-
#560 by TheTraveller on 16 Aug, 2015 05:41
-
1) The force magnitude is a whooping 10,000 times higher than for previous cases.
Prof Yang has shown the end plate bounce Force is very much larger if the H field (TE mode) is doing the bounce than if the E field (TM mode) is doing the bounce.
-
#561 by TheTraveller on 16 Aug, 2015 05:43
-
Quote
Resistivity ohm-m
Pure Copper 1.664 E-08
High Strength Yellow Brass 1.437 E-07
Ratio of resistivities 8.64
Square Root of Ratio 2.94
So using High Strength Yellow Brass implies a quality of resonance Q about 1/3 that of using pure copper
Hmmm...
1 - Possibly a used or damaged instrument would work, seeing as we'd be concerned only with the one end?
2 - Perhaps it would be possible to silver coat the interior?
Brass can easily be electroplated with pure copper, or gold plated. The shape and consistency is what I like about it.
Todd
Prof Yang used brass. Didn't say if the inside was electroplated or not. -
#562 by seggybop on 16 Aug, 2015 07:24
-
The only experimenters so far that have actually measured and reported a mode shape has been NASA Eagleworks that actually measured TM212 mode shape by using a thermal camera pointed at the big base.
Thermal cameras are available pretty cheaply now as smartphone accessories. Seek Thermal and FLIR One make the most prominent ones. I got the Seek unit a while ago for around $150 and it's worked quite well for me. The numerical accuracy of the temperature measurements can't be counted on, but for visualizing the heating patterns created by different mode shapes it should be viable. -
#563 by TheTraveller on 16 Aug, 2015 08:15
-
Here is a public folder with all of Eagleworks publically posted Force measurements. As far as I know this is the 1st time all the known, public Eagleworks EMDrive Force measurement runs are available in one folder to view the results. If I have missed any, please post them so I can add them to the collection.
https://drive.google.com/folderview?id=0B7kgKijo-p0ifnFrZ2V1UmZEY25FXzNrX0hjNXJmQXR5YzRnaVBqcTdMZUhxcjVkMUUtaXc&usp=sharing
Please review. You may find the measured atmo and vac results are more positive than you thought they were and that the Force onset is very rapid.
Also note Eagleworks do a measurement system calibration pulse before and after each EMDrive test run. -
#564 by TheTraveller on 16 Aug, 2015 08:51
-
Looking at all the atmo Force curves I do note the very strange one which occurred when the frustum was enclosed by foam panels. I assume EW did this to shield against air currents.
What I see is the effect of reduced acoustic vibration on the end plates reducing the necessary external vibratory Force needed to move the frustum from IDLE mode into MOTOR mode. The result is a Force generation curve suggesting the EMDrive was struggling to get into MOTOR mode as the required external vibratory Force has been very significantly reduced by the foam panel enclosure.
-
#565 by Rodal on 16 Aug, 2015 10:49
-
1) The force magnitude is a whooping 10,000 times higher than for previous cases. At this point we don't know how much of this greater magnitude is due to the fact that this computer run is for twice as long a time as previous runs (with stresses that are increasing with time) and how much is due to the fact that this force is due to a transverse electric (TE) mode shape while the other ones were for transverse magnetic (TM) modes.
Just to clarify, you mean an increase of 10^4 not 10^10000.
Yes that's what that means. Sorry but I don't understand why it needs to be clarified.10,000 times higher
cannot mean 10^10,000 under any interpretation I am aware of.
"Times" means multiplication: 10,000 x a means 10,000 times a
On the other hand (exponentiation) 10^10,000 reads: 10 to the power of 10,000
Hence it is not possible to confuse "10,000 times higher" with 10^10,000 or any other base raised to any other power.
-
#566 by RERT on 16 Aug, 2015 11:19
-
This seems highly relevant.
http://iopscience.iop.org/1367-2630/17/5/053035/pdf/1367-2630_17_5_053035.pdf
Experimental evidence for Abraham pressure of light
Keywords: momentum of light, light in media, optomechanics, optofluidics
Abstract
The question of how much momentum light carries in media has been debated for over a century.
