Very good info Mike, thanks a lot! I do have a question though...
Tom talked about doing your own Dosimetery in #5....how does one go about this?
Tom said it was easy, just get your own dosimetery coupons, and that the test only
takes a couple of minutes. Well, during lunch time, I looked, and looked, and looked,
and couldn't find any dosimetery coupons. A lot of dosimetery badges etc., but those
all need to be sent somewhere to get read...
Kind regards,
-Steve Gregory-
Senior Process Engineer
LaBarge Incorporated
Tulsa, Oklahoma
(918) 459-2285
(918) 459-2350 FAX
"Green, Mike" <[log in to unmask]>
Sent by: TechNet <[log in to unmask]>
09/30/2005 11:50 AM
Please respond to TechNet E-Mail Forum <[log in to unmask]>; Please respond to "Green, Mike" <[log in to unmask]>
To: [log in to unmask]@SMTP@Exchange
cc: (bcc: Stephen R Gregory/LABARGE)
Subject: [TN] FW: [TN] X-ray inspection damaging components...
-----Original Message-----
From: Clifford, Tom
Sent: Thursday, September 29, 2005 12:14 PM
Mike -
Thanks for the view of all the TechNet thread on X-Ray damage. Very
good discussions. Here's my summary and advice, which you might share
with that forum:
1) "Voltage" settings /thresholds, etc alone mean NOTHING. Maybe worse
than nothing, if somebody uses only voltage values to try to influence
somebody. Total cumulative delivered dose is a function of voltage,
certainly, but also amps (the dose is linear vs. power), and especially
distance, optics, and certainly time, and off-angle position, and use of
aluminum shields. All units differ, most shops' set-ups differ. Only
if everything else is constant and known, only then can somebody say
credible things like 10% higher voltage will deliver 10% greater dose
rate; or double-the-current, double-the-dose; or 120 volts is the right
operating limit, etc. Note that many X-Ray units have interacting
operating parameters, so guidelines on dose rate vs. amps/volts, can get
tricky.
2) The only way to convey credibility or to assure credibility is to
define the situation properly. Steer way clear of folks who make wide
sweeping reassurance statements. Be very wary of facile simplifications.
3) Ask your X-Ray supplier to look at your exact set-up, at your
magnifications/ distances, at several operating levels; and give you
actual received-dose data. Require that your sub-contract assemblers
provide the same data.
4) Be especially mindful of exposure time. That's linear. Cooking a
component under the beam while you futz with the image-enhancement or
print-outs or referee decisions looking at the monitor, etc can all be
silently detrimental. Set up a verification protocol and administrative
controls, even using timers and buzzers, if necessary.
5) Do your own dosimetry. It's real easy. Dosimetry coupons are cheap
and the tests take only minutes. You set the operating parameters and
position the coupons to simulate your procedure, plus off-centerline
positioning. You can get enough data to develop your shop protocols, to
reassure your customers, and/or to guide/qualify your suppliers.
6)Components for sure differ in their vulnerability. I'm not confident
that an assembler (like me) can generate that sort of vulnerability
data. Get credible max-allowable-accumulative-dose values from device
manufacturers, or from NASA or academia, not from X-Ray salesmen. New IC
constructions, with finer geometries and odd new materials may be even
more vulnerable. We use 100 Rad(sil) as a good conservative value ...
that's a factor of 10 to 1000 conservative. but we are reappraising that
threshold.
7) As mentioned in the messages below, the image you get at a given set
of X-Ray conditions is influenced by lids and thermal-management add-ons
and the board construction. The received-dose value is influenced also
by board density and other-side parts' construction if the BGA is
flip-side. All that needs to be modeled and built into your operating
protocol.
8) Damage will not reveal itself real-time. You might think you have not
damaged the part; hopefully, you haven't. Damage can show up later.
Please do not tell your customers or yourselves that your process is OK
because you observe no damage. Parts will not erupt in flames. Damage
can be latent, sporadic, and subtle, and could easily escape simple
continuity or functional tests.
