In the early days of integrated circuits, say the 1960s and 1970s, ultrasonic cleaning was generally not used for cleaning components. The affect was not on the crystals or components themselves, but on the leads going through the packaging.

Repeated, high-powered ultrasonic cleaning simply vibrated the leads too strongly, causing metal fatigue and component failure. This was particularly true for military and aerospace systems, which actually banned ultrasonic cleaning in one of the now-long-forgotten mil specs.

Modern ultrasonic cleaners use "sweep" frequency management to avoid this problem. The machines change frequency constantly, so the energy is never on a harmonic frequency of the lead for very long, and never at the same power setting, which has been proven to minimize damage to components.

At the same time, I think every expert would urge careful and thorough testing on your boards, your components and your systems before certifying the process to be trouble-free.

A better answer might be to look at vapor degreasing. Vapor degreasing is a much gentler form of cleaning that does not need ultrasonics to work well. Check out one source at Micro Care, Why is Cleaning With Vertre Better than Cleaning With Water or Some Other Solvent ? or another option is Does MicroCare Have Any Solvents Suitable for Use in Vapor Degreasers?

If the frequency of the crystal matches the frequency of the ultrasound, then there is resonance and the component can fail.

Regular cleaners are better than ultrasonic cleaners precisely for this reason.

We have several published articles available for download on our web site, click on the "Recommended Reading" button.

The article by B.P. Richards et al. "Does Ultrasonic Cleaning of PCBs Cause Component Problems: An Appraisal" best addresses your question: "Does Ultrasonic cleaning have an adverse affect on crystals or other components on a PCBA?" My paper, "Reducing the Cost of Misprinted PCBs", addresses the parameters that would help prevent potential problems.

I also recommend the IPC-7526, "Stencil and Misprinted Board Cleaning Handbook."

The answer is no. The answer is yes.

Take your pick. There are two schools on this subject. Smart, intelligent, and experienced engineers will tell you that there is significant risk of component damage when exposed to ultrasonic energy. From damaged crystals to wire bond fractures, that fact is that there is the possibility of damage. In fact, several years ago, a major assembler of electronic assemblies published data linking component damage to ultrasonic energy.

In reality, while the above statements are true, the chance of component damage on modern ultrasonic cleaning equipment is rare. Modern ultrasonic cleaning equipment utilizes frequencies of 40 kHz. The frequency is "swept" up and down to reduce potential component damage. Also, if the watts to fluid volume is low enough, damage will be further reduced.

Here's the real issue. It doesn't really which school of thought you subscribe to. What matters is what your customer, QC manager, reliability engineer, and others think.

The fact that there is legitimate belief that ultrasonic energy has the potential to cause component damage normally precludes it from being used to preform a defluxing process.

The fact that you are asking this question, illustrates my point that ultrasonic energy is not openly accepted in the defluxing world, despite compelling arguments to the contrary.

I would strongly recommend spray-in-air technology for your cleaning / defluxing requirements. It is universally accepted as the conventional wisdom for defluxing applications and is never suspected of contributing to component damage.