Hi all,
Please let me add another point that no one has yet mentioned, a little
bit more fundamental. How do ultrasonics work? The transducer, when
moving "upwards" tries to compress an incompressible liquid and thus
displaces it. When moving "downwards", it is easier to create a vacuum
bubble, called a cavity, than to move the whole mass of liquid. These
cavities form where there is a nucleus, which may be any foreign
molecule but is hopefully the a part of the surface to be cleaned. At
the next compression, the cavity does not collapse (why, I know not: it
is like as if the wave passed through a diode), so the cavity gets
larger and larger with each cycle of compression and decompression. Once
it reaches a critical size, it implodes in a nanosecond, compressing the
vapours it contains adiabatically to an astronomically high temperature
(thousands of degrees). This rise in temperature is, of course,
dissipated instantaneously but in so doing, creates an intense shock
wave, which is what enhances the cleaning action of the solvent.
With a fixed frequency system, the waveform of the shock is a spike,
usually repetitive, which is a subharmonic of the U/S frequency and
contains harmonics up to the oscillator frequency, so that it can
augment resonance where a bonding wire happens to have a fundamental or
harmonic natural frequency of approximately the same value. With a
sliding frequency oscillator or multi-frequency system, the implosion of
the cavitation happens at pseudo-random times, so augmented resonance
due to the cavitation will occur less frequently, but it will still
happen, even though the results may be less disastrous.
This explanation also shows why the solvent must be degassed before U/S
cleaning can be effective. If there is a gas molecule dissolved, it will
form a nucleus which will cause cavitation at a random position in the
bath, but usually well away from the articles being cleaned. In the case
of "traditional" halogenated solvents in a vapour degreaser, the
combination of boiling and the U/S cavitation itself will promote
degassing, so that after, say, one hour, the efficiency will be at its
highest. Nevertheless, dissolved dirt will also be deleterious, for the
same reason, so the solvent in the cold tank must always be kept as
clean as possible by an adequate distillation rate. Fluorinated
solvents, however, may dissolve oxygen from the air more readily than
other types and require better degassing.
Water and aqueous solutions are much more difficult to degas, because
the higher surface tension tends to cause the gas bubbles to not escape.
Practical experience has shown that introducing an external degasser
into the DI water stream feeding an ultrasonic tank will increase the
cleaning efficiency (and speed of cleaning) very remarkably. At the same
time, because there is no oxygen or CO2 dissolved, ferrous metals will
not rust in the bath.
In practice, I concur there is ALWAYS a risk of U/S cleaning where there
are components with unsupported wires of a length which may resonate
(i.e. in a cavity). This may include many component types and I have had
more failures with quartz crystals than any other component type. I
don't know whether this is due to the high-Q resonance of the crystals
themselves being set off by a harmonic of the U/S frequency or
cavitation, or the wire soldered to the metallisation resonating. Either
is plausible. Post mortem examination showed the metallisation to be
torn away round the joint. Cleaning without U/S caused no problems. I've
also had a few failures of transistors in TO-18 and TO-5 cases: not
sufficient in number to warrant action and it could be that they were
components that would fail in a short time, anyway. However, we were
afraid that good components may be weakened, so we stopped U/S cleaning
assemblies. I agree that glob-topped components and those otherwise
fully encapsulated should be safe. I have also had a very insidious
case. A client called me in complaining that a certain large ceramic
multilayer capacitor was failing in service. This was before SMD and the
component was, if you like, like a large SMD one with wires soldered at
either end and subsequently encapsulated in a very thin layer of resin.
Examination of the returned assemblies showed a number of these
capacitors were missing a slice of their body, reducing the capacity
pro-rata. My client and I conducted systematic tests to try and find a
cause (the manufacturer was not able to help). We found it in the
ultrasonic cleaner. In fact, almost all the capacitors cleaved, but they
were held together by the metallisation at each end. Only a microscopic
examination revealed the presence of an externally visible crack, with
an estimated width of < 1 um. In time, mechanical transportation
vibration, thermal cycling or whatever caused the metallisation to fail,
reducing the capacity and causing the circuit to malfunction. I
theorised that the capacitor was of a physical size to correspond with a
quarter wavelength of the U/S frequency in the solvent and was therefore
set into oscillation in such a way that the cleavage planes were
stressed. By putting capacitors in the bath in a certain orientation, we
were able to get complete disintegration in about one hour, showing
conclusively that the U/S was responsible, as there was no problem with
the oscillator switched off in the same conditions.
I have always advised my clients that they should use U/S ONLY if they
could PROVE that no damage to any component resulted from long and
systematic testing which would include thermal cycling and shock and
vibration testing, as well as field failure analyses. This is something
that cannot be forecast with precision: some clients clean "forbidden
components" with U/S very successfully: others using the same components
may not be able to, even though their cleaning machines are identical. I
recently visited a client in the Far East who was cleaning silicon
strain gauge diaphragms. He had been using U/S in a halogenated solvent
very successfully for years. A new model, slightly smaller and thinner,
simply shattered, to his total surprise, because in reality it was more
mechanically robust!
IMHO, the answer is a lemon: you have to suck it and see to find out how
sour it is.
FWIW
Brian
Vinit Verma wrote:
>
> Hi Technetters,
>
> I am presently evaluating post reflow PCB cleaning machines, both aqueous
> and ultrasonic. I have a concern regarding ultrasonic cleaning. Read
> somewhere that the ultrasonic frequencies can have an effect on the wire
> bonds inside the packaging. Does anyone have any idea of this?
>
> Thanks in anticipation.
>
> Regards
> Vinit Verma
>
> ---------------------------------------------------------------------------------
> Technet Mail List provided as a free 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 Technet
> To temporarily halt delivery of Technet send the following message: SET Technet NOMAIL
> Search previous postings at: www.ipc.org > On-Line Resources & Databases > E-mail Archives
> Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
> information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
> ---------------------------------------------------------------------------------
---------------------------------------------------------------------------------
Technet Mail List provided as a free 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 Technet
To temporarily halt delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL
To recieve ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search previous postings at: www.ipc.org > On-Line Resources & Databases > E-mail Archives
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
---------------------------------------------------------------------------------
|