9) Note also that many folks are also doing XRF measurements upstream to
characterize lead finish (in the lead-free, tin-whisker hysteria). That
dose can be significant, too; and is additive to what the part might see
later in assembly in-process verification
10)IPC should step in to standardize the discussion, to provide
procedures for characterizing your unit (settings and dosimetry tests)
as well as to offer operating limits re accumulative dose.
11) IPC should also trigger and sponsor data and statements from the
semiconductor/device/JEDEC folks, regarding standard, generic, and
specific damage thresholds.
my three cents worth Tom Clifford
-----Original Message-----
From: TechNet On Behalf Of Joyce Koo
Jack, here is the reference regarding the issues.
jk
-----Original Message-----
From: TechNet Behalf Of Jack Crawford
Nick/Jack/Ken, if you look at the technet archives of the last several
days you'll see a flock of msgs on this topic.
Does JEDEC have any test methods or other guidance in place or proposed
that would assist users with this issue? Because it is so closely
focused on components I believe this would be more a JEDEC issue. Thanks
Jack
-----Original Message-----
From: TechNet On Behalf Of Stadem, Richard
I bet the folks at IPC are scrambling as well.
----Original Message-----
From: TechNet On Behalf Of Franklin Asbell
(chuckle)
-----Original Message-----
From: TechNet On Behalf Of Jason Gregory
They probably all scrambled off to secret meetings as soon as Steve's
Email surfaced.
Jason Gregory
-----Original Message-----
From: TechNet Behalf Of Franklin Asbell
Of course, and I'm thinking much of this would already have been
performed, and data available, from the component suppliers. I never
suggest taking on such a task without involving them, they're a valuable
resource...and it's surprising none have spoken up yet. Franklin
-----Original Message-----
From: TechNet On Behalf Of Stadem, Richard
The correct approach here is to establish a standard for how to test a
particular component's susceptibility to X-ray exposure. The test
methods would need to be established under one standard, the rating
system under another standard. The rating system would need to include a
method of measurement of exposure, both short term maximum exposure in
Rems, and accumulative limits.
-----Original Message-----
From: TechNet On Behalf Of Jason Gregory
Surely one could functionally test a known good part, then subject it to
incrementally increasing amounts of X-rays to eventually arrive a
failure point, then log the results...but could such a test be deemed
accurate? I would imagine that different components (same part number,
package, etc.) would suffer different levels of failure, however slight
or large. Probably is true that the component makers just want to be
sure, so they use substantially lower numbers to ensure that their
products are "safe" and usable. But, at what point can the engineer,
like Steve, say "130 kV is too much, but 90 kV is just fine". Just
gambling. Jason Gregory
----Original Message-----
From: TechNet On Behalf Of Michael Sewell
Not to mention that the .050" core makes a good shield when x-rayed
with the BGA on the top side, which is why we were turning the Kv up
anyway. Now, if its on the bottom side during x-ray and nearly flush
to the x-ray source.....
----Original Message-----
From: TechNet Behalf Of Guy Ramsey
This is one spooky thread. I know that NASA did some work on this
subject,
but I left my references back at ACI. There was nothing in it suggesting
problems with X-ray inspection. They had some evidence that airport
inspection equipment might damage sensitive components.
-----Original Message-----
From: TechNet Behalf Of Bev Christian
And what about the possibility of resulting intermittent failures? Bev
RIM
-----Original Message-----
From: TechNet Behalf Of Franklin Asbell
My first question at such a technical meeting would be
How much exposure will damage the part?
I'm sure if you place the part in the chamber, turn on the x-ray and go
to lunch you might have some problems. But would seconds, cause damage?
What might be happening is component vendors want to protect their
product and obviously so, so they place tighter than required standards
on them, what is the time to failure under 90 kv at 20 mA, and at 50 mA,
and at 90 mA? What is time to failure under 40 kv at 90 mA for 5
seconds? Or 30 seconds?
Just some things I would try to answer during such an IPC technical
activity.
Franklin
-----Original Message-----
From: TechNet On Behalf Of Stadem, Richard
Good Morning, Stephen!
This is good information. So now we have component vendors telling us
some of their components should not be overexposed to X-rays. Was there
any type of notation on the part print? The industry needs a standard.
