-
#2340
by
Gilbertdrive
on 13 May, 2016 20:56
-
Dr. Rodal, could you take a moment out of the current discussion and comment on the Sonny White video specifically re TRL2 to TRL 3, and the sustained 20 Newtons/kilowatt reported. Sonny also stated that EWL had or was in the process of manufacturing and sending their test articles out to other institutions for independent evaluation. Thnx FL
The video in YouTube was posted in ~June 2014? I think it was a conference about 2 years ago or more at a College? Certainly before the NASA Ames conference which was post-EM Drive paper.
Who knows? Maybe it was a Mach-Lorentz device covered by NDA ? Is he referring to Boeing/Darpa/Serrano Field Effect that as described it had the problem that the thrust (unlike steady anomalous force for EM Drive) was skewed bi-directional pulses that would have to be made uni-directional and there is no straightforward way to do that?
20 N/Kw is not useful unless it is unidirectional
I don't recall Star-Drive discussing <<TRL2 to TRL 3, and the sustained 20 Newtons/kilowatt>> while he was posting at NSF EM Drive thread in late 2014, early 2015, which was after that conference presentation. TRL2 to TRL3 20 N/kW definitely does not sound like the EM Drive project results at NASA...
Maybe 20 mN/Kw ?
-
#2341
by
rfmwguy
on 13 May, 2016 20:57
-
OK thanks...now that I'm a bit more focused as my testing will begin shortly, your high level of force does not preclude anything, such as the likely thermal expansion of the waveguide transmission line. This should have been a one-time force that reached a peak somewhere and began to dissipate slowly over time...a single event which is relatively predictable, like the thermal lift of my teeter totter: pretty repeatable and not an instantaneous force. Amazed thermal expansion was not considered during the first testing and that it took 2 years and another paper to resolve...hmmmm.
Regardless, thanks again for the reply...
-
#2342
by
Rodal
on 13 May, 2016 21:12
-
.../...
A copper surface has incident and reflected EM plane waves I and R and a transmitted wave T. The transmitted wave T is exactly that required to create the power loss in the copper. Conservation of energy requires power balance i.e. P(I)=P(T)+P(R). Momentum Balance requires the force on the copper to be M(I)-M(T)+M(R) (in terms of fluxes per unit time). Hence the momentum of the transmitted wave is directly relevant to the force on the copper, and its energy to the heat dissipated.
.../...
I don't get it about what you call transmitted wave and the momentum balance expression : in the end the EM momentum that crosses the outer surface of copper is either absorbed (or reflected ?) a few skin depth later... so it seems to me that looking the copper wall in its entirety there is no transmission (no EM energy is going through) and while M(I)-M(T)+M(R) can be considered valid as measured at the surface of copper, the copper wall in its entirety will absorb the "transmitted" momentum and thus for the copper wall we have (M(I)-M(T)+M(R))+M(T)=M(I)+M(R)=net_momentum. Or said otherwise, if looking to the bulk wall from the start we know that transmission is 0 and that net_momentum=M(I)+M(R) directly. This later expressions that goes for the bulk copper wall makes more sense to me because it can be related to the efficiency of a photon sail (orthogonal to plane wave) : when perfectly absorbing (black) momentum is what is Incident, M(R)=0 => net_momentum=M(I), and when perfectly reflecting momentum is double what is Incident, reflection is of same magnitude (and opposite direction) M(R)=M(I) => net_momentum=2×M(I). I'm trying to relate to known and classic situations, not trying (just can't) to comment on Abraham–Minkowski issues and applicability to EMdrive...
...
"Transmitted" is being transmitted into a lossy medium, hence (due to the skin depth being only 1 micrometer) dissipated within the skin depth and exponentially decaying rapidly to zero by the time one reaches the
exterior surface of the metal.
So, nothing coming out of exterior metal. Everything that went inside got dissipated.
