Author Topic: EM Drive Developments - related to space flight applications - Thread 3  (Read 1876607 times)

Offline mwvp

  • Full Member
  • **
  • Posts: 254
  • Coincidence? I think Not!
  • Liked: 169
  • Likes Given: 30
...Experience I have gained from others says it is critical to be able to tune the Rf amp frequency to the highest Return Loss dB (lowest VSWR) from the cavity as otherwise little or no Force generation will happen.

With respect, trying to play the EMDrive Poker Machine and hope a Rf gen frequency matches what the cavity needs is a good recipe to waste a lot of time and money.

I was going to correct you on your use of "Return Loss". I looked it up: https://en.wikipedia.org/wiki/Return_loss

Everyone I've been around was doing it wrong; we tuned for "low" (as in big-negative) return loss. But wikipedia says that's backwards. Well, you can't believe everything you read there, now can you?

Anyways, if you really want to tune to the center frequency of the cavity, put at tap on it, and put an amplifier between the very loosely coupled cavity output tap and the exciter loop. Then it's the tuning element of an oscillator, and will track the cavity.  ;D Not like you couldn't tune the cavity.

However, it may be advantageous to to just above or below the point it may oscillate at to maximize effects. Something to consider.

Offline TheTraveller

...Experience I have gained from others says it is critical to be able to tune the Rf amp frequency to the highest Return Loss dB (lowest VSWR) from the cavity as otherwise little or no Force generation will happen.

With respect, trying to play the EMDrive Poker Machine and hope a Rf gen frequency matches what the cavity needs is a good recipe to waste a lot of time and money.

I was going to correct you on your use of "Return Loss". I looked it up: https://en.wikipedia.org/wiki/Return_loss

Everyone I've been around was doing it wrong; we tuned for "low" (as in big-negative) return loss. But wikipedia says that's backwards. Well, you can't believe everything you read there, now can you?

Anyways, if you really want to tune to the center frequency of the cavity, put at tap on it, and put an amplifier between the very loosely coupled cavity output tap and the exciter loop. Then it's the tuning element of an oscillator, and will track the cavity.  ;D Not like you couldn't tune the cavity.

However, it may be advantageous to to just above or below the point it may oscillate at to maximize effects. Something to consider.

Using a 1 port example, Return Loss dB, VSWR and reflection coefficient are just different ways to measure the same thing.
http://cgi.www.telestrian.co.uk/cgi-bin/www.telestrian.co.uk/vswr.pl

As example a measured VSWR of 1.04:1 is also

1) Return Loss of: 34.151 dB
2) Reflection coefficient: 0.020

My Rf amp measures the real time VSWR when actively driving the cavity. I can then adjust the freq to get min VSWR or max Return Loss dBs. Doing this at min power of 79mWs will ensures no damage will happen to my 100W Rf amp.

Then knowing the max Return Loss dBs means I can take 3dBs off the peak value:
34.151dBs - 3dBs = 31.151dBs = 1.057 VSWR at the -3dB bandwidth points. Then simple to vary the frequency +- until the VSWR hits 1.057 at each side and measure the unloaded cavity bandwidth and unloaded Q.

Which then gives me an inbuilt 1 port S11 VNA like capability.

This calculator is also useful as it shows delivered power and reflected power (in Watts and dBms) as per the VSWR:
http://www.csgnetwork.com/vswrlosscalc.html
« Last Edit: 07/21/2015 08:14 AM by TheTraveller »
"As for me, I am tormented with an everlasting itch for things remote. I love to sail forbidden seas.”
Herman Melville, Moby Dick

Offline Ricvil

  • Member
  • Posts: 69
  • Liked: 66
  • Likes Given: 24
What about the cavity effective Q with negative resistence added by the magnetron?
One can model the magnetron  amplification by making the dipole antenna with negative loss material in Meep?
« Last Edit: 07/21/2015 10:17 AM by Ricvil »

Offline mwvp

  • Full Member
  • **
  • Posts: 254
  • Coincidence? I think Not!
  • Liked: 169
  • Likes Given: 30
What about the cavity effective Q with negative resistence added by the magnetron?

