Mach Effect progress

[DeltaV]

MONDAY, JULY 30, 2012
...
NFF-03. Future Flight Propulsion Systems II
Chair(s): Gregory Meholic (The Aerospace Corporation) and Marc Millis (Tau Zero Foundation)
...
5:00 PM - 5:30 PM
CONTROL ID: 1284801
TITLE: Recent Results of an Investigation of Mach Effect Thrusters

AUTHORS (LAST NAME, FIRST NAME): Woodward, James F.1
INSTITUTIONS (ALL): 1. California State University, Fullerton, CA, United States.

PRESENTATION TYPE: Technical Paper
CURRENT TOPIC: 48th AIAA/ASME/SAE/ASEE Joint Propulsion Conference & Exhibit
CURRENT SUB-TOPIC: Nuclear and Future Flight Propulsion: Future Flight Propulsion Systems
KEYWORDS: PROPULSION & ENERGY: Energetic Components & Systems, PROPULSION & ENERGY: Electric Propulsion.

[GIThruster]

from another thread:

I hope sincerely that Woodward or March will demonstrate scaling with frequency. The results will I think be incontrovertible, even to those now theoretically opposed.

The problem with demonstrating scaling with frequency is that one gets much better mechanical displacement and its subsequent accelerations on resonance. So even if Jim had a very expensive, custom high speed z-matching network and could sweep the frequency without phase shift, the system would still show a thrust spike at resonance rather than the scaling curve you'd like to see.

Jim has done this kind of work before. When he was testing the MLT he got the sorts of scaling with frequency results theory predicts. He didn't however get the thrust magnitudes he wanted because the MLT is not set up to generate sufficient bulk acceleration.

Since all M-E thrusters require mechanical bulk acceleration, they are all acoustic resonators. Since they're resonators, they are all best operated at a single frequency. IMHO, the scaling data you're suggesting would be convincing, should be sought with other methods, such as scaling the voltage. That's been done and clearly demonstrates thrust scaling as per theory.

[GeeGee]

He didn't however get the thrust magnitudes he wanted because the MLT is not set up to generate sufficient bulk acceleration.

I was reading Nembo's SPESIF paper on mach effects, and he suggests using ferromagnetic materials to produce a strong bulk acceleration:

"In an attempt to get the best of both worlds, the use of ferromagnetic materials is proposed. Using ferromagnetic materials as active materials for producing mass fluctuations has several benefits. Using a divergent magnetic field, it is possible to induce a change in the internal energy of the body and a simultaneous unidirectional strong bulk acceleration of the body itself. The main difference between accelerating a body by means of a mechanical push and accelerating it with electromagnetic fields is that in the former case a pressure wave is generated that is propagating at the sound speed in the body, while in the latter all the ferromagnetic particles of the body are accelerated simultaneously in the same direction. An effect similar to pressure waves, in a sense that it is a macroscopic mechanical effect as well, still remains: tidal forces. Because the divergent magnetic field possesses a gradient, the side of the body facing the high field region will be subjected to a stronger force than the opposite side. However, these tidal forces would contribute additively to the change of internal energy of the body, as they would manifest themselves with the same timing of the other energy storing phenomena (domains alignment,
magnetostriction)."

Is he (or anyone else) currently trying to do the experiment he outlined in his paper?

[GIThruster]

I haven't spoken with Nembo in many months, but last I heard yes indeed, he planned to pursue this line of experimentation, and he is a first rate experimentalist. I imagine if he has some progress we'll find out in Feb. or whenever the next STAIF II is.

[Diogenes]

He didn't however get the thrust magnitudes he wanted because the MLT is not set up to generate sufficient bulk acceleration.

