-
#140 by X_RaY on 02 Dec, 2016 13:49
-
I went back to your original model, solver and mesh configuration.Currently I'm solving monomorphics model with spherical end plates with a modified loop antenna, i.e. position and diameter. The current result is shown below. It shows that using the related dimensions leads to a coupling factor below 1.
Can you please double check the dimensions for the antenna? A radius of 15mm seems large and i'm still showing over coupled. Did you mean a diameter of 15mm? That seems more in line with EW's 13.5mm diameter loop antenna.
I found the same result for the suggested antenna position as you have shown above (upper pic). Till now I'm not sure which of the modifications i did lead to the huge different results as shown.
For further consistency I will use the settings you have send with the model.
What I found is that the overcoupling problem can be solved as suggested before: move the antenna closer to the big end. Using your config, the z-position should be at ~1cm. (lower pic)
Maybe a finer mesh at/near the antenna is needed so close to the wall?
-
#141 by X_RaY on 02 Dec, 2016 18:01
-
I don't change anything else but the antenna position, the configuration is still the one, James has send to me.
I tink thats good enough to go with. In a real world construction due to imperfections, in any case the exact antenna position has to be tuned/verified with a VNA. The basic system impedance will almost never be exact 50Ω and so on...
Nevertheless there will be only a small impedance mismatch that has to tuned out with an external tuner, this minimize the losses that is produced within the tuner (compared to the situation where the impedance difference is much grater).
From a construction viewpoint the position is almost ideal because there are no long wires necessary to feed the loop antenna. That configuration minimize the excitation of other modes inside the frequency range. TE01p is degenerate with TM11p (at least in a cylindrical cavity almost exact at the same frequency, while the conically shape seperates both modes, but not that much regarding to the eigen-frequency)
-
#142 by Monomorphic on 03 Dec, 2016 13:52
-
I don't change anything else but the antenna position, the configuration is still the one, James has send to me.
I tink thats good enough to go with. In a real world construction due to imperfections, in any case the exact antenna position has to be tuned/verified with a VNA. The basic system impedance will almost never be exact 50Ω and so on...
Nevertheless there will be only a small impedance mismatch that has to tuned out with an external tuner, this minimize the losses that is produced within the tuner (compared to the situation where the impedance difference is much grater).
From a construction viewpoint the position is almost ideal because there are no long wires necessary to feed the loop antenna. That configuration minimize the excitation of other modes inside the frequency range. TE01p is degenerate with TM11p (at least in a cylindrical cavity almost exact at the same frequency, while the conically shape seperates both modes, but not that much regarding to the eigen-frequency)
With the antenna in that configuration I get a huge -39dB reflection coeficient with a Q factor of 111,454!
-
#143 by Monomorphic on 03 Dec, 2016 14:38
-
Q sounds a little bit high, how is it calculated? f/df (-3dB down from the baseline for S11)?I don't change anything else but the antenna position, the configuration is still the one, James has send to me.
I tink thats good enough to go with. In a real world construction due to imperfections, in any case the exact antenna position has to be tuned/verified with a VNA. The basic system impedance will almost never be exact 50Ω and so on...
Nevertheless there will be only a small impedance mismatch that has to tuned out with an external tuner, this minimize the losses that is produced within the tuner (compared to the situation where the impedance difference is much grater).
From a construction viewpoint the position is almost ideal because there are no long wires necessary to feed the loop antenna. That configuration minimize the excitation of other modes inside the frequency range. TE01p is degenerate with TM11p (at least in a cylindrical cavity almost exact at the same frequency, while the conically shape seperates both modes, but not that much regarding to the eigen-frequency)
With the antenna in that configuration I get a huge -39dB reflection coeficient with a Q factor of 111,454!
Yes, I used the -3dB (fc/Δf) method. According to the chart that's 2.40565Ghz/2.15841e-5Ghz
There are tools under the Measure tab for setting markers and getting exact frequencies.
-
#144 by Peter Lauwer on 04 Dec, 2016 12:11
-
Thanks X_RaY and Monomorphic. These sims may prove to be vital to understanding the EMDrive behavior.
I plan to test these loops (along the central axis) in a cylindrical cavity first.