Depends what you mean by "effective". It's as you suspect, up to the point the magnetron saturates. You can increase the "effective" Q with either a 1 port negative resistance, or with a 2 port amplifier, if "effective" to you means having a high Q cavity for, say a filter or preselector on your radio and are not going to be saturating or running the magnetron at its maximum output.

I've tested circuits active filters with positive feedback that show increased Q. The downside is higher noise, and the potential for instability as the Barkhausen stability criterion unity-gain is approached.

But if "effective" for you is having the energy-equivalent of a magic-mass that exhibits asymmetrical inertia, or negative inertial impedance, then you want max energy, max mass, max magnetron output and the limiting factor for the max energy/effective cavity mass-energy is going to be the loss and dissipation  of the cavity and max magnetron output. The effective Q drops as the magnetron saturates.

So you go nowhere with this unless you're into filter-sharpening and don't care too much about stability, repeatability or noise.

One can model the magnetron  amplification by making the dipole antenna with negative loss material in Meep?

Lol. Now you have an unstable computer simulation ;D

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
...If I might be pardoned for butting in, the simulations so far, while very important to familiarize with the software, don't inspire my confidence WRT utility.....Use the eigenvalue mode of interest to excite the cavity in the time-domain (meep) and see what it does when the cavity is accelerated. I believe meep can do this. I read a post on the meep mailing list where the Cerenkov effect was being modeled by moving the charge-source between runs. Why not the cavity? Something I hope to look into....

Most important is familiarization with the numerical technique used to solve the partial differential equations and familiarization with what partial differential equations are being solved by the software.

With respect to what Meep can or cannot do, Meep is an open source software program NOT meant to be used as a black box but most people that use Meep in published papers write their own equations (constitutive equations, etc.) into Meep, thus the answer to most newbie questions into Meep is naturally "write your own function".
To write your own procedures, one has to familiarize himself/herself with what the code is actually solving.

With respect to << excite the cavity in the time-domain (meep) and see what it does when the cavity is accelerated. I believe meep can do this. I read a post on the meep mailing list where the Cerenkov effect was being modeled by moving the charge-source between runs. Why not the cavity?>> the answer is that Meep, out of the box, canNOT show you any cavity acceleration whatsoever.

Meep only solves Maxwell's equations.  There are no mechanical equations in Meep, there are no equations there that will allow you to model any Newtonian acceleration, even as a rigid-body (and much less as a deformable body, thermoelastic deformation, etc.).

Writing a procedure to move the charge-source between runs to model the Cerenkov effect is trivial in comparison with re-writing Meep to allow modeling of cavity acceleration.

Meep is not a multi-physics program.  For a multi-physics simulation you can use ANSYS, COMSOL, ADINA, and other codes, etc.  To include mechanical, multi-physics simulations into Meep would mean a major re-write, one would be better off writing the whole code from scratch.  To model cavity acceleration, thermoelastic deformation of the cavity, mechanical vibration, and other effects one needs something much more powerful and capable than Meep.



« Last Edit: 07/21/2015 01:20 PM by Rodal »

Offline Ricvil

  • Member
  • Posts: 69
  • Liked: 66
  • Likes Given: 24
What about the cavity effective Q with negative resistence added by the magnetron?

Depends what you mean by "effective". It's as you suspect, up to the point the magnetron saturates. You can increase the "effective" Q with either a 1 port negative resistance, or with a 2 port amplifier, if "effective" to you means having a high Q cavity for, say a filter or preselector on your radio and are not going to be saturating or running the magnetron at its maximum output.

I've tested circuits active filters with positive feedback that show increased Q. The downside is higher noise, and the potential for instability as the Barkhausen stability criterion unity-gain is approached.

But if "effective" for you is having the energy-equivalent of a magic-mass that exhibits asymmetrical inertia, or negative inertial impedance, then you want max energy, max mass, max magnetron output and the limiting factor for the max energy/effective cavity mass-energy is going to be the loss and dissipation  of the cavity and max magnetron output. The effective Q drops as the magnetron saturates.

So you go nowhere with this unless you're into filter-sharpening and don't care too much about stability, repeatability or noise.

