-
#2960 by SeeShells on 02 Jun, 2016 14:22
-
Also, NASA did NOT use a cone dielectric for the Cannae drive, they used a PTFE ("Teflon") cylinder. You can see the dielectric in this COMSOL FEA model: (it is Shawyer that used a cone dielectric inside a cavity with constant cylindrical cross-section in one of his patents)
This image has always confused me, so I'm not sure I'm seeing exactly what you're saying. The long thin rod down the center of the right pipe extension is the antenna right? Is the PTFE this cyclinder at the end and the antenna runs through that? What are the approximate dimensions and shape of the PTFE insert in Cannae drive?
Also realize that the antenna that they inserted into the cavity more than likely was coaxial cable and many coaxial cables have a center conductor surrounded by a insulator.
In most of those cases the center conductor is copper
The insulator is a dielectric
Then surrounded by a woven copper shield
All surrounded by an outer plastic sheath
https://en.wikipedia.org/wiki/Coaxial_cable
This simple fact tells me they were using a dielectric comprised of the coax insulating dielectric.
Shell -
#2961 by Rodal on 02 Jun, 2016 14:26
-
Also, NASA did NOT use a cone dielectric for the Cannae drive, they used a PTFE ("Teflon") cylinder. You can see the dielectric in this COMSOL FEA model: (it is Shawyer that used a cone dielectric inside a cavity with constant cylindrical cross-section in one of his patents)
This image has always confused me, so I'm not sure I'm seeing exactly what you're saying. The long thin rod down the center of the right pipe extension is the antenna right? Is the PTFE this cyclinder at the end and the antenna runs through that? What are the approximate dimensions and shape of the PTFE insert in Cannae drive?
Also realize that the antenna that they inserted into the cavity more than likely was coaxial cable and many coaxial cables have a center conductor surrounded by a insulator.
In most of those cases the center conductor is copper
The insulator is a dielectric
Then surrounded by a woven copper shield
All surrounded by an outer plastic sheath
https://en.wikipedia.org/wiki/Coaxial_cable
This simple fact tells me they were using a dielectric comprised of the coax insulating dielectric.
Shell
But the report explicitly states:
has a dielectric PTFE slug in the throat
and
the drive antenna coaxial cable and the region around the cable within the PFTE dielectric slug as seen in Fig. 14.
So, I interpret this as Monomorphic: the coaxial cable goes through and inside the PTFE dielectric slug which is a separate item and distinct from the coaxial cable and the PTFE slug is located at the throat
------------------
Also the fact that they used a coaxial cable for the Cannae test,regarding TellMeAgain's posts? didn't TellMeAgain say also that the NASA Cannae tests were subject to Lorentz forces? No Lorentz forces on the antenna coaxial cable...
But the problem with the coaxial cable is that it delivers 100% of the electromagnetic momentum in the axial direction of the cable, and hence it delivers a force in its axial direction, hence it may exhibit this force in whatever direction it is pointed in.
Best way to discriminate this would be to use a battery powered source self-integrated, rather than having power coming from a stationary source.
Has Cannae reported a test with battery-powered Cannae in their torsional pendulum yet? -
#2962 by Monomorphic on 02 Jun, 2016 14:39
-
has a dielectric PTFE slug in the throat
and
the drive antenna coaxial cable and the region around the cable within the PFTE dielectric slug as seen in Fig. 14.
So, I interpret this as Monomorphic: the coaxial cable goes through and inside the PTFE dielectric slug which is a separate item and distinct from the coaxial cable and the PTFE slug is located at the throat
Okay, this helps. If I had to make an educated guess, I would say the PTFE is the length shown here, and slightly smaller diameter as the right pipe extension - so it can fit inside.
-
#2963 by Rodal on 02 Jun, 2016 14:43
-
has a dielectric PTFE slug in the throat
and
the drive antenna coaxial cable and the region around the cable within the PFTE dielectric slug as seen in Fig. 14.
So, I interpret this as Monomorphic: the coaxial cable goes through and inside the PTFE dielectric slug which is a separate item and distinct from the coaxial cable and the PTFE slug is located at the throat
Okay, this helps. If I had to make an educated guess, I would say the PTFE is the length shown here, and slightly smaller diameter as the right pipe extension - so it can fit inside.