-
#145 by Monomorphic on 04 Dec, 2016 13:20
-
X_Ray, have you had any luck with doing a Time Domain Analysis in the 2.4Ghz range? I need to solve for fmin and fmax. Basically we need to figure out how to make the simulated frequency range equal to the time signal bandwidth. In my case that is 2.4054Ghz and 2.4059Ghz.
-
#146 by X_RaY on 04 Dec, 2016 16:31
-
X_Ray, have you had any luck with doing a Time Domain Analysis in the 2.4Ghz range? I need to solve for fmin and fmax. Basically we need to figure out how to make the simulated frequency range equal to the time signal bandwidth. In my case that is 2.4054Ghz and 2.4059Ghz.
To me it looks like a stand alone tool to analyze spectra of a predefined pulse only.
The resulting graphs (mag,phase,real,imag) can be added to cartesian plots.Till now I couldn't find a way to apply it to the data.
Still learning
..second pic is a huge gif-file
http://www.altairuniversity.com/wp-content/uploads/2015/03/UserManual.pdf
section 9.9
-
#147 by Monomorphic on 04 Dec, 2016 22:11
-
Nice! Similar to some results I had a while back. Notice the counting in ns. This is not TE013, but another mode.
-
#148 by WarpTech on 07 Dec, 2016 02:46
-
In a PM with @zellerium, we were discussing the software he uses and I started to think about why I wanted it. It just dawned on me that this would be much faster as a team effort. So I would like to make this proposal to those who have modeling capability. We need organization. Dr. Rodal's recent paper shows us how detailed we need to be in every aspect of this analysis, in order to find the right answers. The following list is the "data" that needs to be modeled and recorded. Starting with a frustum. I propose we use TT's design with the spherical end caps, or model Shawyer's latest design, but I'm open to suggestions. The point is, we choose 1 model, and model it to death, to get all the relevant data into a report. I do not mind preparing that report if the modelers can do the real work.
This list for data is based on much of what we've been doing here already, as much as on my own wish list. It is open for discussion and starts like this;
1. Optimal location for the antenna to excite TE013 and have ~50 Ohm input Z.
2. Optimal length, width, shape of the antenna for 50 Ohms.
3. Optimal location for probes to identify frequency and decay time at big end, small end and center wall. i.e., determine Q from the w*t and do VNA at each port.
4. Relative E, H and A, vector field strengths for different materials, PC, Ag, Cu, Al, etc. as a color plot, values & vectors. Everything consistent so they can be compared.
5. Relative Energy Density for different materials, PC, Ag, Cu, Al, etc. as a color plot and values.
6. Relative surface power dissipation for each material, color plot and values.
7. Relative temperature for each material, color plot.
I would put this together into a report or Smartsheet. http://www.smartsheet.com/for each Frustum shape/design/frequency mode. The key to me is to be consistent and thorough with each design, and repeat this for each mode shape and material. It's a lot of work to be thorough!
Dr. Rodal has done real research reports on the truncated cavity last year and now on the Mach effects in the MEGA drive. This type of research takes a lot of time and effort and needs to be coordinated and documented. That's why I want the software so I can do what is needed to advance the cause, but if we work as a team, we can do it together, under budget and ahead of schedule.
Todd
-
#149 by Peter Lauwer on 07 Dec, 2016 13:59
-
In a PM with @zellerium, we were discussing the software he uses and I started to think about why I wanted it. It just dawned on me that this would be much faster as a team effort. So I would like to make this proposal to those who have modeling capability. We need organization. Dr. Rodal's recent paper shows us how detailed we need to be in every aspect of this analysis, in order to find the right answers. The following list is the "data" that needs to be modeled and recorded. Starting with a frustum. I propose we use TT's design with the spherical end caps, or model Shawyer's latest design, but I'm open to suggestions. The point is, we choose 1 model, and model it to death, to get all the relevant data into a report. I do not mind preparing that report if the modelers can do the real work.
This list for data is based on much of what we've been doing here already, as much as on my own wish list. It is open for discussion and starts like this;
1. Optimal location for the antenna to excite TE013 and have ~50 Ohm input Z.
2. Optimal length, width, shape of the antenna for 50 Ohms.
3. Optimal location for probes to identify frequency and decay time at big end, small end and center wall. i.e., determine Q from the w*t and do VNA at each port.
