Moisture levels when receiving a part.

[johnnyfp]

Hi,

A blog explaining why, if it is, important to have part moisture levels low, and why when ordering certain parts from Digi or element or even the manufactures come with a moisture level warning.

What's all that about. One or two parts come with this but the rest don't! Does it mean an electronic part has a shelf life. Sorta on that line

Jfp

[IanB]

Do you mean LEDs?

Most electronic parts are completely sealed from the environment and moisture levels are irrelevant. However some parts like LEDs have elements like clear transparent domes that are sensitive to humidity. In general that will only happen if it's "necessary", since it is in no-ones interest to make fragile parts.

[Psi]

It's not just leds.
Atmel AVRs from digikey sometimes come in sealed metal foil with a moisture detection card inside.

[IanB]

Atmel AVRs from digikey sometimes come in sealed metal foil with a moisture detection card inside.

That doesn't make sense. The IC package should be hermetically sealed from the environment. Do you have any insight into why moisture might matter? Is it anything to do with oxidation on the leads that might impact soldering efficiency?

[krenzo]

The IC package should be hermetically sealed...

If a part comes with a moisture barrier bag and indicator, then it is not hermetically sealed and is susceptible to damage from moisture during reflow.

You're supposed to bake parts that have moisture warnings for 24 hours before wave soldering them if the moisture indicator is past a certain level.  The idea is that you slowly evaporate any moisture in the chip instead of too quickly which could cause a sudden increase in pressure and damage it.

"Moisture from atmospheric humidity enters permeable packaging materials by diffusion. Assembly processes used to solder
SMD packages to printed circuit boards (PCBs) expose the entire package body to temperatures higher than 200°C. During
solder reflow, the combination of rapid moisture expansion, materials mismatch, and material interface degradation can result
in package cracking and/or delamination of critical interfaces within the package."

[Psi]

Here's a ATMEGA16 i got that came that way.

[johnnyfp]

 Its odd that atmel has this issue but not microchip.  Anyway for us hobbiest electronic engineers what does this mean. Should we only buy one device, or can we reseal the bag? What would really happen if we solder the device? And what does it mean afterwards. Do we have too lacquer these devices?

Jfp

[IanB]

Its odd that atmel has this issue but not microchip.  Anyway for us hobbiest electronic engineers what does this mean. Should we only buy one device, or can we reseal the bag? What would really happen if we solder the device? And what does it mean afterwards. Do we have too lacquer these devices?

Apparently it doesn't matter if you hand solder such devices, it matters only if you solder them in an oven.

I'm kind of surprised about it, but apparently many plastic packages are permeable to moisture. This is why, for instance, the plastic bags for potato chips or crisps have a shiny metal film on the inside surface. Otherwise moisture would be able to diffuse through the plastic and reduce the freshness of the contents.

[cybergibbons]

All ICs suffer from this. Plastic encapsulated ones are permeable to moisture. Ceramic ones allow moisture in along the leads. Microchip is included in this, it's just the supply chain doesn't control it.

http://www.national.com/ms/PL/PLASTIC_PACKAGE_MOISTURE-INDUCED_CRACKING.pdf

[mikeselectricstuff]

AFAIUI, this is only a big deal for wave soldering, due to the rapid temperature rise. With  conventional reflow, the temp rise is slow enough  to not be an issue.
Wave soldering SMDs is not a common process - typically only used on consumer products with mixed TH and SMDs on the TH solder side.
 

[ciccio]

AFAIUI, this is only a big deal for wave soldering, due to the rapid temperature rise. With  conventional reflow, the temp rise is slow enough  to not be an issue.
Wave soldering SMDs is not a common process - typically only used on consumer products with mixed TH and SMDs on the TH solder side.
 

From my limited experience: last week each board assembled with  op-amps (SMD) coming from a single reel that had a moisture warning label showing high moisture exposure were rejected at first test stage because each and every op-amp had at least one pin not soldered, and each board had 21 of them: 50 boards assembled for a total of 8400 pins to be reworked by hand..
All the other boards using op-amps coming from a different reel were soldered perfectly, with the same solder paste and the same temp profile and with PCBs coming from the same production lot.
Not sure, but I think that moisture has something do do with surface oxidation of pins.
Plastic is permeable to humidity, but also to air: this is the reason of ceramic cased ICs, for hi-reliability application.
Best regards

[Mechatrommer]

some opamp i ordered comes with this RH indicator. not sure why though.

