-
#3620 by TheTraveller on 07 Jul, 2016 00:07
-
My point was the decay displacement return to level to use is the level just before the Rf was applied.
.../...
Either I don't understand this phrase or I fail to see how this invalidates my point (that a major part of reported forces are thermal). I know that EagleWorks liked to play around with "floating" baselines but at least they managed to get some pretty decent overshoots at power-off : the first inversion of velocity occurring after crossing the baseline, baseline understood as being the straight line averaging the remaining oscillations until settling. See for instance the negative overshoot here, baseline is at 0V (not because this is 0V but because this is the observed level around which the oscillations settle after the excitation step down at t=1)
Such overshoot is nowhere to be seen with the observed magnetron-off decays, but as expected is clearly visible with horizontal "tap test" when a forcing excitation is actually stepped down. Magnetron-off is not a step down in excitation, it is a slowly decayed excitation with a time constant > natural period of pendulum. Are you seriously objecting that (most of) the excitation decays slowly on the order of minutes ? Are you not seeing that we have something like 50% of whatever made the pendulum deviate in the first place still forcing (in the same direction) about 2 minutes after power was interrupted ?
With respect to Dave, his rig is not an EW class rig. It is big and is easily disturbed by air currents.
There is still lots to characterise. All the tests that I know of have been done small end North facing. Future tests will be done South, Down & Up facing. I'm sure everyone of those tests will reveal issues that need to be sorted and move Dave's test rig closer and closer to the level of the EW test rig.
Additionally there appears to be but not yet proven some possible interaction with beam displacement & the 3 wires connecting the maggie to the remote PSU. For sure it twists and untwists as the beam moves and modifies the natural action of the piano wire torsion section. However the forced thermal action appears to have the same very slow slope up and down as would be expected.
There is work to do to understand the why of the slope differentials and the thermal characterists of the system. -
#3621 by TheTraveller on 07 Jul, 2016 00:30
-
Ok, thanks for the repost. This again illustrate the point I make : it's not a matter of time to "settle" understood as "stop oscillating". It's the matter than on this plot when a forcing excitation is suddenly relieved (clear cut step down) there is no oscillation while decaying overall. All the oscillations during the 5 minutes time to "settle" are taking place around the baseline. This is an under-damped 2nd order oscillator ringing in response to a step down. But for the recorded behavior at magnetron power-off, there is no way to explain the observed oscillation on top of an exp. decay of much longer time constant (than the oscillations) other than a slowly decaying 1st order forcing excitation with its own longer time constant. You can make all the arguments you want about steepness or whatnot at power-on, observed behavior at power-off puts a major part of magnitude of excitation in the long term decay effects, hence thermal in nature (whatever the precise mechanisms of thermal).
The "observed oscillation on top of an exp" actually can be explained. If you have seen my experiment https:
//www.youtube.com/watch?v=UsOee729YBM you will observe an oscillation on top of the exp too. The explanation is that, the sudden force or the sudden removing of the force is applied to the side of the balance. It is more than a pure rotary torque. The balance's hanging wire, the supposed pivot, is not fixed in position. The force or the removing of the force will introduce a pendulum swing on top of the rotary swing. The instant pivot is probably near the oil damper (of his) and near the magnet damper (of mine). The pendulum swing period is shorter thus you see the faster oscillation on top of the exponential decay.
Nicely pointed out. The oil damper needs to be at the centre of the beam rotation and not where it is currently located at the middle of the non EmDrive side of the beam. Have advised Dave.
-
#3622 by Monomorphic on 07 Jul, 2016 00:39
-
A friend said my workbench looked like it was being assimilated by the Borg. I'm not sure what gave him that impression.
-
#3623 by FattyLumpkin on 07 Jul, 2016 00:40
-
Todd, thank you for that....I don't remember where, but I read it: that the reviewers were mostly (or something along that line) satisfied with their (EWL'S) test procedures, but that they really ripped into Sonny's QV theory. Damned if I can't remember where I saw that. The impression that I was left with was something like 'the effect is real, but because you can't explain it to our satisfaction, we can't accept it'---I believe this was inferred rather than stated. Does this go along with what Paul said to you in his mail? Thanks, Kevin
-
#3624 by Rodal on 07 Jul, 2016 00:44
-
,,,Are you not seeing that we have something like 50% of whatever made the pendulum deviate in the first place still forcing (in the same direction) about 2 minutes after power was interrupted ?
