I've been on since the beginning and I agree with Wayne that this is an interesting thread and although I'm a failure mode analysis and not a manager like Wayne I agree with everything he said in his post. Even if something is done which seems to have fixed the problem the reliability of the PCBA will always be in question and the most value would be to do a failure analysis and determine the root cause so it doesn't happen again
George
-----Original Message-----
From: TechNet [mailto:[log in to unmask]] On Behalf Of Wayne Thayer
Sent: Thursday, January 04, 2018 3:14 PM
To: [log in to unmask]
Subject: Re: [TN] BGA pad non-wetting
Hi-
Got on late, interesting thread, but the manager in me says:
Guy, I'm sorry but all of those boards are trash. If you were your own customer, you probably would not accept them given what you know right now.
BUT, if you do whatever it takes to get to "root cause" so that we all know what the probability of a reliability problem is, then we might think differently about whether the modules are acceptable.
I make this point because of your hesitancy to do destructive testing:
Since the boards are presently trash, their only value is for Failure Analysis. If you destroy one to figure out root cause, you didn't lose $45K, you paid $45K (plus analysis services) for an education that gives you a chance of rescuing whatever multiple of $45K the production run is worth! Plus, you'll probably get a little smarter and do something that makes future disasters like this more unlikely.
You have to find out what is under that tin you found in XRF. Trying to abrade the pads or do other "blind" corrective actions is window dressing and leaves the remainder of this important assembly at unknown risk.
Wayne
On Tue, Dec 26, 2017 at 7:38 AM, Steve Gregory <[log in to unmask]> wrote:
> Hi Everyone,
>
> I trust that everybody was on Santa's "nice" list and got what they
> wanted for Christmas. 😀
>
> Anyways, Guy sent some more photos on Friday of the issue we've been
> talking about, so here's another look at things:
>
> http://stevezeva.homestead.com/double_end_mat_bone.jpg
> http://stevezeva.homestead.com/mat_both_signal_and_plane.jpg
> http://stevezeva.homestead.com/mat_on_copper_-_peeled_mask.jpg
> http://stevezeva.homestead.com/matt_bottom_kiss.jpg
>
> Steve
>
> On Tue, Dec 26, 2017 at 8:30 AM, Stadem, Richard D <
> [log in to unmask]
> > wrote:
>
> > Thank you for chiming in Joyce!
> > A rubber, Sulphur-based ordinary pencil erasure will sometimes work,
> > but that is not what I (and apparently Joyce) have used in the past.
> > What we are talking about is a semi-hard polymer containing very
> > fine abrasive particulates. It comes in a variety of Dremel tip
> > geometries, as well as a small flexible disk.
> > It is especially wonderful for cleaning up immersion silver PWBs
> > with slightly tarnished surface. It restores solderability like you
> > would not believe.
> > This is not a high-pressure application, very little pressure is
> required.
> > It does work for certain situations, but is a last resort, and does
> > not always work well.
> > Dave's suggestion to use XrF as a plating analysis tool is an
> > excellent method for determining the root cause of the issue. If XrF
> > determines
> there
> > is a great variance of nickel from pad to pad on an UNSOLDERED PWB,
> > and
> the
> > root cause is nickel skips or some other plating issue, then the use
> > of
> the
> > Dremel tip will not work; there is no solution to the issue except
> > scrap
> or
> > salvage.
> >
> > So Dave's words of wisdom ring true; one must understand the exact
> > nature of the problem in order to determine if there is a solution.
> >
> > -----Original Message-----
> > From: TechNet [mailto:[log in to unmask]] On Behalf Of Yuan-chia Joyce
> > Koo
> > Sent: Friday, December 22, 2017 11:40 AM
> > To: [log in to unmask]
> > Subject: Re: [TN] BGA pad non-wetting
> >
> > respectably disagree. non-sulfur containing rubber eraser are great
> > to clean up plating surfaces (not immersion type, but electro-plated
> > or electroless)... surface oxide layer some times is formed due to
> > lazy
> chaps
> > raise the unit too late (after shut down the electricity) or leave
> > it too long before rince... eraser are great to recover zif, wire
> > bonding pads, etc... (provide your plating pass the 3M tape test
> > - enough adhesion to the under lying metallization). IMHO. as for
> > immersion type, agree with dave - too thin and too soft - you might
> > have debris embedded in the surface layer cause you problem...
> > jk
> > On Dec 22, 2017, at 8:52 AM, David Hillman wrote:
> >
> > > And I'll third that vote! All of the immersion type finishes have
> > > roles to play for our electronic designs but ImSn is the least
> > > robust to assembly processes. Also, I don't recommend ever using
> > > an eraser for "cleaning" a pad. If you examine a pad that has been
> > > erased, you will find eraser contamination embedded in the pad
> > > that can cause solder joint integrity issues later. Using an
> > > eraser to troubleshoot is ok but no further use should be conducted.