Two rivalling theories, one from 1908 by Hermann Minkowski and the other from 1909 by Max
Abraham, predict the exact opposite when light enters an optical material: a pulling force in
Minkowskiʼs case and a pushing force in Abrahamʼs. Most experimental tests have agreed with
Minkowskiʼs theory, but here we report the first quantitative experimental evidence for Abrahamʼs
pushing pressure of light. Our results matter in optofluidics and optomechanics, and wherever light
exerts mechanical pressure. -
#567 by Rodal on 16 Aug, 2015 12:05
-
This seems highly relevant.
http://iopscience.iop.org/1367-2630/17/5/053035/pdf/1367-2630_17_5_053035.pdf
Experimental evidence for Abraham pressure of light
Keywords: momentum of light, light in media, optomechanics, optofluidics
Abstract
The question of how much momentum light carries in media has been debated for over a century.
Two rivalling theories, one from 1908 by Hermann Minkowski and the other from 1909 by Max
Abraham, predict the exact opposite when light enters an optical material: a pulling force in
Minkowskiʼs case and a pushing force in Abrahamʼs. Most experimental tests have agreed with
Minkowskiʼs theory, but here we report the first quantitative experimental evidence for Abrahamʼs
pushing pressure of light. Our results matter in optofluidics and optomechanics, and wherever light
exerts mechanical pressure.
To put some numbers into this, n, the coefficient of refraction
where c is the speed of light in vacuum and v is the phase velocity of light in the medium.
The theories of Abraham and Minkowski differ by the index of refraction (whether it should be a multiplying or a dividing factor).
For Minkowski, the momentum is:
where h is the Planck constant, ν is the frequency of the light and c is the speed of light in vacuum.
while for Abraham, the momentum is
hence the momentum between Minkowski and Abraham differs by n^2
For the EM Drive operating in vacuum,
n=1
(phase velocity equals c) and the copper acts as a reflective surface and not a transparent medium.
Hence, when operating in vacuum, there is no difference between Abraham and Minkowski.
For air,
n=1.000293
(phase velocity in air is 0.029% lower than c)
Hence for air, the difference between Minkowski and Abraham is 0.057%
the paper considers liquids, for example, for water n=1.333
(phase velocity is 33% lower than c)
Hence for a hypothetical EM Drive with water inside the cavity, the difference between Minkowski and Abraham would be 78%
For the dielectric used in the NASA experiments, high density polyethylene (HDPE) the index of refraction n is less than 1.48, so the difference between Abraham and Minkowski is less than 219%
So the Abraham Minkowski controversy could enter when discussing experiments with a dielectric insert (like NASA's) but not for experiments conducted in vacuum without a dielectric insert (unless one wants to consider the tiny effect from the thin coating used on copper to prevent corrosion of copper). -
#568 by Prunesquallor on 16 Aug, 2015 12:40
-
"Like a trombone" I said...
Todd
I'm thinking Baritone at this point... A baritone might be big enough to allow the use of a 2.45GHz magnetron.
Here's a shiny one for 150 bucks
http://www.alibaba.com/product-detail/Baritone_210767159.html?spm=a2700.7724838.35.1.BBzfaa
Hah! I used to play one of those! -
#569 by Rodal on 16 Aug, 2015 12:56
-
Wikipedia
https://en.wikipedia.org/wiki/Abraham%E2%80%93Minkowski_controversy
quote:
"The Abraham–Minkowski controversy is a physics debate concerning electromagnetic momentum within dielectric media. Related theories have been put forward that, should their principles be proven, may allow the design of a reactionless drive" Wikipedia reference 7.
The Wikipedia reference 7 to Brito is particularly deceiving since this is a reference to Brito's old 1999 paper and the article in Wikipedia fails to mention the more recent paper by Brito. Marini and Galian published 10 years later
(
Null Findings on Electromagnetic Inertia Thruster Experiments using a Torsion Pendulum
Hector H. Brito,* Ricardo Marini† and Eugenio S. Galian‡
45th AIAA/ASME/SAE/ASEE Joint Propulsion Conference & Exhibit 2 - 5 August 2009, Denver, Colorado
AIAA 2009-5070
)
where they nullify the Mach Lorentz Thruster type of Woodward and the Minkowski interpretation. Later on, in a peer-reviewed paper Marini and Galian
( Ricardo L. Marini and Eugenio S. Galian. "Torsion Pendulum Investigation of Electromagnetic Inertia Manipulation Thrusting", Journal of Propulsion and Power, Vol. 26, No. 6 (2010), pp. 1283-1290. doi: 10.2514/1.46541 http://arc.aiaa.org/doi/abs/10.2514/1.46541?journalCode=jpp )
proceed to defenestrate the Minkowski interpretation for such a thruster with experiments that nullify such thrust. -
#570 by Silversheep2011 on 16 Aug, 2015 13:38
-
Rodal,
I hadn't realised you where replying to that last post..when I deleted it...