I was reading Nembo's SPESIF paper on mach effects, and he suggests using ferromagnetic materials to produce a strong bulk acceleration:

"In an attempt to get the best of both worlds, the use of ferromagnetic materials is proposed. Using ferromagnetic materials as active materials for producing mass fluctuations has several benefits. Using a divergent magnetic field, it is possible to induce a change in the internal energy of the body and a simultaneous unidirectional strong bulk acceleration of the body itself. The main difference between accelerating a body by means of a mechanical push and accelerating it with electromagnetic fields is that in the former case a pressure wave is generated that is propagating at the sound speed in the body, while in the latter all the ferromagnetic particles of the body are accelerated simultaneously in the same direction. An effect similar to pressure waves, in a sense that it is a macroscopic mechanical effect as well, still remains: tidal forces. Because the divergent magnetic field possesses a gradient, the side of the body facing the high field region will be subjected to a stronger force than the opposite side. However, these tidal forces would contribute additively to the change of internal energy of the body, as they would manifest themselves with the same timing of the other energy storing phenomena (domains alignment,
magnetostriction)."

Very clever idea. The magnetic field would act on all the molecules simultaneously, and thereby possibly alleviate the acoustical coupling problems. Of course inductive reactance increases with frequency, so it would seemingly be harder to take advantage of the higher frequencies if it were constructed as a standard coil design.

I think ferromagnetic ultrasonic tranducers can be driven around 100khz, but I can't say i've heard of ferromagnetic transducers operating in the Mhz range.

Perhaps the ferromagnetic material can be shaped into a resonate form and driven with an efficiently coupled radio wave? That might make it possible to get the thing up into the megahertz and beyond. Might make for a much simpler design.

It could be massively paralleled and spaced at resonance nodes, (on a plane) and driven simultaneously from the same rf source.

[kurt9]

He didn't however get the thrust magnitudes he wanted because the MLT is not set up to generate sufficient bulk acceleration.

I was reading Nembo's SPESIF paper on mach effects, and he suggests using ferromagnetic materials to produce a strong bulk acceleration:

"In an attempt to get the best of both worlds, the use of ferromagnetic materials is proposed. Using ferromagnetic materials as active materials for producing mass fluctuations has several benefits. Using a divergent magnetic field, it is possible to induce a change in the internal energy of the body and a simultaneous unidirectional strong bulk acceleration of the body itself. The main difference between accelerating a body by means of a mechanical push and accelerating it with electromagnetic fields is that in the former case a pressure wave is generated that is propagating at the sound speed in the body, while in the latter all the ferromagnetic particles of the body are accelerated simultaneously in the same direction. An effect similar to pressure waves, in a sense that it is a macroscopic mechanical effect as well, still remains: tidal forces. Because the divergent magnetic field possesses a gradient, the side of the body facing the high field region will be subjected to a stronger force than the opposite side. However, these tidal forces would contribute additively to the change of internal energy of the body, as they would manifest themselves with the same timing of the other energy storing phenomena (domains alignment,
magnetostriction)."

Very clever idea. The magnetic field would act on all the molecules simultaneously, and thereby possibly alleviate the acoustical coupling problems. Of course inductive reactance increases with frequency, so it would seemingly be harder to take advantage of the higher frequencies if it were constructed as a standard coil design.

I think ferromagnetic ultrasonic tranducers can be driven around 100khz, but I can't say i've heard of ferromagnetic transducers operating in the Mhz range.

Perhaps the ferromagnetic material can be shaped into a resonate form and driven with an efficiently coupled radio wave? That might make it possible to get the thing up into the megahertz and beyond. Might make for a much simpler design.

It could be massively paralleled and spaced at resonance nodes, (on a plane) and driven simultaneously from the same rf source.

I have a copy of one of Nembo's earlier experiments. He cites inductive reactance as a frequency limitation in his approach. However, 100KHz should be good enough as proof of concept. I think an actual propulsion device requires MHz frequency assuming that the affect scales to the power of 3 as function of frequency.

[djolds1]

I have a copy of one of Nembo's earlier experiments. He cites inductive reactance as a frequency limitation in his approach. However, 100KHz should be good enough as proof of concept. I think an actual propulsion device requires MHz frequency assuming that the affect scales to the power of 3 as function of frequency.

Metamaterials have been cited as allowing some IMPRESSIVE manipulations of magnetic fields. Sufficient to pop the operating frequencies 1 to 4 orders of magnitude (1 MHz to 1 GHz) over what a ferromagnetic core alone could produce?

[GIThruster]

I just forwarded Jim's technical group some materials on superconducting metamaterials that manipulate Thz frequencies as they'd be handy for certain M-E applications, but I haven't heard back what people's opinions are yet.

