TECHNET Archives

March 2012

TechNet@IPC.ORG

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Steven Creswick <[log in to unmask]>
Reply To:
TechNet E-Mail Forum <[log in to unmask]>, Steven Creswick <[log in to unmask]>
Date:
Tue, 27 Mar 2012 10:15:59 -0400
Content-Type:
text/plain
Parts/Attachments:
text/plain (116 lines)
Inge,

Please do not make that pedestal too tall, for I may fall off.  I would see
what Doug and Dave might also have to say.


I had not previously looked at JEDEC 142 until this morning, but since it
sites all the old Mil Specs here are my limited comments.

I my mind, when speaking of a hybrid, or MCM [multichip module], I only
think of bare die applications, most being true hermetic modules.  I don't
think of soldering, etc. [except for special situations that I will comment
on below].

Since we could not wire bond to Tin, or any other solder alloys, it never
showed up as component terminations, bonding pads, etc.

Never being able to reliably, electrically adhesively bond to solder
terminated components, all our components were bare Si, AuSi, Au, Pd, PdAg,
or PtPdAg.

The packages were typically Kovar, with leads and package body being Ni and
Au plated for wirebonding and/or soldering to the board [after hybrid
assembly and seal].

Other packages were the normal alumina, aluminum nitride, HTCC and LTCC
materials with seal rings attached using either CuAg eutectic [about 780°C],
or AuSn eutectic at 280°C

Sealing was mostly welding with some AuSn eutectic.

In general, one would absolutely never corrupt the interior of a hermetic
package with something as vile as fluxes [and solder].  That would be like
using a chamber pot to make tea...  I don't particularly care how well you
clean it, I would prefer not to use it as a tea pot.  Hermetic means
whatever is in there STAYS in there.

So, from that rather narrow view of the world, them leaving out all
reference to tin is not all that much of a surprise.

Before shipping [after all the burn-in and environmentals were done] the
leads of the modules we would typically be 'tinned' in Sn63/Pb37 - I
believe.  Could this have promoted growth at the point where solder
approaches the glass to metal seal?  Could the customer have 're-tinned' in
something else?


However, I do note that in the section on chip caps [5.2.8] they mention
solder balls, solder coverage, and are dancing around the topic of wetting
characteristics - ewe!  Are they attempting to corrupt my narrow view of a
'true' hybrid...?

My humble suggestion would be that if they are going to include soldering
into the topic, they should indeed caution about the use of tin.  I am in
total agreement with you there.


Yes, there have been situations where solder has been used in hybrids that I
am familiar with.  As in high-power devices where the IGBTs or FETs were
soldered with either Pb100 or Pb95/Sn5 without flux, in at least a reducing
atmosphere/or vacuum.  But in these cases, I would envision the assembly to
be rather immune from whiskers due to the extremely high lead content.
Other situations were when creating 'stacked' assemblies within TO style
packages.  In this case soldering involved NiAu plated leads to thick film
or thin film metallizations on the substrates.

I can remember there being stacked assemblies using 'tin plated' [alloy not
known] headers and solder sealed packages [alloy not known] back in the
early 70's.  These went into avionics that were conformally coated...  That
could easily have been an 'oops' 

I can also remember visiting a large relay manufacturer in the mid 80's when
providing them with arc suppression networks.  They appeared to have tin
plated cans and header bases, but even then, not much thought was given to
tin whiskers....

The last two examples, were likely whiskers waiting to happen!

Also ran into a wireless component supplier that wanted to have RF shields
bright tin plated.  This was back in the 90's.  That was immediately changed
... of course those cell phones are not likely to still be with us.

That is the memory dump for the day.

Steve


-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Inge Hernefjord
Sent: Monday, March 26, 2012 3:20 PM
To: [log in to unmask]
Subject: Re: [TN] NTC RE: [TN] Out of the world tin whiskers

Isn't it remarkable:

JEDEC 142  OBTAINING AND INSPECTING MATERIAL/COMPONENTS FOR HYBRIDS/MCM

Not a word about checking for pure Tin ! With the apparent risk of whiskers,
a line or two about that would not be wrong...or did I miss something? Maybe
I don't have a fresh copy?

Steve, as a specialist on these, what's your comment.

Inge







______________________________________________________________________
This email has been scanned by the Symantec Email Security.cloud service.
For more information please contact helpdesk at x2960 or [log in to unmask] 
______________________________________________________________________

ATOM RSS1 RSS2