4. Relative E, H and A, vector field strengths for different materials, PC, Ag, Cu, Al, etc. as a color plot, values & vectors. Everything consistent so they can be compared.
5. Relative Energy Density for different materials, PC, Ag, Cu, Al, etc. as a color plot and values.
6. Relative surface power dissipation for each material, color plot and values.
7. Relative temperature for each material, color plot.
I would put this together into a report or Smartsheet. http://www.smartsheet.com/for each Frustum shape/design/frequency mode. The key to me is to be consistent and thorough with each design, and repeat this for each mode shape and material. It's a lot of work to be thorough!
Dr. Rodal has done real research reports on the truncated cavity last year and now on the Mach effects in the MEGA drive. This type of research takes a lot of time and effort and needs to be coordinated and documented. That's why I want the software so I can do what is needed to advance the cause, but if we work as a team, we can do it together, under budget and ahead of schedule.
Todd
If I can contribute with a few experiments, I'd be happy to do so. I plan to measure with a central loop, as you simulated, but first in a cylindrical cavity. And at a few locations. I have to make a new loop for every location, so the number will be limited.
Material: semi-rigid (RG402), feed-through in the endplate with the sma bus as in the attached picture. The loop itself can be made from the inner conductor of the semi-rigid or from silvered Cu wire.
Peter
-
#150 by Testingflight on 10 Dec, 2016 15:49
-
Has anyone considered that the thrust from the emdrive found in recent experiments might result from the Coriolis Effect. I have not researched this type of drive and only came across articles about this topic by accident so forgive me if this has been suggested already. In this type drive the microwaves would probably setup vibrations in the drive casing. This vibration might result in an effective force on the support structure as the Earth rotates. This might be interpreted as thrust. The Coriolis Effect is used in vibrating structure gyroscopes and mass flow meters. I haven't thought this out in detail but it seems a simple explanation given the small thrust. It could be tested by changing the drive orientation in relation to the Earth's rotation and by measuring and varying the casing vibration. I really hope this type of drive is real and physics are found to it explain it's operation bUT it seems a reach.
-
#151 by WarpTech on 10 Dec, 2016 17:09
-
In a PM with @zellerium, we were discussing the software he uses and I started to think about why I wanted it. It just dawned on me that this would be much faster as a team effort. So I would like to make this proposal to those who have modeling capability. We need organization. Dr. Rodal's recent paper shows us how detailed we need to be in every aspect of this analysis, in order to find the right answers. The following list is the "data" that needs to be modeled and recorded. Starting with a frustum. I propose we use TT's design with the spherical end caps, or model Shawyer's latest design, but I'm open to suggestions. The point is, we choose 1 model, and model it to death, to get all the relevant data into a report. I do not mind preparing that report if the modelers can do the real work.
This list for data is based on much of what we've been doing here already, as much as on my own wish list. It is open for discussion and starts like this;
1. Optimal location for the antenna to excite TE013 and have ~50 Ohm input Z.
2. Optimal length, width, shape of the antenna for 50 Ohms.
3. Optimal location for probes to identify frequency and decay time at big end, small end and center wall. i.e., determine Q from the w*t and do VNA at each port.
4. Relative E, H and A, vector field strengths for different materials, PC, Ag, Cu, Al, etc. as a color plot, values & vectors. Everything consistent so they can be compared.
5. Relative Energy Density for different materials, PC, Ag, Cu, Al, etc. as a color plot and values.
6. Relative surface power dissipation for each material, color plot and values.
7. Relative temperature for each material, color plot.
I would put this together into a report or Smartsheet. http://www.smartsheet.com/for each Frustum shape/design/frequency mode. The key to me is to be consistent and thorough with each design, and repeat this for each mode shape and material. It's a lot of work to be thorough!
Dr. Rodal has done real research reports on the truncated cavity last year and now on the Mach effects in the MEGA drive. This type of research takes a lot of time and effort and needs to be coordinated and documented. That's why I want the software so I can do what is needed to advance the cause, but if we work as a team, we can do it together, under budget and ahead of schedule.
Todd
Which of these platforms would you prefer we use to store data?
