Author Topic: EM Drive Developments Thread 1  (Read 797926 times)

Offline frobnicat

  • Full Member
  • ****
  • Posts: 518
  • Liked: 500
  • Likes Given: 151
Re: EM Drive Developments
« Reply #1480 on: 10/05/2014 04:26 PM »

Assume that such effect would exist for discussion purposes

A) what would explain flipping the direction (hence the sign) of the thrust force when flipping orientation

B) what would explain that a microwave frequency of 1932 MHz would have twice the measured thrust force as 1936 MHz, with the Q value being 1/4th as much as with the higher frequency?

Would rather first try to know if it is possible, at what magnitudes, then where and how.
I'm not sure I'm understanding well the design but (anomalous thrust...) the dielectrics are encased in a conductive tube, no ? Don't know if there is some small space or not between dielectric and those walls, don't know if it would make any difference. Surely <<wavelength but near field effects ? Anyhow, if any rectifier effect is present in such a situation as to be the cause of a net thrust in a given thruster's orientation (say, a small DC current loop component somewhere) then reversing the orientation could reverse said thrust (relative to a vertical permanent B component). Details (where, how) must be addressed, but I don't know enough to tackle that more than qualitatively.
B) would imply a square root dependence on Q ? Wouldn't 4 times a Q imply linearly 4 times E and B fields and hypothetically 4 times DC current loops, that is 4 times forces ? Needs to know how non linear would be such rectifier effects (if possible at all).

@Mulletron : what I was asking for was not for a known way to do a proper efficient AC->DC conversion, but rather for an "imperfection" of dielectrics known to generate a small (usually unwanted) DC component. This would be a deviation from the ideal linear behaviour. Non linearity would not be enough, it would require non linearity around 0 E field (that is, non forward/backward symmetrical). It would likely be filed under "unwanted distortions" or "deviations from the ideal" for instance in RF signal processing literature (that I'm not familiar with). Maybe it doesn't exist, or if it exists it is marginal and never or not often mentioned...

Now, come on people, the time I write that there is SEVEN new posts ! And a new one while just writing this very sentence ! Sheesh !
Must do some other business, see you later.

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1116
  • Liked: 779
  • Likes Given: 1030
Re: EM Drive Developments
« Reply #1481 on: 10/05/2014 04:29 PM »
.../...
In my job I mostly deal with electronic engineering. I know a lot about radars, communications (satellite mostly), and navigation equipment like gyros. Hence why I have a clue about waveguides ...
.../...

Sorry to skip the rest but, about that, do you know of any possible rectifier effect in dielectrics, that is acting like a fast switching diode to convert (a fraction of) AC RF energy to a significant DC current term ?

Quote
And I want a hovercar.

Yeah. So do I. But you know reality doesn't much care about making achievable all what we want ?


All the rectifiers I'm familiar with use diodes. You can do it old school with tubes.

Do  you agree "dielectric, is acting like a fast switching diode to convert (a fraction of) AC RF energy to a significant DC current term" ?

It's the same problem as axion detection in reverse.  You need the right combination of field, phase and photons.

Ok you can use a RC network as a filter (which is done after you go through a half or full wave rectifier normally) to filter out all but the "tops" of the ac sine wave. You would end up with a dc with a LOT of ripple.
There is a fair amount of "ripple" in the measured response

Well here's a math problem for you and that's your strength. You know the frequency, you have a graph of the response with ripples intact and a time reference on the graph, the sloppy rectifier I explained to you is a 1/2 wave rectifier. You can figure out every aspect of this thing by knowing frequency and time which are in the paper. I think the ripples are noise. There is the same noise ripple when the device is off.

Changed that to half wave.
« Last Edit: 10/05/2014 04:34 PM by Mulletron »
Challenge your preconceptions, or they will challenge you. - Velik

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
Re: EM Drive Developments
« Reply #1482 on: 10/05/2014 04:34 PM »
Well here's a math problem for you and that's your strength. You know the frequency 1, you have a graph of the response with ripples intact and a time reference on the graph, the sloppy rectifier I explained to you is a 1/4 wave rectifier. You can figure out every aspect of this thing by knowing frequency and time which are in the paper. I think the ripples are noise. There is the same noise ripple when the device is off.