Any energy being dissipated is energy that cannot be used for acceleration.ANALOGY: a truck impacting a car. There will be less g-forces on the occupants of a deformable car than those in a rigid carTo protect cars from impact, one makes them crushable so that the impact energy goes into permanent deformation, so as to mitigate the acceleration on the occupants.If a big truck hits a car, it is much worse to be inside a rigid car (momentum from truck goes into high g forces to occupants of car) than to be inside a deformable car (some of the kinetic energy from the truck goes into permanent deformation energy of the car, hence lower g is experience by the occupants of the deformable car)
-
#2343
by
rfmwguy
on 13 May, 2016 21:16
-
.../...
A copper surface has incident and reflected EM plane waves I and R and a transmitted wave T. The transmitted wave T is exactly that required to create the power loss in the copper. Conservation of energy requires power balance i.e. P(I)=P(T)+P(R). Momentum Balance requires the force on the copper to be M(I)-M(T)+M(R) (in terms of fluxes per unit time). Hence the momentum of the transmitted wave is directly relevant to the force on the copper, and its energy to the heat dissipated.
.../...
I don't get it about what you call transmitted wave and the momentum balance expression : in the end the EM momentum that crosses the outer surface of copper is either absorbed (or reflected ?) a few skin depth later... so it seems to me that looking the copper wall in its entirety there is no transmission (no EM energy is going through) and while M(I)-M(T)+M(R) can be considered valid as measured at the surface of copper, the copper wall in its entirety will absorb the "transmitted" momentum and thus for the copper wall we have (M(I)-M(T)+M(R))+M(T)=M(I)+M(R)=net_momentum. Or said otherwise, if looking to the bulk wall from the start we know that transmission is 0 and that net_momentum=M(I)+M(R) directly. This later expressions that goes for the bulk copper wall makes more sense to me because it can be related to the efficiency of a photon sail (orthogonal to plane wave) : when perfectly absorbing (black) momentum is what is Incident, M(R)=0 => net_momentum=M(I), and when perfectly reflecting momentum is double what is Incident, reflection is of same magnitude (and opposite direction) M(R)=M(I) => net_momentum=2×M(I). I'm trying to relate to known and classic situations, not trying (just can't) to comment on Abraham–Minkowski issues and applicability to EMdrive...
...
I think he maybe analyzing the interface vacuum/copper or air/copper and using this terminology for waves hitting a surface:
TRANSMITTED = transmitted electromagnetic momentum from air or vacuum to copper
REFLECTED = reflected electromagnetic momentum from copperback to air or vacuum
You are right, here transmitted, is being transmitted into a lossy medium, hence (due to the skin depth being only 1 micrometer) dissipated.
To me, an analysis of anything that can give a net resultant force to the center of mass, has to be made in terms of forces, which are a result of integrating the stresses, and the derivative of the Poynting vector with respect to time.
Regarding skin depth, shouldn't this be clarified as E field and not B field? I'm a little unclear here regarding the containment of the magnetic field...my assumption is the B field sets up eddy currents that are superimposed on the outside of the copper frustum which in turn sets up its own B field. Whether it replicates, distorts or inverts the internal B field is one area I'm having trouble visualizing.
-
#2344
by
Rodal
on 13 May, 2016 21:21
-
.../...
A copper surface has incident and reflected EM plane waves I and R and a transmitted wave T. The transmitted wave T is exactly that required to create the power loss in the copper. Conservation of energy requires power balance i.e. P(I)=P(T)+P(R). Momentum Balance requires the force on the copper to be M(I)-M(T)+M(R) (in terms of fluxes per unit time). Hence the momentum of the transmitted wave is directly relevant to the force on the copper, and its energy to the heat dissipated.
.../...
