Introduction

• The transducer is a disk-shaped object 25mm dia by 13mm long, with a 100 mm pigtail for connection to the electronics.
• The active face of the transducer is immersed in castor oil; there are other oils that can be used, but this is cheap and relatively good for sonic impedance matching.
• The oil seal is achieved by a neoprene gasket. It is crinkled as shown to allow the transducer to articulate along the axis shown, to at least 45 degrees either side of centre. 30 degrees is shown in the drawing for illustrative purposes only.
• The gasket is sealed to the transducer by a hose clamp and silastic 732 RTV or similar. This seal is along a right-angled flange (not shown) on the gasket. The seal should not be achieved by the transducer mount, but by a separate structure that connects only to the gasket and transducer body. That way, alterations to the transducer axis assembly will not disrupt the oil seal, which can get messy.
• The gasket is sealed to the body of the assembly by a similar arrangement; it is not shown either. I expect that a silastic bond and a rectangular clamp with screws into the body would do the trick.
• The front face of the body has to be relatively transparent to ultrasound, and impedance matched in thickness to allow the best sound transfer. I don't have the details of which material and how thick at present; I understand that lexan (perspex) is one possible choice. For now, consider a 3mm thick rectangle of lexan, curved to the proper arc, and silastic'ed in place, to be the way to go.
• The shaft runs in a bearing, preferably a ball or roller race for best wearing. A plain bearing might do at a pinch. As it is not part of the oil seal, the bearing does not have to be airtight. Earlier envisages of the assembly did require this!
• The oil bath needs an air trap. This is a closeable port which allows air bubbles to be bled off and removed. Air bubble artefacts can seriously downgrade the images produced. The air trap must be accessible to the operator on a daily basis, so its construction must be robust, small, and reliable. Volume changes in the oil bath chamber (due to temperature expansion and contraction) will be accomodated by the neoprene gasket, so there is no real need for an air space in the oil bath chamber.
• The oscillating mechanism is a bit of a mystery at present. Two offerings are being considered - one being a stepper motor from a floppy drive, the other being a constant speed DC motor using a heart-shaped cam to achieve a constant radians per second sweep. Note that the electronics and software for these alternatives are quite different; the stepper motor can be relied upon to do the constant radians per second - but its operating life may be limited, and it may not be able to achieve the necessary oscillations per second. The DC motor needs serious work to minimise speed fluctuations, but it can easily do the 3 sweeps per second (in fact, small DC motors need a gearbox to reduce their shaft RPM to the desired speed).
• The printed circuit boards are still being designed. The preamp board should end up being about 50mm x 70mm x 15mm (thick), and the motor drive board should end up being about 100mm x 70mm x 15mm (thick).
• There are three coaxial cables emerging from the unit, which run to the computer. They are standard RG58 coax (about 4mm dia) and cable support/ strain relief, and termination must be allowed for. I leave the details to your good judgement. One cable is for future use (electronic focussing) and is not presently needed. But it's nice to have the hole for its placement handy.
• A multicore (non-coax) cable emerges from the unit also, it is CAT 5, 6mm diameter for control functions and power supply.