If you ever have the time, you could model this with FEKO, to see whether you get a FEKO solution for the dielectric as a slug surrounding the coaxial cable. This would allow you to discriminate between this interpretation and Shell's interpretation above. (Shawyer's interpretation could also be modeled, but Shawyer's interpretation of a cone dielectric in a pipe with constant cylindrical diameter does not make sense to me, from NASA's writing for Cannae, instead that is Shawyer's old patented design, not Cannae's). -
#2964 by Monomorphic on 02 Jun, 2016 14:48
-
has a dielectric PTFE slug in the throat
and
the drive antenna coaxial cable and the region around the cable within the PFTE dielectric slug as seen in Fig. 14.
So, I interpret this as Monomorphic: the coaxial cable goes through and inside the PTFE dielectric slug which is a separate item and distinct from the coaxial cable and the PTFE slug is located at the throat
Okay, this helps. If I had to make an educated guess, I would say the PTFE is the length shown here, and slightly smaller diameter as the right pipe extension - so it can fit inside.
If you ever have the time, you could model this with FEKO, to see whether you get a FEKO solution for the dielectric as a slug surrounding the coaxial cable. This would allow you to discriminate between this interpretation and Shell's interpretation above. (Shawyer's interpretation could also be modeled, but Shawyer's interpretation of a cone dielectric in a pipe with constant cylindrical diameter does not make sense to me, from NASA's writing for Cannae, instead that is Shawyer's old patented design, not Cannae's).
I have already constructed the geometry (best guess at dimensions). Just wasn't sure about the antenna. It should be a simple matter now. Do you know the relative permittivity and dialectric loss tangent of PTFE? -
#2965 by Monomorphic on 02 Jun, 2016 14:51
-
I also noticed there are a number of ways to define a dialectric in FEKO. See image here. You recognize any of these???
-
#2966 by SeeShells on 02 Jun, 2016 14:53
-
But the report explicitly states:
has a dielectric PTFE slug in the throat
and
the drive antenna coaxial cable and the region around the cable within the PFTE dielectric slug as seen in Fig. 14.
So, I interpret this as Monomorphic: the coaxial cable goes through and inside the PTFE dielectric slug which is a separate item and distinct from the coaxial cable and the PTFE slug is located at the throat
------------------
Also the fact that they used a coaxial cable for the Cannae test,regarding TellMeAgain's posts? didn't TellMeAgain say also that the NASA Cannae tests were subject to Lorentz forces? No Lorentz forces on the antenna coaxial cable...
But the problem with the coaxial cable is that it delivers 100% of the electromagnetic momentum in the axial direction of the cable, and hence it delivers a force in its axial direction, hence it may exhibit this force in whatever direction it is pointed in.
Best way to discriminate this would be to use a battery powered source self-integrated, rather than having power coming from a stationary source.
Has Cannae reported a test with battery-powered Cannae in their torsional pendulum yet?QuoteBut the problem with the coaxial cable is that it delivers 100% of the electromagnetic momentum in the axial direction of the cable, and hence it delivers a force in its axial direction, hence it may exhibit this force in whatever direction it is pointed in.
Dr. Rodal, if that's the case in the coaxial cable delivering the momenta in the Cannae drive (I can see it making sense
) Then the case for powering a loop probe antenna into the sidewall of the cavity or the case of a waveguide into the sidewalls doesn't make sense as the delivered momenta and force is 90o to the directions of thrusts.
Shell
-
#2967 by Monomorphic on 02 Jun, 2016 15:11
-
has a dielectric PTFE slug in the throat
and
the drive antenna coaxial cable and the region around the cable within the PFTE dielectric slug as seen in Fig. 14.
So, I interpret this as Monomorphic: the coaxial cable goes through and inside the PTFE dielectric slug which is a separate item and distinct from the coaxial cable and the PTFE slug is located at the throat
Okay, this helps. If I had to make an educated guess, I would say the PTFE is the length shown here, and slightly smaller diameter as the right pipe extension - so it can fit inside.
