LISTSERV - TECHNET Archives - LISTSERV.IPC.ORG

Date: Fri, 14 Oct 2005 16:58:20 +0200
From: [log in to unmask]
Subject: [LF] From Jennie Hwang--Response--Bi Effects--Selected Points
Sender: Leadfree <[log in to unmask]>
To: [log in to unmask]
Message-id: <[log in to unmask]>
MIME-version: 1.0
Content-type: text/plain; charset=windows-1255
Content-transfer-encoding: 7BIT
Content-class: urn:content-classes:message
Thread-topic: [LF] From Jennie Hwang--Response--Bi Effects--Selected Points
Thread-index: AcXQ0ExnbmxuadsHQkS4FVHPePUOyA==

Roger, Dave:

Thanks for your comments.

Regarding contradicting data and corresponding conclusions, I am confident
that they can be explained once we know the specific parameters and conditions
in the context under a given design, including test conditions and solder
joint-making process, as well as the contacting surfaces and interface (surface
finish, component coating).

There are several fronts in discussing Bi effects.

--Bi adding to 63Sn37Pb solder material: effects on application, physical,
mechanical properties
--Bi in component coating: dissolving into 63Sn37Pb solder joint; estimation
of Bi concentration in solder joint; effects
--Bi in Pb-free solder joint contributed from components: on SnAg, SnCu,
SnAgCu, SnAgIn, SnAgCuIn, etc.
--Bi-containing Pb-free solder alloys by design: application, physical,
mechanical properties
--Fillet-lifting vs. Bi: is it caused by Bi?
--Low temperature BiPbSn phase: presence or absence; general guidelines;
detectable or non-detectable effects

Each has to be clearly discerned. This is not the venue to make a
dissertation on the subject. Nonetheless, a few points are warranted at this time.

Although results are condition-specific, there are areas with certainty:

--Once we know the specific parameters and conditions including test
conditions and solder joint-making process that have made the Bi-containing solder
joint "better or worse," an explanation can be made.
--The effect of surface finish and component coating on a solder joint can be
judged only when a sound solder material is used.
--The effect of surface finish and component coating can be judged only when
a good solder joint is made. That is the process is right and there is no
extraneous factors, such as an inadequate or improper process introduces <
stratified> solder joint, nature of component lead-base-material, improper test
conditions, etc.
--My statement would not be made based on bulk material only, nor on one or
several laboratory-test results, nor on one or more production results, nor on
single accelerated reliability test methodology. It is based on the
combination of lab test results, production, accelerated reliability tests and sustained
filed performance. Further, it has to be in congruence with fundamentals.
This applies to the situation that a small amount of Bi in 63Sn37Pb solder
material (as eluded in this thread of e-mails as Chapter 28 of the book). This doped
material (in solder joint) has been widely used in producing many products
over the years before the Lead-free topic became a common discussion.
--All observed phenomena, positive or negative, are in sync with fundamentals
of materials science and metallurgy in bulk materials and in solder joint (A
great comfort!)
--There are hardly mysteries.

Take one example, if a temperature cycling test of 42Sn58Bi alloy (melting
temperature 138oC), either as a bulk material or in solder joint, is set from
-55oC to +125oC (or +100oC) the results are expected to be definitely negative.
The upper limit temperature is too high for this alloy. And this alloy is not
designed, nor suitable for elevated temperature applications (relative to
63Sn37Pb). But at room temperature, this alloy has higher strength than 63Sn37Pb
over a wide range of strains. But fatigue resistance depends on the strain
amplitude. Take another example on fillet lifting. Bi does not cause
fillet-liftng. Alloy 42Sn58Bi has no fillet-lifting problem, while other Bi-free alloys are
associated with fillet-lifting. However, improperly designed Bi-containing
alloys can severely augment the driving force for fillet-lifting.

Roger's example is right that strength drops, regardless of component
coating, with the continued temperature cycling. Excerpt:<< Pull strength required to
destroy a SJ went from 19.5N to 16N (avg) (1000 cycles) for 63/37 and Sn2Bi
plating, worst case saw a 35% drop in SJ strength.  Interestingly, for a 63/37
and Sn10Pb lead plating, the SJ strength went from 17.5N to 15N (avg; 1000
cycles), worst case saw a 35% drop in SJ strength.>>

Regarding Sn2Bi, Dave is right. The primary purpose of 2% Bi is to mitigate
Sn whisker. In this regard, Bi is one of very few elements, which can <act> in
Sn matrix similarly to Pb.