One can model the magnetron  amplification by making the dipole antenna with negative loss material in Meep?

Lol. Now you have an unstable computer simulation ;D

Snif. So simulate a laser/maser cavity is futile. :)

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
...
My take on it, if it's for real, is that it behaves like an animation I saw on (the only?) youtube lecture by Woodward; a rocket with a spring and brick on back bouncing its way forward, sort of like a squid.

Unlike the Woodward effect which relies on the charged capacitor having more inertia than an uncharged one, the "Shawyer Effect" I call a "Sagnac Ratchet"; the force in the forward, and impedance in the reverse directions are the result the frustrum's asymmetrical dispersion, group velocity, and sum/difference frequency filtering characteristics.

Maybe Shawyer would have gotten a different response if he made it clear it was acting as a ratchet, so CoM wouldn't have been the issue.

Variable inertia (Woodward, etc.) gets around the Conservation of Momentum issue (and brings bigger cosmological problems, I will not get into now) but

"asymmetrical dispersion, group velocity, and sum/difference frequency filtering characteristics"

don't really get around the Conservation of Momentum issue in a frame-indifferent Universe.

Conservation of Momentum issues remain, and as often remarked by Frobnicat and deltaMass, even more disturbing issues related to conservation of energy are brought forth.

Even if it takes a little tap or a little vibration to get it started, it remains an overunity machine if it produces constant acceleration for a given Power Input that result in a free-energy machine.  Just changing the initial conditions does not solve the conservation of momentum and the conservation of energy problems.
« Last Edit: 07/21/2015 01:02 PM by Rodal »

Offline TheTraveller

Maybe Shawyer would have gotten a different response if he made it clear it was acting as a ratchet, so CoM wouldn't have been the issue.

He did make it clear, see attached, just nobody cared to read and think. Much easier to join groupthink, bash the man and his invention.
"As for me, I am tormented with an everlasting itch for things remote. I love to sail forbidden seas.”
Herman Melville, Moby Dick

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
People read statements like this in slide #5 of the previous document

<<An EM wave propagated inside a CLOSED waveguide is an OPEN system
Momentum can therefore be exchanged between the EM wave and the waveguide end walls
Momentum is therefore conserved>>

The man is not bashed.  Statements like these are bashed.  See the difference?

For more on this, and to appreciate that statements like these are bashed, rather than the man, read for example:

https://golem.ph.utexas.edu/category/2006/10/new_scientist_reacts.html

Quote from: Greg Egan
Shawyer’s claim is that the photons bounce against the walls of the cavity, and in so doing exert a force on it. In this he is perfectly correct. He further claims, however, that the net force on the cavity is asymmetrical, without any photons leaving the cavity, and in this he is not correct. The centre of mass of a closed system can not accelerate (or resist gravity either, since that amounts to the same thing) as a result of interactions between the parts of the system. That has been understood theoretically since the time of Newton, and confirmed experimentally in thousands of contexts and probably billions of individual events.   

Quote from: Greg Egan
Well, firstly Shawyer doesn’t actually share your belief that there is “a big new field yet to be discovered”, because he thinks his claims are a consequence of two old, established and respected theories: Maxwell’s theory of electromagnetism and Einstein’s theory of special relativity. He didn’t build his gizmo, measure this mysterious thrust, and then ask “What new science can explain this?” Rather, he invented the “drive” first on (he says) purely theoretical grounds, based on established theory.

This is why I (and a few million other people) know with mathematical certainty that his theoretical claims are wrong. Maxwell and Einstein’s theories can be expressed mathematically, and it can be proven that they will always without exception respect the relativistic law of conservation of momentum. In other words, whatever Shawyer has measured in the lab, if he thinks he has deduced theoretically from Maxwell and Einstein some consequence that violates that law, he is making a statement about mathematics, and it is trivially easy to prove that it is wrong.

