TECHNET Archives

1996

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:
Date:
Thu, 31 Oct 96 7:44:49 CST
Content-Type:
text/plain
Parts/Attachments:
text/plain (806 lines)
Gary

I believe that popcorning causes two failure modes,
1.	damage to bond wires and other features, or
2.	moisture condensation causes electrical failure.

I believe that you have the latter case that's why baking cure the failure. 
 However, the failure will re-occur once moisture permeate and condense 
again.

Kuan-Shaur Lei
Senior Member of Technical Staff
Compaq Computer Corporation
(713) 518-8099
[log in to unmask]
-------------
Original Text
>From [log in to unmask], on 10/30/96 6:36 PM:
To: <[log in to unmask]>

------------------------------

Content-Type: text/plain

TechNet-d Digest				Volume 96 : Issue 101

Today's Topics:
	 Re: DES and fab?? Back Pressure
	 RE: pwb-cca : conformal coating
	 DES, ASSY: Fine Pitch SMT
	  RE: Relative issues concerning the tenting of vias -Reply
	 ASSY:  Passive Devices under PLCC-44
	 FAB: Post-Lam Bake
	 Re: Vision System
	 Re[2]: pwb-cca : conformal coating
	 RE: pwb-cca : conformal coating
	 job posting
	 bismuth based solder
	 ASSY: GEN: Popcorning during reflow

Administrivia:
**********************************************************************
**		       IPC TechNet Digest List			    **
**********************************************************************

------------------------------

Date: Wed, 30 Oct 1996 14:08:08 -0500
From: [log in to unmask] (Doug McKean)
To: [log in to unmask]
CC: [log in to unmask]
Subject: Re: DES and fab?? Back Pressure
Message-ID: <[log in to unmask]>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

[log in to unmask] wrote:
>      Has anybody heard of back pressure signaling??????
>      It has to do with electrical performance characteristics.
>      I do know that much.
>      Pweeze Halp?
>      Groovy

GrooveMan, 

Don't know that much about it but here's an attempt... 

Has to do with "traffic control" in Ethernet (I think) to prevent 
congestion 
on the network.  It's to prevent overloading in network nodes. 
Literally, electrical constipation. 

Back pressure signaling is used to notify a sending node to reduce its 
flow. 
A back pressure signal is sent 'upstream' to the first upstream node to 
slow 
the xmit speed to match the recieve speed of the first 'downstream' node on 
a net.  This can obviously lead to all sorts of delays on a net. 

I 'think' this is what you're asking. 

Regards, Doug

*******************************************************
Doug McKean
[log in to unmask]
-------------------------------------------------------
The comments and opinions stated herein are mine alone,
and do not reflect those of my employer.
-------------------------------------------------------
*******************************************************

***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************

------------------------------

Date: Wed, 30 Oct 1996 13:13:26 -0600
From: [log in to unmask] (Mary Davis)
To: [log in to unmask] (CINDY KEMP ORLANDO ISC *8-306-6),
        [log in to unmask] (Technet)
Subject: RE: pwb-cca : conformal coating
Message-ID: <[log in to unmask]>
Content-Type: text/plain; charset="US-ASCII"

Thought I would add my two cents on the subject.

We are on a steep learning curve with respect to plastic parts.  One of   
several problems we encountered was dewetting at conformal coat. After   
working the issue for a while, we have concluded that plastic parts are   
inherently harder to wet than other parts and that some conformal coating   
products are inherently better at wetting than others.  A formula to   
predict wetting would be nice but we so far we have progressed by trial   
and error.

The first coating that we tried was our   
 'good-old-stand-by-coat-anything', single component, solvent based,   
MIL-I-47058 UR.  Wetting was spotty and unpredictable.

We were also working with the manufacturer of our solvent based coating   
to evaluate their line of UV cure, single component, 100% solids,   
MIL-I-47058 coatings.  We tried their 100% solids coatings on the plastic   
parts and had even less success. The dewetting of the plastic parts was a   
text book example of 'crawling' or  'retraction'.  Wetting was good   
everywhere except on the plastic parts.  Cleaning the plastic parts did   
not improve wetting.
    

