straightened flat during pin contact.

Hand soldering and rework has also been shown to damage parts if correct
procedures are not followed or poorly trained staff are used. It is common where
components come into direct contact with a solder bit rather than using the
solder to provide effective heat transfer. During any hand soldering operation
using soldering irons the solder is used to transfer the heat to the board and
component not the bit. 

Today it is more common to use hot air pencils to rework surface mount
component, this minimises physical contact with parts. If the components are
correctly specified with a solder coating and solder coated boards are used then
normally only flux is required to make a joint. As sufficient solder is already
present to make a connection on the board and termination. This is by far the
best process. 

A lot has been said about pre heating components prior to rework, this is no
going to happen in the real world. If you can't rework parts using standard
technique don't design it into the product its called GOOD DFM, design for
manufacture.

Board flexure during board handling, break out, in-circuit test or final
mounting of the board into the product is more likely to cause failure. This is
again true if the component is not correctly specified and tested during initial
assessment of the supplier. 

Most of the cases currently being examined by myself for customers are flexure
failures. It is considered that board flexure is one of the main cause of
failure. If you break out multi panels using bolt cutters across a bench or just
snap V scored boards what stress is being applied???

Alternative laminates or higher Tg materials may help to provide a flat surface
for assembly. This is only of value if it coincides with balanced layout during
design and correct specification of the solderable finish. The more stable
laminate will not help if the board is flexed during assembly.

Nine times out of ten look for board flexure and you find the cause of the
problem.

Bob Willis
Process Engineering Consultant
Electronic Presentation Services
2 Fourth Avenue, Chelmsford, Essex CM1 4HA. England.
Tel: (44) 01245 351502
Fax: (44) 01245 496123
Home Page: http://our world.compuserve.com/homepages/bwillis
Email: [log in to unmask]


***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************