Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Françoise von Trapp (00:00):
This
episode of the 3D Insights
podcast is sponsored by NordsenTest and Inspection.
Leaders in acoustic, opticaland x-ray inspection and
metrology systems for thesemiconductor and SMT
manufacturing markets.
Nordsen's next-generation 3Dx-ray inspection software,
dynamic Planar CT, revealsincredible details for Nordsen's
(00:20):
zero-defect strategy.
Nordsen's advanced software istwo times faster than planar CT,
with enhanced detail quality,superior reconstruction and a
larger field of view forincreased coverage and shorter
cycle times all to improveyields, processes and
productivity.
Learn more at Nordsoncom.
Hi there, I'm Francoise vonTrapp, and this is the 3D
(00:55):
Insights Podcast.
Hi everyone, in this week'sepisode of the 3D Insights
Podcast we're continuing a deepdive into the importance of 100%
inspection in the semiconductormanufacturing environment, and
this time we're going to focuson advancements in x-ray
inspection and specifically for3D imaging.
(01:15):
So I am speaking with AndrewMathers of Nord's Intestine
Inspection to learn about what'sgoing on in this area and pick
his brain on his knowledge.
So welcome to the podcast,andrew.
Andrew Mathers (01:28):
Thank you so
much for having me on.
Françoise von Trapp (01:30):
Before we
dive in.
I understand you're new to thecompany at Nordson.
Can you tell me a little bitabout yourself and the role that
you've taken there?
Andrew Mathers (01:38):
Sure, I joined
Nordson about six months ago as
a principal product line managerfor their automated x-ray
inspection product line and mybackground is really in x-ray.
I've been in the x-ray worldfor about 10 years now, as
starting out as a postdoctoral,postgraduate at university using
(01:59):
x-ray to look at plants, tolook at them in terms of their
root development underground andtheir leaf development above
ground, and then moved heavilyinto the industry side and have
been in the industry for abouteight years working for
different x-ray systemmanufacturers and recently
joined Nordson to look atelectronics.
Françoise von Trapp (02:20):
So you're
bringing this really varied
background in.
Are you finding the needs ofelectronic inspection a lot
different than the other areasthat you've studied?
Andrew Mathers (02:33):
No, I don't
think so.
I think there's a commonalitybetween what a lot of customers
want from x-ray inspection.
They want high resolution, sothey need to be able to see very
, very small things down to themicron level.
They obviously want it to bedone quickly, they want it to be
cost effective and, most of all, they want it to help drive
(02:56):
better products or betterunderstanding of what they're
imaging their imaging, and thatwe know is becoming very
important in this industry, asthe devices that we're
inspecting are getting more andmore complex and therefore more
and more.
Françoise von Trapp (03:13):
I don't
want to say valuable, but when
you capture the defects can meanthe difference down the road of
having to start from scratch orbeing able to repair something
and the loss there right.
Andrew Mathers (03:28):
Absolutely, and
I think this brings us purely
around to 3D imaging, because inthe past many of our customers,
certainly many people operatingin this space, have
historically relied on 2Dradiographic imaging for
inspection.
As manufacturing requirementsbecome ever more stringent and
the quality and the reliabilityof devices has to continue to
(03:51):
increase to meet the demandsfrom the suppliers, from the end
customers and the market ingeneral, 3d imaging, driving
that higher performance,inspection, that better
resolution, better quality, isbecoming a much, much more
common requirement to supportthese efforts.
Françoise von Trapp (04:08):
So how
long has 3D imaging been an
option in the electronic spacethen?
Andrew Mathers (04:13):
3D imaging has
been around for a while in
electronics imaging.
However, the quality of it hasalways been something that's
quite difficult to achieve.
One of the problems with thisstyle of imaging is you're
looking at a planar object,you're looking at something
that's typically very flat, likea semiconductor wafer or a
printed circuit board, and toimage these objects you use
(04:36):
planar scanning, or laminographyas it's also known, and this
can suffer from lots of imagingartifacts, which are features or
distortions that are present inyour images that aren't truly
present in reality in the sample, for example, in electronic
samples.
These make it very, verydifficult to clearly visualize
(04:57):
between the different layers andthe devices.
However, the 3D imaging side ofthings allows us to very, very
clearly discriminate betweenthese layers and see things in
clear detail, which isincredibly valuable.