Shawyer in fact insists that the claimed operation of his drive would not violate the law of conservation of momentum. Unfortunately for him, it’s even clearer that this is false. Before the drive is switched on, the momentum of the spacecraft in its rest frame is zero. If the drive is used for some time and then switched off, the momentum of the spacecraft in that same frame is now non-zero. The only way to make the “before” and “after” momenta equal is if there’s something else hanging around in the “after” scenario, such as photons emitted as exhaust. But Shawyer himself denies that there is any such exhaust (and even if there was, it would actually carry thousands of times too little momentum to balance his claimed effect). So he is claiming that zero equals something other than zero. I’m all for healthy scepticism against scientific orthodoxy, but that’s just farcical.

Then we come to his experiments, which conveniently support an effect that Shawyer “predicted” with this garbled non-theory. Can you really think of no other explanations for small forces appearing in a machine with a big power supply and a cooling system, other than your suggestion of exotic new physics (which is not Shawyer’s own claim at all), or Shawyer’s suggestion that zero equals not zero? 

The problem people have is NOT so much with the invention and it is certainly NOT with the man, it is the fact that the man from the start advanced a theory, and continues to support a theory that goes against known Physics.  It is his counter-Physics theory that is bashed, not the man.

What many people also ask, is whether it was wise for the man to have advanced and continue to defend this theory, when the man is not an academic, instead of concentrating on proving his invention and making his invention a commercial success.

In stark contrast, Edison did not advance and continue to push theories that went against known physics.  Edison invented and concentrated his efforts in making his inventions useful to mankind and a commercial success.
« Last Edit: 07/21/2015 02:04 PM by Rodal »

Offline Ricvil

  • Member
  • Posts: 69
  • Liked: 66
  • Likes Given: 24
Another thougth.

Offline mwvp

  • Full Member
  • **
  • Posts: 254
  • Coincidence? I think Not!
  • Liked: 169
  • Likes Given: 30
...With respect to what Meep can or cannot do, Meep is an open source software program NOT meant to be used as a black box..."write your own function".

Of course; the C and scripting allows the user to run it iteratively, change variables, to optimize parameters.

...excite the cavity in the time-domain (meep) and see what it does when the cavity is accelerated. I believe meep can do this.

the answer is that Meep, out of the box, canNOT show you any cavity acceleration whatsoever.

Meep only solves Maxwell's equations.  There are no mechanical equations in Meep, there are no equations there that will allow you to model any Newtonian acceleration
...
Writing a procedure to move the charge-source between runs to model the Cerenkov effect is trivial in comparison with re-writing Meep to allow modeling of cavity acceleration.

I didn't think I would have to re-write Meep, just write a script to either modify the position of the frusturm WRT the field variables (or vice-versa), or shift the phase of the fields on the frustrum and enclosed space, for each iterative step.

A far less elegant, conclusive, but far faster executing hack (plan-B) is to integrate the fields, and derive a lumped impedance equivalent network to create a spice model from, then use a phase shifting component in spice to model the doppler shift. The forces on the frustrum would be derived from the voltage and currents in the equivalent network.

You're no doubt the expert WRT FEA & such. Hope you're wrong about Meep.
« Last Edit: 07/21/2015 02:27 PM by mwvp »

Offline deltaMass

  • Full Member
  • ****
  • Posts: 955
  • A Brit in California
  • Liked: 671
  • Likes Given: 275
Maybe Shawyer would have gotten a different response if he made it clear it was acting as a ratchet, so CoM wouldn't have been the issue.

He did make it clear, see attached, just nobody cared to read and think. Much easier to join groupthink, bash the man and his invention.
My advice to you is to cut down on your theoretical "thinking" and continue with building your experiment. This strategy maximises your chances of success.

Offline deltaMass

  • Full Member
  • ****
  • Posts: 955
  • A Brit in California
  • Liked: 671
  • Likes Given: 275
A far less elegant, conclusive, but far faster executing hack (plan-B) is to integrate the fields, and derive a lumped impedance equivalent network to create a spice model from, then use a phase shifting component in spice to model the doppler shift. The forces on the frustrum would be derived from the voltage and currents in the equivalent network.
How did Doppler shift suddenly creep in to the discussion? What is the physics basis for that statement?