I don't get it about what you call transmitted wave and the momentum balance expression : in the end the EM momentum that crosses the outer surface of copper is either absorbed (or reflected ?) a few skin depth later... so it seems to me that looking the copper wall in its entirety there is no transmission (no EM energy is going through) and while M(I)-M(T)+M(R) can be considered valid as measured at the surface of copper, the copper wall in its entirety will absorb the "transmitted" momentum and thus for the copper wall we have (M(I)-M(T)+M(R))+M(T)=M(I)+M(R)=net_momentum. Or said otherwise, if looking to the bulk wall from the start we know that transmission is 0 and that net_momentum=M(I)+M(R) directly. This later expressions that goes for the bulk copper wall makes more sense to me because it can be related to the efficiency of a photon sail (orthogonal to plane wave) : when perfectly absorbing (black) momentum is what is Incident, M(R)=0 => net_momentum=M(I), and when perfectly reflecting momentum is double what is Incident, reflection is of same magnitude (and opposite direction) M(R)=M(I) => net_momentum=2×M(I). I'm trying to relate to known and classic situations, not trying (just can't) to comment on Abraham–Minkowski issues and applicability to EMdrive...
...
I think he maybe analyzing the interface vacuum/copper or air/copper and using this terminology for waves hitting a surface:
TRANSMITTED = transmitted electromagnetic momentum from air or vacuum to copper
REFLECTED = reflected electromagnetic momentum from copperback to air or vacuum
You are right, here transmitted, is being transmitted into a lossy medium, hence (due to the skin depth being only 1 micrometer) dissipated.
To me, an analysis of anything that can give a net resultant force to the center of mass, has to be made in terms of forces, which are a result of integrating the stresses, and the derivative of the Poynting vector with respect to time.
Regarding skin depth, shouldn't this be clarified as E field and not B field? I'm a little unclear here regarding the containment of the magnetic field...my assumption is the B field sets up eddy currents that are superimposed on the outside of the copper frustum which in turn sets up its own B field. Whether it replicates, distorts or inverts the internal B field is one area I'm having trouble visualizing.
No !
The other way around. The skin depth is due to the Magnetic Field.
There is NO microwave magnetic field escaping outside, we already discussed that. The only magnetic field that can escape is LOW FREQUENCY (60 Hz or DC, because the skin depth becomes larger than thickness at very low frequency, much lower than microwave frequency).
The eddy currents are due to the Magnetic Field, and not to the electric field.
The skin depth in TE Modes in truncated cone cavity is completely due to the H field in the metal.(
No microwave frequency B field makes it outside since skin depth is due to magnetic field~1 micrometer and wall thickness ~ mm which is 1,000 times more, hence exponential decay over 1,000 times skin depth extinguishes the B field ).
In TE modes (the mode shape preferred by Shawyer and by Yang, and many builders: TheTraveller) there is NO E field at the walls of the EM Drive.Proof here:
http://forum.nasaspaceflight.com/index.php?topic=39214.msg1526577#msg1526577
-
#2345
by
rfmwguy
on 13 May, 2016 21:28
-
E field or electric field is stopped by skin depth is what I was trying to say, no E fields through the copper...my questions involve the B/H magnetic field's depth and transference externally...the open system.
-
#2346
by
Monomorphic
on 13 May, 2016 22:20
-
Torsion pendulum modified at top clamp to improve stickiness.
What clamp are you using to hold onto the piano wire? Right now I simply have a knot tied and run the wire through a small hole.
-
#2347
by
FattyLumpkin
on 13 May, 2016 22:38
-
rfmwguy , would you give us the dims of your frustum again? thnx , FL
10 inch large diameter, 6.25 inch small diameter, 8 inch height.
-
#2348
by
rfmwguy
on 13 May, 2016 23:04
-
Torsion pendulum modified at top clamp to improve stickiness.
What clamp are you using to hold onto the piano wire? Right now I simply have a knot tied and run the wire through a small hole.
Get a small crimp on connector, cut back the plastic insulator. Clamp tightly around piano wire especially at departure end. Leave enough wire to loop a couple of times around and thru holes in connector. Make sure wire does not contact feed thru hole sides. Crimped end should be slightly smaller than through hole.
-
#2349
by
Monomorphic
on 14 May, 2016 00:28
-
Work continues on the tortional pendulum. Balancing will be critical - if you look closely, you can see i'm using a wrench as a counter-weight on the frustum side.