Françoise von Trapp (05:14):
Okay.
So I want to step back just oneminute because I want to make a
clear distinction when you'retalking about planar, you're
talking about using planarimages to create a 3D image
versus planar for a 2D image.
Andrew Mathers (05:27):
So when I say
planar, I mean using planar CT,
which is designed for flatsamples.
Ok, so you are building a 3Dvolume of your images at the end
, but because of the nature ofthe sample being so flat and the
way that you have to rotatearound it to get that computed
tomography approach, it tends tobe quite difficult to avoid
(05:49):
having quite nasty artifacts inthe image.
Françoise von Trapp (05:52):
Okay, and
the artifacts aren't actually on
the sample that you're using.
These are actually introducedbecause of the imaging process.
Andrew Mathers (06:00):
Absolutely.
It's generated features ordistortions that are present in
the image that aren't there inreality in the sample.
But you're quite right, theycome from the imaging approach
that you're using, but it is thebest imaging approach for these
planar samples.
Françoise von Trapp (06:18):
So with
laminography they're taking
different slices of images andthen putting them together using
software.
Andrew Mathers (06:25):
So laminography
, you tend to have a static
sample and you rotate yoursource and detector, or just
your detector, in a motionaround the point that you're
inspecting your region ofinterest within the sample and
you still reconstruct and buildup a 3D volume.
However, you tend to inspectthis in slices in the Z-axis.
(06:47):
So you're looking top down.
You have very good clarity inthose slices, but once you look
at a vertical slice and you lookat it from side on, you tend to
see quite bad artifacts thatdisrupt the data.
Françoise von Trapp (06:59):
So for a
3D x-ray, what are we getting as
the output?
Andrew Mathers (07:04):
You're getting
data which is non-destructively
looking at the structure ofsomething, so you won't just get
surface detail.
Yes, indeed, you're getting thesurface externally and internal
surfaces and anything inbetween.
Françoise von Trapp (07:16):
Okay, and
the image that's output is a 3D
image.
Andrew Mathers (07:22):
The
reconstruction of the scan is a
3D volume, but you can intersectthat volume at any angle.
You like to create a 2D imageas a cross-section.
Françoise von Trapp (07:33):
How has
Nordson addressed these
challenges that you've beendescribing about planar CT?
Andrew Mathers (07:40):
So in the past
we've had a solution to acquire
images called planar CT, whichtakes images by moving our
detector around and keeping thesample and the x-ray source
static.
But, as I mentioned, there is alot of artifacts that these
types of scans can suffer from.
So one thing that we'vereleased more recently and we're
(08:01):
really excited about is dynamicplanar CT, and this is our next
generation of 3D inspection.
It's still a planar approach,it's still laminography, but it
has some major advancements andimprovements over planar CT
solution.
Françoise von Trapp (08:17):
So how
does it work?
What does it do?
Andrew Mathers (08:19):
It's a
software-based product and it
offers high-speed,high-resolution 3D imaging of
electronic devices to identifycritical defects during the
production, particularly in thesemiconductor market and the SMT
market.
And it's a lot, lot faster thanplain SET, so it acquires
(08:42):
images at twice the speed, whichobviously has huge benefits to
the customer because it has muchhigher throughput and much
higher units per hour.
You certainly don't want thebottleneck in your manufacturing
process to be waiting for yourx-ray system to check the
quality of what you've produced.
You want to keep your line oryour fab moving at high speed to
(09:05):
get your products out the door,and that's what this enables.
Françoise von Trapp (09:15):
So with
planar CT that you were
describing, you were talkingabout laminography.
Does the dynamic planar CTstill use laminography or is it
a different image captureprocess?
Andrew Mathers (09:23):
It does still
use laminography, but it is a
slightly different image captureprocess.
It's far smoother and it uses amuch wider field of view, so
you can acquire images of a muchwider area, which means it can
be done a lot more quickly.
And it also means that,critically, you are dosing your
electronic samples up to a muchlower level, and X-ray dose is
(09:46):
really quite critical for a lotof x-ray inspection components
in the electronics world, likehigh bandwidth memory.
So we take an orbit with ourdetector which sweeps around the
point that we're inspecting andwe can take many, many more
projections, many, many more 2Dradiographs.
