Absolutely! I love destruction too. I use to be rather violent,
diligently using diamond saw, pliers, drills, any tools to have a look
on inside. However, ain't possible in this case. Each substrate costs
1,500 USD and we have a whole bunch of them for deciding: useful or
rejection. And I have the baton. Want to change over, but noone wants
it.
You have a point, Wayne. The gold is extremely dense to be such thin a
thickfilm. If the adhesion was indisputable, we'd went on and not been
so fussy about the thickness. Now, there is another aspect too, which
you are familiar with, namely the adhesion vs. thickness vs. firing
conditions. I've studied internal reports from Tanaka, duPont, ESL,
Heraeus and some more, and the curves demonstrate, clearly, the
functions for the thickness. Varies a lot depending on what mixture and
what manufacture. But there is a general tendency that the adhesion has
a lower value at a minimum thickness, has an increase, a maximum, and
lowers when you pass a maximum thickness. I was afraid, that we had to
do with such a catastrophic minimum caused by too thin a paste. I know
what you will say now, the spread isn't so dramatic, that you get a
magnitude or more lower adhesion. You are maybe right. There is probably
another parameter that has a stronger impact. Or it may be a combination
of wrong firing and that parameter.
It's a mixed binder, and the frit is the dominant one, according to our
substrate maker.
I'm sorry for bla-bla-bla a lot, but this matter is kind of emergency
case, and I have just a few days to come with something intelligent.
Now, if I compare with others, e.g. Trikeman's VSOP, there is a
significant difference (if not the very parameter I search) and it's the
MECHANICAL binding mechanism. Brand X, or Y or Z have zillions of
microsized glass 'hooks' from the tape, glass protrusions that stretch
into any void or opening in the thickfilm, thus working as very strong
bonds. The thickfilm that we talk about as problematic has no such glass
'hooks' or at least very few. I've demonstrated this to the substrate
maker, but they don't seem to know what I'm pointing at, or do they
dislike my ambition to find a process anomaly.
Thanks Wayne. Even if I didn't get a solution, at least you pushed me in
the right direction. Good.
Inge
-----Ursprungligt meddelande-----
Från: TechNet [mailto:[log in to unmask]] För Wayne Thayer
Skickat: den 30 mars 2007 12:18
Till: [log in to unmask]
Ämne: Re: [TN] Cofired gold
Hi Inge!
I always prefer destruction. Something about the finality of the job.
Anyway, I think you are probably going in the wrong direction. If the
deposits are not porous, then the coverage is probably OK, and as Steve
points out, the print process doesn't have a big window for thickness
you can deposit, anyway--certainly not the factor of 3 you are
suggesting you may be thin by.
If you are ball bonding to LTCC, generally the ball goes on the part
you've mounted to it and the stitch goes on the LTCC. Are you just
doing some kind of qualification test?
LTCC is magical stuff! It's a miracle that the stuff holds together
during processing at all, especially during the densification phase of
the firing process. I'd be looking for a chemistry cause for your
trouble, not print thickness. If the firing environment got just a tad
low in oxygen, other phases of the process may have stolen the oxygen
molecules necessary for the bond between the metal and the dielectric.
That's exactly what happened to me several years ago with a big LTCC
job: I fired on low temperature copper on one side, and on the other
side the previously fired and well-adhered gold actually fell off!
So it may be that the LTCC manufacturer was lying about the adhesion you
should have, or it may be that the substrate manufacturer messed up the
firing environment slightly.
Wayne
>>> [log in to unmask] >>>
Ingemar,
Hmmm. 2 microns huh! Do you think this is a 'direct write' process?
Sounds awful thin for a stencil don't you think?
I was using a Sonix acoustic microscope. With a 200+ MHz transducer
[think it was 280 MHz] I could reasonably well pick out the outline of
major circuitry features on the underside of a flip chip, if I had it
gated properly.