> > >
> > > Dave
> > >
> > > On Thu, Dec 21, 2017 at 7:42 PM, George Wenger
> > > <[log in to unmask]>
> > > wrote:
> > >
> > >> Tan Geok,
> > >>
> > >> The one think I absolutely agree with is that I also hate
> > >> immersion tin surface finish
> > >>
> > >> George
> > >>
> > >> -----Original Message-----
> > >> From: TechNet [mailto:[log in to unmask]] On Behalf Of Tan Geok Ang
> > >> Sent: Thursday, December 21, 2017 8:15 PM
> > >> To: [log in to unmask]
> > >> Subject: Re: [TN] BGA pad non-wetting
> > >>
> > >> Believe it is Immersion Sn surface finish. I usually use a
> > >> pencil eraser to slightly rub against those pads and it works
> > >> well. I hate immersion Sn finish. Take care of ESD.
> > >>
> > >> ________________________________
> > >>
> > >> From: Guy Ramsey <[log in to unmask]>
> > >> Date: 22 December 2017 at 6:01:45 AM SGT
> > >> To: [log in to unmask] <[log in to unmask]>
> > >> Subject: Re: [TN] BGA pad non-wetting
> > >>
> > >> Wow, thanks
> > >>
> > >> On Thu, Dec 21, 2017 at 4:58 PM, Stadem, Richard D
> > >> <[log in to unmask]> wrote:
> > >>
> > >>> I agree with George that these pictures exhibit classic black
> > >>> pad or immersion tin intermetallics of some type, but even with
> > >>> those some wetting normally takes place on any given pad.
> > >>> I do not think it is a thin soldermask film on the pads, but it
> > >>> could be.
> > >>> The reason I don't think it is soldermasking is because if you
> > >>> look carefully at the picture, some pads appear to be fully
> > >>> wetted, and some do not, but no pad has any partial wetting.
> > >>> It is either "on" or "off". However, the pads that appear
> > >>> wetted may not really be wetted. As Werner Englemaier once
> > >>> pointed out, you can achieve complete "wetting" of solder on a
> > >>> toothpick if you know how.
> > >>> But at $45,000 per assembly, it would be a career-enhancing move
> > >>> to find a way to clean up the oxidation, whatever the cause. I
> > >>> have a process that has worked for me, tell you in a minute.
> > >>> But before you attempt to clean up any oxidation, please make
> > >>> one more check. See if the pads that never wetted are connected
> > >>> to power, ground, or reasonably heavier traces than the pads
> > >>> that DID wet, or at least appeared to wet.
> > >>> The root cause may be a perfect storm of insufficient heat in
> > >>> the reflow profile, right on the edge of getting the board hot
> > >>> enough to reflow unattached pads or pads not severely
> > >>> heatsinked, along with some finish-related oxidation issues, and
> > >>> perhaps a flux not quite active enough to remove the oxides at
> > >>> the current reflow temperature.
> > >>>
> > >>> Otherwise, I have used very fine Dremel #463 blue-oxide abrasive
> > >>> tips impregnated with abrasive to remove oxides, rough areas,
> > >>> small burrs and scratch marks. See one here:
> > >>>
> > >>> https://urldefense.proofpoint.com/v2/url?
> > >>> u=https-3A__www.google.com_se
> > >>> arch-3Frls-3Dcom.microsoft-253Aen-2DUS-253AIE-2DSearchBox-26ei-3
> > >>> D&d=
> > >>> Dw
> > >>> IFaQ&c=nvOEfClv88r3To4eOyYtQN0213v_-JGWbujap2hCee4&r=-
> > >>> MAMt3MrjiRvYJiwV
> > >>> eFoas7eghr5IF6QpUBreyGR1qs&m=ZtT8m__jsmKXx-
> > >>> NxB6DAVryCAm8iznmmCFPq9PL_Q
> > >>> qE&s=80pAGk1ZIDMwG9Yl4yta_g7WK1A6ibC-_9lxLJZVSmM&e=
> > >>> LCI8WrLkFIiJ0gK3p6YCw&q=Dremel+blue+rubber+polishing+
> > >>> tips&oq=Dremel+blue+rubber+polishing+tips&gs_l=psy-ab.3..
> > >>> .3284.20303.0.21161.40.39.0.1.1.0.263.5212.0j26j4.30.0....0.
> > >>> ..1c.1.64.psyab..9.28.4637...0j0i10k1j0i131i67k1j0i131k1j0i
> > >>> 67k1j0i22i30k1j0i22i10i30k1j33i22i29i30k1j33i21k1j33i160k1.0.Sug
> > >>> 1ul9
> > >>> 9S
> > >>> sE
> > >>>
> > >>> They have a tiny yet firm point to clean pads of oxides or other
> > >>> contamination. It takes a very fine touch to do it, is difficult
> > >>> on pad diameters smaller than .020", but it can be done and for
> > >>> a single component it takes less than a minute or two on the
> > >>> unwetted pads only. An overhead ionizer is used to prevent or
> > >>> minimize any ESD charge buildup, and the area to be reworked is
> > >>> masked off prior to the polishing such that only the BGA pads
> > >>> are exposed. A vacuum with a grounded tip is used to clean up the debris as you go along.
> > >>> Then you wipe the area with a drop of IPA on a foam swab, and
> > >>> you apply a thin veneer of a good tacky flux like Alpha WS820
> > >>> immediately after the last vacuuming. You then manually solder
> > >>> coat the pads, leave the flux on but add a little more, place
> > >>> the part and reflow with a hot nitrogen reflow machine. When
> > >>> finished, unmask under the ionizer, prewash by directing a low
> > >>> pressure spray under all four sides of the BGA, then follow that
> > >>> with a pass through the in-line or batch cleaner, blow off the
> > >>> water at 30 PSI max (but thoroughly), bake at
> > >>> 105 C for just
> > >>> 20 minutes, visually inspect using a good scope and prism, and
> > >>> re-test. If it passes, touchup the coating if present.
> > >>> If this seems to work, perform some type of thermal cycling that
> > >>> makes sense for the final use environment on two or three CCAs
> > >>> to ensure the oxidation or reaction does not re-occur over time.
> > >>> Get the customer buy-in if selling those units as downleveled CCAs.
> > >>> If that all seems to work, then continue with the rework on a
> > >>> larger sample of CCAs and closely monitor the BGA performance at
> > >>> test and final visual inspection using a good microscope and prism.
> > >>> No problems? Increase the sample size of the rework and continue.
> > >>> Everything still good? Then walk upstairs and barge into the
> > >>> Upper Management Meeting and tell them you deserve at least a
> > >>> 50% increase in salary, a corner office, and a company car with
> > >>> expense account if they want you to continue to keep the entire
> > >>> company from washing down the toilet just when Donald was about
> > >>> to give them a huge tax break.
> > >>> Let us know how it goes!
> > >>>
> > >>> Good Luck, and Merry Christmas!
> > >>>
> > >>> Odin
> > >>>
> > >>> -----Original Message-----
> > >>> From: TechNet [mailto:[log in to unmask]] On Behalf Of George
> > >>> Wenger
> > >>> Sent: Thursday, December 21, 2017 2:05 PM
> > >>> To: [log in to unmask]
> > >>> Subject: Re: [TN] BGA pad non-wetting
> > >>>
> > >>> Steve,
> > >>>
> > >>> I just went back and re-read your initial post and looked at the
> > >>> pictures again. I originally thought we were looking at the
> > >>> pads on the BGA but now I think we're looking at the pads on
> > >>> your PCBA. If they are the pads on the PCBA then my suggestion
> > >>> is to look at a new non-soldered PCBA and let us know if the pads look like gold or tin.
> > >>> If they look like gold you have an ENIG Black Pad problem if
> > >>> they look like tin then you have an oxidized Cu-Sn IMC. The
> > >>> real problem in either case there really isn't a reasonable fix.
> > >>> The only think I would try is to use a very aggressive flux and
> > >>> try to pre-tin the pads with solder and then clean off the flux
> > >>> residue and go back through your
> > >> normal repair process.
> > >>>
> > >>> George
> > >>>
> > >>> -----Original Message-----
> > >>> From: TechNet [mailto:[log in to unmask]] On Behalf Of George
> > >>> Wenger
> > >>> Sent: Thursday, December 21, 2017 2:58 PM
> > >>> To: [log in to unmask]
> > >>> Subject: Re: [TN] BGA pad non-wetting
> > >>>
> > >>> Steve,
> > >>>
> > >>> I totally agree with John. My first thought was that the pad
> > >>> looked
> > >>> like thin immersion tin where the IMC came to the surface an
> > >>> oxidized in which case solder will not stick to the pad. But I
> > >>> don't know of any BGA where they use Immersion tin so I still
> > >>> suspect you have an ENIG Bad Pad issue.
> > >>>
> > >>> George
> > >>>
> > >>> -----Original Message-----
> > >>> From: TechNet [mailto:[log in to unmask]] On Behalf Of John Burke
> > >>> Sent: Thursday, December 21, 2017 2:51 PM
> > >>> To: [log in to unmask]
> > >>> Subject: Re: [TN] BGA pad non-wetting
> > >>>
> > >>> Assuming it's not ENIG???
> > >>>
> > >>> If not ENIG I would suspect immersion tin where the
> > >>> intermetallic has come to the surface.
> > >>>
> > >>> If an ENIG problem or an immersion Tin problem it is not
> > >>> realistically recoverable without impacting the assembly with
> > >>> some serious reliability issues.
> > >>>
> > >>> Sent from my iPhone
> > >>>
> > >>>> On Dec 21, 2017, at 11:39 AM, Steve Gregory
> > >>>> <[log in to unmask]>
> > >> wrote:
> > >>>>
> > >>>> Hi all,
> > >>>>
> > >>>> I've been asked to post this to the Technet by a dear friend:
> > >>>>
> > >>>> *We installed this little processor. At test we identified
> > >>>> opens on the device. When we removed it we saw a number of pads
> > >>>> that were matte gray with no evidence of solder adhered to the
> > >>>> pad. * *We attempted to bump the pads with solder, they did not
> > >>>> wet easily. In some cases we had to scrape through the matte
> > >>>> finish to a shiny metal.
> > >>>> But, we did get all the pads to appear wetted.*
> > >>>>
> > >>>> *We cleaned the site well and then fluxed the bumps with a
> > >>>> tacky flux ROLO designed for POP, placed the BGA and reflowed
> > >>>> it on a rework
> > >>>> station.*
> > >>>>
> > >>>> *The assembly failed test with similar result, opens under the
> > >>>> processor.*
> > >>>>
> > >>>> *The attached photo shows that the bump process only appeared
> > >>>> to wet the pads. The condition returned when we removed the BGA
> > >>>> second
> > >>>> time.*
> > >>>>
> > >>>> *
> > >>>> https://urldefense.proofpoint.com/v2/url?
> > >>>> u=http-3A__stevezeva.homest
> > >>>> ead.com_2017-2D12-2D21-5F11.21.17.jpg&d=DwIFaQ&c=nvOEfClv88r3To
> > >>>> 4eOy
> > >>>> Y
> > >>>> tQN0213v_-JGWbujap2hCee4&r=-
> > >>>> MAMt3MrjiRvYJiwVeFoas7eghr5IF6QpUBreyGR1
> > >>>> qs&m=ZtT8m__jsmKXx-
> > >>>> NxB6DAVryCAm8iznmmCFPq9PL_QqE&s=Du_vGyBf05tjpSvFe
> > >>>> t5EPytyLuSbCfhNQ4wlPVTfIK8&e=
> > >>>> <https://urldefense.proofpoint.com/v2/url?
> > >>>> u=http-3A__stevezeva.homes
> > >>>> tead.com_2017-2D12-2D21-5F11.21.17.jpg&d=DwIFaQ&c=nvOEfClv88r3T
> > >>>> o4eO
> > >>>> y
> > >>>> YtQN0213v_-JGWbujap2hCee4&r=-
> > >>>> MAMt3MrjiRvYJiwVeFoas7eghr5IF6QpUBreyGR
> > >>>> 1qs&m=ZtT8m__jsmKXx-
> > >>>> NxB6DAVryCAm8iznmmCFPq9PL_QqE&s=Du_vGyBf05tjpSvF
> > >>>> et5EPytyLuSbCfhNQ4wlPVTfIK8&e= >*
> > >>>>
> > >>>> *Have you seen this before? Is there any recovery? This
> > >>>> assembly is worth about $45K. *
> > >>>>
> > >>>> Thanks,
> > >>>>
> > >>>> Steve
> > >>>>
> > >>>> --
> > >>>> Steve Gregory
> > >>>> Kimco Design and Manufacturing
> > >>>> Process Engineer
> > >>>> (208) 322-0500 Ext. -3133
> > >>>>
> > >>>> --
> > >>>>
> > >>>>
> > >>>> This email and any attachments are only for use by the intended
> > >>>> recipient(s) and may contain legally privileged, confidential,
> > >>>> proprietary or otherwise private information. Any unauthorized
> > >>>> use, reproduction, dissemination, distribution or other
> > >>>> disclosure of the contents of this e-mail or its attachments is
> > >>>> strictly prohibited.
> > >>>> If you have received this email in error, please notify the
> > >>>> sender immediately and delete the original.
> > >>>
> > >>
> >
>
>
>
> --
> Steve Gregory
> Kimco Design and Manufacturing
> Process Engineer
> (208) 322-0500 Ext. -3133
>
> --
>
>
> This email and any attachments are only for use by the intended
> recipient(s) and may contain legally privileged, confidential,
> proprietary or otherwise private information. Any unauthorized use,
> reproduction, dissemination, distribution or other disclosure of the
> contents of this e-mail or its attachments is strictly prohibited. If
> you have received this email in error, please notify the sender
> immediately and delete the original.
>
|