When we do this, and with abrand new reconstruction
(10:09):
algorithm to build these alltogether, we get a much higher
quality 3D model and data out ofthe other end.
Françoise von Trapp (10:17):
So how
does this impact the image
quality?
Andrew Mathers (10:23):
So it's greatly
improved.
As I mentioned, one of thebiggest challenges is these
artifacts that you can't seethat makes it very difficult to
see between different layers.
It also means if you've gotquite a dense component on top
of something else in yourcircuit board within your
semiconductor sample, it can bevery, very hard to see
underneath it.
And in 3D we can separate outthe different layers and we can
(10:48):
inspect each one individually,so we can see details that were
previously obscured or masked bythose dense or really complex
components.
Françoise von Trapp (10:57):
If it's
software based, does it remove
artifacts or does it preventthem from happening altogether?
Andrew Mathers (11:07):
The artifacts
to a certain degree with this
type of imaging are alwayspresent, but we take the images
in a slightly different way.
We take far more images.
We take far more individualshots from different angles, far
more projections and we have inthe past used a more iterative
approach to reconstruction,Whereas now, because we've taken
(11:27):
far more images, we canleverage different
reconstruction techniques andtherefore it builds a much
higher quality, detailed modelin 3D of the sample.
Françoise von Trapp (11:38):
Are you
implementing AI for this?
Andrew Mathers (11:42):
No, not
currently, not this product.
We are looking into AI in allsorts of areas, but I'm
currently not able to speakabout those.
Françoise von Trapp (11:50):
Okay this
tool.
Is it something that's usedmore in development, or is it
actually used in themanufacturing side of things?
Andrew Mathers (11:59):
It's used more
in the manufacturing side with
my product line with automatedX-ray inspection.
So it needs to happen extremelyquickly.
We're talking about capturing ascan of a specific area in
under three seconds typicallycloser to two seconds and it
needs to reconstruct veryquickly and perform the analysis
very quickly as well.
So this is for inlinemanufacturing very, very high
(12:21):
speed, short tack time andreally high throughput to keep
the samples moving down the lineor keep samples being analyzed
at high speed.
Françoise von Trapp (12:33):
So how do
you distinguish between when
something is automated versus, Iguess, manual?
Andrew Mathers (12:44):
I guess manual
Sure.
So in the manual x-ray worldyou'll typically have an
operator that loads samples on atray or individually into an
inspection area, into a systemand they will kick the scan off
themselves with the software,wait till the scan's completed
and then exchange the sample.
The difference here withautomated x-ray inspection is
samples are typically fed intothe systems on a conveyor belt
or via a loader, like a magazineloader which you might have in
(13:07):
a fab environment forsemiconductor, and the samples
are analyzed and then taken outthe other side or they come out
the same side back into theloader and then the next sample
is put in.
So it's completely automated.
There's no human interactionneeded.
It's all done autonomously.
Françoise von Trapp (13:24):
Okay, so
this is specifically designed
for use in volume manufacturingsemiconductor fabs and also in
SMT and surface mount technology.
Okay.
Andrew Mathers (13:36):
And what?
Françoise von Trapp (13:37):
about in
the advanced packaging
facilities.
Is it also suitable for there?
Andrew Mathers (13:41):
No, so that's a
different product line.
That's our metrology x-rayproduct line.
Anything that's a wafer that'spopulated, we would do with AXI.
But AXM, which is very much thewafer fab very, very beginning
level of building asemiconductor wafer, that would
(14:04):
be what would be used for thatand they don't use dynamic
planar CT for that.
Françoise von Trapp (14:09):
What are
some of the applications that
this tool is ideally suited for,and where in the manufacturing
process would you find the tool?
Andrew Mathers (14:18):
So this is
really good for the
semiconductor back end or SMTline in a factory.
It's really useful for any 3Dapplication with electronics
X-ray inspection.
But focus on those two markets.
It's really really useful forlooking at voiding present in
(14:39):
ball grid arrays, imprintedcircuit boards in SMT or looking
at voiding in micro bumps inflip chip devices in the
semiconductor backend.
Being able to visualize eachlayer of the sample and clearly
discriminate these voids in highresolution without artifacts
from the neighboring or dense,complex components around is
(15:00):
really really valuable for ourcustomers and is something they
desperately need to ensure highquality of their products.
