GIThruster wrote:Despite this, since the device has to oscillate at two frequencies, one of them will not be optimized. There's no way around this, even with more than one element in a stack.
Not following you here. I can see where the piezo device needs to be driven at mechanical resonance, but I do not see why the drive for the capacitor ought have an effect on the drive of the piezo device.
I would think you could drive them both with optimized frequencies.
Mach Effects are generated at the 2w of the M-E generating signal, so if you drive a piezo capacitor with high electromechanical linking properties, you are charging/discharging and generating acceleration at 1w or the frequency of the drive signal, but your fluctuation occurs at twice that frequency. You then need to "rectify" that fluctuation into useful force, by pushing it when it's heavy and pulling it when it's light, so that is a 2w drive. The exception is when you have a material that is both highly piezoactive (1w mechanical response) and electrostrictive (2w mechanical response). In such instances you can drive with a simple sine wave though, you'd be better off with sawtooth. In this instance, where the material has both mechanical responses, the phase between the 1w and 2w is a function of the material and it cannot be changed. You cannot alter the phase between the 1w and 2w components because there is no 2w drive signal component.
Ideally for proof of science, you want a material with only one of these two different electromechanical linking qualities, because when you're forced to drive with both 1w and 2w signals, you have control over the phase between them. By altering the phase, you can eliminate the thrust, which makes the thruster its own perfect dummy load, or reverse the thrust. Since there are no other phenomena that predict reversible thrust based on the phase between the 1w and 2w drive components, reversing thrust is a superb protocol in demonstrating the science. This is the main reason Woodward went to PMN--it has no piezo response. He has to provide both 1w and 2w drive signals but he can then demonstrates what's he's got by altering the phase angle between the drive components.
Regardless of how you get your 2 mechanical oscillations, if you do this with one element--meaning you don't have one to accelerate at 1w and another at 2w but for practical reasons you have just 1 element as in all Woodward's work the last decade--then one of those oscillations has to be off resonance since each assembly has only one resonance.
Now certainly the work-around is to have at least two elements driven on resonance at two different frequencies, and you get the benefit of each of these operating on their natural resonance. Woodward has never explained why he stopped doing this years ago. He just likes the setup he's using. For ultrasonic iterations this is trivial to realize since you have to clamp your components together anyway. For VHF and UHF, multiple elements thrusters are problematic because they're typically not clamped and the bonds between elements cannot be glued. They require the more advanced bonding techniques I mentioned the other day. Woodward has never availed himself to such. He doesn't even sputter electrodes. He uses epoxy and brass foil. These create acoustic reflections and account for the accelerometer noise he's always wrestling with.
There is utility in single element designs, especially as regards generating acceleration noise. If what you want is to get out of the noise floor, you really don't want internal reflections and places where you can get destructive interference because your stack isn't driven as a single unit.
"Courage is not just a virtue, but the form of every virtue at the testing point." C. S. Lewis