Precisely the point I was making to SeeShells -
#3625 by SeeShells on 07 Jul, 2016 01:43
-
And you made a good point although with Dave's setup it's very tough to discern what the anomaly is showing on the beam. He will be taking some suggestions into his next run that should make it easier to see what is going on.,,,Are you not seeing that we have something like 50% of whatever made the pendulum deviate in the first place still forcing (in the same direction) about 2 minutes after power was interrupted ?
Precisely the point I was making to SeeShells
I hated the way my beam would show up the pitch and the yaw of the beam on the laser affixed to the beam, trying to narrow out any meaningful data from the bouncing ball of laser light was very hard. Similar to Dave's pitching and rocking beam.
I came upon simple solution that will negate the pitch and minimize the roll just leaving the yaw or horizontal rotational moment on the laser and metered ruler.
It is a simple laser level outputting a line instead of a laser dot. The laser level is mounted on the center line of the torsion wire directly on the beam.
Shell
-
#3626 by zen-in on 07 Jul, 2016 03:49
-
More thanks to Dave for helping me through the installation of the laser displacement sensor. I made a couple of novice mistakes that cost me hours of frustration. It's working now!
The only problem i'm seeing now is in the chart below. The little squiggles is from when I tapped the beam, but you will notice a slow rise across the span and a sudden periodic drop. That is not from the beam moving. I think that may be the cheap resistor i'm using. It is 270 Ohm with 5% tolerance. Dataq recommends a Precision 250 Ω resistor, 0.1% tolerance - which I have on the way.
There really are big differences in different types of leaded resistors when you are going for minimal noise. In building low noise sensor amplifiers I found that carbon comp and carbon film (==cheap) were the worst. Metal film 1% resistors are much quieter and when you get your hands on a .1% or .01% resistor the accuracy and temperature stability is better still. If you are using surface mount resistors the same applies. -
#3627 by keithpickering on 07 Jul, 2016 05:45
-
...
I'm in the very private email discussion group that is working with Dave. He doesn't know when the top ring magnet cracked other that it was not working when he did his last test run. The bad maggie is being replaced and a few other suggested tidy ups implemented before restarting the test series.
New items will be using a phase change substance to limit the maggie temp increase as Dave has lost 2 maggies to overheat plus better monitoring of the maggie anode voltage to see if it is possible, as I believe it is, to detect when maggie to frustum lock occurs and a better feed of the high current to the maggie that should not cause any issues with the filament current.
There is a lot of work yet to be done.
But I say again, none of the measured thermal effects with the frustum small end pointing UP has shown anything like the very rapid initial force generation when the maggie locks to the frustum.
I should add, the observed delay time with just a side flick of a pen on the side of the beam and the active power off decay time are around the same value of 5 minutes, 20 seconds, which suggests there is not much thermal in the power off decay.
I don't believe it is practical to do tests with a battery if a 1 kW magnetron is used. For that reason I believe we have learned as much as we need to about magnetron emdrive experiments. The noise and thermal affects completely swamp out any real emdrive thrust if it exists. For a battery powered, no umbilical test a solid state amplifier driven with a compact stable oscillator may be the best option. A wireless (300 MHz garage door type might work best) control for switching on the amp and oscillator could be used. If the amp was class C it would not heat up much on short runs. However class C amps are non-linear and have fixed power outputs.
It is possible to construct a device that sharply minimizes thermal effects, compared to most DIY setups seen so far. Things that could be done are:
1. Covering the frustum with high-R insulating foam.
2. Putting the entire device inside a reasonably airtight box.
3. Isolating the magnetron and power supply from the frustum via waveguide.
4. Running the magnetron at less than full power.
-
#3628 by meberbs on 07 Jul, 2016 06:07
-
While Dave is waiting for his new magnetron, I think it would be a good time to characterize and find the source of the displacement in the last run with the broken magnetron. This is a good null test with this "broken" setup (assuming there are no safety issues with continuing to run it this way).
The current displacement is purely due to some undesired (probably thermal) effect. Based on the previous runs, this thermal effect is dominating the entire response even when the magnetron is working properly. The only way to get really clear data will be if this effect can be effectively eliminated, and I think it would be good to see the current setup modified to do so before the working magnetron is installed.