Offline mwvp

  • Full Member
  • **
  • Posts: 254
  • Coincidence? I think Not!
  • Liked: 169
  • Likes Given: 30
...Maybe Shawyer would have gotten a different response if he made it clear it was acting as a ratchet, so CoM wouldn't have been the issue.

"asymmetrical dispersion, group velocity, and sum/difference frequency filtering characteristics"

don't really get around the Conservation of Momentum issue in a frame-indifferent Universe.

It isn't a matter of getting around CoM, it's a matter of how its conforming with it.

...Conservation of Momentum issues remain, and as often remarked by Frobnicat and deltaMass, even more disturbing issues related to conservation of energy are brought forth.

Yes, I followed that exchange a few days back, and felt like a naive sucker for trusting the rocket-scientists at Eagleworks' figures - 4N/kW, 2 week mars trips, et. I haven't had the heart, patience or time to do the math myself, but I get specifying constant force/power can result in CoE violation. Shame on them.

And I know EW is testing several devices with several different theories and they have a few theories of their own. But don't fling Eagleworks' 4N/kW rottting, dead, CoM-violating cat at Shawyer, because he does point out he believes his gadget looses power with acceleration and conforms with CoE and CoM. This adds to his credibility. Eagleworks credibility should be questioned for sighting such figures as they have without qualification.

Offline TheTraveller

Maybe Shawyer would have gotten a different response if he made it clear it was acting as a ratchet, so CoM wouldn't have been the issue.

He did make it clear, see attached, just nobody cared to read and think. Much easier to join groupthink, bash the man and his invention.
My advice to you is to cut down on your theoretical "thinking" and continue with building your experiment. This strategy maximises your chances of success.

My theoretical understanding of how an EMDrive works is fine as is the build progress. There is no doubt of anything but success as Shawyer and the team at SPR has already travelled this pathway.

I note Dr. Rodal didn't comment on slides 6 nor 7 but quoted slide 5 out of context of the other slides.

Dr Rodal and others here will need to learn to think outside the square a bit more. More on that when I have my experimental data showing how microwave Power (I can directly measure and record power supply Power consumed, Forward and Reverse Power, try doing that with a magnetron), EMDrive generated Force, rotary table Acceleration and Velocity all link together.
"As for me, I am tormented with an everlasting itch for things remote. I love to sail forbidden seas.”
Herman Melville, Moby Dick

Offline TheTraveller

But don't fling Eagleworks' 4N/kW rottting, dead, CoM-violating cat at Shawyer, because he does point out he believes his gadget looses power with acceleration and conforms with CoE and CoM. This adds to his credibility. Eagleworks credibility should be questioned for sighting such figures as they have without qualification.

The SPR published rotary table data, attached, supports Shawyer's claim. Plain to see in the attachment that Power consumption drops as Velocity increases. End Velocity was 2cm/sec and the 100kg mass was moved 1.85m in 80 seconds.

Shawyer's comment on the chart:
Quote
An electrical reaction occurs between the EM wave and the reflector surfaces of the resonator, resulting in an input impedance change with acceleration. This is seen in the power curve in fig 10.
« Last Edit: 07/21/2015 03:03 PM by TheTraveller »
"As for me, I am tormented with an everlasting itch for things remote. I love to sail forbidden seas.”
Herman Melville, Moby Dick

Offline mwvp

  • Full Member
  • **
  • Posts: 254
  • Coincidence? I think Not!
  • Liked: 169
  • Likes Given: 30
A far less elegant, conclusive, but far faster executing hack (plan-B) is to integrate the fields, and derive a lumped impedance equivalent network to create a spice model from, then use a phase shifting component in spice to model the doppler shift. The forces on the frustrum would be derived from the voltage and currents in the equivalent network.
How did Doppler shift suddenly creep in to the discussion? What is the physics basis for that statement?

Oh no, I can see where this is going. Of course your right; there is no absolute velocity detector or relativity violation. I use doppler-shift WRT the relative velocity of the frustrum WRT stale, old energy (the vast majority according to Q) in the space it encloses. New energy is applied to the frustrum by a probe that is in the same inertial frame as the frustrum. The ~10^4 greater old energy is what is responsible for the thrust/reaction forces. If you don't like doppler shift, how about accelerated photons scattered off the accelerated frustrum?