Google Drive
Dropbox.com
Smartsheet.com
emdrive.wiki
For me, Smartsheet would be the easiest to organize and maintain. Dropbox would be my second choice. The wiki is my last choice because it's difficult to edit equations and such in HTML.
-
#152 by X_RaY on 10 Dec, 2016 17:29
-
Has anyone considered that the thrust from the emdrive found in recent experiments might result from the Coriolis Effect. I have not researched this type of drive and only came across articles about this topic by accident so forgive me if this has been suggested already. In this type drive the microwaves would probably setup vibrations in the drive casing. This vibration might result in an effective force on the support structure as the Earth rotates. This might be interpreted as thrust. The Coriolis Effect is used in vibrating structure gyroscopes and mass flow meters. I haven't thought this out in detail but it seems a simple explanation given the small thrust. It could be tested by changing the drive orientation in relation to the Earth's rotation and by measuring and varying the casing vibration. I really hope this type of drive is real and physics are found to it explain it's operation bUT it seems a reach.
That's one of several reasons why the experiments are done in different directions.
This force would present also when the microwave generator is switched off.
Fc=(4•pi•m•v•siny)/T
-
#153 by Testingflight on 10 Dec, 2016 22:40
-
I didn't explain my thought well. When the drive is active the drive case or support structure might vibrate.The vibrations induced in the drive casing might be sinusoidal. If so the rotation of the Earth would cause a separate force on the casing in addition to the normal Coriolis effect. Vibrating structure gyroscopes and mass flow meters use this principle in different ways. The effect would be small and dependent more on the vibration frequency than power input. The casing would have specific frequency at which it resonated so the effect might not be greater at more power. This additional Coriolis effect would only occur when the system is active. To calculate the effect you would first need to know the vibration frequency of the drive casing if it is, in fact, vibrating steadily. To calculate the extra Coriolis effect other system specific information would also be needed. I am not sure the necessary data would even be recorded in the tests. I certainly don't have the expertise to calculate the potential force from this Coriolis effect in a vibrating system but NASA should be able to if they think this effect even possible in the test system. Thanks.
-
#154 by Peter Lauwer on 14 Dec, 2016 13:49
-
Nice! Similar to some results I had a while back. Notice the counting in ns. This is not TE013, but another mode.
Hi Monomorphic and X_RaY,
Why don't you summarize your results in a pdf? It will be better preserved and it will be easier to refer to it.
Btw, was there a consensus about the best position for the loop?
regards, Peter -
#155 by X_RaY on 16 Dec, 2016 20:20
-
Due to a pleasure of FattyLumpkin, I did some short calculations on his suggested frustum dimensions (actually it had taken ~2 full days of CPU time using my old weak PC*). This is not the max possible mesh size, therefore the frequence should be a little lower in real world, but it should be enough for a first shot.
I used a loop wire near the big end as an antenna (dimensions are also attached in the pic, units are in mm!)
In fact the suggested dimensions is the Brady cone with dielectric HDPE dics but scaled to reach TE012 at ~1GHz (for the exact dimensions, please take a look to the attached pic).The used antenna position is still sub-optional with a linear reflection coefficient of ~0.18. Frustum material is Cu.
FattyL please note that your suggested, scaled values are correct.
But the problem is still the same as described by EW(...) i.e. there are other modes very, very close to the field pattern you would like to excite. This can lead to problems.
Best Regards
Christian
*I will catch a new one next time to reduce the huge amount of time that's needed to get some results
EDIT:
pic was updated an corrected
-
#156 by X_RaY on 26 Dec, 2016 17:58
-
Quote from: copy from emdrive thread 9
Chen Yue's RF resonant cavity thruster may be not a frustum, but a semicylinder!
Short calc results using the given dimensions, source was a electric dipole.