As Andrew Palfreyman put it, "there are lots of ways to skin this M-E cat".

[GIThruster]

"If the power source (with no outside input) is not part of the masses apparatus, they are NOT performing a closed experiment, and are getting a local effect from moving mass in the system.

Like I said, one objection after the next until he finds one that sticks.

Those who have examined the system, including PhD EE's from Oxford and Penn State; know that the kind of linking you are presuming, is not occurring. There have been plenty of tests and protocols to ensure this does not occur. In short, tests using a dummy load are the proper controls to remove this possibility above. Rotating the UFG to vertical is the perfect dummy load because in that configuration all the other variables are identical, and yet the system should not produce a thrust signature. Rotating the test item is only one of many such controls used to remove from possibility any spurious sources.

Again, what we have are the complaints/accusations of a nameless person posting anonymously on a blog, about an experiment he is not familiar with, and disparaging a professional with 30 years experience in his field. It's pretty hard even on the face of it to take Wiz seriously.

[GeeGee]

Who are you responding to? I don't see that post in this thread.

Also, has anyone conducted a self-contained experiment? Would it be harder to do than the current set-up?

[GIThruster]

That was copied from WizWom in another thread.

Yes, a self contained experiment was done with the MLT some years ago, but it did not generate thrust. That was one of the experiments that led up to Nembo's Bulk Acceleration Conjecture, which was to note that from the maths, the entire dielectric need to be accelerated, not just the mobile ions in a the BaTiO3 lattice as with an MLT. It's important to note though, that the Bulk Acceleration Conjecture was in the math. It was not an empirical discovery. The problem with self-contained apparatus is if they don't work, you really don't know why.

Self contained is much, much harder to do and delivers far less data, because unless you have a huge research budget, you have to give up things like monitoring the accelerations of the active mass. It is the ability to measure such things that makes one able to test various issues. You might have noted on the mail list just these last few days, Jim mentioned he was doing a spectral analysis of the 1w and 2w accelerations to rule out another possible source of error. You can't do things like that with a normal self-contained experiment.

Of course you could design a self-contained thruster and power system with pll active feedback, high speed z-matching for when the ceramic goes though inductance and capacitance changes with temperature, the ability to sweep the frequency and voltage, monitor accelerations, etc, all sending back real time data through a wireless link that is specially formed to penetrate all the HV and RF noise, but that's not something anyone is going to do without a team of engineers and several million dollars in funding. This is why it's important to note the distinctions between research and development. Self contained setups are Technology Readiness Level 6 or 7 activities. Jim's work is TRL-3.

[paulmarch]

I just forwarded Jim's technical group some materials on superconducting metamaterials that manipulate Thz frequencies as they'd be handy for certain M-E applications, but I haven't heard back what people's opinions are yet.

As Andrew Palfreyman put it, "there are lots of ways to skin this M-E cat".

Ron:

Food for thought. Pure barium titanate (BaTiO3) has a d33 piezoelectric response of ~78 picometers/Volt. I tested new batches of the Vishay/Cera-Mite Y5U, Z5U, Y5R and N4700 BaTiO3 based alloy ceramic dlielectric caps on the Eagleworks new d33 piezoelectric meter last month and found that all save the Y5R matieral had a d33 piezoelectric response of less than 2.2 picometer/Volt with no +/- polarity of note. The new Y5R based caps on the otherhand had a polarized d33 piezo reposne of approximately +9/-15 picometers/Volt, which is only ~15% of pure BaTiO3 piezo response, but it is considerably larger than the Y5U caps used by Jim in his MLT experiments. This could readily explain why the MLT-2004 performed as well as it did and the near null results that Nembo obtained in his STAIF-2006 test series while testing Jim's Mach-6 MHz test articles. This fact could also explain why Nembo only observed turn-on thrust spikes during these same tests due to the much larger dP/dt signals during these turn-on transients.

