 |
| Charles Leech, Director of Engineering, Innovative Drying Company LLC
|
Moisture Sensitive Device (MSD) package delamination, due to expansion of ingested moisture, isn't new. It first surfaced in the early '80's. Everyone agrees with this. But when you start talking about solutions, there seems to be as many opinions, misconceptions and myths as there are people.
With that in mind I have come up with "6 Menaces, 9 Myths and 1 Horror Story" that are costing the electronics industry literally millions of dollars every year. Check these out and save yourself some time and money.
Six Menaces that you have to take seriously
Menace 1: Not taking the MSD issue seriously. Delaminated packages are a source of lowered "first pass yield" and will increase rework expense. It is estimated that 30% of EMS firms have yet to put a MSD program in place.
Menace 2: Not having a formal MSD training program for your organization. A comprehensive training program is essential to avoid MSD yield-losses.
Menace 3: Not adjusting your MSD program to conform to IPC/JEDEC J-Std-033B.1. No lead solders have radically raised reflow solder temperatures. This significantly increased the internal pressure of expanding ingested moisture. This resulted in a serious increase in the Moisture Sensitivity Level (MSL) of all Surface Mount Device(s) (SMD) packages. The current revision reflects this change.
Menace 4: Believing that an assembled MSD package that passes electrical test with an internal delamination is OK. "Minor" cracks in a package can to propagate over the life of the PCB assembly leading to product failure. These failures will be traced back to you. It's not good for your company's reputation.
Menace 5: Attempting to dry-bake MSD packages at below 100°C in an atmosphere that has a relative humidity greater than 5%. Below the boiling point of water, drying efficiency is dependent on the temperature and a low humidity level in the drying chamber.
Menace 6: Not understanding the capabilities and limitations of dry-cabinets. You should review IPC/JEDEC J-Std-033B.1.
- Placing SMD packages, which have been exposed to factory ambient conditions for greater than one hour, in a dry cabinet or dry pack does not necessarily stop/pause the floor life clock.
- For moisture sensitivity Levels 2, 2a, 3 with floor life exposure not greater than 12 hours, a minimum desiccating period of 5X the exposure time is required to dry the SMD packages enough to reset the floor life clock. This can be accomplished by a dry cabinet that is capable of maintaining not greater than 10% RH. For components exposed anytime less than their stated floor life; dry packing or placing the components in a dry cabinet, maintaining not greater than 10% RH, will stop/pause the floor life clock.
- For moisture sensitivity Levels 4, 5, 5a with floor life exposure not greater than eight hours, a minimum desiccating period of 10X the exposure time is required to dry the SMD packages enough to reset the floor life clock. This can be accomplished by dry pack according to 3.3 or a dry cabinet that is capable of maintaining not greater than 5% RH.
Now the 9 Myths - Take heed
They say that you can never take anything for granted. We are all too easily led astray by doing things the way we have always done them and expecting different results. Take a look at these 9 Myths and ask yourself what you believe.
Myth 1: "A 24-hour 125°C oven dry-bake is the default cycle." J-Std-033B.1 now requires a 48-hours dry-bake period.
Myth 2: "Your 125°C dry-baking is limited to one 24-hour cycle." J-Std-033B.1 currently limits 125°C dry-baking to 96 hours (two 48-hour 125°C cycles).
Myth 3: "The Moisture Barrier Bag (MBB) which MSD packages are shipped in is 100% effective in protecting their contents from moisture." Errors at packaging can negate the bags' effectiveness. (1) Excessive package vacuum can stretch the bag material making it porous and more susceptible to moisture leakage. (2) Inserting the wrong number of desiccant pouches can leave the contents vulnerable to moisture infusion. (3) It is not unusual to find MBB's punctured during shipping and receiving. The moisture level shown by the Humidity Indicator Card (HIC) is the only reliable gauge of the moisture content of packages in a MBB.
Myth 4: "The baking schedules published in Table 4.1 of J-Std-033B.1 are the only acceptable temperatures and durations for driving moisture from MSD packages." Footnote 1 - "Table 4-1 is based on worst-case molded lead frame SMD packages. Users may reduce the actual bake time if technically justified (e.g., absorption/desorption data, etc.). In most cases it is applicable to other nonhermetic surface mount SMD packages." Processes like the IDC 70°C - 24-hour Thermal-Wall" process that can meet this justification conform to J-Std-033B.1.
Myth 5: "The rate of moisture diffusion from a package is linear over the dry-bake cycle." 50% to 70% of the ingested moisture in a package is removed within the first 24-hours of drying. The remainder of the bake duration only assures that the package is totally dry. This remains relatively constant regardless of the temperature of the dry-bake cycle.
Myth 6: "The upper temperature limit of commercial polyester, polystyrene tapes and high-impact propylene reels is 40°C." A study by Advantek Corp was a portion of an IDC Proof of Process Experiment. It showed that these tapes and reels remained completely functional after a 24-hour exposure to 70°C. Caution: Special tooling and strict temperature control is required for 70°C baking.
Myth 7: "Vacuum accelerates the rate of moisture removal from a plastic encapsulated package." Flextronics and IDC experiments have refuted this. Vacuum is only a cost effective and reliable means of lowering the chamber relative humidity to near zero.
Myth 8: "A 70°C vacuum-assisted drying process is augmented by a nitrogen backfill." Experiments have shown that a N2 atmosphere is not required for moisture removal.
Myth 9: "A 24-hour, 70°C vacuum-assisted drying process is more effective than a 24-hour 125°C factory atmosphere baking process." Recent studies have shown that a 70°C drying process can remove only greater than or equal to 50% to 70% of the absorbed moisture. This is less than is removed by a 24-hour, 125°C convection dry-bake process.
The Horror Story
I swear this is true. A small EMS firm told me that their "MSD Policy" was to "reflow solder" parts until they "find a package fracture-failure" and then send the remainder of these components back to the distributor as "defective." I had two reactions; (1) Their distributor must love them and (2) I want to avoid anything that comes out of their shop. Please tell me this isn't you!
Biography
Charles S. Leech Jr. is Director of Engineering for Innovative Drying Co. LLC of Glendale, Arizona, USA. The firm has been aggressively involved in the issue of removing moisture from I.C. packages and I.C. package components for ten-years. He has over forty-years of electronics manufacturing management experience and has been issued two U.S. Patents for advanced drying processes.