If you ever have the time, you could model this with FEKO, to see whether you get a FEKO solution for the dielectric as a slug surrounding the coaxial cable. This would allow you to discriminate between this interpretation and Shell's interpretation above. (Shawyer's interpretation could also be modeled, but Shawyer's interpretation of a cone dielectric in a pipe with constant cylindrical diameter does not make sense to me, from NASA's writing for Cannae, instead that is Shawyer's old patented design, not Cannae's).
I have already constructed the geometry (best guess at dimensions). Just wasn't sure about the antenna. It should be a simple matter now. Do you know the relative permittivity and dialectric loss tangent of PTFE?
Okay, that was a simple google search: relative permittivity for PTFE is 2.0 - 2.1 and the loss tangent is 0.00028 @ 3 GHz.
And am I right that PTFE has a mass density of 2,200 kg/m3? It's a LOT heavier than HTPE at 970! -
#2968 by Rodal on 02 Jun, 2016 15:22
-
..
It may not make sense to you, but it makes sense to me.
Dr. Rodal, if that's the case in the coaxial cable delivering the momenta in the Cannae drive (I can see it making sense
) Then the case for powering a loop probe antenna into the sidewall of the cavity or the case of a waveguide into the sidewalls doesn't make sense as the delivered momenta and force is 90o to the directions of thrusts.
Shell
It is incorrect to model the EM Drive powered by a coaxial cable ignoring the effect of the tapered frustum as to the direction of the expressed force. It has to satisfy the equations of conservation of energy and momentum.
In the case of the Cannae drive this problem is simplified because the coaxial cable is directed along the same axis as the axis of axisymmetry of the Cannae drive.
The electromagnetic momentum and force are delivered as per direction of the coaxial cable (I previously posted a paper by Kirk McDonald from Princeton proving this point). Once the electromagnetic momentum is inside a tapered frustum of a cone, the direction of the electromagnetic momentum will abide by the conservation of energy and conservation of momenta in the whole system, as per derivative with respect to time of the energy density and divergence of the Poynting vector in the conservation of energy, and as per derivative with respect to time of the Poynting vector and divergence of the stress tensor in the conservation of momentum.
The direction of the force is a result of the equations of motion of the system, as in any dynamic system.
==> I don't know what will happen when one runs with batteries, it will be exciting to get positive results as Hackaday Aachen team self-assesses in their report. To this date, I think that the best experiments are the ones run by NASA. Look forward to NASA conducting battery-powered tests.
-
#2969 by Rodal on 02 Jun, 2016 15:30
-
has a dielectric PTFE slug in the throat
and
the drive antenna coaxial cable and the region around the cable within the PFTE dielectric slug as seen in Fig. 14.
So, I interpret this as Monomorphic: the coaxial cable goes through and inside the PTFE dielectric slug which is a separate item and distinct from the coaxial cable and the PTFE slug is located at the throat
Okay, this helps. If I had to make an educated guess, I would say the PTFE is the length shown here, and slightly smaller diameter as the right pipe extension - so it can fit inside.
If you ever have the time, you could model this with FEKO, to see whether you get a FEKO solution for the dielectric as a slug surrounding the coaxial cable. This would allow you to discriminate between this interpretation and Shell's interpretation above. (Shawyer's interpretation could also be modeled, but Shawyer's interpretation of a cone dielectric in a pipe with constant cylindrical diameter does not make sense to me, from NASA's writing for Cannae, instead that is Shawyer's old patented design, not Cannae's).
I have already constructed the geometry (best guess at dimensions). Just wasn't sure about the antenna. It should be a simple matter now. Do you know the relative permittivity and dialectric loss tangent of PTFE?
Okay, that was a simple google search: relative permittivity for PTFE is 2.0 - 2.1 and the loss tangent is 0.00028 @ 3 GHz.
And am I right that PTFE has a mass density of 2,200 kg/m3? It's a LOT heavier than HTPE at 970!