We need to have some type of rating system for components that can be
damaged by X-rays. Jack, is there any activity going on with this within
IPC?
-----Original Message-----
From: TechNet Behalf Of Dehoyos, Ramon
I would not be surprised if the X ray does damage some
components depending on the intensity and the duration of the exposure.
Part of the process in building a chip or IC is ion implantation. A
material is doped positive or negative with respective ions, i.e. Ion
implanter. Layer thickness is getting smaller every year, in the
Angstroms, (1 micrometer = 10000 Angstroms = 0.001 millimeters =
0.0000394 inches ). The length of a transistor gate is 60 nanometers. I
have also heard not to take some sensitive electronic equipment through
X ray at the air ports due to the intensity of the exposure since
several years ago from people that work with me and take electronic
equipment all the time.
Does anyone know someone that can do that type of experimentation or
know someone that already have? Ramon
-----Original Message-----
From: TechNet On Behalf Of Charles Caswell
Steve, I have heard of X-Ray erasing some memory but not actually
damaging the component.
-----Original Message-----
From: Stephen R Gregory
Hi Phil! The BGA in question is a small one, only 100-balls, and a
plastic part with no heat-sink. It's on the bottom side of the board,
and has to go through reflow one more time when the topside is built.
As I said, everything looks fine at x-ray...it is inspected pretty
closely too. Our customer is the one who is worried about x-ray, not us.
They can't get the device to accept a program on some boards, and they
are thinking that possibly x-ray damaged the device.
Kind regards,
========================================
To: TechNet@IPC Stephen R Gregory/LABARGE)
Hi Steve, I haven't ever heard of or experienced any issues with an
x-ray inspection system causing damage to a chip. It is hard to
believe that x-rays momentarily exposed to a chip could cause damage. I
have seen some Altera BGAs that have difficulty getting good solder
joint reflow due to the heat sinks on top of them, the PWBs 24 layers,
and the sheer number of solderballs (1500+). With your .050" copper
core, I would look at the thermal reflow and make sure that all the
balls on the BGAs have fully
collapsed. Was this BGA on the second side or first side reflowed? It
is tough sometimes to tell where the problem is if you don't have a
board level test in place.
Phil
-----Original Message-----
From: TechNet Behalf Of Stadem, Richard
I know of a few cases where over-exposure to X-rays has damaged certain
components, most often CMOS components. Some CMOS EEPROM components can
even be damaged by overexposure to UV light.
It is not very likely, but it is definitely quite possible. It has
happened, and it is a concern.
If you are looking at a particular area on a PWA, the adjacent areas are
also being exposed. If you continue to go back and look at that area,
the total accumulated dose can hurt certain CMOS components (and other
types as well).
It is not very likely, but there are steps you can take to prevent
inadvertent damage from overexposure.
Again, read Tom Clifford's article on the subject.
-----Original Message-----
From: TechNet On Behalf Of Steve Gregory
I submitted a question to Altera via their email support center about
x-ray inspection possibly damaging a EPM7256EEFI100-7 device. The device
is a EEPROM made using an advanced CMOS process. Here is our exchange so
far: Request No: 10515187 Status: Close-Pending Date Opened (PDT):
9/26/2005
>Description: I am doing x-ray inspection of a EPM7256AEFI100-7 device
after board assembly.
>The PCB has a 1.27 mm thick copper core in the center of it, so it
>requires
that a high kv x-ray must
>be done in order to see the solder connections of the device. The x-ray
source power is around
>120-130 kv. Will this x-ray power level damage the device?,
===================
>To Customer:
>Hi Stephen, Typically assembly uses 90 KV using 10 to 20 mA with a
few second for inspection of IC package. Our recommendation will be a
pro-rated 130 KV with 7 mA to 13 mA for a few second. Note that Prolong
exposure under X-ray will definitely damage the device. I hope it's
helpful to you. Best Regards, Daniel Cao
My question is; doesn't the figure of 90 KV at 10-20 mA seem pretty low?