As to Dave's remarks about the book, the first Lead-free book: "
Environment-friendly Electronics:Lead-free Technology", which was composed during 1999, is
the culmination of solid ten-year sustained and systematic R&D without
legislative "incentives." The book focuses on material properties. As a matter of
fact, the book only houses a small portion of overall research results conducted
due to the space limitation and the relevancy to the SMT manufacturing sector.
Before considering the interface effects and contact surfaces, the intrinsic
materials properties go first. An inferior material does not deliver a high
performance solder joint with any surface finish and coating. It's a sort like
preparing a premium steak, it starts with premium beef. Then a good chef can <
interface> this premium beef with the grill and spices (PCB surface finish and
component lead coating) under right temperature (process) to achieve the
result.

Thus, with a good start in material, one can expect to achieve solder joint
integrity by considering the process, the interface and interfacial phenomena
during the soldering process and afterwards.

The second book: "Implementing Lead-free Technology--A manufacturing Guide",
designed for actual production considering the assembly system (material,
solder joint interface, process, Pb-contamination, etc...) is based on actual
production and reliability test results. It should be noted that each successful
production of the frontier lead-free electronics producers was not lightly put
into implementation. Rather, they were verified and validated through a
thorough analysis and comprehensive testing program, particularly for higher
reliability products, with accelerated conditions above and beyond what the industry
has done for 63Sn37Pb,

For the ultimate effects of Bi, how much and how Bi is incorporated in a
solder joint and under what conditions determine positive or negative results. And
sometimes how the data is interpreted needs to be calibrated. This is not any
different from how much and the limit, albeit to different extents, of Ag or
Cu or Au or In or Pd or other elements can be introduced into any solder joint
without jeopardizing its integrity.

The same criteria (properties vs. dosage) apply to the elemental composition
of an originally designed alloy as well as to the resulting makeup of the
elemental constituents and composition of a solder joint after having additional
elements introduced to a solder joint.

Best Regards,
Jennie
Dr. Jennie Hwang
H-Technologies Group, Inc.
Office: 216-839-1000
Personal E-mail: [log in to unmask]

----------------------------------------------------------------------------------------------------------
Subj:   Re: [LF] Leadfree Processing temp requirement
Date:   10/13/05 12:35:04 PM Eastern Daylight Time
From:   [log in to unmask]
Reply-to:   [log in to unmask]
To: [log in to unmask]
Sent from the Internet (Details)
Hi Dave,

I'll bite on this one...
I should have said also that there are concerns that at high temps, SJs
containing Bi have less strength. (possibly due to the increased
plasticity of the solder?)
You bring up some good points (as always). Interesting story:
We have a product that uses a LQFP-144 that originally had Sn10Pb lead
plating.  Unbeknownst to us, the vendor changed to a Sn2Bi lead plating.
Well, created quite a firestorm here on that one.  We asked them to
provide test data to demonstrate what impact the lead plating would have
on long-term reliability.  End result was a fairly thorough test of
wetability and SJ strength testing.  The stress testing was perform with
a total of 1000 cycles from -40degC to +125degC at 30 min each and a
pre-soak of 105degC for 4hrs.  Their testing showed negligible
difference in SJ strength, with the exception that with a Cu-alloy frame
(not Alloy 42), there was a greater incidence of joint failure at the
board more so than the joint itself.  Pull strength required to destroy
a SJ went from 19.5N to 16N (avg) (1000 cycles) for 63/37 and Sn2Bi
plating, worst case saw a 35% drop in SJ strength.  Interestingly, for a
63/37 and Sn10Pb lead plating, the SJ strength went from
17.5N to 15N (avg; 1000 cycles), worst case saw a 35% drop in SJ
strength.  We have had no field failures as yet, and this part is used
in a device mounted inside the cabs of bulldozers, motorgraders, and the
like.
Would you give me a link to all of the data you mentioned? Seems the
jury may still be out on this topic, and I gather all the info I can on
Bi in 63/37 joints. Can't wait for the final results.