As on re-reading those reference's it seemed to be adding 'clutter' to the discussion
Anyway thanks for good reply.
-
#571 by SeeShells on 16 Aug, 2015 16:05
-
That is exactly what it looks like Dr. Rodal and yes you would have a rotational mish- mashed asymmetry of the modes with one loop.2) It is evident that the parallel 2-Dipole antenna severely affects the mode. What we learn from this is that to excite a perfect TE012 mode, a circular loop antenna is needed
Many books and articles talk about square and rectangular antennas. This is fine for rectangular cross-section cavities, but to excite a perfect mode in a circular cross-section cavity it appears that a circular loop-antenna is needed.
Like this?
NASA must have the RF near the big end (correct me if I'm wrong) because they have the dielectric at the small end blocking the ability to put an antenna there.
It appears (correct me if I'm wrong) that NASA has the circular loop near the conical lateral surface. From the Meep runs I have learned that this produces all kinds of asymmetric waves that decrease the Q. It would be much better to have the circular loop in the very center with the axis of axi-symmetry of the cone going through the center of the circular loop antenna instead of being offset.
I imagine that a bigger loop antenna with its center on the axis of axi-symmetry of the cone would be ideal.
I wonder whether this placement of the NASA loop feed being so asymmetric is one of the reasons that NASA was not able to robustly reproduce TE modes (the only successful TE test they were able to conduct with only 2 watts input power was the mode that produced the most force/InputPower NASA ever reported). NASA has been running TM212 ever since.
Instead of the center placement, let's force the TE012 modes by placing 4 loops around the base circumference thereby pushing the modes to develop inline with the loops. 2 in phase and 2 out of phase by 180 degrees. This should almost act like an active filter locking out any other modes except for the small end where they might try to develop and then be pushed into the small end.
If we still wanted to maintain TE012 and have a rotational mixer (for you rfmwguy) angle the loops 11.25 degrees (multiples of 360 measured it in the pic) close like the loop in the EW photo.
Morning coffee thoughts,
Shell
-
#572 by Rodal on 16 Aug, 2015 16:27
-
...That is exactly what it looks like Dr. Rodal and yes you would have a rotational mish- mashed asymmetry of the modes with one loop.
Shell,
Instead of the center placement, let's force the TE012 modes by placing 4 loops around the base circumference thereby pushing the modes to develop inline with the loops. 2 in phase and 2 out of phase by 180 degrees. This should almost act like an active filter locking out any other modes except for the small end where they might try to develop and then be pushed into the small end.
If we still wanted to maintain TE012 and have a rotational mixer (for you rfmwguy) angle the loops 11.25 degrees (multiples of 360 measured it in the pic) close like the loop in the EW photo.
Morning coffee thoughts,
Shell
I noticed that the image "ezz30-t04.png" attached in your message corresponds to the latest set calculated by Meep, inside the cavity, at longitudinal location "30".
I don't know at what time (from RF being turned ON) are these images calculated, do you know whether they correspond to a Meep total run time of 640 time slices = "64 periods" = 0.026 microseconds or whether they correspond to 320 time slices ="32 periods" = 0.013 microseconds ? -
#573 by aero on 16 Aug, 2015 17:05
-
...That is exactly what it looks like Dr. Rodal and yes you would have a rotational mish- mashed asymmetry of the modes with one loop.
Shell,
Instead of the center placement, let's force the TE012 modes by placing 4 loops around the base circumference thereby pushing the modes to develop inline with the loops. 2 in phase and 2 out of phase by 180 degrees. This should almost act like an active filter locking out any other modes except for the small end where they might try to develop and then be pushed into the small end.
If we still wanted to maintain TE012 and have a rotational mixer (for you rfmwguy) angle the loops 11.25 degrees (multiples of 360 measured it in the pic) close like the loop in the EW photo.
Morning coffee thoughts,
Shell
I noticed that the image "ezz30-t04.png" attached in your message corresponds to the latest set calculated by Meep, inside the cavity, at longitudinal location "30".