I've also included an image of the system i'm using to suspend the piano wire. By turning it, this works pretty well in centering the pendulum. If you are wondering what all the holes are, I used the old air-track as the top suspension beam.
-
#2350
by
spupeng7
on 14 May, 2016 03:40
-
Thanks to Phil in Australia who found Paul Stansell's public science report he used with his ISEF science experiment. It is attached below.
Thanks Rfmw,
this is how an experimental report should be written. It is still scant on method and could benefit from more photos but it raises the bar for the rest of us. Clear and unreserved descriptions of what, how and why. Well done Mr Stansell, please keep up the good work. JMN..
PDF added because I couldn't make the link work
-
#2351
by
TheTraveller
on 14 May, 2016 05:08
-
Thanks to Phil in Australia who found Paul Stansell's public science report he used with his ISEF science experiment. It is attached below.
Thanks Rfmw,
this is how an experimental report should be written. It is still scant on method and could benefit from more photos but it raises the bar for the rest of us. Clear and unreserved descriptions of what, how and why. Well done Mr Stansell, please keep up the good work. JMN..
PDF added because I couldn't make the link work
Paul's build, photos, comment & UpDate history can be read here.
https://www.reddit.com/user/PaulTheSwag?count=25&after=t1_cvnxfe7If you are interested in his process & progress, reading all his comments and the involved threads are a good investment.
Should add that Pauls' s mN/pixel calc in the paper is 10x too high. So all the stated thrust measurements are 10x too high. That said his EmDrive generated 8.9 & 11.8 mN peak thrust and 4.0mN averaged thrust.
Suspect his magnetron antenna position is good but the frustum length is not ideal.
EmDrives need to be tuned while under full power to achieve the best thrust. VNA is helpful to get close but no guarantee of achieving thrust generation.
DIYers please note: EVERY EmDrive Roger & SPR have built were tunable under full power to achieve max thrust generation. The Demonstrator & Flight Thrusters went the next step & adjust the initial tune to continually track frustum & drive system changes that could effect thrust generation as the entire system warmed.
-
#2352
by
Gilbertdrive
on 14 May, 2016 08:42
-
Should add that Pauls' s mN/pixel calc in the paper is 10x too high. So all the stated thrust measurements are 10x too high. That said his EmDrive generated 8.9 & 11.8 mN peak thrust and 4.0mN averaged thrust.
Suspect his magnetron antenna position is good but the frustum length is not ideal.
EmDrives need to be tuned while under full power to achieve the best thrust. VNA is helpful to get close but no guarantee of achieving thrust generation.
DIYers please note: EVERY EmDrive Roger & SPR have built were tunable under full power to achieve max thrust generation. The Demonstrator & Flight Thrusters went the next step & adjust the initial tune to continually track frustum & drive system changes that could effect thrust generation as the entire system warmed.
Your message made me think to something.
That is about the fact, mentionned earlier in these threads, that is some cases, reduction of vibration could reduce the force.
If a system is not correctly tracked and tuned, and that the resonance is lost, should it possible that the resonance is attained again during a short period of time for each vibration ?
When the frustrum vibrates, it deforms, and it could get, during a small time for each vibration, the exact dimensions to resonate. Since this deformation is several ten of thousands time slower than the speed of light, there should be a small amount of time during each vibration when the frustrum can resonate almost perfectly, during thousands of microwave trips
Maybe it could explain differences of result following vibrations. A non vibrating frustrum would need to be perfectly tuned to achive resonnance, when a vibrating frustrum, even not perfectly tuned, would achive resonnance a fraction of the time, at regular intervals, and get thrust
-
#2353
by
rfmwguy
on 14 May, 2016 13:37
-
Thanks to Phil in Australia who found Paul Stansell's public science report he used with his ISEF science experiment. It is attached below.
Thanks Rfmw,
this is how an experimental report should be written. It is still scant on method and could benefit from more photos but it raises the bar for the rest of us. Clear and unreserved descriptions of what, how and why. Well done Mr Stansell, please keep up the good work. JMN..
PDF added because I couldn't make the link work
Paul's build, photos, comment & UpDate history can be read here.
https://www.reddit.com/user/PaulTheSwag?count=25&after=t1_cvnxfe7
If you are interested in his process & progress, reading all his comments and the involved threads are a good investment.
Should add that Pauls' s mN/pixel calc in the paper is 10x too high. So all the stated thrust measurements are 10x too high. That said his EmDrive generated 8.9 & 11.8 mN peak thrust and 4.0mN averaged thrust.
Suspect his magnetron antenna position is good but the frustum length is not ideal.
EmDrives need to be tuned while under full power to achieve the best thrust. VNA is helpful to get close but no guarantee of achieving thrust generation.
DIYers please note: EVERY EmDrive Roger & SPR have built were tunable under full power to achieve max thrust generation. The Demonstrator & Flight Thrusters went the next step & adjust the initial tune to continually track frustum & drive system changes that could effect thrust generation as the entire system warmed.
Paul achieved something probably none of us have, taking a science fair project to an international competition after winning top honors in his continent! While its disappointing that the judges chose stepper motor enhancement as the top winner in Engineering Mechanics, I'm sure it was well deserved albeit not what I would consider breakthrough technology; it was more mainstream which seems to be the thing for much of academia. Understandable given the conservative nature of institutions. A win would have surprised me.
I think somebody should probably update the emdrive.wiki page with Paul's results and perhaps a link to his paper.
-
#2354
by
Monomorphic
on 14 May, 2016 14:31
-
FEKO run of Paul Stansell's frustum. In my humble opinion, his issue is the same as some DIYers I have seen. They used NASA's frustum dimensions with a ~2.45Ghz magnetron. Which may sound good, but NASA's frustum does not resonate at 2.445Ghz - 2.45Ghz - the operating frequency range of most household microwaves. I have not run a sim of his 5cm tuning add-on yet. Hope to do that soon.
-
#2355
by
Rodal
on 14 May, 2016 14:43
-
FEKO run of Paul Stansell's frustum. In my humble opinion, his issue is the same as some DIYers I have seen. They used NASA's frustum dimensions with a ~2.45Ghz magnetron. Which may sound good, but NASA's frustum does not resonate at 2.445Ghz - 2.45Ghz - the operating frequency range of most household microwaves. I have not run a sim of his 5cm tuning add-on yet. Hope to do that soon.
For builders to use NASA's dimensions makes sense because:
1) Only NASA has provided the exact internal dimensions of the frustum used for their tests. Neither Shawyer nor Yang ever provided the exact internal dimensions of the frustums they used, their dimensions having to be guesstimated from diagrams, photos or other expressions (like the "Design Factor").
2) NASA provided a COMSOL FEA run showing natural frequency without polymer insert occurring at 2.4575GHz for mode shape TM212: the only mode shape that has been verified in any testing of the EM Drive by anyone.
What mode shape are they trying to excite in NASA's frustum without a polymer insert?
What did Paul use to excite the natural frequency and where was it located?
NASA's Frank Davies showed with COMSOL Finite Element Analysis that NASA's frustum has a natural frequency at 2.4575 GHz in mode shape TM212. This is a transverse magnetic mode shape, whose excitation is different from transverse electric mode shapes as used by Shawyer, and Yang.
I recall verifying Frank Davies COMSOL FEA calculations, and my conclusion was that Frank Davies analysis was excellent, well within 1% of the exact natural frequency.
Frank Davies shows the sensitivity of the natural frequency to geometrical dimension imperfection: greatest sensitivity is to the dimensions of the small diameter of the frustum of a cone:
∂NaturalFrequency/∂SmallDiameter = - 110 ΔMHz per Δinch of small diameter
Followed by the sensitivity to changes in dimensions to the big diameter. Observe that smallest sensitivity is to changes in length dimensions, which is the only dimension that Shawyer attempted to change with his Demonstrator EM Drive, because the easy dimension to change is the longitudinal one:
∂NaturalFrequency/∂Length = - 77 ΔMHz per Δinch of length
Suggestion: problem in achieving resonance is due to geometrical imperfections of EM Drive builders, and to excitation probe, as well as to magnetron.
-
#2356
by
Monomorphic
on 14 May, 2016 14:43
-
Here is the NASA frustum modes chart that shows the frequency range of a standard magnetron. Closest resonance is TM212.
-
#2357
by
rfmwguy
on 14 May, 2016 14:44
-
Highly recommended paper for those building and testing with a torsion pendulum...a portion of it follows:
Recommended Practices in Thrust Measurements
IEPC-2013-440
The 33rd International Electric Propulsion Conference, Washington, DC October 6-10, 2013
B. Controlling Systematic Errors
The rate of thrust measurement drift due to thermal effects can be minimized with sufficient cooling along the
thermal path between the thruster and the stand as well as through the use of materials with high thermal
resistance. Placing the stand within an actively cooled shroud enclosure will block plasma impingement
and radiative heat transfer to the stand. An effective electrical cable waterfall design can prevent cable
expansion and contraction from pushing or pulling on the stand. Additionally, aligning the connection of
electrical cables from the base of the stand to the moving assembly orthogonal to the direction of thrust
prevents undesired variable forces on the stand. Active leveling control will eliminate changes in the canting
of a thrust stand resulting from thermal fluctuations. Where possible, measuring electrical current in control
electronics rather than voltage is advisable in order to avoid errors due to varying voltage drops across cabling
caused by temperature fluctuations. Utilizing thermocouples in several locations on the thrust stand can
aid in tracking thermal drift effects. As thermal drift is generally unavoidable, frequent stand recalibration
is recommended.
Errors caused by friction in the system can be reduced through the use of adequate linear and torsional
springs in the thrust stand design. Ensuring that all moving parts are clear of draped electrical cables and
other stationary objects through their entire range of motion will minimize intermittent or variable friction
and blockage. Attention to cleanliness can prevent unwanted debris from obstructing the motion of the
stand. Frequent calibration will also minimize error resulting from gradual changes in friction.
Where applicable, the use of vibrational damping material will minimize measurement error due to external
sources of vibration. For null balance thrust stands, a second electromagnetic coil, commonly referred to
as a damper coil, can be used to separate vibrational noise from the steady-state thrust signal. Attaching
physical stops which restrict the range of motion of the stand will prevent unexpected vibration or impulses
from pushing the stand out of range and possibly damaging components.
Where possible, the use of non-ferrous materials in the thrust stand design will mitigate measurement error
due to undesired electromagnetic interaction. Maximizing the distance between any magnetic field sources
and affected thrust stand components will further reduce measurement errors. Using coaxial or twisted-pair
cabling will also reduce interference. If the interaction is unavoidable, characterizing the interaction across
the range of possible settings can allow the error to be accounted for during data reduction.http://mwalker.gatech.edu/papers/2013_IEPC_Polk.pdf
-
#2358
by
rfmwguy
on 14 May, 2016 16:00
-
Note on the above paper - this was a paper presented at a conference in the initial attempt to establish a recommended set of conditions for low thrust devices per the AIAA...i.e. Standards.
Interesting the paper goes on to say many of the things that have been discussed on NSF, namely thermal lift. They recommend leveling actuators, but seemed to miss out on a self-leveling technique. If the beam is located far enough below the lower attachment point of a torsion wire, it will naturally inhibit vertical deflection of the beam end. In other words, a solid rod from the lower clamp point to the beam attenuates the fulcrum (teeter-totter) action at the clamping point.
-
#2359
by
Monomorphic
on 14 May, 2016 16:05
-
See here rfmyguy's spectrum analysis of one of his magnetrons. It starts very close to 2.45Ghz and then drifts to 2.445Ghz as the magnetron heats. I do not think the frequencies emitted by a magnetron are close enough to 2.4575Ghz to excite TM212, even when the dirty magnetron signal is accounted for.