I think that the ripples are the dynamic response (due to the mechanical degrees of freedom and their derivatives) of the inverted torsional pendulum which is not completely damped out.  I need to have actual damping value supplied by magnetic damping but that's what my model shows.  Observe that highest intensity of ripple is actually in response to the application of the "on" impulse (and it decays thereafter) and to the "off" impulse (and it decays thereafter).  Clearly dynamics of the pendulum to me.
« Last Edit: 10/05/2014 04:35 PM by Rodal »

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1116
  • Liked: 779
  • Likes Given: 1030
Re: EM Drive Developments
« Reply #1483 on: 10/05/2014 04:38 PM »
Well here's a math problem for you and that's your strength. You know the frequency 1, you have a graph of the response with ripples intact and a time reference on the graph, the sloppy rectifier I explained to you is a 1/4 wave rectifier. You can figure out every aspect of this thing by knowing frequency and time which are in the paper. I think the ripples are noise. There is the same noise ripple when the device is off.

I think that the ripples are the dynamic response (due to the mechanical degrees of freedom and their derivatives) of the inverted torsional pendulum which is not completely damped out.  I need to have actual damping value supplied by magnetic damping but that's what my model shows.  Observe that highest intensity of ripple is actually in response to the application of the "on" impulse (and it decays thereafter) and to the "off" impulse (and it decays thereafter).  Clearly dynamics of the pendulum to me.

Well before and after each rf on event they did a cal pulse. Also if there wasn't a feedback loop to the damper, the unit would surge and dip upon first rf on.
Challenge your preconceptions, or they will challenge you. - Velik

Offline frobnicat

  • Full Member
  • ****
  • Posts: 518
  • Liked: 500
  • Likes Given: 151
Re: EM Drive Developments
« Reply #1484 on: 10/05/2014 04:39 PM »
Ok you can use a RC network as a filter (which is done after you go through a half or full wave rectifier normally) to filter out all but the "tops" of the ac sine wave. You would end up with a dc with a LOT of ripple.
There is a fair amount of "ripple" in the measured response

Ok, last one :
my assumption is a small part of RF AC is rectified somewhere, this would be an unwanted effect, no fancy filtering, but the ripples are at RF also, so if a current loop (in the permanent B field) has a DC component + RF ripples, it transfers a force with the same time signature but the varying component is filtered by the mechanical inertia which has time constants many orders of magnitude higher than the variability : so in effect the balance measures only the DC component.

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
Re: EM Drive Developments
« Reply #1485 on: 10/05/2014 04:44 PM »
Since @frobnicat is taking a break to "have a life" and reduce his cholesterol  :)   , I think this is an appropriate moment to recapitulate:


1) Calculation based on Unruth radiation got closest to Shawyer results.  Problem: it hasn't been tested for other experimental results.  It depends highly on Q value and the "back of the envelope formula" does not include mode shape hence not able to deal well [at this point] with NASA tapered cavity results for 4 MHz lower frequency with 4 times lower Q having 2 times higher thrust force

2) Calculations on dark matter can get close based on extremely optimistic assumptions.  Average assumptions put it orders of magnitude off.

3) Coupling of magnetic damping with power cable and with dielectric has not been analyzed at this point where we have numbers: so it is not in the same level of analysis as the the other two above.

~~~~~~~~~~~~~~~~
« Last Edit: 10/05/2014 05:25 PM by Rodal »

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1116
  • Liked: 779
  • Likes Given: 1030
Re: EM Drive Developments
« Reply #1486 on: 10/05/2014 04:59 PM »
Ok you can use a RC network as a filter (which is done after you go through a half or full wave rectifier normally) to filter out all but the "tops" of the ac sine wave. You would end up with a dc with a LOT of ripple.
There is a fair amount of "ripple" in the measured response

Ok, last one :
my assumption is a small part of RF AC is rectified somewhere, this would be an unwanted effect, no fancy filtering, but the ripples are at RF also, so if a current loop (in the permanent B field) has a DC component + RF ripples, it transfers a force with the same time signature but the varying component is filtered by the mechanical inertia which has time constants many orders of magnitude higher than the variability : so in effect the balance measures only the DC component.

As I mentioned, you can do a sloppy rectifier by using a filter. Dielectrics are by definition inductive and capacitive. I mentioned RC networks, but maybe LC or RLC network would have been a better choice to put out there because those take advantage of the properties of the dielectric.
Challenge your preconceptions, or they will challenge you. - Velik

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1116
  • Liked: 779
  • Likes Given: 1030
Re: EM Drive Developments
« Reply #1487 on: 10/05/2014 05:04 PM »
Since @frobnicat is taking a break to "have a life" and reduce his cholesterol  :)   , I think this is an appropriate moment to recapitulate:


1) Calculation based on Unruth radiation got closest to Shawyer results.  Problem: it hasn't been tested for other experimental results.  It depends highly on Q value and the "back of the envelope formula" does not include mode shape hence not able to deal well [at this point] with NASA tapered cavity results for 4 MHz lower frequency with 4 times lower Q having 2 times higher thrust force

2) Calculations on dark matter can get close based on extremely optimistic assumptions.  Average assumptions put it orders of magnitude off.

3) Coupling of magnetic damping with power cable and with dielectric has not been analyzed at this point where we have numbers: so it is not in the same level of analysis as the the other two above.