I don't know at what time (from RF being turned ON) are these images calculated, do you know whether they correspond to a Meep total run time of 640 time slices = "64 periods" = 0.026 microseconds or whether they correspond to 320 time slices ="32 periods" = 0.013 microseconds ?
That's for the 64 cycle run. It matches the image, "ezz30-t04.png" and anyway, the 64 cycle run is the only one for which I have generated the z-30 slice data so far. -
#574 by RERT on 16 Aug, 2015 17:07
-
Ref: Abraham - Minkowski
It seems to me useful to recall two things. One is that there is no argument that classical EM theory is consistent with a time varying arrangement of currents creating a non-zero net Lorentz force on matter, once retarded fields are considered. Apart from anything else, a photon rocket shows a net force on matter.
One problem with the EM drive is that there is nothing like enough radiation emitted to balance COM if it is true that p=E/c.
I find it highly encouraging that despite the protestation that the subject is well understood and the EM Drive impossible, there is ongoing controversy over what the momentum of the EM field actually is, under certain circumstances!
Finally, I understand the arithmetic on the refractive index of air bring close to one. However, it seems a reasonable hypothesis that the skin currents in the copper will create oscillating fields into, as well as out of, the copper, which might generate interesting effects.
The paper I cited gives measurements to the effect that the EM field carries differing momentum in different cases, and suggests that the difference corresponds to whether it behaves as a wave or as a particle. If the probability of such behaviour were different at each end of the frustrum, it might help explain what's observed. -
#575 by X_RaY on 16 Aug, 2015 17:11
-
At 180 deg phase shift all the antennas would work against the next antenna nearby.
That is exactly what it looks like Dr. Rodal and yes you would have a rotational mish- mashed asymmetry of the modes with one loop.2) It is evident that the parallel 2-Dipole antenna severely affects the mode. What we learn from this is that to excite a perfect TE012 mode, a circular loop antenna is needed
Many books and articles talk about square and rectangular antennas. This is fine for rectangular cross-section cavities, but to excite a perfect mode in a circular cross-section cavity it appears that a circular loop-antenna is needed.
Like this?
NASA must have the RF near the big end (correct me if I'm wrong) because they have the dielectric at the small end blocking the ability to put an antenna there.
It appears (correct me if I'm wrong) that NASA has the circular loop near the conical lateral surface. From the Meep runs I have learned that this produces all kinds of asymmetric waves that decrease the Q. It would be much better to have the circular loop in the very center with the axis of axi-symmetry of the cone going through the center of the circular loop antenna instead of being offset.
I imagine that a bigger loop antenna with its center on the axis of axi-symmetry of the cone would be ideal.
I wonder whether this placement of the NASA loop feed being so asymmetric is one of the reasons that NASA was not able to robustly reproduce TE modes (the only successful TE test they were able to conduct with only 2 watts input power was the mode that produced the most force/InputPower NASA ever reported). NASA has been running TM212 ever since.
Instead of the center placement, let's force the TE012 modes by placing 4 loops around the base circumference thereby pushing the modes to develop inline with the loops. 2 in phase and 2 out of phase by 180 degrees. This should almost act like an active filter locking out any other modes except for the small end where they might try to develop and then be pushed into the small end.
If we still wanted to maintain TE012 and have a rotational mixer (for you rfmwguy) angle the loops 11.25 degrees (multiples of 360 measured it in the pic) close like the loop in the EW photo.
Morning coffee thoughts,
Shell
2 of them produce a magnetic field in one direction the other 2 in the opposite. That is against the field directions/vectors of the TE01p mode.
All of them in phase will work as well
EDIT:
Its the same like the 2 antennas "look" in the opposite direction (magnetic North Pole/South Pole of the loop).
It would work for TE012 if 2 antennas are near the small end and 2 (180°Phase shifted)antennas are near the big end if they have the same orientation.
magnetic H field vectors are blue in the pic, E is red
-
#576 by SeeShells on 16 Aug, 2015 17:24
-
64 cycle Shell 2D loop ant › z-30-pngs...That is exactly what it looks like Dr. Rodal and yes you would have a rotational mish- mashed asymmetry of the modes with one loop.
Shell,
Instead of the center placement, let's force the TE012 modes by placing 4 loops around the base circumference thereby pushing the modes to develop inline with the loops. 2 in phase and 2 out of phase by 180 degrees. This should almost act like an active filter locking out any other modes except for the small end where they might try to develop and then be pushed into the small end.