[SeanB]

The moisture indicator is a means to detect excess humidity in shipping and storage. The leads corrode if exposed to moisture, so this is a good way to show poor storage ( stored in open packages, stored in high humidity in a warehouse, left in rain) which will make the product have poor solderability during manufacture.

The packages themselves are not totally impervious to water, and will slowly diffuse water vapour through, especially bad on a small SMT device where the actual die is barely under the epoxy, and where any voids will accumulate moisture, which can be bad if a lead wire or bonding pad is part of the cavity, as that will eventually corrode and go open.

Driving off the moisture will not help remove existing corrosion, but will keep it from spreading. Best is to store the units ( if you are going to store for period of 6 months or more) in the sealed original unopened packaging, and additionally place this in a further bag with a desiccant sachet inside as well, and check this at regular intervals for moisture ingress. With a mylar bag that has a waterproof lining, and properly sealed, this can last decades. I often got spares that had been stored for 40 odd years, still perfect with the desiccant showing no moisture, as military stored spares. These had been in a warehouse with no climate control other than a tin roof keeping rain off, exposed to temperature swings every day.

[alm]

My impression is that moisture is mainly an issue during soldering, eg. corrosion of the leads or popcorning. If moisture in the air would damage the die, what about storage after production? Do you expect your customers to store their appliances in climate controlled rooms or sealed in mylar bags?

[IanB]

AFAIUI, this is only a big deal for wave soldering, due to the rapid temperature rise. With  conventional reflow, the temp rise is slow enough  to not be an issue.
Wave soldering SMDs is not a common process - typically only used on consumer products with mixed TH and SMDs on the TH solder side.

Evidently moisture induced damage is a unique issue with SMT parts soldered in reflow ovens. A few quotes from the IPC/JEDEC document linked above:

The advent of surface mount devices (SMDs) introduced a new class of quality and reliability concerns regarding package
damage ‘‘cracks and delamination’’ from the solder reflow process. This document describes the standardized levels of floor
life exposure for moisture/reflow-sensitive SMD packages along with the handling, packing and shipping requirements necessary
to avoid moisture/reflow-related failures. Companion documents J-STD-020 and JEP113 define the classification procedure
and the labeling requirements, respectively.

Moisture from atmospheric humidity enters permeable packaging materials by diffusion. Assembly processes used to solder
SMD packages to printed circuit boards (PCBs) expose the entire package body to temperatures higher than 200°C. During
solder reflow, the combination of rapid moisture expansion, materials mismatch, and material interface degradation can result
in package cracking and/or delamination of critical interfaces within the package.

The solder reflow processes of concern are convection, convection/IR, infrared (IR), vapor phase (VPR) and hot air rework
tools. The use of assembly processes that immerse the component body in molten solder are not recommended for most
SMD packages.

The purpose of this document is to provide SMD manufacturers and users with standardized methods for
handling, packing, shipping, and use of moisture/reflow sensitive SMD packages that have been classified to the levels
defined in J-STD-020. These methods are provided to avoid damage from moisture absorption and exposure to solder reflow
temperatures that can result in yield and reliability degradation. By using these procedures, safe and damage-free reflow can
be achieved, with the dry packing process, providing a minimum shelf life capability in sealed dry-bags of 12 months from
the seal date.

This standard applies to all nonhermetic SMD packages subjected to bulk solder reflow processes
during PCB assembly, including plastic encapsulated packages and all other packages made with moisture-permeable polymeric
materials (epoxies, silicones, etc.) that are exposed to the ambient air.

This standard applies to bulk solder reflow assembly by convection, convection/IR, infrared (IR), and
vapor phase (VPR) processes. It does not apply to bulk solder reflow processes that immerse the component bodies in molten
solder (e.g., wave soldering bottom mounted components). Such processes are not allowed for many SMDs and are not
covered by the component qualifications standards used as a basis for this document.


The document goes on to explain that moisture sensitive parts need to be packed with desiccant in moisture barrier bags with a moisture indicator card. If the moisture indicator reads high the parts need baking in a special drying oven before they can be used in a reflow process. If the parts are removed from their dry packaging under factory conditions for more than an hour, the clock starts ticking on their "floor life" and they may also require re-drying if they are exposed to the air for too long.

One gets the impression that the impact of moisture had not been fully recognized in the past and that many defects during reflow soldering that might previously have gone unattributed could in fact have been due to poor moisture control.

[IanB]

My impression is that moisture is mainly an issue during soldering, eg. corrosion of the leads or popcorning. If moisture in the air would damage the die, what about storage after production? Do you expect your customers to store their appliances in climate controlled rooms or sealed in mylar bags?