~~~~~~~~~~~~~~~~

So I introduce another classical physics explanation, not yet investigated (forgive me if my memory is incorrect):

4) Thermal radiation pressure.  For several years the Pioneer anomaly http://en.wikipedia.org/wiki/Pioneer_anomaly remained unexplained.  Very exotic physics explanations were offered from different camps.  Including the Unruth radiation explanation.  Bottom line: a finite element analysis from JPL (http://journals.aps.org/prl/abstract/10.1103/PhysRevLett.107.081103   ) conclusively showed that it was due to Thermal Radiation Pressure. Although we eliminated thermal expansion, shifting of Center of Mass as an explanation, we have not examined thermal radiation pressure.  The problem with thermal expansion is that the transient heat response is related to the heat capacity and density of the Teflon, which results in too slow a response (compared to measurements).

Thermal radiation pressure would  be related to mode shape, as it is thermal radiation from the walls, and specifically their asymmetry that would be responsible for thrust effect.

I can rule out thermal radiation pressure right now by virtue that heating and cooling are not instantaneous. We did see thermal effects, what I mentioned before about the rising slope over 30 seconds with rf on, then a dropping slope over 30 seconds with rf off.

I think we need to combine the forum's conclusions in a living document. Even though science is not democratic, I don't want to keep beating dead horses that have been decided on as fact (not to say any of my ideas are generally accepted or anything). Do you agree? If so, we could implement it using this forum or a shared google doc.
Challenge your preconceptions, or they will challenge you. - Velik

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
Re: EM Drive Developments
« Reply #1488 on: 10/05/2014 05:05 PM »
Ok you can use a RC network as a filter (which is done after you go through a half or full wave rectifier normally) to filter out all but the "tops" of the ac sine wave. You would end up with a dc with a LOT of ripple.
There is a fair amount of "ripple" in the measured response

Ok, last one :
my assumption is a small part of RF AC is rectified somewhere, this would be an unwanted effect, no fancy filtering, but the ripples are at RF also, so if a current loop (in the permanent B field) has a DC component + RF ripples, it transfers a force with the same time signature but the varying component is filtered by the mechanical inertia which has time constants many orders of magnitude higher than the variability : so in effect the balance measures only the DC component.

As I mentioned, you can do a sloppy rectifier by using a filter. Dielectrics are by definition inductive and capacitive. I mentioned RC networks, but maybe LC or RLC network would have been a better choice to put out there because those take advantage of the properties of the dielectric.

Can the dielectric together with the rest of the system result in a van der Pool oscillation http://en.wikipedia.org/wiki/Van_der_Pol_oscillator  like in a triode ?

Such a circuit produces relaxation oscillations. For a given set of properties the relaxation response looks almost like a rectangular wave
« Last Edit: 10/05/2014 05:30 PM by Rodal »

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1116
  • Liked: 779
  • Likes Given: 1030
Re: EM Drive Developments
« Reply #1489 on: 10/05/2014 05:05 PM »
I really want some feedback concerning my comments trying to shut down dark matter. Good science is trying to break things.
« Last Edit: 10/05/2014 05:23 PM by Mulletron »
Challenge your preconceptions, or they will challenge you. - Velik

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
Re: EM Drive Developments
« Reply #1490 on: 10/05/2014 05:08 PM »
...

I can rule out thermal radiation pressure right now by virtue that heating and cooling are not instantaneous. We did see thermal effects, what I mentioned before about the rising slope over 30 seconds with rf on, then a dropping slope over 30 seconds with rf off.

I think we need to combine the forum's conclusions in a living document. Even though science is not democratic, I don't want to keep beating dead horses that have been decided on as fact (not to say any of my ideas are generally accepted or anything). Do you agree? If so, we could implement it using this forum or a shared google doc.

I completely agree on the living document and I support implementing it on this forum

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1116
  • Liked: 779
  • Likes Given: 1030
Re: EM Drive Developments
« Reply #1491 on: 10/05/2014 05:19 PM »

Assume that such effect would exist for discussion purposes

A) what would explain flipping the direction (hence the sign) of the thrust force when flipping orientation

B) what would explain that a microwave frequency of 1932 MHz would have twice the measured thrust force as 1936 MHz, with the Q value being 1/4th as much as with the higher frequency?

Would rather first try to know if it is possible, at what magnitudes, then where and how.
I'm not sure I'm understanding well the design but (anomalous thrust...) the dielectrics are encased in a conductive tube, no ? Don't know if there is some small space or not between dielectric and those walls, don't know if it would make any difference. Surely <<wavelength but near field effects ? Anyhow, if any rectifier effect is present in such a situation as to be the cause of a net thrust in a given thruster's orientation (say, a small DC current loop component somewhere) then reversing the orientation could reverse said thrust (relative to a vertical permanent B component). Details (where, how) must be addressed, but I don't know enough to tackle that more than qualitatively.
B) would imply a square root dependence on Q ? Wouldn't 4 times a Q imply linearly 4 times E and B fields and hypothetically 4 times DC current loops, that is 4 times forces ? Needs to know how non linear would be such rectifier effects (if possible at all).

@Mulletron : what I was asking for was not for a known way to do a proper efficient AC->DC conversion, but rather for an "imperfection" of dielectrics known to generate a small (usually unwanted) DC component. This would be a deviation from the ideal linear behaviour. Non linearity would not be enough, it would require non linearity around 0 E field (that is, non forward/backward symmetrical). It would likely be filed under "unwanted distortions" or "deviations from the ideal" for instance in RF signal processing literature (that I'm not familiar with). Maybe it doesn't exist, or if it exists it is marginal and never or not often mentioned...

Now, come on people, the time I write that there is SEVEN new posts ! And a new one while just writing this very sentence ! Sheesh !
Must do some other business, see you later.

Never heard of something like that. A dielectric is open to DC. Dc current doesn't flow through capacitors. It is used to actively isolate DC from other things in circuits. As soon as the cap is charged up and fully polarized, no more current flow. Where is this leading?
Challenge your preconceptions, or they will challenge you. - Velik

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
Re: EM Drive Developments
« Reply #1492 on: 10/05/2014 05:26 PM »
...

I can rule out thermal radiation pressure right now by virtue that heating and cooling are not instantaneous. ..

I agree. I just did  a quick calculation.   I edited the record accordingly.

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
Re: EM Drive Developments
« Reply #1493 on: 10/05/2014 05:29 PM »
I really want some feedback concerning my comments trying to shut down dark matter. Good science is trying to break things.

Agreed, but good science also takes time to enable thoughtful responses. Just like thermal effects  :)

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
Re: EM Drive Developments
« Reply #1494 on: 10/05/2014 05:34 PM »
Ok you can use a RC network as a filter (which is done after you go through a half or full wave rectifier normally) to filter out all but the "tops" of the ac sine wave. You would end up with a dc with a LOT of ripple.
There is a fair amount of "ripple" in the measured response

Ok, last one :
my assumption is a small part of RF AC is rectified somewhere, this would be an unwanted effect, no fancy filtering, but the ripples are at RF also, so if a current loop (in the permanent B field) has a DC component + RF ripples, it transfers a force with the same time signature but the varying component is filtered by the mechanical inertia which has time constants many orders of magnitude higher than the variability : so in effect the balance measures only the DC component.

As I mentioned, you can do a sloppy rectifier by using a filter. Dielectrics are by definition inductive and capacitive. I mentioned RC networks, but maybe LC or RLC network would have been a better choice to put out there because those take advantage of the properties of the dielectric.

Can the dielectric together with the rest of the system result in a van der Pool oscillation http://en.wikipedia.org/wiki/Van_der_Pol_oscillator  like in a triode ?

Such a circuit produces relaxation oscillations. For a given set of properties the relaxation response looks almost like a rectangular wave

To make electrical circuits like a van der Pool oscillator, active circuit elements with a cubic (or quintic, etc.) nonlinear property, i=ϕ(v)=γv^3−αv , are required, where i and v are current and voltage, respectively.  van der Pol built the oscillator using the triode or tetrode.  Also with a tunnel diode ( http://en.wikipedia.org/wiki/Tunnel_diode ) instead of the triode, the van der Pol oscillator can take place as well.
« Last Edit: 10/05/2014 05:38 PM by Rodal »

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1116
  • Liked: 779
  • Likes Given: 1030
Re: EM Drive Developments
« Reply #1495 on: 10/05/2014 05:39 PM »
Ok you can use a RC network as a filter (which is done after you go through a half or full wave rectifier normally) to filter out all but the "tops" of the ac sine wave. You would end up with a dc with a LOT of ripple.
There is a fair amount of "ripple" in the measured response

Ok, last one :
my assumption is a small part of RF AC is rectified somewhere, this would be an unwanted effect, no fancy filtering, but the ripples are at RF also, so if a current loop (in the permanent B field) has a DC component + RF ripples, it transfers a force with the same time signature but the varying component is filtered by the mechanical inertia which has time constants many orders of magnitude higher than the variability : so in effect the balance measures only the DC component.