If we still wanted to maintain TE012 and have a rotational mixer (for you rfmwguy) angle the loops 11.25 degrees (multiples of 360 measured it in the pic) close like the loop in the EW photo.
Morning coffee thoughts,
Shell
I noticed that the image "ezz30-t04.png" attached in your message corresponds to the latest set calculated by Meep, inside the cavity, at longitudinal location "30".
I don't know at what time (from RF being turned ON) are these images calculated, do you know whether they correspond to a Meep total run time of 640 time slices = "64 periods" = 0.026 microseconds or whether they correspond to 320 time slices ="32 periods" = 0.013 microseconds ?
-
#577 by SeeShells on 16 Aug, 2015 17:31
-
I believe they are the same but the CSV run is little different with the last set cleaned up in the post filtering showing a little better mode action, if I have the info correct. It is a little hard to keep it all straight....
So what is then the difference between the contents of this folder
That's for the 64 cycle run. It matches the image, "ezz30-t04.png" and anyway, the 64 cycle run is the only one for which I have generated the z-30 slice data so far.
new-csvs
with the contents of this folder
64-cycle-run ...
Both of them are in the 64 cycle Shell 2D loop ant folder
The pic was for just to show a mode to explain the idea I was trying to get across and not anything else.
Shell -
#578 by X_RaY on 16 Aug, 2015 17:56
-
You think about and post your ideas for discussion. I like that
I believe they are the same but the CSV run is little different with the last set cleaned up in the post filtering showing a little better mode action, if I have the info correct. It is a little hard to keep it all straight....
So what is then the difference between the contents of this folder
That's for the 64 cycle run. It matches the image, "ezz30-t04.png" and anyway, the 64 cycle run is the only one for which I have generated the z-30 slice data so far.
new-csvs
with the contents of this folder
64-cycle-run ...
Both of them are in the 64 cycle Shell 2D loop ant folder
The pic was for just to show a mode to explain the idea I was trying to get across and not anything else.
Shell
-
#579 by SeeShells on 16 Aug, 2015 18:03
-
Ok i see what you're trying to say and that would work.
At 180 deg phase shift all the antennas would work against the next antenna nearby.
That is exactly what it looks like Dr. Rodal and yes you would have a rotational mish- mashed asymmetry of the modes with one loop.2) It is evident that the parallel 2-Dipole antenna severely affects the mode. What we learn from this is that to excite a perfect TE012 mode, a circular loop antenna is needed
Many books and articles talk about square and rectangular antennas. This is fine for rectangular cross-section cavities, but to excite a perfect mode in a circular cross-section cavity it appears that a circular loop-antenna is needed.
Like this?
NASA must have the RF near the big end (correct me if I'm wrong) because they have the dielectric at the small end blocking the ability to put an antenna there.
It appears (correct me if I'm wrong) that NASA has the circular loop near the conical lateral surface. From the Meep runs I have learned that this produces all kinds of asymmetric waves that decrease the Q. It would be much better to have the circular loop in the very center with the axis of axi-symmetry of the cone going through the center of the circular loop antenna instead of being offset.
I imagine that a bigger loop antenna with its center on the axis of axi-symmetry of the cone would be ideal.
I wonder whether this placement of the NASA loop feed being so asymmetric is one of the reasons that NASA was not able to robustly reproduce TE modes (the only successful TE test they were able to conduct with only 2 watts input power was the mode that produced the most force/InputPower NASA ever reported). NASA has been running TM212 ever since.
Instead of the center placement, let's force the TE012 modes by placing 4 loops around the base circumference thereby pushing the modes to develop inline with the loops. 2 in phase and 2 out of phase by 180 degrees. This should almost act like an active filter locking out any other modes except for the small end where they might try to develop and then be pushed into the small end.
If we still wanted to maintain TE012 and have a rotational mixer (for you rfmwguy) angle the loops 11.25 degrees (multiples of 360 measured it in the pic) close like the loop in the EW photo.
Morning coffee thoughts,
Shell
2 of them produce a magnetic field in one direction the other 2 in the opposite. That is against the field directions/vectors of the TE01p mode.
All of them in phase will work as well
EDIT:
Its the same like the 2 antennas "look" in the opposite direction (magnetic North Pole/South Pole of the loop).
It would work for TE012 if 2 antennas are near the small end and 2 (180°Phase shifted)antennas are near the big end if they have the same orientation.
