I found the knee of the solder joint to be the weak point.
On cross section, there would be a pad but the plated area at the
intersection my be thin to non existent.

It was very evident on the PCA's that had a memory module in the rework
Pb/Sn solder fountain for a minute.
On occasion the pads would be gone, on others there would be disconnected
rings.

It happened much faster on the ImSn PB free assemblies.

The high flow rate, temperatures and Sn content as stated contribute to the
problem.
Often the root cause is the badly designed PCB where there are no thermal
reliefs.

Selective solder is an extremely flexible process.
It is a light weight tool in comparison to a wave solder machine in the
heat department.
Often the tool ends up misapplied because it is the only hammer available.

When using the tool in a production environment where seconds matter, care
must be taken to minimize the heat requirement.
Lead not lead free, SAC305 not SN100, Alcohol not VOC free flux.
Maximize contact area using larger or custom nozzle.
Heat the solid copper plane the solder joint is in.
Anything to reduce the amount of heat that has to be transmitted from the
fountain to the barrel of the PCB to make a good top side fillet.
Add preheat above PCA during soldering process if over 2 minutes.

Limit rework by repeated exposure to solder fountain.
I have seen the pins in a connector erode after two dips.

I have not selective soldered Lead Free HASL.
I have run it down the SMT line.







On Mon, Nov 21, 2016 at 1:12 PM, Joyce Koo <[log in to unmask]> wrote:

> in addition, (1) HASL usually 30-1500 micro inch thick coating compare to
> the immersion Sn of 8-15 micro inch thick...(2) HASL is Pb/Sn and solder
> is Pb/Sn, reaction by physical exchange, limited chemical reaction...
> where as immersion sn and Cu pad in contact with solder of SAC, you got
> continues change chemistry at boundary of the pad/solder interface...
> (selective solder the volume of SAC is low assume...) plus extra heat and
> repeat... does not look good ;-).
>  jk
> > Thanks Dave I hadn't seen that data before
> >
> > ----- Original Message -----
> >
> > From: "David Hillman" <[log in to unmask]>
> > To: "TechNet E-Mail Forum" <[log in to unmask]>, "George Wenger"
> > <[log in to unmask]>
> > Sent: Monday, November 21, 2016 11:51:06 AM
> > Subject: Re: [TN] Immersion Tin - problems
> >
> > Hi George - Rockwell Collins/Celestica did some copper dissolution work
> > with the SAC305 alloy as part of the NASA DoD project. For a temperature
> > of 260C, the dissolution rate of a QFP pad was 0.043 mils/second. With
> > that rate, I can dissolve a 2.5 mil thick copper pad in about 60 seconds.
> > We published the data at the 2013 SMTA ICSR in Toronto.
> >
> > Dave Hillman
> > Rockwell Collins
> > [log in to unmask]
> >
> > On Mon, Nov 21, 2016 at 10:24 AM, George M. Wenger <
> > [log in to unmask] > wrote:
> >
> >
> > Juliano,
> >
> > You suggestion would be to cross section one of the pads before anything
> > is soldered to the pad to see how much copper there is below the
> immersion
> > tin.
> >
> > The first step in the immersion tin plating process is to microetch the
> > copper (which removes some copper) and then during the immersion tin
> > plating more copper is removed and replaced with tin. The thicker the
> > immersion tin is applied the more copper is removed. So I think you need
> > to determine how much copper there is left so that when you go through
> the
> > soldering process you should be able to determine if there will be any
> > copper left after soldering.
> >
> > I haven't seen any specific dissolution rate info for Pb-Free solders but
> > I know the dissolution rate for copper into SnPb solder at 435F given in
> > an old Wally Bader paper is 4 u"/sec. so if your solder is molten for a
> > minute and a half you'll dissolve 360 microinches of copper.
> >
> > George
> >
> > ----- Original Message -----
> >
> > From: "Richard D. Stadem" < [log in to unmask] >
> > To: [log in to unmask]
> > Sent: Monday, November 21, 2016 8:47:54 AM
> > Subject: Re: [TN] Immersion Tin - problems
> >
> > Thanks for your input Juliano.
> > If others have also seen this, it would be interesting to see just how
> > widespread this issue is. IPC should have guidelines on what to do to
> > prevent it (ENIG finish, 1/2 oz copper minimum, etc).
> > As I posted earlier, I have seen it happen many times at many client
> > companies, and I am simply trying to get the word out. Many have no idea
> > what may have happened, and be unfairly blaming the PWB fabricator.
> > It is a process/design issue.
> > dean
> >
> > -----Original Message-----
> > From: TechNet [mailto: [log in to unmask] ] On Behalf Of Francesco Di Maio
> > Sent: Friday, November 18, 2016 8:12 AM
> > To: [log in to unmask]
> > Subject: Re: [TN] Immersion Tin - problems
> >
> > Hello Juliano,
> > I think you are facing a problem of copper dissolution as Mike wrote.
> > This problem become more evidentemente with lead free solder alloy, also
> > it may be presente with Sn/Pb alloy.
> > In our experience the use of Sn100 alloy may help to mitigate the effect
> > of the annullar ring partial or complete dissolution if compared to
> > SAC305 or SAC405 alloy.
> > We noted this problem during solder pot (solder fountain) rework used to
> > remove and resolder pth connector.
> > The selective soldering with long contact time, higher solder
> temperature,
> > higher solder flow rate in my opinioni may cause the problem you noted.
> > Try to reduce these variables and problem should be mitigate.
> > Preheating the assembly before or during the selective wave soldering,
> may
> > help.
> >
> > Regards
> > Francesco Di Maio
> > SEM Comunicazione & Gestlabs
> >
> >
> > -------- Messaggio Inoltrato --------
> > Oggetto: Re: [TN] RES: [TN] RES: [TN] Immersion Tin - problems
> > Data: Thu, 17 Nov 2016 16:25:03 +0000
> > Mittente: Mike Fenner < [log in to unmask] >
> > Rispondi-a: TechNet E-Mail Forum < [log in to unmask] >, Mike Fenner
> > < [log in to unmask] >
> > A: [log in to unmask]
> >
> >
> >
> > There could be a number of reasons. I offer the following to add to your
> > list of things to investigate.
> >
> > The immersion tin will dissolve very rapidly into the Solder which will
> > then start to dissolve the Cu pad underneath to form a solder joint. If
> > your pads/tracks are small/thin then it is possible the Pb-free solder
> > could dissolve the pad entirely. Pb-free solders can be more aggressive
> in
> > this respect than Sn/Pb due to the higher tin content. The Sn is more
> > concentrated/less dilute compared to Sn/Pb. Processing temps are also
> > higher. The problem can be increase by a turbulent or fast wave.
> > Reducing the exposure time may help.
> > I suggest you might want to consider your Pb-free finish for unrelated
> > reasons. Immersion tin is a variable process and has a relatively short
> > storage life. It's hard to think of any advantages it has, save possibly
> > for cost.
> >
> > ---
> >
> >
> > Regards
> >
> >
> > Mike
> > www.chrisfennerfund.org
> > www.facebook.com/chrisfennerfund/
> >
> > -----Original Message-----
> > From: TechNet [mailto: [log in to unmask] ] On Behalf Of Datacom - Juliano
> > Ribeiro
> > Sent: Thursday, November 17, 2016 11:46 AM
> > To: [log in to unmask]
> > Subject: [TN] RES: [TN] RES: [TN] Immersion Tin - problems
> >
> > Many pads
> >
> > _____________________________
> > Juliano Bettim Ribeiro
> > DATACOM
> > +55 (51) 8446-2135
> > +55 (51) 3933-3000
> > Ramal: 3484
> >
> >
> > -----Mensagem original-----
> > De: Stadem, Richard D. [mailto: [log in to unmask] ] Enviada em:
> > sexta-feira, 11 de novembro de 2016 11:40
> > Para: TechNet E-Mail Forum; Datacom - Juliano Ribeiro
> > Assunto: RE: [TN] RES: [TN] Immersion Tin - problems
> >
> > One, or many?
> >
> > -----Original Message-----
> > From: TechNet [mailto: [log in to unmask] ] On Behalf Of Datacom - Juliano
> > Ribeiro
> > Sent: Friday, November 11, 2016 5:21 AM
> > To: [log in to unmask]
> > Subject: [TN] RES: [TN] Immersion Tin - problems
> >
> > Damaged = "Pad disappeared" the track has broken
> >
> > _____________________________
> > Juliano Bettim Ribeiro
> > DATACOM
> > +55 (51) 8446-2135
> > +55 (51) 3933-3000
> > Ramal: 3484
> >
> >
> >
> > -----Mensagem original-----
> > De: TechNet [mailto: [log in to unmask] ] Em nome de George Wenger Enviada
> > em: quinta-feira, 10 de novembro de 2016 17:32
> > Para: [log in to unmask]
> > Assunto: Re: [TN] Immersion Tin - problems
> >
> > What do you mean by "damaged"? Do you mean solder doesn't wet the pad or
> > is there physical damage?
> >
> > Sent from my iPhone
> >
> >> On Nov 10, 2016, at 1:16 PM, Vladimir < [log in to unmask] >
> >> wrote:
> >>
> >> Hi Juliano,
> >>
> >> The board finish most probably has nothing to do with the problem. It
> >> must
> > be your process.
> >>
> >> Regards,
> >>
> >> Vladimir
> >> SENTEC
> >>
> >> Sent from my BlackBerry 10 smartphone on the Rogers network.
> >> Original Message
> >> From: Datacom - Juliano Ribeiro
> >> Sent: Thursday, November 10, 2016 13:13
> >> To: [log in to unmask]
> >> Reply To: TechNet E-Mail Forum
> >> Subject: [TN] Immersion Tin - problems
> >>
> >> Hello to all,
> >>
> >>
> >>
> >> We recently are changing to Lead Free process to specific products and
> >> we need to change the finishes boards from (HAL Tin Lead) to Immersion
> > Tin.
> >>
> >> But during the Selective Soldering process the component pad is damaged.
> >>
> >> Are there limitation about Immersion Tin about Max. Temperature, Max.
> >> Reflow/Wave process, Max Soldering Contact Time, etc.???
> >>
> >> In this case the HAL Lead Free is better?
> >>
> >>
> >>
> >>
> >>
> >> Thank you
> >>
> >> Juliano Ribeiro
> >
>