Seeing internal traces within an LTCC body was always dependent upon how
many internal ground planes and RF shields were placed inside the body
as well. Vias above the target region were always bad news. You are on
the surface. You may have a chance, but am not so sure you will be able
to differentiate the top and bottom surface of the conductor when it is
that thin. Definitely go for the reflected, high-frequency image.
Give the challenge to a really good acoustic microscope guy with a good
machine and decent software :-)
Steve
-----Original Message-----
From: Hfjord [mailto:[log in to unmask]]
Sent: Thursday, March 29, 2007 12:34 PM
To: 'TechNet E-Mail Forum'; Creswick, Steven
Subject: SV: [TN] Cofired gold
24 layer stack with cofired AuAg top. Lifts at ball bond pull tests,
insufficient adhesion, I would say. Should take 4,000 psi. Smashed ball
approx. 2 mil diameter. If the gold lifts at 10 grams, then you are >
one magnitude below 4,000 psi. SEM'd edges of lifted lines= 2
micrometers thick. Paste maker recommends >7 micrometers as fired.
Substrate maker denies, that's why I'm asking for method to check all
incoming parts without destruction of the gold.
Abcd recommended confocal laser, don't think that will work. VHF
transducer, not bad idea! Good. Will test that next week. Thanks.
Inge
-----Ursprungligt meddelande-----
Från: TechNet [mailto:[log in to unmask]] För Creswick, Steven
Skickat: den 29 mars 2007 16:14
Till: [log in to unmask]
Ämne: Re: [TN] Cofired gold
Ingemar,
Are you speaking of a standard, single print of gold which has been
co-fired into the body of the part, or possibly of a stack of inks which
has been co-fired in at one time?
XRF comes to mind, but it will require special 'standards', and the prep
of the standards would result in the destruction of a sample of parts to
construct them ... and even then I am not sure how successful you will
be.
It sounds as though a simple profilometer won't get you what you desire
- especially since the bulk of the conductor is below the surface. I
occasionally did internal inspection of LTCC parts back in the olden
days - but that was looking more for omissions and commissions.
I believe that acoustically, with a very high frequencurface of the
conductor.
The margin of error [due to surface imperfections] may be quite large,
however.
Can you get them to tell you what their stencil thickness was and based
on solids content have some idea what you may have ended up with?
Will ponder further
Steve
-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Hernefjord Ingemar
Sent: Thursday, March 29, 2007 9:39 AM
To: [log in to unmask]
Subject: [TN] Cofired gold
Trikeman or other gnu,
Anyone know of a non destructive method to check cofired topgold
thickness (Au over LTCC)?
Inge
THE INFORMATION CONTAINED IN THIS E-MAIL MESSAGE AND ANY ATTACHMENTS
SENT FROM GENTEX CORPORATION IS GENTEX CONFIDENTIAL INFORMATION INTENDED
ONLY FOR THE PERSONAL USE OF THE INDIVIDUAL OR ENTITY NAMED ABOVE. If
you are not the intended recipient, you are hereby notified that any
review, distribution, or copying of this communication is strictly
prohibited. If you have received this communication in error, please
immediately notify the sender by return e-mail, and delete this e-mail
message and any attachments from your computer.
---------------------------------------------------
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
-----------------------------------------------------
THE INFORMATION CONTAINED IN THIS E-MAIL MESSAGE AND ANY ATTACHMENTS
SENT FROM GENTEX CORPORATION IS GENTEX CONFIDENTIAL INFORMATION INTENDED
ONLY FOR THE PERSONAL USE OF THE INDIVIDUAL OR ENTITY NAMED ABOVE. If
you are not the intended recipient, you are hereby notified that any
review, distribution, or copying of this communication is strictly
prohibited. If you have received this communication in error, please
immediately notify the sender by return e-mail, and delete this e-mail
message and any attachments from your computer.
---------------------------------------------------
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
-----------------------------------------------------
|