We then tried a Dymax product, a single component, 100% solids, UV cure,   
acrylic copolymer (meets AR, ER, and UR.)  We had much better luck.  So   
far this product coats plastic parts without a problem.  It seems very   
robust with respect to plastic parts.

Wetting is a function of the surface tension of the liquid and the free   
energy of the surface.  If I understand the physics, wetting will not   
occur unless the surface tension of the liquid is less than the critical   
free energy of the solid.   I assume that these parameters vary over a   
wide range for the materials in question - but there is very little data   
available.  Even without the mold release variable, plastic is a low   
energy surface relative to metal and ceramic and, is therefore, more   
difficult to wet. I assume that solvent in a conformal coating improves   
wetting and, also, that the wetting characteristics of 100% solids   
products are greatly affected by the selection of monomers.

Hope this information is of some help

Mary Davis
Sr. Material & Process Engineer
Alliant Techsystems
206-356-3311
[log in to unmask]
 ----------
From:  CINDY KEMP ORLANDO ISC *8-306-6[SMTP:[log in to unmask]]
Sent:  Wednesday, October 23, 1996 5:44 AM
To:  Technet
Cc:  GSPLASV
Subject:  FWD: pwb-cca : conformal coating

Technet,

I'm forwarding this from a colleague.

Cindy Kemp
Lockheed Martin
Orlando, FL
**************************************************************************  
*****
*

From: [log in to unmask]
Date: Tue, 22 Oct 96 16:35:22 EDT
Message-Id: <[log in to unmask]>
To: [log in to unmask]
Subject: pwb-cca : conformal coating
Sender: [log in to unmask]
Precedence: bulk

hi group;

We are finding that in designs using conformal coating (UR) over plastic   
parts
that the cc is dewetting on the parts.

Question - is this seen as a problem ?

I am proposing that a note should be added to our designs as follows:

 CONFORMAL COAT DEWETTING ON PLASTIC ENCAPSULATED MICROCIRCUIT
 COMPONENT BODY IS ACCEPTABLE.

If there is a better way please let me here from you.

What is the real would doing ?


   thanx; Skip Greb
========================================================================
To learn more about this email exploder (majordomo), including how to
unsubscribe from this list, send e-mail to: [log in to unmask]
and place the command "help" in the body of the message - subject may be
left blank. Help can also be obtained on the WWW at
http://www.epic.lmco.com/Majordomo
========================================================================

**************************************************************************  
*
* TechNet mail list is provided as a service by IPC using SmartList v3.05   
*
**************************************************************************  
*
* To unsubscribe from this list at any time, send a message to:   
          *
* [log in to unmask] with <subject: unsubscribe> and no text.   
       *
**************************************************************************  
*



***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************

------------------------------

Date: 	Wed, 30 Oct 1996 11:26:12 -0800
From: "John Parsons" <[log in to unmask]>
To: "IPC TechNet" <[log in to unmask]>
Subject: DES, ASSY: Fine Pitch SMT
Message-Id: <[log in to unmask]>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit

A customer of ours has the following question(s):

1.  We are currently working mostly on surface mount cards and a concern
has arisen about solder flowing thru via's during the reflow process and
shorting surface mount pads together as well as shorting vias's together.
Via spec is currently 0.020 hole on a 0.040 pad.

a)  Does anyone seen this sort of problem associated particularily with
20mil pitch devices.
b)  If yes, does simply covering the vias with mask (LPI) eliminate this
problem, or must the vias be "Capped".

Any insight into this would be appreciated.

Regards
John Parsons
Pre-Production Engineering
Circuit Graphics Ltd.

