Thank you. Regards Steve Kelly

-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Steven Creswick
Sent: August-04-09 1:01 PM
To: [log in to unmask]
Subject: Re: [TN] Wire bond

Steve,

I likely read too much into your question, but let me take a shot. 

As you said, we will assume the plating is correct for whatever type of
wirebonding technique is utilized.


In theory, if the bond will fit on the pad, you can use whatever bonding
technique you desire.  The wire bonder tech/engineer has multiple options at
his/her disposal with respect to bonding tools, temperature, frequency, etc.


General Statement - Although not necessarily ABSOLUTELY true, in general,
the narrower the pad, the more suited it is to wedge bonding [the caveat
here is that with the right tool and the right bonding conditions, the
second bond of a ball bonder does not have to be extremely wide either...
It is just that one does not have to struggle as much with the wedge bonding
technique to get it to 'fit'.  I hope that makes some sense.]  

The problem that comes along with wedge bonding to narrow/tightly spaced
pads is that the major axis of the pads MUST essentially be coincident with
the line between the respective two bond pads.

Steve, I hope this is not too much info

Steve C



Additional - If one were aluminum wedge bonding on a 2 mil [0.002"] wide pad
with a 1.25 mil dia aluminum wire, it would be nice to have a pad which was
at least 2-4 mils long [although one can get away with a very short bond
foot if necessary, if just makes process control all that much more
difficult].  If bonding with the 1.25 mil dia wire, the deformed [as-bonded]
wedge can be ~1.5 mil wide.  There is a great deal of internal preference
here.  The Mil Std allowed for some [IMHO] horrendously over-bonded wedge
bonds.  

Bottom line being that so long as your wire bond did not hang over the edge
of the pad - you should be good to go.  

Now then, this same pad would not be desired for 2, 3 or 5 mil wire due to
the mechanical 'fit' problems.

Switching over to thermosonic gold ball bonding.  It may be a bit iffy to
place the ball for a 1 mil wire on this pad as the ball size will be on the
order of 2-2.5 mils in dia.  As with everything, there are caveats.  One can
definitely run with smaller balls, but a lot depends on the surface, the
machine and the process control.  One could also place the stitch
bond/second bond for that same 1 mil gold wire on your 2 mil wide pad.
Again the Mil Std allowed for horrendously over-bonded wedges, but one does
not need/desire to go there to get a reliable bond.  With the right bonding
tool, I could easily get both thermosonic gold ball and wedge bonds on your
2 mil wide pads.

So, in the ideal summary, one might say that these pads somewhat dictate the
wedge bonding approach [but one could switch over to 0.0007" dia gold, and
ball bond that!  But that is a hassle too!  It just depends on our
assumptions].


Okay - we've had enough of the ideal world scenario.  Now the real world!

When you get down to splitting hairs like this, you had better make darn
sure the pattern is etched VERY uniformly and consistently [I am thinking of
automatic wire bonding]. The bonder plans on the center of the bonding pad
[assuming you have a multiplicity of pads around a device] to be at location
X,Y,Z - if it isn't, due to etching irregularities, stretch, etc. - the bond
may be off the pad.  This applies to wedge bonding, or ball bonding.  If you
are speaking of manual bonding, then it just goes back to the operator's
technique - should be able to do without too much difficulty assuming good
lighting and scope set-up.

Most wire bonders can place bonds to within 4-5µm accuracy, so it [bond
placement accuracy] typically is not a wire bonder problem.

In a nutshell - I've seen some pretty wild etch variation over the years
which would suggest you are walking on thin ice here [with respect to
automatic wire bonding.  Not single point references.].

Now then, if your pattern were what one might call a 'single point', the
wire bonder's vision system could perform a pattern rec on that site and
locate it smack-dab dead-on regardless of [within reason] where it lay.
Typically, however, we would have a central die attach pad with multiple
peripheral bonding pads. As alluded to earlier, the bonder mathematically
has these pads located in memory.  Their locations are all referenced to
usually two master identifying features.  If the little pads move around
independently of the master identifying features - all hope is lost for
accurate bond placement [the machine will attempt to place the bonds to
where the pads mathematically SHOULD be].  Naturally, one uses features that
are etched at the same time as the bonding pads as the master identifying
points [reference point/fiducials].
 
With thermosonic gold wedge, or gold ball bonding, very tiny pads can
seemingly 'swim' during the bonding process allowing for an ultrasonic
dampening effect as the bond pad [which should be rigidly clamped during
bonding] seems to move around a bit.  Bonding on flex can present
similar/worse issues.  I have played with arrays of various size pads on a
given laminate, on the same bonder, at the same time and found that below
some size, the bonding process becomes erratic.  It may not always show up
as a problem at the bonder, but will typically manifest as a 'tail' on a
Weibull plot, etc.  Indicating that something is amiss.  Large pads have no
problem!  Moral of the story would be to run with the largest possible pads,
at the lowest possible temperature, or use the highest Tg laminate....


Surface - Many times tiny pads have funky edges, rounded shoulders, etc
which minimize the flat bondable area which a wire bond would like to see
for reliable bonding.


Strength - One could be somewhat naïve and place a 2, 3, 5 mil aluminum wire
on your pads [okay, so it overhangs a bit...].  If I were to perform a wire
pull test, there is a great chance I will lift your pad and trace from the
laminate!!



So Steve, to some extent it boils down that tried and true "Deweyism" - it
all depends....

  ....  On the accuracy and consistency of etching, etch quality, Tg, and
how familiar the person is with the wire bonding equipment, and just how
capable that equipment happens to be.  

For a manual Al [RT] wedge bonding approach I would have no problem with
getting wedge bonds on your pads.  

For a manual thermosonic bonding approach, I could easily get them there,
but may not necessary guarantee they are good bonds.



-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Steve Kelly
Sent: Tuesday, August 04, 2009 11:36 AM
To: [log in to unmask]
Subject: [TN] Wire bond

Good Morning,

Does the size of a wire bond pad (in this case with a perfect etch .002
wide) have any effect on whether the wire bonder does Al wedge or
Thermosonic gold( I know the gold plating is different) - just asking about
the process. Thanks. Steve Kelly



Steve Kelly (PLEASE NOTE NEW E-MAIL ADDRESS)

(416) 750-8433 (work)

(416) 750-0016 (fax)

(416) 577-8433 (cell)




---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 15.0
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/contentpage.asp?Pageid=4.3.16
for additional information, or contact Keach Sasamori at [log in to unmask] or
847-615-7100 ext.2815
-----------------------------------------------------

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 15.0
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/contentpage.asp?Pageid=4.3.16
for additional information, or contact Keach Sasamori at [log in to unmask] or
847-615-7100 ext.2815
-----------------------------------------------------

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 15.0
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/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-615-7100 ext.2815
-----------------------------------------------------