I am not sure exactly what kind of effect is causing it (air currents, thermal expansion, something else), but it would be very informative to have this isolated down to a demonstrable cause. Good luck to those who are working on doing this, since finding and eliminating the source of this signal will likely be difficult. -
#3629 by FattyLumpkin on 07 Jul, 2016 06:26
-
Perhaps the strangest (stupidest?) question I've ever asked (about Emdrive) Here goes: Can a frustum that resonates well in at a given frequency and mode also resonate well at a higher "harmonic" of the fundamental resonant frequency? Obviously not mechanical/sound.
Leading to a sort of opposite of the question: If a frustum is reduced in size by 1/2 can is resonate in some sort of elemental "part" of the lower waveform/frequency?
There is a reason for this question: will disclose after response(s) Please respond!
-
#3630 by FattyLumpkin on 07 Jul, 2016 06:32
-
Re Dave's gen 2 build, sure would like to see what we'd/he'd get (simulations), if he were to mount dialectic, say in the top, like the NASA EW build. I could do a volume calc. and come up with a +/- size and shape.
-
#3631 by frobnicat on 07 Jul, 2016 09:26
-
Ok, thanks for the repost. This again illustrate the point I make : it's not a matter of time to "settle" understood as "stop oscillating". It's the matter than on this plot when a forcing excitation is suddenly relieved (clear cut step down) there is no oscillation while decaying overall. All the oscillations during the 5 minutes time to "settle" are taking place around the baseline. This is an under-damped 2nd order oscillator ringing in response to a step down. But for the recorded behavior at magnetron power-off, there is no way to explain the observed oscillation on top of an exp. decay of much longer time constant (than the oscillations) other than a slowly decaying 1st order forcing excitation with its own longer time constant. You can make all the arguments you want about steepness or whatnot at power-on, observed behavior at power-off puts a major part of magnitude of excitation in the long term decay effects, hence thermal in nature (whatever the precise mechanisms of thermal).
The "observed oscillation on top of an exp" actually can be explained. If you have seen my experiment https:
//www.youtube.com/watch?v=UsOee729YBM you will observe an oscillation on top of the exp too. The explanation is that, the sudden force or the sudden removing of the force is applied to the side of the balance. It is more than a pure rotary torque. The balance's hanging wire, the supposed pivot, is not fixed in position. The force or the removing of the force will introduce a pendulum swing on top of the rotary swing. The instant pivot is probably near the oil damper (of his) and near the magnet damper (of mine). The pendulum swing period is shorter thus you see the faster oscillation on top of the exponential decay.
This explanation is not compatible with some known facts :
The swinging (center of mass oscillating around its rest position) whether along the axis of pendulum arm or the orthogonal axis are governed by a period T=2π√(L/g) where L is length of suspension wire. To get to around or above 1 minute period for those modes requires L=894m on Earth (T=60s). Whatever the setup, T for those modes will be << 1 minute.
The swinging in roll (axis parallel to pendulum arm) will depend on how far the center of mass is below the attachment point of suspension wire to pendulum arm (assuming this is roughly the point of free pivot for this mode) and of the moment of inertia (rotational inertia) around the axis parallel to pendulum arm and going through center of mass. Seen from this axis the moment of inertia is not tremendous. Later on...
The swinging in pitch (horizontal axis orthogonal to pendulum arm) will depend on how far the center of mass is below the attachment point of suspension wire to pendulum arm (assuming this is roughly the point of free pivot for this mode) and of the moment of inertia (rotational inertia) around the axisparallelhorizontal and orthogonal to pendulum arm and going through center of mass. Seen from this axis the moment of inertia can be important, much higher than for the swinging in roll, meaning that apart from the torsional swinging (Yaw around the vertical suspension wire axis) of the 4 mentioned "spurious modes candidates" that will be the one with the longer period. Because of a high moment of inertia around this axis, this period can be made arbitrarily long without having km sized suspension wire, actually by lowering the distance of center of mass below the pivot (roughly, attachment point of suspension wire to pendulum arm) and lowering the gravity induced equivalent of stiffness of a callback spiral spring for a inertia wheel pendulum. So we need more quantitative modelling or experimental data here...
