Yes, it is. This is the difference between ultrasonic force transducers like Jim's thruster as well as the Langevin transducer (transducers you can buy prebuilt on EBay for $20), and displacement actuators which are spring loaded. Langevin's generally have mechanical Q, which is the best numeric measure of the extension of the crystal; of about 700. In order to get this, you have to have a reaction mass which is really just a single layer Bragg reflector or acoustic mirror, that is 1/4 the wavelength of the frequency you use.Diogenes wrote:9 is a long way from 30k. Perhaps there is a way to bind one of these higher K materials.
Perhaps i'm not grasping something. How do you get decent movement when you clamp all these disks together in a stack? Is the bolt stretching and relaxing at 30khz?
If you put an actuator on each side of the active mass, you'll get very close to sinusoidal motion. In order to stop that from being damped too badly, you want to use as light an active mass as possible, hence the ALTAS material we planned to try. You can purchase ALTAS type material caps online very easily, and glue a series of them together. Put them in series and you add the voltage and divide the capacitance, and they'll be much easier to drive.What I am envisioning is a free floating crystal
with a capacitor bonded to one or both sides. The crystal will both push and pull the capacitor, and if you make one with a capacitor on each side, the crystal would be separating them when one is more massive and the other is less massive while it would be pulling them when the opposite is true.
What you need to be wary of in this are two issues. First, if you put active mass on each side of the actuator, then the thrusts will cancel. That's no help. Second, you need to find a way to have the two drive signals perfectly in phase, or you won't get thrust. The cap needs to drive at 1w, and the actuators at 1w+2w.n (This is unless you can confirm that the caps themselves generate sufficient accelerations to provide the 1w acceleration, in which case the actuators can do just the 2w. To make such a confirmation, I think you'd need a laser doppler vibrometer and I don't know of any rentals services. Let me know if you find one.) Note that you cannot drive the actuators at 1w=the natural resonance R, because at 2R they will go into piezoelectrical antiresonance, where the impedance goes to infinity and no current can enter the stack. 2W cannot = 2R. Jim tried this for ages and it was a failure because of the electrical antiresonance issue he had not understood. 1w needs to be 1/2R.
If you take something like the ALTAS caps and glue them between a pair of cheap piezo actuators, there is a fair chance the setup won't disintegrate. While piezo materials need to be clamped in place at ultrasonic frequencies using what is called a "preload" that is normally about 1/10 the force generated by the actuator, you do not need to preload these materials at all at VHF. If you look at the construction of an ultrasonic cleaning transducer like the Langevin, they're clamped and the ceramic loses about 1/3 its k value from the clamping. If you look at the construction of a higher frequency cleaner, such as a contact lens cleaner operating at 2-5 Mhz, you'll see the ceramic is not clamped. When I discovered this I camped on it and studied it for a long time, eventually even speaking with a medical ultrasound designer who I wanted to do some fabrication work for me. He confirmed that at VHF and above, these materials do not need to be clamped.
So you could try it with and without the clamping but the clamp adds weight. First thing is take some cheap PZT from Steminc and power it in the VHF range to see if it disintegrates. And wear safety glasses.
These things create millions of gees acceleration and the tiny bits are sharp.So ideally for a shuttler, you want to test an actuator, can be any shape if the stuff between a pair; and see it doesn't fly apart unclamped. That will cost you $10. No problem. You can go rectangular, round or annular if you're concerned with heat buildup. Next is bond the caps to each other electrically--probably very careful soldering. Next is bond the caps and actuators together, probably a very hard epoxy bonding is called for--use whatever is the most rigid. You don't then HAVE to use an acoustic mirror, but you will get better results with than without because it is the 1/4 wave reflection that can give mechanical Q's above 1. I'm not sure how you'd create a Bragg reflector as a hobbyist, but FYI, the standard is 7 layers and you want one on each side. But as I said, Jim just uses the one reaction mass and Langevin's get Q's up to 700. Thing is with a proper Bragg, you get mechanical Q's up to 2,500. So its worth investigating after you have a resonant frequency for the active assembly, if you can have reflectors sputtered at a reasonable price. Sputtering services are pretty competitive so the prices are low. Bragg construction at these frequencies is problematic, however. In any case, you want an empirical finding for the frequency rather than calculate it since epoxy makes it impossible to predict what the thing will resonate at when complete.
Perhaps no reasonable size of capacitor can be affixed to a resonator in this manner.
The ALTAS caps are tiny because of their huge capacitance. You need to check to see what frequency they're rated for and work enough below that. They're not Chinese so they are probably properly derated.
