What damage does the assembly process do to a pcb? (part 5)

In part 4 of this series of posts I discussed the effect moisture has on the printed circuit board at soldering temperatures. I explained the material properties of FR-4 laminate and how they are hygroscopic. We also covered an acceptable practice known as dry baking used to force moisture from the product just prior to being exposed to soldering temperatures. I received some very good comments and feedback regarding part 4 of this blog series. I thought it appropriate to share this feedback in another post.

With regards to sources of moisture, in part 4 of this series I comment upon a practice where some printed circuit board manufacturers shall dry bake the printed circuit boards prior to shipping them. The important thing to add here is that common packaging materials used by the industry do not act as a 100% vapor barrier. The shrink wrap film used only slows down the process of moisture absorption by the product. Moisture in the environment outside the bag shall migrate through the protective film but at a much slower rate. Things to consider are as follows…

  1. What types of materials are used for shrink wrapping film and how much moisture resistance shall they provide? Different manufacturers use different materials that offer different levels of protection.
  2. How well was the product shrink wrapped and sealed? Some boards have edges routed or scored to a sharp angle at the corner of the board or the pallet/array. The sharp corner may puncture the film thus compromising the vacuum seal.
  3. How well are the shrink wrapped bundles boxed up? In the past I’ve received board shipments from suppliers where the box had burst open at the corners. The shrink wrapped bundles were not tightly packed and were loose within the box. Weight placed on top of my delivery crushed the box and split it open. Since the shrink wrapped bundles were loose they could easily move around and ultimately the vacuum seal burst open.
  4. Does the product need to pass through customs? Product from sources located in another country may be randomly selected for inspection by a customs official. Boxes and packages are opened and the containing environment are compromised as a result.
  5. Are moisture indicators and packs of desiccant provided within the shrink wrapped bundles? In the event the vacuum seal is compromised the moisture indicator shall at least let you know if you have a problem or not.

Short of having a customs agent open the box and shrink wrapped bundle there are several ways in which a printed circuit board may be exposed to moisture while in transit. The following are some recommendations that may be specified as part of a shipping specification. Keep in mind that everything you specify is technically considered an extra option and shall increase cost. Are the increased shipping costs worth the added value? If you don’t have a problem now then anything extra is wasted money. If you currently receive compromised packaging then consider the following…

  1. Shrink wrapped bundles should be packaged in a way that prevents the bundles from shifting around within the shipping box. Either styrofoam peanuts or extra bubble wrap should be placed around the shrink wrapped bundles for support.
  2. Double walled shipping boxes may be used.
  3. Strapping reinforcement may be added to the outside of the shipping box.
  4. Specify a weight limit on the boxes. Approximately 35 to 40 Lbs. (16 to 18 kg) may be a good range. Something to consider for weight limit determination is what are your people comfortable lifting. Also, the more weight allowed in a single box the more mass it has. Commercial shipping companies are not known for being gentle. When a shipper tosses a heavy box with a lot of mass onto a stack of boxes the heavy box wants to keep going.
  5. Ask for each shrink wrapped bundle to be provided with a moisture indicator and desiccant appropriate for the solderable surface finish.

Are there shipping products available that offer a 100% vapor barrier?

There are packaging materials available that offer a 100% vapor barrier. For example, packaging materials that meet the requirements of MIL-PRF-81705 shall provide a 100% vapor barrier. The draw back to specifying packaging materials to this shipping standard is cost. MIL-PRF-81705 packaging is very expensive. From my experience 100% vapor barrier packaging is rarely used. When it is used, its only on very expensive and sensitive products.

With regards to the damaging effect moisture has on the printed circuit board there is an added expansion stress factor. When materials are heated they expand. Copper, glass and epoxy all expand at different rates of expansion. This is referred to as the Coefficient of Thermal Expansion, aka CTE. Copper and glass have relatively low rates of thermal expansion. Below the laminate Tg the epoxy has a low rate of thermal expansion. Above the laminate Tg the epoxy has a very high rate of thermal expansion. That the copper, glass and epoxy do not expand at the same rates of expansion adds stress upon the printed circuit board at assembly temperatures.

The woven fiberglass weave within a printed circuit board helps to keep the levels of thermal expansion low in the X and Y axis of a printed circuit board, even above the Tg. In the Z axis (board thickness) there is no woven glass. When the temperature rises above the Tg, the epoxy expands at a much greater rate. The only opposing force applied against the expansion of the epoxy in the Z axis is from the copper in the plate through holes. Keep in mind that the copper expands at a low rate (approximately 19 ppm/°C) where above Tg the epoxy may expand at a much higher rate (potentially 250 ppm/°C). This places a tremendous amount of strain on the copper

Copper has elastic properties. When copper is stretched a little, the copper shall spring back to its original length. When the force applied passes the copper’s point of elasticity the copper does not spring back to its original length and remains stretched. When a printed circuit board is exposed to assembly temperatures the copper is stretched in the Z axis. When the strain applied passes the point of elasticity the pads on the outer layer may appear lifted from the surface of the printed circuit board. When the strain on the copper passes the material limit of the copper, the hole ruptures and may fail in one of the following ways…

  1. Post hole separation, aka inner layer separation. The barrel of the plated through hole disconnects from the inner layer copper pad at the hole wall.
  2. Hole wall pull away. On a multilayer this could lead to a field failure since the pull away shall continue to propagate till a post hole separation forms.
  3. Barrel cracks. When the copper is stressed beyond its limit the copper cracks and forms a front to back disconnect.

When the epoxy of the printed circuit board contains moisture an extra stress factor is introduced. When the assembly temperature exceeds 100°C the moisture vaporises and expands at a much greater rate. The expansion stress is referred to as the Coefficient of Hygroscopic Expansion, aka CHE. It may also be referred to as the Coefficient of Moisture Expansion, aka CME. What this means is thats the epoxy shall expand at a greater rate above and beyond the standard CTE that is expected. The greater the moisture content, the higher the strain placed against the barrel of the plated through hole. The higher the strain placed upon the copper in the hole, the greater the potential of a plated through hole rupture or failure to occur.

Tags: , , , , , , , ,

One Response to “What damage does the assembly process do to a pcb? (part 5)”

Leave a Reply