If it's about accelerating simulations, couldn't we theoretically try and make something like the SETI client and distribute prepared packages to those willing to donate CPU time to simulate a ton of EM drive configurations - even over longer periods of time and operation?
We could make a MEEP Docker package that gets the data either from the Google drive or the Git repo. Anybody with  a Windows, Mac, or Linux box can run it.

Offline notarget

  • Member
  • Posts: 18
  • San Carlos, CA
  • Liked: 17
  • Likes Given: 9

Right now, turning off the source after 0.013 microseconds would be much more informative than running Meep up to 1 microseconds and beyond.

This is something that Todd asked many days ago: just turning off the source and seeing what happens.
I've attached a movie file showing a Ez spatial slice just off of the input source for one source period at the end of the simulation (30 source periods), courtesy of Aero's ctl file). If we're looking for amplification / resonance ISTM that it takes a lot more than 30 time periods of duration to get there based on this E-field animation...

WRT meep - I'm puzzling out how an input labeled "gaussian pulse" appears to translate to a continuously modulating source.  Also - just feeling things out, this is a cartesian coordinate simulation tool, so the boundaries are stairsteps that are sub-grid fitted and the whole simulation is in a cartesian block as compared with a curvilinear fitted/transformed space - ugly IMO, but I'm not a CEM person...

To "stop the source" will introduce some noise into the sim one way or another.  My immediate thought was to just restart the sim from a previous result by reading in the E/B fields and having the source absent.  But I think this will create a lot of spurious transient noise.  Alternatively, I think I can mod the source class to drop the amplitude over some periods, which seems preferable.

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
...
I've attached a movie file showing a Ez spatial slice just off of the input source for one source period at the end of the simulation (30 source periods), courtesy of Aero's ctl file). If we're looking for amplification / resonance ISTM that it takes a lot more than 30 time periods of duration to get there based on this E-field animation...
...
I understood from aero that the simulation for rfmwguy had:

32 time periods = 320 time slices @ ~10 time slices per period

Yes, there is no doubt that 32 time periods is relatively nothing.  I estimated it will take 100 to 1,000 longer time to get to steady state standing waves. 


... this is a cartesian coordinate simulation tool, so the boundaries are stairsteps that are sub-grid fitted and the whole simulation is in a cartesian block as compared with a curvilinear fitted/transformed space - ugly IMO, but I'm not a CEM person......
Ugly indeed.  Initially images with contour plots were given that had the maxima at the maximum color for each image.  Naturally this showed fractals that were a complete numerical artifact of the staircase boundary condition and the near zero level of the initial frames. 
The use of Cartesian Coordinates for a problem that has spherical coordinates as their natural coordinates is ugly, but typical of this kind of code.

Mesh convergence should not be taken for granted, particularly near the boundary condition.  Anything looking like a fractal should be suspected to be a numerical artifact due to the coarse mesh.  Meep meshes are suspect until a convergence study is made to study solution convergence.

This Meep mesh appears to be favoring lower m modes, for example rfmwguy appears as a TM114 mode, instead of higher m modes: I have not seen a Meep result showing the TM212 mode that was verified by NASA (at a lower frequency because they used a dielectric, without a dielectric the TM212 mode appears near 2.4GHz).  To model higher m modes, a finer, less coarse mesh is needed.  So the actual mode shape shown by Meep is also suspect when a lower mode is shown, as it may be due to the use of a coarse mesh.  Coarser meshes favor lower m modes.

Coarser meshes are a result of:  finer meshes take much longer amount of time (this is a 3D problem so it goes like the cube of the sides) and even now for just 32 periods the calculations were running 1 hour.  Running of Meep at longer times/overnight was precluded.

___________________

On the positive side:

1) Meep has shown the importance of antenna location and its effect.  This remains to be further exploited: studying other antenna locations, antenna geometries, types of antenna, a magnetron as a source, a waveguide as a source, etc.


« Last Edit: 07/21/2015 06:13 PM by Rodal »

Tags: