Mach Effect progress

[GIThruster]

Its a newish program. Corresponds to the new spring loaded gadget that shoots them out the side of ISS.

[hanelyp]

Since NASA does loft nanos for free...

One benefit of testing in orbit is a thruster that delivers micro-newtons of net thrust under extensive noisy vibration could still deliver an overwhelmingly unambiguous signal. Coming anywhere close to that level of isolation from sticktion by any other way is HARD.

[GIThruster]

Coming anywhere close to that level of isolation from sticktion by any other way is HARD.

True, but you don't need to actually isolate to identify and account for sticktion or "Dean Drive" effects. You can make small changes to the system that would affect such effects and thereby eliminate them. If you place an attenuator into the system and the signal changes, you know you have a false signal. Constant thrust survives such things. This is why all the methods and protocols are so very important in identifying thrust signatures.

But yeah, the whole point of TRL7 is you don't need to think about the details any longer--you can see the thing does or does not work as intended. TRL7 is arguably the most important milestone in the TRL progression.

[AcesHigh]

question is: which is cheaper and more possible at short term? Making a nano sattelite with ME propulsion to test it in space?

or make a more powerful version that can prove the concept, maybe optically, beyond any margin of error (that means much better thrusts), at ground?

[GIThruster]

The second is necessary for the first, so it's not a choice. However, there are excellent reasons that any future thruster ought to be so designed as can be placed in a nanosat, so you stay on the proven path to get TRL7.

[vnbt4]

I hope i am not interrupting the current train of thought in the thread at this time but I have a relatively stupid question. In the impulse term of the Transient mass fluctuation equation there is a value i am having a hard time finding out how to solve for. \rho 0 or the proper density of the dielectric, I hope those on the forum with a background in mathematics could assist me. Thank you

[GIThruster]

Are you trying to get a tool for thrust magnitude with given values? I'm asking because Andrew Palfreyman did this back in 2004 or '05 and there are several hidden variables in the mix that make this impractical without the ability to observe any given thruster/setup. For example, when you preload the ceramic to stop it from explosive decomposition, you reduce both the k and the electromechanical linking coefficient, and importantly, you reduce the excursions of the ceramic, and thus the accelerations it experiences. Since both the M-E generation, and the thrust generated scale linearly with excursion and acceleration, preload ends up contributing an inverse quadratic effect on thrust. That's not something even good FEM can predict well, since even then you need estimated values. So to test for an accord with theory for thrust magnitude, you really need a laser interferometer vibrometer and look to see what the actual excursions are, as well as to test the ceramic's dielectric constant at the frequency and preload used.

Too many loose variables for sensible predictions, save general ones saying "we ought to do better with this preload than that" etc.

[hanelyp]

Related to dielectric density, how should ME device performance with a vacuum capacitor compare?

[GIThruster]

M-E thrust scales with the square of the dielectric constant k, but requires there be some mass to experience a Mach Effect, or percent mass fluctuation. Vacuum caps can't generate Mach Effects at all since there is no active mass.

The requirements to create a Mach Effect are:

1) matter that includes "squishy bonds" that can store the mass fluctuation. The particles don't actually change mass. The molecular bonds between them do.

2) the matter experience an internal change in energy--the energy of the bonds

3) the matter accelerate with respect to the rest of the universe and especially the Far Off Active Matter or FOAM as per Mach's Principle.

If there's no mass in the cap, it can't meet any of these criteria.

[vnbt4]

I am trying to calculate the potential thrust for a MLT device. After my visit to Dr. Woodward's lab I bought his book and decided to build a test device. I was hoping to calculate the thrust from (starting with one cap and two inductor coils) the device and find out exactly how much the caps contribute to the thrust under the acceleration form the inductors. I have already taken into account the need for phase control in the circuits i am working on but they are very basic. I also have a 2Mhz frequency generator which I intend to run my 1Kv 2200pf caps off of. My knowledge in the electronics and math are very basic and i doubt i have accounted for all the variables in the circuit or in the calculations. Because of this i was hoping to calculate the thrust of the device before construction so that i could optimize the thruster itself allowing me to tweak the circuit if I missed some aspect of the physics or electronic controller.

[GIThruster]

Well, there are a handful of difficulties with the approach you sound to be using so I'll just try to run over them quickly.

In general, perovskite caps such as PZT and BaTiO3 include an acoustic attenuator in their bonding or sintering agent, so they don't develop significant extensions and accelerations. What you want to use is more like a PZT disc or some such intended for force transducers or displacement actuators. The softer PZT blend also has significant electrostriction which is what Jim has been using, and you can get that stuff cheap on EBay from Steminc. http://www.steminc.com/ Using something that has both a piezo and an electrostrictive excursion allows you to use a sine, triangle, square or preferably sawtooth wave, as opposed to the 1w+2w wave necessary when you have only one of these two actions.

Something needs to then "preload" the devise so the material doesn't fly apart (since its lacking the above attenuator influence). The preload requires significant design since the higher the frequency, the less efficient the preload becomes. A badly designed preload can thwart 99% of an actuator's extension in dynamic mode, so this is a very important part of the design process.

The active mass has to be secured to a passive mass or preferably, to an acoustic mirror that will magnify the mechanical Q by reflecting the acoustic energy back into the active mass to promote extension. This requires a 1/4 wave resonator and 1/4 wave Bragg Reflector layers. You can look up Bragg on wiki. The devise will not operate at all, unless the passive mass or reflector causes the active mass to act as a 1/4 wave resonator. This is NOT a simple issue.

Just as you've mentioned, the RLC tank circuit you create has to be run on resonance, meaning the the capacitance and inductance need to be matched. Since the capacitance of the active mass changes as it heats up, you'll need a variable inductor to tune the circuit in real time. If you don't do this, the current and voltage will arrive at different times, and you won't have the dE/dt into the device necessary to generate Mach Effects.

Also as you seem to be aware, it will help you greatly to have a PLL circuit to adapt itself to the changing mechanical resonance of the thruster as it heats.

Finally, PZT, BaTiO3 and PMN all have their dielectric constant drop off dramatically well before 2Mhz. I doubt you can get useful results from the setup you have in mind. Even if you can get the thing running like a top at 2Mhz, the capacitance will then be so low as to thwart worthy investigation. My suggestion is use a different material, or go to lower frequency. There is a PLL amp that works quite well to 120 Khz, and that might be a better solution to the dielectric issue. That, or if you have a friend with a crystal growth tank who won't mind loaning it out for 8 months. . . Here's a cheap PLL generator, IIRC it is 20-120Khz:

http://www.ebay.com/itm/Sono-Tek-06-051 ... 485d79beac

http://www.ebay.com/itm/Sono-tek-Part-0 ... 4d187ad1c8

And of course the thruster is not the harder part of the process. The instrumentation and measurement apparatus is the more difficult part. Though if you have reason to suspect you have something, you could possibly get Jim to put it on his balance.

If this stuff were simple as it seems at first, you can be assured we would have very different data to date.

BTW, there are simpler M-E experiments than thrusters one can do if you find this daunting. For example, if you hang a big chunk of PZT suspended by an acoustic attenuator like a rubber band on a balance and power it at resonance, it will not produce thrust because it will resonate in 1/2 wave mode (both ends free). This allows it to resonate at twice the frequency at any given thickness and mass as a 1/4 wave resonator. The ends of the stack are then accelerating in opposite directions and you can't get thrust because the thrusts will cancel, however, you can measure for loss of mass. Given a large enough mass and a large enough fluctuation, you could use something cheap like a Mettler H20 balance for this and might get an interesting result. I happen to own a Mettler H20 so if you or anyone else here takes an interest and design a useful experiment let me know and I'll have it forwarded to you. For truly useful results you need to add vacuum, but if you had a really interesting result, there are places where you might find vacuum available. As with the above, the preload is a major challenge here. The best solution is very likely to use an "antagonistic" geometry where the 2 free ends of a pair of 1/4 wave resonators meet in the middle and the entire thing is put i a cage that is not intended to change size/shape, but that has the limitations of lower frequency all 1/4 wave resonators have. Another solution is to buy an assembled actuator from a provider that has already preloaded it and tested it under dynamic conditions, such as PiCeramics. And there are other options. . .

On a positive note, every dielectric material one would have an interest in, should have a "sweet spot" where it can be run sans preload. For example, PZT has been run in thin films at 40 Ghz in TFBAR's and these do not include preload despite they are in fact, 1/4 wave resonators. The issue is, that when the frequency goes up, the accelerations go up, but the mass and force goes down, and at some frequency there comes a time when 1/4 wave resonators no longer need to be preloaded. (Yes, this seems odd since both acceleration and force are linear with mass, but it is true nonetheless or there would be no unloaded MEMs.)

If you like, you can purchase 2Mhz piezo discs from Steminc and run them suspended as 1/2 wave resonators and just see if they shatter. They likely will, but who knows? Even given the greatly reduced k of PZT at 2 Mhz, it is possible you could construct a working MET with it at 2 Mhz. (Actually a 2 Mhz disc will resonate at 1 Mhz when in 1/4 wave mode.)

So there are things to try here and learn from, but in any case, you'll need a broadband generator. You can't run discs that change mechanical resonance as they heat with a single frequency generator. You can add active cooling with a cheap Peltier device also from EBay and stabilize the resonance, but you'll still need a broadband PLL equipped generator to search for and find the proper frequency of any given resonator.

[vnbt4]

Are you referring to the work of Andrew Palfreyman? Also what would you suggest as starter transducers from the website you suggested.

[GIThruster]

Yes, Andrew did the thrust analysis but it was flawed as I noted above. There are too many "loose variables" for the equation to be useful.

I can recommend all sort of different kinds of approach but without the details of what you're planning, I'm at a loss. If for example you want to use a conventionally preloaded stack and accelerometer, the largest diameter is likely the best choice. The thickness is determined by the voltage you can develop. While there are significant advantages to using a single disc for a stack and removing internal reflections, this requires a higher voltage source than if you use a stack of thinner discs. There are many such trades to consider.

If you just want to run some PZT at 1/2 wave for loss of mass, they have a 2Mhz disc 43mm dia. For about $10 you could at least find out if it will crack when powered sans preload. If not, then you can test it for its true k value at 2Mhz and do the calculation of how much mass it ought to lose.

[vnbt4]

May i ask how much you know about the trust balance used by Doctor Woodward?

[GIThruster]

I know lots. I've been involved in his work for more than 8 years, since before Tom built the balance. There's lots of info about it in his book but if you have any simple questions I'll be happy to answer them.