REGISTER | LOGIN
Home    Bloggers    Blogs    Article Archives    Messages    About Us   
Tw  |  Fb  |  In  |  Rss
Barry Harvey

Those Darn Plastic Packages

Barry Harvey
Page 1 / 2   >   >>
Erickk
Erickk
4/3/2013 12:51:55 PM
User Rank
Newbie
Re: Old ICs
The use of humidity control measures (silica gel, sealed bags, humidity indicator cards) is due to the shift to surface mount assembly.  In the old days (through hole assembly), it was the PC board that was heated by the solder wave, and the IC package itself experienced only a mild thermal shock.  In SMT assembly, the reflow oven heats the IC package itself comparatively rapidly.  If the plastic has absorbed moisture, the package can rupture due to the rapid heating, just as popcorn does due to moisture trapped in the corn kernel.  Therefore, it is necessary that parts be stored and handled under controlled humidity.  The data sheet for each part now lists a Moisture Sensitivity Level (MSL) that defines the necessary precautions.  If parts have been exposed to too much humidity, they can be salvaged by a slow bake process, which gradually removes the absorbed water without danger to the part.

 

The modern epoxy packages (black material) are actually quite moisture resistant.  Those of us who are older will remember packages for ICs and transistors molded with a gray silicone pastic.  Those devices exhibited terrible moisture moisture absorbtion, and would fail in humid conditions, due to swelling of the plastic and internal corrosion  I believe it was National Semicondutor that solved this problem, by promoting the Epoxy-B system that is the forerunner of modern package materials.

50%
50%
Brad Albing
Brad Albing
4/2/2013 10:07:19 AM
User Rank
Blogger
Re: Old ICs
I also noticed that a lot of my old DIP ICs' package material seemed to be a different material - rather more tough and impervious to soldering.

50%
50%
antedeluvian2
antedeluvian2
4/1/2013 2:23:22 PM
User Rank
Newbie
Old ICs
Barry,

fascinating post!

I was wandering if there has been a "recent" change in the plastic/epoxy used. It seems to me that ICs only started arriving packed together with silica-gel (and humidity sensitive dots) in perhaps the last 6-7 years. I have some pretty old ICs (~1970s) that seem to work fine every time I use one when I am prototyping and need a part quickly. Of course they never are used for anything in production, not least beacuse they are through-hole technology. They have been through dry and humid climates, including being shipped by sea. 

50%
50%
Barry Harvey
Barry Harvey
4/1/2013 1:11:54 PM
User Rank
Blogger
Re: humidity related shift
Sorry, I'm not much of an academic.  I mostly survey the people around me (like packaging engineers) to get some collective wisdom.  Sometimes this doen't work, like when no one agrees. 

50%
50%
amrutah
amrutah
3/31/2013 11:21:53 PM
User Rank
Master
humidity related shift
@Barry, Thanks for the post.

"We have determined that our references drift only single ppms per 1000hr. when humidity is controlled, several times more than when not."

  I come from a place very near to coast where the humidity levels are of the order of 85%-90% most of the year... The effects of humidity you have covered in the blog explaining how it effects the package and how it effects the shift of reference due to humidity is just good.  Any reference papers you recommend on this??

50%
50%
amrutah
amrutah
3/31/2013 11:16:17 PM
User Rank
Master
Package shift...

@Barry, thanks for the blog post.

   "We trim our op-amps at wafer probe to as low as 10uV offset voltage..."  

I have faced this kind of problem in one of my designs.  For Reference system, when I probe at wafer level and trim the device it operates fine, but once packaged we saw a shift of the voltage for the same trim settings...  The mechanical stress induced offset or Package shift.

50%
50%
SunitaT0
SunitaT0
3/31/2013 10:16:53 PM
User Rank
Master
Re : Those Darn Plastic Packages
To keep the package dimensions stable, about 50 percent of the plastic's volume is filled with silica particles, essentially quartz bits.

@Barry, very informative post. I never knew that we face so much of challenge to maintain the package dimiension.  Learnt a lot about packaging through your blog.


50%
50%
Barry Harvey
Barry Harvey
3/30/2013 5:10:38 PM
User Rank
Blogger
Re: Lead free solder temps
Fun...like more delamination, shifts, ugly surfaces, having to re-learn how to solder...No sir, I didn't like the change to lead-free.  Not at all.

50%
50%
Barry Harvey
Barry Harvey
3/30/2013 5:07:56 PM
User Rank
Blogger
Re: What about oxide isolated components?
Thanks, Scott

Silicon is very hard;  silicon oxide is even harder, even deposited.  Everything in the die is hard, excepting aluminum wires.  Thus I figure any external stress travels right through the unyielding layers with very little compressive relief.  So all components, in the silicon or above it, are exposed to package and die attach stress.

So yes, your isolated poly resistors will demonstrate piezo changes.  High sheet-rho poly is the worst, low sheet-rho poly being less sensitive.

Even capacitors have a modest stress sensitivity.  So sorry, you could well have ~100ppm oscillator shifts after packaging.

50%
50%
Brad Albing
Brad Albing
3/30/2013 10:58:39 AM
User Rank
Blogger
Re: Darn
This is why should read all the good things that are posted here. Always something new to learn - as an engineer, keep learning or fall way behind and be left in the dust. There are a lot of people posting things here and they are from a very wide range of backgrounds (by which I refer to industries, life experiences, and even geographical).

50%
50%
Page 1 / 2   >   >>
More Blogs from Barry Harvey
There are analog 'scopes and digital 'scopes. There are good 'scopes and bad 'scopes. There are several manufacturers. How do these categories intertwine?
Bad events occur -- it's inevitable. Sometimes, it's best just to have the cacatological event happen and move on with the design's evaluation.
Just as stated in the corollaries to Murphy's Law, self-starting oscillators don't oscillate and very wideband amplifiers do oscillate. Finding and stopping that amplifier oscillation requires some clever techniques and smart sleuthing.
flash poll
educational resources
 
follow Planet Analog on Twitter
Planet Analog Twitter Feed
like us on facebook
our partners
Planet Analog
About Us     Contact Us     Help     Register     Twitter     Facebook     RSS