TECHNET Archives

July 2003

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:
Ryan Grant <[log in to unmask]>
Reply To:
TechNet E-Mail Forum.
Date:
Fri, 11 Jul 2003 14:46:51 -0600
Content-Type:
text/plain
Parts/Attachments:
text/plain (185 lines)
Hi Joe,

I hesitate to say micro vias in pad guarantee voiding, but...  For most
solder pastes, the flux vehicle becomes liquid during the flux
activation stage of reflow.  The liquid will wet where it can, AND also
fill up the "trough" a via naturally makes, before the solder becomes
liquid and can fill the trough.  Driving the flux sideways out of the
trough is impossible, and there is now a lot of flux to force by weight
of liquid solder, to the top of the pad where it has a chance of
escaping the joint.  However, remember that "hydraulics" can support a
lot of weight, so when the solder becomes liquid over the top of the
liquid flux, it becomes semi trapped in the joint.  At this point, it
acts much like the shock absorber in your car.  In order for the piston
to move, the fluid in the shock absorber must exchange from one side of
the piston to the other side.  (otherwise you have hydraulic lock).  In
the same way, the liquid solder must "exchange" from on top of the flux
to the bottom while it is in the cylinder of the via.  Getting this
exchange to happen BEFORE the BGA ball becomes liquid
is...well...difficult.  I won't say its impossible to achieve void free
joints with micro vias in pad, but I've never seen one.  

However, if the via is filled and plated over so you have a flat pad,
then of course it would be no different than a normal pad. 

Ryan Grant

-----Original Message-----
From: Macko, Joe @ IEC [mailto:[log in to unmask]] 
Sent: Friday, July 11, 2003 1:50 PM
To: [log in to unmask]
Subject: Re: [TN] BGA Voids


Mr. Ryan,

Great explanation of void formation.

Does blind via's in pad always guarantee some voiding from your
experience?

joe

-----Original Message-----
From: Ryan Grant [mailto:[log in to unmask]]
Sent: Thursday, July 10, 2003 5:07 PM
To: [log in to unmask]
Subject: Re: [TN] BGA Viods


Hi Carl,

I'm a little late in replying to your question, but anyways... BGA voids
occur because the flux vehicle becomes trapped in the liquid solder.
Remember of course that the flux vehicle is intermixed with the solder
powder, and when the powder becomes liquid, it does not become liquid
all at the same time.  In fact, it may take several seconds for a paste
deposit to become fully liquid.  In the time between the powder changing
from solid to liquid, the flux must be driven from the center of the
paste deposit to the outside.  Since the liquid will fall to the bottom
of the paste deposit, and it is more dense than the flux, the flux will
rise above the liquid solder.  Surface tension will try to minimize the
surface area of the solder that is liquid, helping to close any
"tunnels" of flux that may exist within the liquid and drive the flux to
the outside of the paste deposit.  However, if the tunnel becomes
closed, there will not be any forces to "push" the flux (and now the
trapped volatiles) in any particular direction except straight up
towards the BGA ball.  This is the reason "flux only" BGA processing
never has voids.

As you can imagine, there are several factors that can help or hinder
the entrapment of flux.  First is the viscosity of the "liquid" flux in
the reflow oven.  (The viscosity of the flux during the dwell time has
nothing to do with the viscosity of the paste during printing.)  Reflow
profile can certainly have an affect with the viscosity of the flux in
the oven, however, each flux vehicle has its own unique way of
responding.  Some will gum up the longer it is heated, some will
dissipate a little more, but most stay pretty consistent.  This makes
some solder pastes worse at voiding than others.  Also, keep in mind
that for many flux vehicles, as the flux removes oxidation, it turns
into a gummy sludge.  This is why nitrogen helps reduce voids; less
oxidation to remove, less gummy flux residue.