This leads to a problem though. My MLT-2004 used eight Y5R caps that did NOT have their piezoelectric polariity aligned becuase I didn't know about that issue at the time, yet I saw a thrust signature measured in milli-Newtons. I could have just been lucky or there was enough piezo polarity alignments to make things work up to a point. However, Jim's and my Faraday shielded Mach-2MHz test article used just two of the same Y5U Cera-Mite BaTiO3 alloy ceramic caps that Jim used in his Mach-6 test articles, but they were an older batch than what I tested last month, yet it typically demonstrated a milli-Newton of thrust before its thrust die off occured. If correlated d33 piezoelectric bulk accelerations along with a concurrent and phased alighed dP/dt signal are required to express the M-E, then why did the Mach-2Mhz perform as well as it did when it's bulk accelration was near zero? Unless of course that particular batch of Y5U caps did have some non-zero piezo response. There are still just too many uncontrolled variables in all this business to make much sense of it yet...

Best,

[GIThruster]

My MLT-2004 used eight Y5R caps that did NOT have their piezoelectric polariity aligned becuase I didn't know about that issue at the time, yet I saw a thrust signature measured in milli-Newtons.

Paul, allow I remind you of a conversation we had about 6 years ago. I have a strong recollection of something you said back then, but since haven't been able to verify or recall--namely that you said you suspected that your highest thrust test results were taken when the caps had been overheated, and you suspected they might have gone into cubic phase.

Several years ago I researched this and it appeared back then that BaTiO3 is more highly piezoactive in it's cubic phase, than even in highly polarized tetrahedral. So if indeed you did heat those caps sufficient to go to cubic, you would have created the conditions necessary to have a much higher bulk acceleration in those experiments than in any other MLT work to date.

That could indeed explain mN thrusts. . .

[Ric Capucho]

Hi Paul, All,

So what do you think would be the logical steps needed to move from away from formal scientific experimentation and more to proof of concept engineering? Say a small radio-controlled buggy? Could we get enough effect using Y5Rs (or Y5Us) powered by an array of cellphone batteries to demonstrate tangible movement on a nice smooth surface?

I'm thinking of scaring the cat.

Ric

[GIThruster]

Ric, IMHO, proof of concept engineering is to build something that produces commercial grade thrusts. 20mN with just a couple hundred watts is more than enough to replace all Hall thrusters, given the thrust die-off issue is addressed. You just need to be able to do this reliably and repeatably, under the proper control conditions such as vacuum.

IMHO, the only way to do this cheaply is to use a power system and Z-matching in the Ham range so you can use cheap, used EBay stuff, especially including a Johnson matchbox style tuner. You then are restricted to ceramics the thickness that corresponds to the frequency you have the power system for.

After that you identify the ceramic you can afford. You might for instance decide to use relatively cheap BaTiO3 single crystal targets heated to cubic phase. This would give you very high power density and frequency. and a very large active mass.

You might instead choose to use a phased array or multiple half wave resonator such as a monolythic actuator, where each layer of ceramic in a large stack corresponds to 1/4 wave of the frequency you drive at. This has the advantages that the ceramic from PiCeramics is co-fired ceramic encapsulated, and you have reasons to expect the lifetime of the ceramic to be improved several orders magnitude.

A viable third choice is to use PMN-PT from TRS technologies for it's extremely high electrostrictive electromechanical linking coefficient, extremely high k value and extremely high electrical standoff. The trouble with materials like PMN-85 is that it only works well in the narrow temperature bandwidth of 85*c, where it is in cubic phase and has a k value of 40,000. Out of cubic phase, it still has a k value far above PZT (what Jim uses) and BaTiO3 (what Paul used in the MLT), but its bulk acceleration drops off hugely. It's great stuff if you can get it, and you have to design around the fact it is primarily an electrostrictor rather than a Piezoactive ceramic, so provides no 1W mechanical response. You thus need a design for this material to provide another means of bulk acceleration to generate the M-E. I believe this is the direction Jim's next gen work with Heidi Fern will take.

In any case, even the simplest experiment that can be hopeful of success is very complex, and not best used to scare the cat, and if you can remove many of the hurdles by using someone else's measurement apparatus, vacuum and protocols, you can focus on just the thruster and power system itself, which is often the smaller part of the task.

The first question to ask is, do you have 1,000 hours to spend on your hobby over the next couple years? If so, there are several sources of help available. If with help you can design and build something that you can measure thrust on a Mettler H20, I'll ship you mine. If that produces results, best is then try to get it in the vacuum chambers of others who do this work.