In your video you link to I notice that the half life of return to rest around Yaw axis is between 12:33:40 and 12:34:18, wall-clock time of t1/2=38s (give or take a few secs), or a time constant τ=55s. On the other hand the oscillations seen (not much) when you manipulate and release the pendulum arm are clearly on the order of a second of period or less for the various swingings save perhaps the swinging in pitch that could be 2s in duration (hard to tell, guessing some upper bound from half a period). Visual tends to confirm what I say above, apart from the slow dynamics t1/2=38s in Yaw (driven by suspension wire stiffness in torsion, restoring torque battling against inertia of pendulum arm perpendicular to it) the next longer mode is for the swinging in pitch, with a period about 20 times less than the half time for Yaw dynamics.
All right, so, your setup is not rfmwguy's, different geometry etc... but I hope you get the point : the half life of the slow decay at power off (for rfmwguy's plots) is about 2 or 3 minutes, the oscillations around are between 1 or 2 minutes. Not an order of magnitude shorter. Given what I said above this is hardly compatible with your explanation unless the center of mass below attachment point is darn close to attachment point of suspension wire on pendulum arm.
But what is decisive (in case of rfmwguy's plots) is the nice return to rest from a side excitation step down. The oscillation period here is clearly that of the Yaw axis. So we know already that the 1 or 2 minutes typical period is that of the Yaw, can't be that of the pitch (and even less that of roll). So we know that the 2 or 3 minutes exp. decay is not an intrinsic mechanical parameter of the measuring system, it must be a characteristic of the forcing function.
edit : in red -
#3632 by rfmwguy on 07 Jul, 2016 12:06
-
.Ok, thanks for the repost. This again illustrate the point I make : it's not a matter of time to "settle" understood as "stop oscillating". It's the matter than on this plot when a forcing excitation is suddenly relieved (clear cut step down) there is no oscillation while decaying overall. All the oscillations during the 5 minutes time to "settle" are taking place around the baseline. This is an under-damped 2nd order oscillator ringing in response to a step down. But for the recorded behavior at magnetron power-off, there is no way to explain the observed oscillation on top of an exp. decay of much longer time constant (than the oscillations) other than a slowly decaying 1st order forcing excitation with its own longer time constant. You can make all the arguments you want about steepness or whatnot at power-on, observed behavior at power-off puts a major part of magnitude of excitation in the long term decay effects, hence thermal in nature (whatever the precise mechanisms of thermal).
The "observed oscillation on top of an exp" actually can be explained. If you have seen my experiment https:
//www.youtube.com/watch?v=UsOee729YBM you will observe an oscillation on top of the exp too. The explanation is that, the sudden force or the sudden removing of the force is applied to the side of the balance. It is more than a pure rotary torque. The balance's hanging wire, the supposed pivot, is not fixed in position. The force or the removing of the force will introduce a pendulum swing on top of the rotary swing. The instant pivot is probably near the oil damper (of his) and near the magnet damper (of mine). The pendulum swing period is shorter thus you see the faster oscillation on top of the exponential decay.
This explanation is not compatible with some known facts :
The swinging (center of mass oscillating around its rest position) whether along the axis of pendulum arm or the orthogonal axis are governed by a period T=2π√(L/g) where L is length of suspension wire. To get to around or above 1 minute period for those modes requires L=894m on Earth (T=60s). Whatever the setup, T for those modes will be << 1 minute.
The swinging in roll (axis parallel to pendulum arm) will depend on how far the center of mass is below the attachment point of suspension wire to pendulum arm (assuming this is roughly the point of free pivot for this mode) and of the moment of inertia (rotational inertia) around the axis parallel to pendulum arm and going through center of mass. Seen from this axis the moment of inertia is not tremendous. Later on...
The swinging in pitch (horizontal axis orthogonal to pendulum arm) will depend on how far the center of mass is below the attachment point of suspension wire to pendulum arm (assuming this is roughly the point of free pivot for this mode) and of the moment of inertia (rotational inertia) around the axis parallel to pendulum arm and going through center of mass. Seen from this axis the moment of inertia can be important, much higher than for the swinging in roll, meaning that apart from the torsional swinging (Yaw around the vertical suspension wire axis) of the 4 mentioned "spurious modes candidates" that will be the one with the longer period. Because of a high moment of inertia around this axis, this period can be made arbitrarily long without having km sized suspension wire, actually by lowering the distance of center of mass below the pivot (roughly, attachment point of suspension wire to pendulum arm) and lowering the gravity induced equivalent of stiffness of a callback spiral spring for a inertia wheel pendulum. So we need more quantitative modelling or experimental data here...
