Phil
Ceramic capacitor cracking is NOT a phobia, it is for real with large capacitors. The
condition required for it to happen is that the "thickness of the sandwich" is about
one-quarter wavelength of the ultrasonic frequency(ies), taking into account the
propagation velocity of the pressure waves in the fluid and the direction. Unlike other
components, which may be damaged by the energy release of cavitation, the cleavage of
capacitors is purely mechanical caused by the compression stresses. I had a client,
umpteen years ago, who suffered from this with one given capacitor type. The
manufacturers researched the cause and, having found it, made some 2 mm lower but wider
capacitors and the problem was solved.
Other components on which I have seen ultrasonic damage occurring include: metal case
semiconductors, ceramic case semiconductors (i.e. those with unencapsulated bonding
wires), quartz crystals, glass diodes, vacuum tubes (valves), especially those with low
current filament cathodes and LCD displays. This list is only my experience and is not
necessarily exhaustive. As a general rule, I have seen no problems with semiconductors
where the bonding wire is supported by encapsulation (moulded plastic casing or
glob-tops), with one exception: a high current (50 A) SCR COB with a very thick bonding
wire.
Most of the above damage has occurred with 40 - 50 kHz agitation. I have one client
using a triple frequency system with the lowest energy at 50 kHz, higher energy at 80
kHz and highest energy at 110 kHz. Although it is less energetically efficient, the
three frequencies seem to give a synergistic cavitation which is gentler yet more
penetrating into blind cavities. He has not experienced any component degradation with
it.
Brian
Phil Zarrow wrote:
> I agree with Mike completely. Too many people make the mistake of simply
> wiping wtih IPA or some other solvent and calling it "clean". Solder spheres
> become trapped in between the solder mask and the pads as well as in vias.
> These spheres are contaminated with the IPA and generally do not meld with
> the new solder paste. The result- solder balls (and sometimes worse).
> When a misprint occurs, wipe the majority of the solder off the board with
> IPA. Then, either run the board through an aqueous cleaner with spray and
> immersion or put it in an ultrasonic stencil cleaner. My research a number
> of years ago uncovered the same wisdom Mike pointed out. Ultrasonics will
> not harm the component per se. One interesting phobia is the fear of
> cracking ceramic capacitors. I learned from my friend and colleague John
> Maxwell (he had done numerous comprehensive studies) that if the capacitor
> was manufactured improperly - specifically if the dielectric had been
> cofired incorrectly, there may be a tendency to crack in ultrasonics.
> However, this is a component that will likely fail quickly in the field
> anyway (assuming it makes it past ICT) with or without ultrasonics. The
> ultrasonics act, in this case, as ESS !
>
> Phil Zarrow
> ITM, Inc.
> Durham, NH USA
> www.ITM-SMT.com
>
> ##############################################################
> TechNet Mail List provided as a free service by IPC using LISTSERV 1.8c
> ##############################################################
> To subscribe/unsubscribe, send a message to [log in to unmask] with following text in
> the body:
> To subscribe: SUBSCRIBE TECHNET <your full name>
> To unsubscribe: SIGNOFF TECHNET
> ##############################################################
> Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
> information.
> If you need assistance - 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 1.8c
##############################################################
To subscribe/unsubscribe, send a message to [log in to unmask] with following text in
the body:
To subscribe: SUBSCRIBE TECHNET <your full name>
To unsubscribe: SIGNOFF TECHNET
##############################################################
Please visit IPC web site (http://www.ipc.org/html/forum.htm) for additional
information.
If you need assistance - contact Keach Sasamori at [log in to unmask] or
847-509-9700 ext.5315
##############################################################
|