***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************

------------------------------

Date: Wed, 30 Oct 1996 11:53:04 -0800
From: Bob Metcalf <[log in to unmask]>
To: [log in to unmask], [log in to unmask]
Subject:  RE: Relative issues concerning the tenting of vias -Reply
Message-Id: <[log in to unmask]>

We have many customers who routinely use 50 micron dry film solder
mask. We are even supplying some customers 25 micron dry film solder
mask. The limiting factor is circuit height. Traditional pattern plating 
will
often produce high conductors requiring additional mask thickness. In
Japan the reason dry film mask works so well is for the most part they
are panel plating with conductor heights usually not exceeding 50-60
microns.

The other thing to consider is application of the dry film solder mask. We
have new applications that allow the mask to conform to circuitry more
efficiently than traditional dry film solder mask. A 50 micron film can
encapsulate what formally required a 100 micron dry film solder mask.

Bob Metcalf
Morton Electronic Materials

>>> Goldman, Patricia J. <[log in to unmask]> 10/30/96 11:24am >>>

Now, maybe I'm wrong here, but I do believe that 50 microns = 2 mils. 
That  doesn't seem very thin to me....  The problem I believe with
extremely thin  dry films is usually pinholes and some sort of  difficulty
with the carrier. 
 The other problem with a dry film solder mask is getting conformity over 
relatively high traces with a very thin film.  You may hear from some df 
solder mask people on this one.

There were some recent lengthy discussions on plugging vias.  Best go
to the  web site and browse on that subject.

Patty
 ----------
From: TechNet-request
To: TechNet
Subject: Relative issues concerning the tenting of vias
Date: Monday, October 28, 1996 11:12PM

As the industry heads towards a point where surface mount technology
and dense circuit topologies are the norm it becomes imperative to revisit
the issue of to tent or not to tent.
With liquid photoimageable solder masks still being an unreliable means of
providing a uniformly conformal coating over copper features, is dry film
the answer?  Have strides been made of late in this technology which
allay the inherent problems of process control, media thickness, and
contamination due to oily residue buildup and its subsequent promise of
sundry other problems such as delimitation?  Has MIL SPEC 2000 (is rev
C out yet?) acquiesced in the use of either the dry film or lpi tenting of
vias as an acceptable practice?
I have seen boards produced in the orient with dry film thicknesses as
small as 50 microns.  Are these available in the US or is their use
shunned as being too thin?  The suppliers I have spoken with cite
minimum thicknesses in the >1 mil range.  This lays bare other problems,
such as potential tomb stoning while providing the opposite benefit of
creating a solder well around SMT lands.  The latter stated benefit is of
dubious value if paste stencils and application coupled with fine tuned
temperature control bogs down production and outweighs the reliability
bonus, if any.
It is a confusing issue and all the literature I can find seems to be 
horribly
out of date (the latest being an excerpt from PC FAB magazine dated
1991.  I am reasonably certain things have progressed since then, but
where is the data?
If you have any insights into this well masked dilemma, I for one could
stand the relief. [log in to unmask]

***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.       
*
***************************************************************************

***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.       
*
***************************************************************************



***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************

------------------------------

Date: Wed, 30 Oct 1996 13:25:31 -0600
From: [log in to unmask]
To: [log in to unmask]
Subject: ASSY:  Passive Devices under PLCC-44
Message-ID: <[log in to unmask]>
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Description: cc:Mail note part

     Address,
     
     I asked this question once before but probably didn't send correctly.  
     Is there anyone out there currently assembling chip resistors 
     underneath a J-leaded PLCC-44?  IPC-SM-782 does not spec the contact 
     lead to package distance.  I may be placing any of following 0603, 
     0805 and 0402 under the PLCC-44.
     
     Concerns I have:
     
     Without running a Thermal Profile with thermocouples will there be 
     enough heat to reflow the underlying chip or will the PLCC-44 device 
     rob it of heat?
     
     I'm I correct to assume the chip may crack should the pressure from 
     the head be too strong and obviously if there is enough room under the 
     device.  The assembler does not have a variable pressure control for 
     pick and place heads.
     
     And again is it feasible.  I have other options but none that are as 
     efficient as the one described.
     