In your video you link to I notice that the half life of return to rest around Yaw axis is between 12:33:40 and 12:34:18, wall-clock time of t1/2=38s (give or take a few secs), or a time constant τ=55s. On the other hand the oscillations seen (not much) when you manipulate and release the pendulum arm are clearly on the order of a second of period or less for the various swingings save perhaps the swinging in pitch that could be 2s in duration (hard to tell, guessing some upper bound from half a period). Visual tends to confirm what I say above, apart from the slow dynamics t1/2=38s in Yaw (driven by suspension wire stiffness in torsion, restoring torque battling against inertia of pendulum arm perpendicular to it) the next longer mode is for the swinging in pitch, with a period about 20 times less than the half time for Yaw dynamics.
All right, so, you setup is not rfmwguy's, different geometry etc... but I hope you get the point : the half life of the slow decay at power off (for rfmwguy's plots) is about 2 or 3 minutes, the oscillations around are between 1 or 2 minutes. Not an order of magnitude shorter. Given what I said above this is hardly compatible with your explanation unless the center of mass below attachment point is darn close to attachment point of suspension wire on pendulum arm.
But what is decisive (in case of rfmwguy's plots) is the nice return to rest from a side excitation step down. The oscillation period here is clearly that of the Yaw axis. So we know already that the 1 or 2 minutes typical period is that of the Yaw, can't be that of the pitch (and even less that of roll). So we know that the 2 or 3 minutes exp. decay is not an intrinsic mechanical parameter of the measuring system, it must be a characteristic of the forcing function.
Hello Professor.
I have taken note of your well thought out commentary and commend you for taking the time. I know what its like to devote time/thoughts to a project and you've done this like me, without remuneration; just for the love of science. Thank you.
Pitch, roll and yaw are a classic problem with a torsion beam. The balance beam eliminates some directional components and was superior in that regard, less useful in others.
My system has the 23 inch long piano wire terminating at the top of an 22 inch aluminum mast. From there, there are 2 guy wires terminating 14 inches from each end of a 114 inch total length wooden beam. The center of mass of the emdrive is about 30 inches from center towards one end. The center of mass of the dampener is roughly the same distance towards the opposite end. The dampener is on an 26 inch aluminum rod that can pivot only along the perpendicular axis of the torsion beam. (swings towards and away from center if needed). The LDS pickup is at the far end of the beam opposite the emdrive.
Total suspended weight is about 26lbs. The emdrive can rotate 360 degrees on a parallel axis to the torsion beam and is locked into place during testing.
Its difficult to describe without a chalkboard. I'll check back from time to time, but have put in a lot of long hours...time is a precious commodity. Thanks again. -
#3633 by rfmwguy on 07 Jul, 2016 12:08
-
A friend said my workbench looked like it was being assimilated by the Borg. I'm not sure what gave him that impression.
Resistance is futile. You must continue your hard work.
-
#3634 by Tellmeagain on 07 Jul, 2016 12:44
-
Perhaps the strangest (stupidest?) question I've ever asked (about Emdrive) Here goes: Can a frustum that resonates well in at a given frequency and mode also resonate well at a higher "harmonic" of the fundamental resonant frequency? Obviously not mechanical/sound.
Leading to a sort of opposite of the question: If a frustum is reduced in size by 1/2 can is resonate in some sort of elemental "part" of the lower waveform/frequency?
There is a reason for this question: will disclose after response(s) Please respond!
The following are just my thoughts. Surely can be wrong.
"Can a frustum that resonates well in at a given frequency and mode also resonate well at a higher "harmonic" of the fundamental resonant frequency?"
I think it depends on whether the frequency of the higher modes are accidentally integer times that of the lower modes. Only simulation can tell because there is no analytical solution to the equations. But guessing from what I learned from musical instrument physics, probably the answer is no.
