I find ball shear to be more telling than plucking, but maybe I'm all
plucked up :-)
Back to a more serious note, I've always found accelerated aging using the
foundation of the Ahrenius equation's time/temp equivalencies to be the MOST
telling for product reliability. I view initial wire pull, shear, pluck as
only a very small part of the overall reliability picture.
Interpreting the data on a Weibull plot is also quite helpful.
Forgive me if I repeat some of what has already been said. I admit to not
taking the time to read all the previous posts. Time constraint.
Steve Creswick
Sr Associate - Balanced Enterprise Solutions
http://www.linkedin.com/in/stevencreswick
616 834 1883
-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Ahmad, Syed
Sent: Thursday, April 24, 2014 10:22 AM
To: [log in to unmask]
Subject: Re: [TN] Wire-bonding AOI
Lot of information has been brought forward in this exchange. TechNet is a
source of a lot of learning at least for me. I can see the value of AOI in
production.
Another test: As wire pull may not lift the ball or stitch bonds every time,
in place of wire pull, ball pluck can also be done. An attachment can grab
the ball and pull it with a wire pull machine.
-----Original Message-----
From: Stadem, Richard D. [mailto:[log in to unmask]]
Sent: Thursday, April 24, 2014 9:13 AM
To: TechNet E-Mail Forum; Ahmad, Syed
Subject: RE: [TN] Wire-bonding AOI
Well spoken, Syed!
The AOI will only provide process control over the gold wire routing, the
bond placement location, and can detect missing bonds or damaged bondwires
and variation in studbump or ball size. But it does that very well. Here at
GD we perform AOI on various devices and standard assemblies using a MVP. At
one of the other clients I work for, there are many wirebonders and
diebonders and there we manufacture literally millions of components, and
that is where we are using AOI as a method of process control for bond
routings, hybrid assembly, BGA assembly and ball attach, and other
technologies. The AOI is a great method for inspecting for ball size and
presence and spacing, as well as for wirebond presence, routing, bump ball
size, and location away from adjacent conductors, just prior to
encapsulation.
Pull testing is performed on a sample basis for strength of wirebonds. Shear
testing is performed for solderball and microcomponent attachment process
control.
The MVP AOI is used in all of these processes. I strongly recommend you
contact a MVP sales rep and discuss with them what you want it to do. More
than likely, you can find several other things to check with the AOI. I even
use it for epoxy staking inspection. In the past I also used it for
connector pin inspection, where all of the pins needed to be located a
certain distance from both a local reference point, but also with respect to
each other. I was able to pretend the top of each pin was an 00105 chip, and
the machine found 100% of any pins out of position. As far as I know, the
MVP is the only AOI that has programmable focus in the Z-axis, but I could
be wrong about that.
-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Ahmad, Syed
Sent: Thursday, April 24, 2014 8:42 AM
To: [log in to unmask]
Subject: Re: [TN] Wire-bonding AOI
AOI will only tell the correct placement of a wire bond.
In the semiconductor packaging world, the degradation of the Au-Al bond is
measured in a number of ways.
Mechanically, it is measured by pulling the wire or by shearing the Au-Al
joint.
The results of the wire pull test are given in the form of the pull strength
in gf (gram-force) values and the corresponding failure modes.
In a wire pull test, when the wire fails in the mid span, the only thing it
tells is that the gold ball bonds to the aluminum bonding pads on the die
and to the bonding pads on the substrate or leadframe are stronger than the
wire itself. As-made unstressed bonds made using an optimized ball bonding
process are expected to fail at mid-span as the gold ball bond area to the
bonding pad on the chip is twice or larger than the cross sectional area of
the wire itself. Failure at other locations may only occur if the process is
not set up properly.
If the failure is at the neck of the bond to the die or the substrate or
frame, again it does not provide much insight into the condition of the
Au-Al bond and just indicates weakness at the neck of the ball bond or at
the neck of the stitch bond. Reasons for such failures have been studied
extensively over the years.
The only time the pull test provides information about the Au-Al interface
is when the Au-Al bond lifts off the aluminum bonding pad on the chip. What
it means is that either it was an un-optimized bond to begin with, or the
Au-Al bond had deteriorated to the extent that its interfacial area or
strength had been reduced significantly.
Additionally, mechanical testing is destructive.
The ball shear test can cause failure through Au or Al metals, can rip
silicon out from under the Al, or shear through the intermetallic eliciting
the weakest part of the structure. This test again is destructive though it
is more useful than the wire pull test in determining the condition of the
Au-Al interface.
Electrically, the health of the joint is looked at by measuring resistance
changes in the interconnect. As the Au-Al interface degrades, the resistance
of the bond increases. This method is non-destructive and can be used for
monitoring the Au-Al bond continuously.
The initial resistance and later changes in the Au-Al wire bonds are
influenced by more than one factor.