Evidently moisture only becomes an issue during reflow soldering when the whole package gets heated up to high temperatures.

[Zeta]

Moisture can cause popcorn efect on most SMD parts.

Moisture Sensitivity Level (MSL) is always indicated in the label. MSL 2 devices can be exposed to room ambient around 100 days. MSL 3 devices can be exposed a couple of days, MSL 4 just a few hours.  So keep your MSL >1 parts sealed with a sack of disicant and moisture indicator.

most DIP parts and some sot23 are rated MSL 1, so they have no issues with moisture.

And I agree with op. It would make a good a vblog

[itdontgo]

You may have an issue if you're looking for a really high yield on mass produced boards made in reflow ovens but for most people it wont be an issue.

If you're making just 100s of boards and your parts haven't been sat around for years you should not see any problems if you dont bake the parts.  In fact baking the parts is not without it's problems as it oxidizes the pins and makes them hard to solder.  Plus 8 hours at 125C is pretty impractical.

[free_electron]

Two bits of information :

ALL pacakges are moisture sensitive , but THIN packages are most affected by the effect when heating.
Even a standard DIP package absorbs moisture. But, since the thickness is high it can easily with stand the stress. things like TQFP's , led's with gel-domes cannot withstand the pressure buildup when the moisture is cooked out and the popcorn.. worst case you get total delamination of the casing .. cut bondwires and intermittent contacts in the package itself.

parts that have been exposed to ambient air for longer than 48 hours need to be baked before sending through reflow. Baking is a slow rampup from ambient to about 80 degrees c for a couple of hours under a dry air atmosphere ( typically the ovens are flooded with nitrogen from large steel dewar's. that contains no moisture at all. after that : vacuum seal with silica gel packs and an indicator strip. Someon suggested 125 for 8 hours .. that is shooting mosquitos' with an elephant gun ! get a real oven for this work that uses nitrogen ! if you can't afford that : stay out of the kitchen and subcontract it so it is done right.

Another factor is the epoxy itself . there are different cmppounds out there and the Rohs directive has forced new epoxies.. That happen to be more sensitive ...
So if digikey ships parts that are susceptible , yes the do treat them correctly and vacuum pack them. Like zeta said : check the MSL. not every chip manufacturer uses the same epoxies. Heck ,most chip makers don't package parts. they subcontract to Amkor Anam ...

Point 2 : RohS. moisture is one of the inducers for the formation of tin whiskers ... combine them with mechanical stress and you have a golden recepy for whisker growth ...
and it is the evaporating moisture that causes the stress..

point 2a : pin corrosion. humidity and a pure tin ( whtout lead pin ) very quickly decays the wettability of the pin to the point that fluxes cannot fully perform the cleaning.

Point 3 ; BGA's are EXTREMELY sensitive sincetheir substrate is a pcb itself. and that stuff is hygroscopic as hell. leave a BGa in ambient air with 40% humidity for a week and it will popcorn the moent it goes into the rampup  ( rampup is the reflow phase between flux activation (120 .. 140 degree c ) and the actual reflow phase where the solder goes liquid ). The time in reflow oven is too short to safely evaporate the moisture.

The same goes for PCB's as well. tose tooneed vcuum and dessicant packaging ! or you resk delamination and via breaking. Boards become more susceptible as the number of layers increases.

So ,yes , expect to see more and more the silica gel packs.

[AlphZeta]

Like others said, moisture is a big problem for "smaller" SMD parts. When the pads are small, oxidation could be the an issue during re-flow. Leads in larger parts (i.e. through hole DIP packages) also suffer from the oxidation issue, but the soldering process is less stringent than that for SMD parts. In general though, it's always a good idea to keep your parts storage area's humidity low.

I order my parts from AVNet and they seem to follow stricter packaging standards then some other suppliers. Not only the QFN parts come with humidity absorbers and indicator cards, almost all SMD parts (e.g. TSSOP, MSOP, etc.) come with moisture control measures as standard practice.

[itdontgo]

I dont think Farnell know what they're doing:



Really?  Bake a coil?

I'd love to know how they store their moisture sensitive parts though.  It must be very impractical.

[G7PSK]

I dont think Farnell know what they're doing:



Really?  Bake a coil?

I'd love to know how they store their moisture sensitive parts though.  It must be very impractical.

Wire insulating varnish absorbs moisture, which is why alternators, transformers and motors come with drying instructions in case of long periods of storage, usually done by applying a low voltage current for a set time with these.