"If a frustum is reduced in size by 1/2 can is resonate in some sort of elemental "part" of the lower waveform/frequency? "
I think not for the truncated cone type of frustum， because mechanically a 1/2 frustum can not be part of a full size frustum, if you have to cut the full size frustum in a symmetric way. But maybe part of the smaller frustum can be part of the full size frustum. I am not very sure about this.
-
#3635 by Rodal on 07 Jul, 2016 12:59
-
Perhaps the strangest (stupidest?) question I've ever asked (about Emdrive) Here goes: Can a frustum that resonates well in at a given frequency and mode also resonate well at a higher "harmonic" of the fundamental resonant frequency? Obviously not mechanical/sound.
Leading to a sort of opposite of the question: If a frustum is reduced in size by 1/2 can is resonate in some sort of elemental "part" of the lower waveform/frequency?
There is a reason for this question: will disclose after response(s) Please respond!
The following are just my thoughts. Surely can be wrong.
"Can a frustum that resonates well in at a given frequency and mode also resonate well at a higher "harmonic" of the fundamental resonant frequency?"
I think it depends on whether the frequency of the higher modes are accidentally integer times that of the lower modes. Only simulation can tell because there is no analytical solution to the equations. But guessing from what I learned from musical instrument physics, probably the answer is no.
"If a frustum is reduced in size by 1/2 can is resonate in some sort of elemental "part" of the lower waveform/frequency? "
I think not for the truncated cone type of frustum， because mechanically a 1/2 frustum can not be part of a full size frustum, if you have to cut the full size frustum in a symmetric way. But maybe part of the smaller frustum can be part of the full size frustum. I am not very sure about this.
1) Yes, there is an exact solution to calculating natural frequencies for a electromagnetically resonant truncated cone cavity with spherical ends, see for example: http://gregegan.customer.netspace.net.au/SCIENCE/Cavity/Cavity.html
we have shown (in previous threads) excellent comparison between the NASA COMSOL Finite Element numerical analysis and the exact solution, the difference being within 1% or less of the exact solution. Such exact solutions to the problem have been known for more than 80 years, since the 1930's, due to Dr. Schelkunoff from Bell Labs (https://en.wikipedia.org/wiki/Sergei_Alexander_Schelkunoff), shown here with conical shaped microwave waveguides (henceforth falsifying those that keep writing that no analysis or experimentation with conical waveguides or cavities was done before the attempted explanations by Shawyer):
2) Resonant systems have as many natural frequencies as the number of degrees of freedom. You can think of the degrees of freedom as particles, connected by springs (potential energy) to other particles. Since the number of degrees of freedom is a very large number, the number of natural frequencies is very large. A continuum field idealization (like Maxwell's equations) results in an infinite number of natural frequencies.
3) The natural frequencies (eigenfrequencies) are associated with mode shapes (eigenmodes): they are the solution to the eigenvalue problem. The eigenvalue problem can be formulated, for transverse magnetic mode shapes as:
Curl[Curl[ E]]] = λ E
∇x(∇xE)= λ E
where λ = ω2/c2 and ω are the eigenfrequencies of resonance
and E is the electric field.
The eigenvalue problem can be solved as a closed-form solution only for a few particular domains, e.g.,for a rectangular box or a circular cylinder. It is educational to see the solution for the natural frequencies in TE mode shapes for a cylinder, because it is close in geometry to the geometry of EM Drives:
Notice that the natural frequency dependence on the longitudinal direction wavepattern goes like the square root of the square of the number "p" times a coefficient dependent on length plus another function that depends on the diameter.
Notice that the natural frequencies do NOT go up linearly even for a cylinder. (However, for mode shapes with very high "p" and low "m" and "n" such that that the term with "p" is much larger, they do go up almost linearly with "p". They never go up linearly with "m" or "n" because they depend on the Bessel function which is not linear with m or n).
For the truncated spherical cone, the eigenvalue problem has to be solved numerically by iteration.
These are the TEmnp and TMmnp mode shapes that we have been discussing and shown in numerous graphics. The distance between natural frequencies is not proportional , it is governed by Spherical Bessel functions in the longitudinal direction, that's why the "wavelength" of the wavepattern changes continuously in the longitudinal direction. And it is governed by associated Legendre functions in one of the perpendicular directions. Therefore the answer to your question is a resounding
NOT in general
because the distance between natural frequencies (in the frequency spectrum) for the EM Drive is not proportional.