Françoise von Trapp (15:06):
So,
basically, your customers are
able to identify these defectsand pull off the line the wafers
that are impacted, so that theydon't go through continued
processing before they end upbeing packaged into some device
that later fails.
Andrew Mathers (15:26):
Absolutely,
that's the entire.
Purpose is to ensure thatproducts that aren't of the
necessary quality to perform ata high level will be identified,
removed from the manufacturingprocess, to ensure that the
quality of what is going out tothe end customer is of the
highest quality and will performto the top level.
Françoise von Trapp (15:47):
Does the
tool provide feedback to help in
process development furtherdown the road?
If you keep finding the samevoid or defect at a certain
stage of the process, can theyuse that information later to,
for instance, improve processesand yields?
Andrew Mathers (16:07):
Absolutely.
And one of the big movements inthe manufacturing world at the
moment, and in particularelectronics, is something called
Quality 4.0 or Industry 4.0.
And this isn't just somethingwhere you're looking at the same
defect over time and reporting.
This is reporting in a veryconnected smart factory back to
the manufacturing defect overtime and this is a reporting in
a very connected smart factoryback to the manufacturing
process to try and remove thatdefect or critical problem in
(16:31):
the samples that are beingcreated.
So as a customer, not spendinga lot of time making that
adjustment, you can make it onthe fly and you can quickly
correct it and ensure theproduct coming through is of the
quality needed to make thefinal cut and be sent to the end
customer.
Françoise von Trapp (16:48):
Okay, so
to just wrap things up, tell me
about Nordson's automated x-raysolution.
Andrew Mathers (16:53):
So Nordson's
current portfolio of automated
x-ray solutions can be splitinto three categories.
We have three different systemseries that serve different
markets.
We have our XS series, whichfocuses on the semiconductor
market, the X series for SMT andthe X hash series, and that
supports two markets the powerhybrid market and the final
(17:17):
assembly, test and pack market.
But the thing that all thesesystems, these different markets
, have in common is that theyall provide high speed, high
resolution and, most critically,high quality imaging with
critical defect inspectionanalysis driven by our advanced
in-house algorithm library.
We have a direct presence inover 35 countries worldwide,
(17:41):
working with the biggest namesin the wide world of electronics
, and we have an install base inexcess of 2000 AXI systems
alone.
Françoise von Trapp (17:52):
What would
you say are the three key
takeaways you'd like listenersto come away with after
listening to this episode?
Andrew Mathers (17:58):
I think the
first one is that there is a
real movement going on at themoment away from 2D radiographic
imaging alone to judge whethersamples are of the highest
quality, and there is a movetowards 3D imaging to give more
detail and higher performance.
So, I would say, is the firstone.
Secondly, what this 3D imagingprovides is greater speed of
(18:19):
inspection, is greater speed ofinspection, so it's faster cycle
time, higher units per hour andthroughput, whilst it's also
reducing the dose given to thesamples while they're under
x-ray.
So this gives you moreconfident detection of defects
due to increased imaging qualityand greatly reduced presence of
artifacts.
And finally, it's a lower costof ownership to customers,
(18:44):
because we don't now needspecialized clamping apparatus
for these samples, which had tobe held in place previously,
because of the new, smoother waythat we acquire the images.
And, fundamentally, dynamicPlanar CT helps drive better
quality from our customers'products, supporting their
development and their success.
Françoise von Trapp (19:05):
Okay, well
, where can people go to learn
more then?
Andrew Mathers (19:09):
They can go to
our website, they can go to
nortoncom.
We also have a YouTube channeland some very nice new videos
just created on the DynamicPlanar CT we discussed today.
Françoise von Trapp (19:21):
So on any
of our socials, our websites and
just Google around on theinternet, Great, and you know
what we will do that Googlingfor them and put some links in
the show notes.
So thank you so much forjoining me today, Andrew.
It's been a pleasure.
Andrew Mathers (19:37):
You're very
welcome.
Thank you for having me on them
Françoise von Trapp (19:49):
Next time
on the 3D InCites podcast, join
me and my guest Pradyush Kamalfrom Siemens, as we discuss the
evolution of EDA since the dawnof 3D integration.
There's lots more to come, sotune in next time to the 3D
InCites podcast.
The 3D InCites podcast is aproduction of 3D InCites LLC.