I might send Dr. Hwang a note asking about any additional testing she
may have done to emulate real-world use...

Regards,
Roger
------------------------------------------------------------------------------
--------------------------
In a message dated 10/13/05 10:55:54 AM Eastern Daylight Time,
[log in to unmask] writes:
> Hi Roger! Well, we had sooo much fun with the ENIG topic yesterday, why not
>
> have some fun with the bismuth topic! There are two camps of thought in our
> industry: 1) those folks who don't believe that bismuth causes issues; 2)
> those folks who believe the bismuth cause issues. The reality is that your
> use environment carries the greatest weight for determining the impact of
> bismuth on solder joint integrity. Dr. Hwang's book is a great book - very,
> very good bulk solder alloy data but a very limited component solder joint
> reliability data set (only a LCC component) in chapter 28. The folks at
> NIST have published a set of very good papers detailing the interaction of
> bismuth with tin/lead solder which shows that the solder joint integrity is
> impacted. Also, the JCAA/JGPP consortia published preliminary data/results
> at the 2005 SMTAI Conference for a wide variety of components (QFP, BGA,
> TSOP, LCC, etc) in Class 3 high performance testing that small quantities
> of bismuth in tin/lead solder do negatively impact the solder joint
> integrity. The real question is "does that reduced solder joint integrity
> cause a problem for your use environment?". Many folks in a Class 2 use
> environment may not have issues with bismuth whereas folks in a Class 3
> environment would have big issues (hence the two different industry camps).
> The JCAA/JGPP consortia will be publishing their final report in February
> which will make the data available for folks to use. Hope this helps.
> Dave Hillman
> Rockwell Collins
> [log in to unmask]
------------------------------------------------------------------------------
-----------------------
> Roger Stoops              <[log in to unmask]>      Sent by:
> [log in to unmask]                                                          Subject:
> Re: [LF] Leadfree Processing temp requirement              10/12/2005 02:08  PM
>
> Hi Rick and all,
>
>   I have heard this for some time, and assumed it to be true.  However,
> after reading Dr. Jennie S. Hwang's book "Environment-Friendly
> Electronics: Lead-Free Technology," chapter 28, it seems that any
> dangers of Bi in a eutectic SnPb system may be greatly exaggerated.
>   She notes that the "low temperature eutectic phases are not expected
> to be formed in the alloys containing a low dosage of Bi."  Also, the
> fatigue life and strength increases with small amounts of Bi, suggesting
> that there is not enough Bi to form the SnPbBi eutectic.
> (pg 757, para. 1)
> Seems that the addition of a small amount of Bi (<5%) only lowered the
> melting point by 3degC or so.  Dr. Hwang provides very convincing data.
>
> Excellent book to have in your library...
>
> Anyone have other thoughts?
> Regards,
> Roger
>
> P.s.: did I mention it's an excellent book?
> -----------------------------------------------------------------------------
> ----------------------------
> -----Original Message-----
> From: Leadfree [mailto:[log in to unmask]] On Behalf Of Smith, Rick
> Sent: Wednesday, October 12, 2005 2:47 PM
> To: [log in to unmask]
> Subject: Re: [LF] Leadfree Processing temp requirement
>
> Check and see if On Semi is using Bismuth on the leads, Bismuth is
> undesirable in the Pb/Sn process because it makes a very low temp alloy
> that could melt at normal operating temperatures depending on circuit
> current.
> Rick SmithSenior Product/ Component Engineer
> Phone: (512)652-3544
> Fax:     (512)652-3545
> Cell:     (512)299-6925
> Email:   [log in to unmask]
> ClearCube Technology, Inc
> 8834 Capital of Texas Hwy N
> Austin, TX 78759
> www.clearcube.com
>
> -----Original Message-----
> From: Leadfree [mailto:[log in to unmask]] On Behalf Of Gilbert, Doug
> Sent: Wednesday, October 12, 2005 1:32 PM
> To: [log in to unmask]
> Subject: Re: [LF] Leadfree Processing temp requirement
>
> RoHS warriors...
>
> While talking about temperature requirement for Pb-free, I was asked why
> semiconductor vendors are not temp rating their Pb-free DIP packages. I
> believe these are considered 'not MSD' but do they never see 260 deg.C?