The relative permittivitty and loss tangent values you have above are correct. The density of PTFE is over two times higher than HDPE. -
#2970 by Mulletron on 02 Jun, 2016 15:48
-
I briefly touched on this a few months ago. I'm not in a position where I can experiment right now to verify because I'm in a state of flux of my own. Studying this thing for the past couple of years has me thinking that the resonant rf isn't the cause of the thrust (thousands of times better than a photon rocket), rather it's the perturbation needed. The non Newtonian motion of the matter in the cavity is the final cause. So I propose adding this. A visual:
-
#2971 by Rodal on 02 Jun, 2016 16:05
-
I briefly touched on this a few months ago. I'm not in a position where I can experiment right now to verify because I'm in a state of flux of my own. Studying this thing for the past couple of years has me thinking that the resonant rf isn't the cause of the thrust (thousands of times better than a photon rocket), rather it's the perturbation needed. The non Newtonian motion of the matter in the cavity is the final cause. So I propose adding this. A visual:
That video shows the mechanical vibration of a plate (with eigenfrequencies and eigenmodes governed by the flexural modulus D of the plate, the dimensions, including thickness and side lengths of the plate, and the density of the material). It can be solved using finite element packages like ANSYS, ADINA, ABAQUS, etc.
Timonshenko has analytical solutions for some simple cases (Kirchoff-Love assumption). Reissner and Mindlin had further solutions adding the effects of transverse shear and rotary inertia.
https://en.wikipedia.org/wiki/Vibration_of_plates -
#2972 by Gilbertdrive on 02 Jun, 2016 16:26
-
Is it possible to use something to measure force with its breaking point?
Maybe for solidstate generators. With a Magnetron, I am afraid that the vibrations would break it even if there is no thrust. Also, it is very hard to predict the exact force needed to break a graphite tip, and we have very low thrust. I am not sure that it could help.
Mechanical pencil graphite tips can be used to nullify noise in resting position.
If there is force with power on, it will break the graphite tip.
Just an idea. -
#2973 by SeeShells on 02 Jun, 2016 16:30
-
I briefly touched on this a few months ago. I'm not in a position where I can experiment right now to verify because I'm in a state of flux of my own. Studying this thing for the past couple of years has me thinking that the resonant rf isn't the cause of the thrust (thousands of times better than a photon rocket), rather it's the perturbation needed. The non Newtonian motion of the matter in the cavity is the final cause. So I propose adding this. A visual:
Beautiful.
The fun part is to envision it in 3D down a asymmetrical cavity.
Shell -
#2974 by Rodal on 02 Jun, 2016 16:41
-
1) the electromagnetic pressure is way too small and the GHz frequency too high to envision significant mechanical vibrations of the EM Drive walls. Shawyer claims a maximum of less than 200 milliNewtons, many NASA experiments, particularly in vacuum, and also by Tajmar are in the microNewton range. One can readily show that the mechanical vibrations produced by such forces are of extremely small magnitude. Besides that, the frequency of oscillation of the microwaves is several orders of magnitude higher than typical mechanical vibrations. Therefore the mode shapes would be extremely high and therefore the video is unrepresentative of these extremely tiny vibrations (they would not be flexural vibrations as shown in the video but extremely small motions)I briefly touched on this a few months ago. I'm not in a position where I can experiment right now to verify because I'm in a state of flux of my own. Studying this thing for the past couple of years has me thinking that the resonant rf isn't the cause of the thrust (thousands of times better than a photon rocket), rather it's the perturbation needed. The non Newtonian motion of the matter in the cavity is the final cause. So I propose adding this. A visual:
Beautiful.
The fun part is to envision it in 3D down a asymmetrical cavity.
Shell
2) the video shows mechanical vibrations, at natural frequencies that are several orders of magnitude smaller than the GigaHertz electromagnetic oscillations. The video only shows mechanical vibrations due to flexure of the plate at much lower frequencies.
3) If one is inclined to envision flexural motion of the EM Drive walls, the attention should rather go towards thermal effects, either as dynamic buckling (that I earlier considered, see: https://www.researchgate.net/publication/268804028_NASA%27S_MICROWAVE_PROPELLANT-LESS_THRUSTER_ANOMALOUS_RESULTS_CONSIDERATION_OF_A_THERMO-MECHANICAL_EFFECT ) or as time-dependent thermoelastic flexural motions (of course such time dependence is governed by the Fourier number due to diffusivity and hence much slower than vibrations).
-
#2975 by Stormbringer on 02 Jun, 2016 16:45
-
battery powered/solar powered portable microwave the size of a thermos bottle.
http://nextbigfuture.com/2016/05/solid-state-rf-technology-enables-cheap.html
This may be of interest to future DIYer EM drive projects -
#2976 by SeeShells on 02 Jun, 2016 16:48
-
After 4 iterations of designing the torsion pendulum I've settled on this aluminum carbon fiber very lightweight platform.