However, since as you go up in frequency, there are more natural frequencies next to each other at higher natural frequencies, it is quite possible that at much higher natural frequencies than the fundamental, that by pure luck and chance you will be next to a harmonic of a low natural frequency.
I attach the study done with COMSOL Finite Element Analysis by Frank Davis at NASA showing the spectrum of natural frequencies for the EM Drive in the frequency bandwidth of interest. Showing that the distribution of natural frequencies is not proportional.
This early study by NASA Eagleworks, that neatly answers your question, shows once again that the quality of analysis done by NASA demonstrates a quality of research that is far above anybody else that has reported on the EM Drive.
We only have access to this information thanks to Star-Drive, and it is too bad that he is no longer able to communicate further information. Those times, when Star-Drive was able to openly discuss information such as this were, in some sense, "the best of times" for public discussion of this topic.
-
#3636 by WarpTech on 07 Jul, 2016 13:48
-
Todd, thank you for that....I don't remember where, but I read it: that the reviewers were mostly (or something along that line) satisfied with their (EWL'S) test procedures, but that they really ripped into Sonny's QV theory. Damned if I can't remember where I saw that. The impression that I was left with was something like 'the effect is real, but because you can't explain it to our satisfaction, we can't accept it'---I believe this was inferred rather than stated. Does this go along with what Paul said to you in his mail? Thanks, Kevin
I really can't say. IMO his QV theory is very close to my own, it's not wrong. That's why I'm preparing a response and an offer to help in that regard. -
#3637 by TheTraveller on 07 Jul, 2016 14:45
-
Its difficult to describe without a chalkboard. I'll check back from time to time, but have put in a lot of long hours...time is a precious commodity. Thanks again.
-
#3638 by FattyLumpkin on 07 Jul, 2016 18:36
-
Dr. Rodal, Tellmeagain, thank you for your responses! I asked this question because recently I was in communication with a want-to-be builder who indicated that certain "parts": amps, antennae, etc. 2.45 GHz in particular, were more available and consequently less expensive than others. I'm not a builder (yet), and would not know about availability/cost of parts et al. That aside, this builder wants to build a small frustum: 2-3 times smaller than frustums built, yet resonate in the regions of 1.5 - 2.5 GHz. That is what lead to my question.
Dr. Rodal, do you have a frequency/resonant mode chart >2.5GHz for our asymmetric resonant cavities (frustums)? I don't believe I saw even one resonant mode "repeat" in your entire chart, (nor would I expect to), based on your overall response.
I did tell this fellow, that he needed to study this ENTIRE site prior to proceeding any further as he had already built a small frustum, signal generator, antenna etc., but did not know what a resonant mode is/was.
After that, I said to myself "we still really don't know what makes this work or for that matter if it works" (Albeit, I leaning toward it being a real phenomenon) , Kevin
-
#3639 by X_RaY on 07 Jul, 2016 18:45
-
Dr. Rodal, Tellmeagain, thank you for your responses! I asked this question because recently I was in communication with a want-to-be builder who indicated that certain "parts": amps, antennae, etc. 2.45 GHz in particular, were more available and consequently less expensive than others. I'm not a builder (yet), and would not know about availability/cost of parts et al. That aside, this builder wants to build a small frustum: 2-3 times smaller than frustums built, yet resonate in the regions of 1.5 - 2.5 GHz. That is what lead to my question.
This is an example based on explicit dimensions (big diameter and height). It shows what happens if the small diameter and therefore the cone half angle θ changes.
Dr. Rodal, do you have a frequency/resonant mode chart >2.5GHz for our asymmetric resonant cavities (frustums)? I don't believe I saw even one resonant mode "repeat" in your entire chart, (nor would I expect to), based on your overall response.
I did tell this fellow, that he needed to study this ENTIRE site prior to proceeding any further as he had already built a small frustum, signal generator, antenna etc., but did not know what a resonant mode is/was. After that, I said to myself "we still really don't know what makes this work or for that matter if it works" (Albeit, I leaning toward it being a real phenomenon) , Kevin
After that, I said to myself "we still really don't know what makes this work or for that matter if it works"