     Please advise,
     
     John Gulley - QA
     
      

***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************

------------------------------

Date: Wed, 30 Oct 96 14:10:58 MST
From: [log in to unmask]
To: [log in to unmask]
Subject: FAB: Post-Lam Bake
Message-Id: <[log in to unmask]>

     What is a typical cycle for baking FR4, rigid, multilayers after 
     lamination?
     Why is this bake necessary?
     
     Any information would be greatly appreciated.

***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************

------------------------------

Date: Wed, 30 Oct 96 09:38:41 PST
From: "MohitGujral" <[log in to unmask]>
To: [log in to unmask]
Subject: Re: Vision System
Message-Id: <[log in to unmask]>

Hi Paresh....it depends how much u want to spend on a system like this,
there are a lot of them out there....according do ur needs i would say 
try out 

1> The theta group...seems to be a good machine....can do a bunch of stuff
they have three models Vista, Verifier and Optima....all these m/c have
the capability of real time SPC 
contact : ken Gribble
1 800 877 4874

2> The other company would be CR Technologies checks and verifies
Component presence/Orientation/Correct part and a bunch of other stuff..
Ph 714 448 0443
Hope this helps...i don't work for them and the standard disclaimer
applies....

mohit gujral

______________________________ Reply Separator 
_________________________________
Subject: Vision System
Author:  [log in to unmask] at uucpmail
Date:    10/30/96 8:53 AM

Good morning,

We have a concern regarding potential of  higher DPU at auto insertion 
when verifier is non-functional and operator loads one or more tubes of 
IC's or reels of components in the feeder. There is also a second 
related issue.  Even when the verifier is working well, operator is 
required to do a TQC on the PCB's as final check at that operation. 
This TQC is very time consuming which in turns reduces productivity.

I need your help in gaining knowledge and gathering information on any 
available vision system that would check for correct component in 
place.  An efficient system should free us from worry when verifier is 
non-functional and increases productivity.

Thanks for your help,
Paresh Patel
Oneac Corp.
847-816-6000 x-704.

*************************************************************************** 
* TechNet mail list is provided as a service by IPC using SmartList v3.05 * 
*************************************************************************** 
* To unsubscribe from this list at any time, send a message to:           * 
* [log in to unmask] with <subject: unsubscribe> and no text.        * 
***************************************************************************


***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************

------------------------------

Date: Wed, 30 Oct 96 13:03:55 
From: "Lynch, Lyn" <[log in to unmask]>
To: [log in to unmask] (CINDY KEMP ORLANDO ISC *8-306-6),
        [log in to unmask] (Technet), [log in to unmask]
Subject: Re[2]: pwb-cca : conformal coating
Message-Id: <[log in to unmask]>

Here we go again. All polymers, including plastics used to encapsulate 
electronics, as well as all conformal coatings, have well documented 
moisture 
vapor transmission rates. So, it makes no practical difference if the body 
of 
the device is coated, or not. Although your customer amy demand complete 
coating
of all surfaces on the board, if all exposed, conductive elements (leads 
and 
traces) are properly coated you will see no performance differences between 
boards with fully coated packages, and those with poorly coated packages.
If, on the other hand, you rely on the coating to "stake" the device to the 
board, to reduce vibration induced stress, then the device should be clean 
and 
coated. 

By the way, if you are not dipping your boards in the conformal coating,
I can assure you that you have uncoated surfaces somewhere on your board.

The opinions stated herein are not necessarily those of my employer.

Lyn R. Lynch
602.276.7361
[log in to unmask]
______________________________ Reply Separator 
_________________________________
Subject: RE: pwb-cca : conformal coating
Author:  [log in to unmask] at internet
Date:    10/30/96 12:23 PM


Thought I would add my two cents on the subject.
     
We are on a steep learning curve with respect to plastic parts.  One of   
several problems we encountered was dewetting at conformal coat. After   
working the issue for a while, we have concluded that plastic parts are   
inherently harder to wet than other parts and that some conformal coating   
products are inherently better at wetting than others.  A formula to   
predict wetting would be nice but we so far we have progressed by trial   
and error.
     