This is a test wire and a little heavier for just testing.
You'll notice the threaded rods on the sides that will be used to beam ballance and the wire goes through the center of a small hole. The wire will be adjusted with weights in the Z and Y dimension on the platform to be in the center of the hole. This assures beam balance.
Underneath the U channel is where the wire is captured 1/2 way between the top and bottom and it's centered to the middle section of the carbon fiber tubes.
The wire tension is set by a guitar Tuner Tuning Peg Machine Head.
This setup even with the heavy wire is extraordinary sensitive and when I finish off the brackets for the frustum and sliding balance weight on the carbon fiber tubes I'll be finishing off the electronics to drive my frustums.
Just a update, back to work.
Shell -
#2977 by Monomorphic on 02 Jun, 2016 16:55
-
battery powered/solar powered portable microwave the size of a thermos bottle.
http://nextbigfuture.com/2016/05/solid-state-rf-technology-enables-cheap.html
This may be of interest to future DIYer EM drive projects
Of particular interest is:
"The technology platform includes NXP’s MKW40Z Kinetis microcontroller (MCU) with a 2.4 GHz signal generator, a MMA25312 pre-driver, MHT1008 driver and MHT1004 final stage amplifier into a single, integrated closed-loop module."
The MKW40Z Kinetis microcontroller comes in a BLE and wireless radio version. If they plan on making this work with your iphone... that would be even better. -
#2978 by Gilbertdrive on 02 Jun, 2016 17:02
-
..
It may not make sense to you, but it makes sense to me.
Dr. Rodal, if that's the case in the coaxial cable delivering the momenta in the Cannae drive (I can see it making sense
) Then the case for powering a loop probe antenna into the sidewall of the cavity or the case of a waveguide into the sidewalls doesn't make sense as the delivered momenta and force is 90o to the directions of thrusts.
Shell
It is incorrect to model the EM Drive powered by a coaxial cable ignoring the effect of the tapered frustum as to the direction of the expressed force. It has to satisfy the equations of conservation of energy and momentum.
In the case of the Cannae drive this problem is simplified because the coaxial cable is directed along the same axis as the axis of axisymmetry of the Cannae drive.
The electromagnetic momentum and force are delivered as per direction of the coaxial cable (I previously posted a paper by Kirk McDonald from Princeton proving this point). Once the electromagnetic momentum is inside a tapered frustum of a cone, the direction of the electromagnetic momentum will abide by the conservation of energy and conservation of momenta in the whole system, as per derivative with respect to time of the energy density and divergence of the Poynting vector in the conservation of energy, and as per derivative with respect to time of the Poynting vector and divergence of the stress tensor in the conservation of momentum.
The direction of the force is a result of the equations of motion of the system, as in any dynamic system.
==> I don't know what will happen when one runs with batteries, it will be exciting to get positive results as Hackaday Aachen team self-assesses in their report. To this date, I think that the best experiments are the ones run by NASA. Look forward to NASA conducting battery-powered tests.
I may have a suggestion for all the DIYERS that are already testing their builds with external power supplies. When your tests are completed, if you want to improve the results, you can use a second torsional pendulum, with on it batteries and inverter, and the power cable between the two items. So you do not have to modify your original build, and you do not overcharge your piano wires. You could be able to mesure separately the thrust on the Emdrive, and the trust on the power package.
The problems of the batteries heating, and inverter cooling is not so big. What is important it that it must not be too noisy. The variation of the thrust from the power package has to be an order of magnitude below the thrust of the Emdrive, so that the calculated thrust is still valuable. It maybe a way of testing a battery powered emdrive without modifying the entire setup. I propose it, but I am not sure that there is no leak in this idea. If anybody sees a bias that would occur if separate torsional pendulum are used for the power package, and for the Emdrive, do not hesitate
-
#2979 by Monomorphic on 02 Jun, 2016 17:05
-
Shell,
What hardware are you using to thread the wire through? I don't need to pass through completely like you, but I need something better to grab onto the wire. Please let me know.