> I am not knowledgeable in assembly processing in detail hence my
> questions.
> For mixed assembly I think the SMD parts are reflowed first then any
> thru-hold parts are wave soldered.  The parts don't have to withstand
> high temp since only the leads are exposed?
> Another mystery.  I know that BGA type parts are not backward
> compatible.
> But ON Semi just issued a PCN saying their Pb-free TO-3 and TO-92
> packages are also not backward and should be used for Pb-free soldering
> only.
> Why would that be?
> Thanks,
> Doug
> -----Original Message-----
> From: Leadfree [mailto:[log in to unmask]]On Behalf Of James, Chris
> Sent: Wednesday, October 12, 2005 6:55 AM
> To: [log in to unmask]
> Subject: Re: [LF] Leadfree Processing temp requirement
> Only to add that yes it is not simple.............
> If you have an old 4 zone oven you might have probs and need more zones.
> It depends on component mix on the board to a certain extent - if you
> have some parts with large thermal mass then all the tiddlers are going
> to get hotter!
> Look carefully at specs on electrolytics (these can distort if over
> cooked) and Leds where the lens will discolour or dull.
> You certainly need to be looking for RoHS process compliant parts with peak
> reflows of at least 240 through to 260C. All this also has an effect on MSL
> rating requirements so storage
> considerations may need to change. Look for parts meeting JSTD-020C.
> Regards,
> Chris
> -----Original Message-----
> From: Leadfree [mailto:[log in to unmask]] On Behalf Of Smith, Rick
> Sent: 12 October 2005 14:39
> To: [log in to unmask]
> Subject: Re: [LF] Leadfree Processing temp requirement
> Ken,
> I'm going to open a can of worms with this one, and I'm sure will get
> plenty of responses.
> The target peak for SAC305 is 253 C for most applications. (245 C is too
> cool).
> You should be above 217 C for 45 to 90 sec.
> But 217 C is not the temp you should worry about, it's really somewhere
> around 235/7 C which is where the solder melts to other metals, (wetting
> temp). If you are not over the wetting temp long enough you will
> experience poor wetting, especially with OSP, where you need to be near
> the "longer" end. You should stay over 235 C for 30 to 45 seconds.
> Now the kicker here is, that if your oven has poor thermal capacity, you
> can see a temp difference across the board of as much as 30 to 35
> degrees C if you are soldering a component like the 775 Socket T and
> trying to get the underside to melt.
> That means other solder joint temps in a bad oven could hit 285 C,
> (don't ask about the component body) which is unacceptable for most
> components that are rated at 260 C.
>
> If your oven cannot maintain at least +/- 10 degrees C across the PCB on
> your heaviest boards/components (+/- 5 degrees C on regular LF product)
> you should replace it. 10 years ago I would have said +/- 3 degrees C,
> but I have been soldering some weird stuff lately, :<)
> --Original Message-----
> From: Leadfree [mailto:[log in to unmask]] On Behalf Of Kanaiyalal Patel
> Sent: Tuesday, October 11, 2005 7:10 PM
> To: [log in to unmask]
> Subject: [LF] Leadfree Processing temp requirement
> All,
> Has IPC modified the document related to component spec from process
> point of view such as reflow and wave. I forgot the document number.
>
> This document is being used by supplier/manufacturer to make parts that
> can withstand needed Leadfree reflow/wave temperature. Few of my
> suppliers' parts are rated around 240c max and I think that is too low
> for leadfree process. Can someone provide the temp numbers withour
> buying a IPC spec for urgent requirements?
> Any help will be appreciated.
> re,
> ken Patel
> or 847-615-7100
> ext.2815
> -

-------------------------------------------------------------------------------Leadfee Mail List provided as a service by IPC using LISTSERV 1.8d
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Leadfree
To temporarily stop/(start) delivery of Leadree for vacation breaks send: SET Leadfree NOMAIL/(MAIL)
Search previous postings 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 is processed by the PrivaWall Email Security Server.



---------------------------------------------------
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
-----------------------------------------------------