Breaking wafers? We saw it also today

Mary Tang mtang at snf.stanford.edu
Thu Sep 16 07:46:06 PDT 2004


Hi Wei & Ben --

I'm sorry this happened...  I saw another note from you later in the day,
saying this happened again, is that correct?  Did you do your street alignment
in the center of the wafer (as opposed to the flat) according to Eric's note?
I have to admit that I have not experienced this problem, but feedback from
other wafersaw users suggests that he is correct.  Are you using a slower
entry speed?  And do you have to align to the bottom of the wafer?  If no to
both questions, then I would suggest aligning at the middle of the wafer
(Eric's suggestion) and using 200 mm as the diameter for the top 1/3 of the
wafer (Alan's suggestion.)  Either of these suggestions seem to work for users
who have experienced this and tried.  Both might give you yet more security.
Jim Kruger had another suggestion:  to use square dimensions instead of the
round ones.  So far, he has found this to work.  All these suggestions make
sense, because the problem appears to be in how the wafersaw computer maps out
the area to be diced -- if it is a circle that is off-center (i.e., off-set in
the Y-direction), then the machine thinks it has a narrower area to cut
towards the top of the wafer, when it is actually positioned near the middle.

I'll definitely add these suggestions to the training and will add them to the
operating procedures as well.  Please let me know if these procedure changes
do not work for you, as then we'll need to investigate further.  (I am
somewhat reluctant to expend a lot of time investigating now, because I
understand now that this was a problem from the beginning, when the machine
was installed, but that K&S and the SNF staff at the time gave up on it
because they could not find the cause and the computer was no longer supported
by K&S -- and because specifying a 150 mm or larger substrate eliminated the
problem.  We could consider upgrading the computer -- I understand from Len,
who looked into this, that this requires a substantial rebuilding of the
system and runs about $60K.)

Again, please let me know if any of these suggestions works for you.

Thanks,

Mary




Eric Perozziello wrote:

> It goes to the last place aligned, so if you re-align at the center,
> it will start going to the center.
>
> -Eric
>
> On Wed, 15 Sep 2004, Ben Jian wrote:
>
> > Hi Mary,
> >
> > Today Wei was doing wafer docing and we saw the same phenomenon, for the
> > first time ever. We were trying to dice the top-most part of a four inch
> > wafer. The blade came down on our wafer and the blade was broken.
> > Luckily, our wafer is OK.
> >
> > We suspect that this problem might have something to do with two other
> > things. First, maybe a sensor is defective (we did see some warning
> > message like "sensor bad" when the problem occurred but did not write it
> > down carefully, sorry). Second, the fact that the machine does not go to
> > the true center of the wafer stage when you hit "Align" but to the
> > bottom part of the wafer chuck. Maybe one of these two is the source of
> > the problem?
> >
> > Regards.
> >
> > Ben Jian
> >
> > -----Original Message-----
> > From: Mary Tang [mailto:mtang at snf.stanford.edu]
> > Sent: Tuesday, September 07, 2004 3:06 PM
> > To: Myers, Alan M
> > Cc: wafersaw at snf.stanford.edu; rcrane; Paul Rissman; jhaydon
> > Subject: Re: Breaking wafers?
> >
> >
> > Thanks,
> >
> > I'd absolutely no idea about this.  I'd been told (by Len, the saw
> > expert) that setting the diameter larger was helpful for the occasional
> > problem in which the saw didn't cut all the way across, but maybe this
> > is related to starting in the middle...  This is easy to do in
> > principle, but does mean we wouldn't be able to take advantage of a slow
> > entry speed (though perhaps small insurance against breaking wafers...)
> >
> > I'll try it out and see if I can likewise induce/cure the problem.
> >
> > And unless I hear otherwise, I'll plan on adding Alan's method to the
> > operating procedures and training...
> >
> > Mary
> >
> > "Myers, Alan M" wrote:
> >
> > > Mary,
> > >         I used to have this problem every time I use the saw.  I
> > > learned the secret is to set the wafer diameter 200 mm whenever I get
> > > near the top of the wafer.  It may take a little longer, but at least
> > > the blades and the wafers remain intact.
> > >
> > > Alan
> > >
> > > -----Original Message-----
> > > From: Mary Tang [mailto:mtang at snf.stanford.edu]
> > > Sent: Tuesday, September 07, 2004 2:37 PM
> > > To: wafersaw at snf.stanford.edu
> > > Cc: rcrane; Paul Rissman; jhaydon
> > > Subject: Breaking wafers?
> > >
> > > Hi everyone --
> > >
> > > "rosti" dropped by my office to say that he was thinking about using
> > > the wafersaw, but was concerned about a problem he had experienced a
> > > couple of times in which the saw BROKE his wafer.  He says that
> > > instead of starting on the edge, in these instances, the saw blade
> > > dropped onto the middle of the wafer and thus broke his wafers.  The
> > > last instance that he observed was some time back last October or so.
> > >
> > > Now I'm worried!  Usually, when a wafer is ready for dicing, you've
> > > expended a lot of time and energy into it -- you simply can't afford
> > > to have it smashed.  I'm also really concerned because I've been
> > > training on the wafersaw for a couple of years now, but have not
> > > observed this problem nor had I heard about it from anyone (and here I
> >
> > > thought I knew all the machine's quirks).  I had heard that it used to
> >
> > > be a problem, but not seen since the system was basically rebuilt a
> > > few years ago. Now, I'm hearing otherwise...
> > >
> > > Has anyone else had this problem?  If so, could you let me know
> > > approximately when it happened and how often (if you were "lucky",
> > > like rosti, and got hit more than once?)  And maybe even the
> > > circumstances when it occurred (if you remember).  Your feedback will
> > > help us decide what to do (if the darned thing is breaking wafers
> > > intermittently, we may need to take it off-line...)
> > >
> > > And, most importantly, if you observe this or any other problem,
> > > please, please, please report it on Coral!  (I'm going back several
> > > years, and don't happen to see anything about the machine breaking
> > > wafers...)
> > >
> > > Thanks,
> > >
> > > Mary
> > >
> > > --
> > > Mary X. Tang, Ph.D.
> > > Stanford Nanofabrication Facility
> > > CIS Room 136, Mail Code 4070
> > > Stanford, CA  94305
> > > (650)723-9980
> > > mtang at stanford.edu
> > > http://snf.stanford.edu
> >
> > --
> > Mary X. Tang, Ph.D.
> > Stanford Nanofabrication Facility
> > CIS Room 136, Mail Code 4070
> > Stanford, CA  94305
> > (650)723-9980
> > mtang at stanford.edu
> > http://snf.stanford.edu
> >
> >
> >
> >

--
Mary X. Tang, Ph.D.
Stanford Nanofabrication Facility
CIS Room 136, Mail Code 4070
Stanford, CA  94305
(650)723-9980
mtang at stanford.edu
http://snf.stanford.edu





More information about the wafersaw mailing list