Another factor is how fast your spike to reflow occurs.  As the solder
powder changes from solid to liquid, it draws heat away from the
surrounding powder, requiring additional heat to be input into the
system.  The faster you put heat into the system, the faster the solder
deposit can go completely liquid.  However, this also allows less time
for the flux to escape to the outside of the solder deposit.  Of course,
once the flux is entrapped, it may still volatize and cause a small void
to become large.  This is why extended dwell above reflow has been known
to increase the size of the void, and move them from being positioned at
the PCB pad level to the BGA side.

Another factor is the distance the flux must travel to get out of the
solder deposit.  Yet another factor is the volume of flux that must be
displaced.  Those two factors are the "size of pad and amount of solder
paste" mentioned by your colleagues.

Honestly, reflow profiles and paste deposit sizes are minor in
controlling voids.  The biggest impact on voids is the flux vehicle
itself.  If you have the ability to change solder paste models, that is
where you will find the biggest change in void size.

Ryan Grant

-----Original Message-----
From: Carlr Ray [mailto:[log in to unmask]]
Sent: Wednesday, July 02, 2003 9:19 AM
To: [log in to unmask]
Subject: [TN] BGA Viods


Hello All,
  I have a question (more like a poll I guess), concerning voiding of
BGA devices.
  I have PBGA 480 IO paremeter grid device and I am seeing lots of
voiding. Per the x-ray I am well within the IPC spec of 35% but I am
still having failures at ICT (OPENS). Once I reviewed the profile I
noticed we were not hitting the dewell time for flux activiation
recommended by the solder paste vender (min 60 seconds) and we were
seeing an extended time above 183c. We have changed the profile but some
of my companions feel that the size of the pad and amount of solder
paste is effecting the voiding. I disagree with them and feel that not
activiating the fluxes and extended time above liquidous is a major
contibutor to the voids.
  So now my question, what impact does the pads have with the voiding of
the spheres on the BGA?

Carl Ray
Sr. Manufacturing Engineer
Huntsville, AL 35807
Phone: 256-882-4800 ext. 8845
Cell: 256-990-1990

---------------------------------------------------
Technet Mail List provided as a free service by IPC using LISTSERV 1.8e
To unsubscribe, send a message to [log in to unmask] with following text
in the BODY (NOT the subject field): SIGNOFF Technet To temporarily halt
or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET
Technet NOMAIL or (MAIL) To receive ONE mailing per day of all the
posts: send e-mail to [log in to unmask]: SET Technet Digest Search the
archives of previous posts at: http://listserv.ipc.org/archives Please
visit IPC web site http://www.ipc.org/html/forum.htm for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700
ext.5315
-----------------------------------------------------

---------------------------------------------------
Technet Mail List provided as a free service by IPC using LISTSERV 1.8e
To unsubscribe, send a message to [log in to unmask] with following text
in the BODY (NOT the subject field): SIGNOFF Technet To temporarily halt
or (re-start) delivery of Technet send e-mail to
[log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to
[log in to unmask]: SET Technet Digest
Search the archives of previous posts at:
http://listserv.ipc.org/archives Please visit IPC web site
http://www.ipc.org/html/forum.htm for additional information, or contact
Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
-----------------------------------------------------

---------------------------------------------------
Technet Mail List provided as a free service by IPC using LISTSERV 1.8e
To unsubscribe, send a message to [log in to unmask] with following text
in the BODY (NOT the subject field): SIGNOFF Technet To temporarily halt
or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET
Technet NOMAIL or (MAIL) To receive ONE mailing per day of all the
posts: send e-mail to [log in to unmask]: SET Technet Digest Search the
archives of previous posts at: http://listserv.ipc.org/archives Please
visit IPC web site http://www.ipc.org/html/forum.htm for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700
ext.5315
-----------------------------------------------------

---------------------------------------------------
Technet Mail List provided as a free service by IPC using LISTSERV 1.8e
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives
Please visit IPC web site http://www.ipc.org/html/forum.htm for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
-----------------------------------------------------

ATOM RSS1 RSS2