As I mentioned, you can do a sloppy rectifier by using a filter. Dielectrics are by definition inductive and capacitive. I mentioned RC networks, but maybe LC or RLC network would have been a better choice to put out there because those take advantage of the properties of the dielectric.

Can the dielectric together with the rest of the system result in a van der Pool oscillation http://en.wikipedia.org/wiki/Van_der_Pol_oscillator  like in a triode ?

Such a circuit produces relaxation oscillations. For a given set of properties the relaxation response looks almost like a rectangular wave

http://scholarpedia.org/article/Van_der_Pol_oscillator
You have to engineer a circuit like this. I can't imagine how it would happen by accident.

"To make electrical circuits described by equation (1), active circuit elements with the cubic nonlinear property, i=ϕ(v)=γv3−αv , are required, where i and v are current and voltage, respectively. "

These things sound cool, why don't I see these in pseudorandom number generators? I just realized we brought chaos theory to the table, sheesh.
« Last Edit: 10/05/2014 05:43 PM by Mulletron »
Challenge your preconceptions, or they will challenge you. - Velik

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
Re: EM Drive Developments
« Reply #1496 on: 10/05/2014 05:46 PM »
Ok you can use a RC network as a filter (which is done after you go through a half or full wave rectifier normally) to filter out all but the "tops" of the ac sine wave. You would end up with a dc with a LOT of ripple.
There is a fair amount of "ripple" in the measured response

Ok, last one :
my assumption is a small part of RF AC is rectified somewhere, this would be an unwanted effect, no fancy filtering, but the ripples are at RF also, so if a current loop (in the permanent B field) has a DC component + RF ripples, it transfers a force with the same time signature but the varying component is filtered by the mechanical inertia which has time constants many orders of magnitude higher than the variability : so in effect the balance measures only the DC component.

As I mentioned, you can do a sloppy rectifier by using a filter. Dielectrics are by definition inductive and capacitive. I mentioned RC networks, but maybe LC or RLC network would have been a better choice to put out there because those take advantage of the properties of the dielectric.

Can the dielectric together with the rest of the system result in a van der Pool oscillation http://en.wikipedia.org/wiki/Van_der_Pol_oscillator  like in a triode ?

Such a circuit produces relaxation oscillations. For a given set of properties the relaxation response looks almost like a rectangular wave

http://scholarpedia.org/article/Van_der_Pol_oscillator
You have to engineer a circuit like this. I can't imagine how it would happen by accident.

"To make electrical circuits described by equation (1), active circuit elements with the cubic nonlinear property, i=ϕ(v)=γv3−αv , are required, where i and v are current and voltage, respectively. "

These things sound cool, why don't I see these in pseudorandom number generators? I just realized we brought chaos theory to the table, sheesh.

The van der Pool circuit with a a tunnel diode ( http://en.wikipedia.org/wiki/Tunnel_diode ) does not look too complicated.  Can the Teflon dielectric act as a tunnel diode ?

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1116
  • Liked: 779
  • Likes Given: 1030
Re: EM Drive Developments
« Reply #1497 on: 10/05/2014 05:48 PM »
The more far fetched we get here, creating more and more complex explanations, probably won't help us. Seems the most simple explanations are more likely correct. Occam's razor and all. I'm ready to start shutting down theories. I also intend to make my critical analysis of the paper more known in detail as soon as I can.
Challenge your preconceptions, or they will challenge you. - Velik

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1116
  • Liked: 779
  • Likes Given: 1030
Re: EM Drive Developments
« Reply #1498 on: 10/05/2014 05:50 PM »
...

I can rule out thermal radiation pressure right now by virtue that heating and cooling are not instantaneous. ..

I agree. I just did  a quick calculation.   I edited the record accordingly.

Where is the record/living document? How did you do it so everyone can share?
Challenge your preconceptions, or they will challenge you. - Velik

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
Re: EM Drive Developments
« Reply #1499 on: 10/05/2014 05:50 PM »
The more far fetched we get here, creating more and more complex explanations, probably won't help us. Seems the most simple explanations are more likely correct. Occam's razor and all. I'm ready to start shutting down theories. I also intend to make my critical analysis of the paper more known in detail as soon as I can.
And the simplest coherent explanations, per your understanding are ...?

Tags: