Back when I used to work in the Department of Radiation Oncology at Stanford University from 1987 to 1992 we made several high powered Ultrasound transducers to heat cancer tumors for Hypertheremia treatments.
I have some of my old papers on line here: http://www.dnull.com/stanford/
Some of this data was never published or I am now unable to find it and would like to share it now.
We had 2 transducers that we wanted to map the fields patters on.
Both were 10 cm PZT disks approx 1mm thick.
One was optimal at around 1 Mhz, and the other was 2 Mhz.
The disks were mounted in an aluminum housing and had nickle platting on both sides of the PZT disks electrical connection was around the periphery and made using silver impregnated conductive rubber O-Rings.
I worked on all the the machining, software and some of the electronics under the direction of Dr. Stavros Prionas.
We made a large acrylic hexagon shaped water tank about 3 feet across.
In it was degassed water and the transducer was mounted in to one of the walls of the tank, on the far side was a rubber acoustic absorber to prevent reflections.
The transducers could output 100 Watts of acoustic power.
You can click on the image to enlarge.
Our sensor was a small 1 mm bead of silicon rubber mounted on a Type T (copper–constantan) thermocouple. We would measure the SAR, Specific absorption rate, basically the slope of temperature rise.
The thermocouple was mounted on the end of a 1 mm OD stainless steel tube, that was attached to a 3 axis robotic stepper motor positioner.
Capturing the images below would take days for just a single image. Power would be applied, the temperature rise would be measured, power off, then we would wait 2 minutes to cool down, move 1 mm, then run the test again at the next position. Between each measurement we would turn our RF generator on and recalibrate against a dummy load to make sure that our power levels were consistent as the amplifiers temperature changed.
The end result is like looking at the beam from a flashlight but these are images of the ultrasound beam we were generating.
In the images you can see patters of constructing and destructive interference.