The first coating that we tried was our   
 'good-old-stand-by-coat-anything', single component, solvent based,   
MIL-I-47058 UR.  Wetting was spotty and unpredictable.
     
We were also working with the manufacturer of our solvent based coating   
to evaluate their line of UV cure, single component, 100% solids,   
MIL-I-47058 coatings.  We tried their 100% solids coatings on the plastic   
parts and had even less success. The dewetting of the plastic parts was a   
text book example of 'crawling' or  'retraction'.  Wetting was good   
everywhere except on the plastic parts.  Cleaning the plastic parts did   
not improve wetting.
     
     
We then tried a Dymax product, a single component, 100% solids, UV cure,   
acrylic copolymer (meets AR, ER, and UR.)  We had much better luck.  So   
far this product coats plastic parts without a problem.  It seems very   
robust with respect to plastic parts.
     
Wetting is a function of the surface tension of the liquid and the free   
energy of the surface.  If I understand the physics, wetting will not   
occur unless the surface tension of the liquid is less than the critical   
free energy of the solid.   I assume that these parameters vary over a   
wide range for the materials in question - but there is very little data   
available.  Even without the mold release variable, plastic is a low   
energy surface relative to metal and ceramic and, is therefore, more   
difficult to wet. I assume that solvent in a conformal coating improves   
wetting and, also, that the wetting characteristics of 100% solids   
products are greatly affected by the selection of monomers.
     
Hope this information is of some help
     
Mary Davis
Sr. Material & Process Engineer
Alliant Techsystems
206-356-3311
[log in to unmask]
 ----------
From:  CINDY KEMP ORLANDO ISC *8-306-6[SMTP:[log in to unmask]] 
Sent:  Wednesday, October 23, 1996 5:44 AM
To:  Technet
Cc:  GSPLASV
Subject:  FWD: pwb-cca : conformal coating
     
Technet,
     
I'm forwarding this from a colleague.
     
Cindy Kemp
Lockheed Martin
Orlando, FL
**************************************************************************  
*****
*
     
From: [log in to unmask]
Date: Tue, 22 Oct 96 16:35:22 EDT
Message-Id: <[log in to unmask]> 
To: [log in to unmask]
Subject: pwb-cca : conformal coating 
Sender: [log in to unmask]
Precedence: bulk
     
hi group;
     
We are finding that in designs using conformal coating (UR) over plastic   
parts
that the cc is dewetting on the parts.
     
Question - is this seen as a problem ?
     
I am proposing that a note should be added to our designs as follows:
     
 CONFORMAL COAT DEWETTING ON PLASTIC ENCAPSULATED MICROCIRCUIT 
 COMPONENT BODY IS ACCEPTABLE.
     
If there is a better way please let me here from you.
     
What is the real would doing ?
     
     
   thanx; Skip Greb
======================================================================== 
To learn more about this email exploder (majordomo), including how to 
unsubscribe from this list, send e-mail to: [log in to unmask]
and place the command "help" in the body of the message - subject may be 
left blank. Help can also be obtained on the WWW at 
http://www.epic.lmco.com/Majordomo
========================================================================
     
**************************************************************************  
*
* TechNet mail list is provided as a service by IPC using SmartList v3.05   
*
**************************************************************************  
*
* To unsubscribe from this list at any time, send a message to:   
          *
* [log in to unmask] with <subject: unsubscribe> and no text.   
       *
**************************************************************************  
*
     
     
     
*************************************************************************** 
* TechNet mail list is provided as a service by IPC using SmartList v3.05 * 
*************************************************************************** 
* To unsubscribe from this list at any time, send a message to:           * 
* [log in to unmask] with <subject: unsubscribe> and no text.        * 
***************************************************************************
     