The quality of initial bond affects the resistance of the interconnect.
The form of the interaction area depends upon the process. The amount of
interaction area also influences the resistance of the bond.
The shape of the interaction area affects the way the intermetallics and
voids will progress at the bond site.
After the bond has been made, changes in resistance are affected by the
gradual formation of the intermetallics, their by-products and their
interaction with materials present in its surroundings.
Various intermetallics have different rates of formation and differ in their
resistivity and strength until they reach an equilibrium.
The diffusion of the metals occurs at their interfaces.
The intermetallic formation process also generates Kinkendall voids. These
voids contribute to changes in the resistance of the interconnect.
Intermetallics and voids formation is also affected by the gases present in
the environment depending on the level of their access to the system.
The gases affecting the intermetallic and void rates can come from the
environment of an un-encapsulated bond.
The sources of gaseous elements are also provided by the encapsulant
materials around the bonds.
When the gases are entering from the sides of interface, the void
proliferation tends to spread laterally.
-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Joyce Koo
Sent: Thursday, April 24, 2014 6:09 AM
To: [log in to unmask]
Subject: [TN] Fw: [TN] Wire-bonding AOI
"All professional skills are mastered by critiqued practice" - by Schon.
Original Message
From: Joyce Koo <[log in to unmask]>
Sent: Wednesday, April 23, 2014 11:14 PM
To: Stadem, Richard D.
Subject: Re: [TN] Wire-bonding AOI
There are developing process regarding the wire bonding. Usually, you do
ball size inspection for consistency, ball share for initial strength, aging
and wet chemical etch plus cross section for intermetallic coverage and
thickness (depend upon the die metallization and thickness, composition of
both your wire and pads). Table tear usually good enough for second bond
strength, the modern machine will automatically identify "no stick". Wire
loop is based on visual, with current state of the art bonds per device
reach to 1000 or more and multi-teir stack type, aoi? With speed of bonder,
very difficult. As for process monitoring, bond 15 at every 4 hour interval,
and do wire pull (stitch peel) and ball shear, will be the same guide after
a fully developed process. (or you can ask for 5 bonds per direction, just
in case you have problem optimize different travel directions). My 1.8
cents.
"All professional skills are mastered by critiqued practice" - by Schon.
Original Message
From: Stadem, Richard D.
Sent: Wednesday, April 23, 2014 5:27 PM
To: TechNet E-Mail Forum; Joyce Koo
Subject: RE: [TN] Wire-bonding AOI
IMC is pull-tested on a sample basis. Bondwire routing and location is 100%
machine inspected (AOI). You use the correct combination of sampling and
visual inspection to ensure that both the wire routing is correct, and the
bond meets the strength requirement.
-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Joyce Koo
Sent: Wednesday, April 23, 2014 12:46 PM
To: [log in to unmask]
Subject: Re: [TN] Wire-bonding AOI
Why you need that? For al or Au? It is self testing by the bonder, if I am
missed something.ý Besides, you always need to do imc. Inspection, aoi is
not good enough. (can't see imc) .
"All professional skills are mastered by critiqued practice" - by Schon.
Original Message
From: Ioan TEMPEA
Sent: Wednesday, April 23, 2014 11:46 AM
To: [log in to unmask]
Reply To: TechNet E-Mail Forum
Subject: [TN] Wire-bonding AOI
Dear Technos,
I need info regarding AOI machines with wire bonding inspection capability.
Efficiency of the technology, suppliers, price tags, table top vs. inline,
personal experiences, can they also do regular SMT AOI, anything you can
tell me, on-line or off-line.
Any particular experience with this one?
http://www.viscom.com/asia/products/wire-bond-inspection-aoi/s6053bo-vs6056b
o-inline-aoi/
Thanks,
Ioan Tempea, P. Eng.
Manufacturing Engineer
Satellite Systems
MDA
21025 TransCanada Highway
Ste-Anne-de-Bellevue, Qc, Canada H9X 3R2
Tel: +1-514-457-2150-3556
www.mdacorporation.com<http://www.mdacorporation.com>
This e-mail, and any attachments, are intended solely for the use of the
intended recipient(s) and may contain legally privileged, proprietary and/or
confidential information. Any use, disclosure, dissemination, distribution
or copying of this e-mail and any attachments for any purposes that have not
been specifically authorized by the sender is strictly prohibited. If you
are not the intended recipient, please immediately notify the sender by
reply e-mail and permanently delete all copies and attachments.
______________________________________________________________________
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]
______________________________________________________________________
______________________________________________________________________
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]
______________________________________________________________________
______________________________________________________________________
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]
______________________________________________________________________
______________________________________________________________________
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]
______________________________________________________________________
______________________________________________________________________
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]
______________________________________________________________________
______________________________________________________________________
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]
______________________________________________________________________
|