My current Glenbrook RTX-113 operates at 35-52 KV at 20-50 mA, and
higher-end Glenbrook systems will pump 70-90 KV at 90-100 mA. Right now,
I'm using a Fein Focus Tiger 160 for x-ray inspection of some copper
cored assemblies that we're building because I have to, that's the only
way we can inspect the BGAs. The copper core in the PCB won't allow
x-ray of the BGAs unless you use a system that can penetrate the core.
But if these devices can be damaged at such a low voltage (if indeed
Daniel's figures are correct), shouldn't Altera issue some sort of
warning to their customers about x-ray damaging the devices? After all,
there will be a lot of customers using x-ray for inspection. -Steve
Gregory-
-----Original Message-----
From: TechNet Behalf Of Guy Ramsey
A couple of years ago, NASA published a warning. They required anyone
transporting via air to hand carry them onboard. The warning had to do
with
high energy X-ray used at some facilities to luggage inspection. They
did
not, as I recall, establish any acceptable level of exposure. I haven't
heard anything lately. But, you might bounce this one off the EMPF
Helpline.
-----Original Message-----
From: TechNet Behalf Of Joyce Koo
http://www.sandia.gov/media/rhp.htm
-----Original Message-----
From: TechNet Behalf Of Stadem, Richard
I have had to address the same concerns for Aerospace programs.
I have not been able to find any formal standards for radiation exposure
limits. All I found was a good paper on the subject from Tom Clifford at
Lockheed Martin ("X-Ray Inspection, Easy Does It" in EP+P magazine)
I also talked to two X-ray vendors and they tell me that even if the
typical X-rays were cranked up to 120 kv or about 3-5 watts, a component
would have to be exposed for a full day or longer to accumulate enough
total Rems such that it could hurt the component, and that the only
types of components that could be possibly damaged would be CMOS memory.
I have found some information on protecting components in space systems
from natural X-rays, but nothing on process-generated X-rays. Everybody
tells me it is not a problem. This is not enough for me, so I asked the
folks at IPC but have not heard a response as yet.
Anyone else know of any standards for limitations of X-ray emissions on
electronic components?
--Original Message-----
From: TechNet Behalf Of Steve Gregory
Sent: Monday, September 26, 2005 11:17 AM
To: [log in to unmask]
Subject: [TN] X-ray inspection, can it damage devices?
Good Mornin' Everyone!
I hope you all had a good weekend! I worked Saturday, but had
Sunday off. Sat around Sunday and watched the NASCAR race
at the "Monster Mile" in Dover Delaware. 20-year old Kyle Busch
gave Jimmie Johnson a run for his money finishing second at the
green/white/checkered ending....but I digress.....
Anyways, we built an assembly here that has a .050" copper core
in the center of it...not the same assembly as the one I did my "okie
cross-section" on).
It is double-sided SMT, and has BGAs on both sides. I'm doing x-ray
inspection with a Fein Focus Tiger 160, and in order to penetrate the
core and do a proper inspection of the solder joints and spheres of the
BGAs, the x-ray power is up around 120-130 kv.
Our customer received a number of the assemblies from us (without test
from us, as they are prototypes and test hasn't been developed yet), and
they are having problems with one of the BGAs.
It's a 100-ball Max-7000 device from Altera, a CMOS EEPROM, and our
customer has been trying to program the device via on-board programming,
but some boards wont program. The x-ray inspection turned out fine. All
BGAs were inspected using oblique angles, so if there was an open joint,
it would have been seen.
They are worried that maybe the x-rays may have damaged the device.
Altera has told our customer that the x-rays shouldn't hurt the device.
I've got a case number started via email in to Altera (since I can't
talk to anyone
on the phone), asking them myself if x-rays will damage the device. I'm
also
going to call Fein Focus and talk to them as well. But have any of you
ever heard of x-ray damaging BGAs?
Kind regards,
---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e
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
-----------------------------------------------------
__________________________________________________________________
This message may contain information that is privileged and confidential to LaBarge, Inc. It is for use only by the individual or entity named above. If you are not the intended recipient, you may not copy, use or deliver this message to anyone. In such event, you should destroy the message and kindly notify the sender by reply e-mail.
---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e
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
-----------------------------------------------------
|