***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************

------------------------------

Date: Wed, 30 Oct 1996 16:30:03 -0500
From: [log in to unmask] (Howard Feldmesser)
To: [log in to unmask] (Technet), [log in to unmask] (Mary Davis)
Subject: RE: pwb-cca : conformal coating
Message-Id: <[log in to unmask]>
Content-Type: text/plain; charset="us-ascii"
Content-Length: 956

Technetters:
        I've been reading the mail about coating plastic parts with mostly
a scientific curiosity type of interest level.  Now the issue will smack me
in the face as we try to coat PEMs for spacecraft use.  We're planning to
use Parylene to coat the assemblies for many reasons.  Does anyone have any
experience with Parylene on PEMs to help me along?  I will share our
results when I get some.
        Howard Feldmesser
        The Johns Hopkins University Applied Physics Laboratory


***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************

------------------------------

Date: Wed, 30 Oct 96 16:42:00 CST
From: "Schreiber, Carole" <[log in to unmask]>
To: "[log in to unmask]" <[log in to unmask]>
Subject: job posting
Message-ID: <[log in to unmask]>

Hi,
I am a recruiter for Electronic Assembly Corporation in Neenah, WI and I 
have some open positions.   How would I go about getting a job posted on 
the 
technet?  What are the costs and all other details? We are a member of IPC.
I would appreciate it if you could send me the details.

I can be e-mailed at [log in to unmask]

Thank you,
Carole S.

***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************

------------------------------

Date: Wed, 30 Oct 96 14:47:17 PST
From: "Bob Bickers" <[log in to unmask]>
To: [log in to unmask]
Subject: bismuth based solder
Message-Id: <[log in to unmask]>

     I have an assembly that requires a second solder process over a large 
     area of the assembly, thus the need for a lower temp solder.  I'm 
     looking for the risks associated with the use of bismuth solder 
     systems.  Thanks in advance for your help.
     
     Bob Bickers

***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************

------------------------------

Date: Wed, 30 Oct 1996 16:12:26 -0700
From: [log in to unmask] (Gary Peterson)
To: [log in to unmask]
Subject: ASSY: GEN: Popcorning during reflow
Message-Id: <[log in to unmask]>

Has anyone else observed the following?

We have a CMOS PQFP that was "popcorned" during reflow soldering.  Cross
sectioning clearly shows that the die, along with its mounting epoxy, has
separated from the leadframe by approximately .002" (2-mils).  There were
no visible cracks in the plastic overmold material at 500X magnification.

The problem is...some parts don't fail until after a few thermal cycles
of 0 to 50 degrees C during live circuit testing at-speed.  And...we can
make them recover after failing by merely baking the entire board from
1.5 to 24 hours at 125 degrees C.  The baking fix is not permanent and the
parts eventually fail again.  Some parts don't work initially but do work
after the above baking treatment (we discovered, quite by accident, that 
baking 
"fixes" the parts).


My question is...is the popcorning likely to be the only problem with
                 these parts?

Has anyone else been able to anneal a part so that it will recover from
popcorning?  Doesn't sound likely to me!

Gary P.
---
                                  Gary D. Peterson
    _/_/_/   _/    _/  _/        SANDIA NATIONAL LABORATORIES     _/_/_/
   _/       _/_/  _/  _/        P.O. Box 5800, M/S 0503            _/_/
  _/_/_/   _/ _/ _/  _/        Albuquerque, NM 87185-0503     _/_/_/_/_/_/
     _/   _/  _/_/  _/        Phone: (505)844-6980           _/  _/_/  _/
_/_/_/   _/    _/  _/_/_/_/  FAX: (505)844-2925             _/  _/_/  _/
                            E-Mail: [log in to unmask]        _/_/_/



***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************

--------------------------------
End of TechNet-d Digest V96 Issue #101
**************************************

***************************************************************************
* TechNet mail list is provided as a service by IPC using SmartList v3.05 *
***************************************************************************
* To unsubscribe from this list at any time, send a message to:           *
* [log in to unmask] with <subject: unsubscribe> and no text.        *
***************************************************************************



ATOM RSS1 RSS2