Excerpts from "An electromagnetic propulsion system and method" CN 105947224 A
• English translation
• Original PDF in Chinese with picturesQuote from: Chen YueBRIEF DESCRIPTION
[0029] FIG. 1 a typical electromagnetic propulsion resonator body diagram;
[0030] FIG. 2 (a) is a typical view of the XY plane electromagnetic propulsion of resonance cavity, (b) is a typical view of the XZ plane electromagnetic propulsion of resonance cavity, (c) is a typical resonant cavity electromagnetic propulsion YZ plane view, (d) is a typical perspective view of the electromagnetic propulsion of resonance cavity;
[0031] FIG. 3 electromagnetic waves in the resonant cavity of the uneven distribution diagram;
[0032] FIG. 4 block diagram of an electromagnetic propulsion program.
detailed description
[0038] The electromagnetic propulsion power module input signal is less than equal to the maximum power capacity of electromagnetic propulsion module to obtain electromagnetic propulsion module to work required thrust power ratio; the electromagnetic propulsion module frequency of the input signal to the electromagnetic propulsion module the resonant frequency is preferably in the center of the 3dB bandwidth of the electromagnetic propulsion module to work properly.
[0039] The electromagnetic propulsion module includes a resonant cavity inside an asymmetric structure, the use of electromagnetic propulsion module inside the resonator cavity asymmetric structure, produce uneven microwave radiation pressure, and then in the resonant cavity be unbalanced electromagnetic force to external output thrust. Asymmetric structure is preferable to adopt a resonant cavity electromagnetic propulsion resonant cavity, electromagnetic propulsion system around the resonant cavity electromagnetic propulsion structures, electromagnetic propulsion resonant cavity shown in FIG. 1, respectively. As can be seen, electromagnetic propulsion resonant cavity is divided into four faces: plane Sa, surface Sb, plane Sc, plane Sd, as shown in FIG. Sa mounted on a plane microwave power input device on a plane Sc plane fitted with microwave power extraction apparatus. Input to the feedback power control module from the microwave power extraction means to extract microwave power as a feedback power. Electromagnetic propulsion thrust output of the resonant cavity Preferred conditions: input microwave power frequency electromagnetic propulsion within 3dB bandwidth of the center frequency of the resonant cavity. Under the operating conditions of the effect of microwave power, electromagnetic propulsion resonant cavity can be unbalanced microwave radiation pressure, and thus in the resonant cavity be unbalanced electromagnetic force, thrust externally output, as shown in FIG. Select the lowest electromagnetic propulsion mode resonator center frequency f〇 electromagnetic propulsion system frequency. Semicylindrical cavity of the radius R (meters), length L (in meters) and preferably the relationship between the center frequency F0 (Unit GHz) between the semi-cylindrical cavity of:
[0040]
⑴
[0041] Preferably R = 86 mm, L = 117.7 mm, using the formula (1) f0 were solver to 2.45 GHz.
...
-
#157 by X_RaY on 01 Jan, 2017 22:24
-
Added for quick find reasons:
----------------------------------------------
Chinese periodic cavity structure:
https://forum.nasaspaceflight.com/index.php?topic=41732.msg1624391#msg1624391
http://forum.nasaspaceflight.com/index.php?topic=42978.msg1721229#msg1721229
----------------------------------------------
Q seems to depends on so called cutoff rule (for circular WG):
https://forum.nasaspaceflight.com/index.php?topic=41732.msg1624968#msg1624968
https://forum.nasaspaceflight.com/index.php?topic=41732.msg1626451#msg1626451
https://forum.nasaspaceflight.com/index.php?topic=41732.msg1626572#msg1626572
https://forum.nasaspaceflight.com/index.php?topic=41732.msg1628948#msg1628948 with EMPro -
#158 by X_RaY on 12 Jan, 2017 21:02
-
The quote attached is a copy of my post in the EMDrive main thread #9.
The simulations are related to the discussion of the {Q}uality -factor when the small end of a truncated conical cavity resonator is below, at or above the cutoff diameter of a cylindrically waveguide.
To use the related value is frequently stated by TheTraveller(TT) (he says he quotes Shawyer in this regard) and have suggested to take it as a rule, i.e. to make the small diameter equal to the cutoff diameter.
Several explanations where given by TT but nothing conclusive.
For example, that there is no reflection at all if the small end plate diameter is below this cutoff rule. This was debunked as nonsense by Dr.Rodal and others.
As one of lastest "explanations" TT stated that the Q for such a cavity is much smaller ( or even: "...below anything usefull..") than for a cavity that fits the so called cutoff rule.
The results shown here debunk this to be nonsense also.
Please note that this tells nothing about differences related to (possible) thrust generation.
https://forum.nasaspaceflight.com/index.php?topic=41732.msg1626572#msg1626572
Can EmPro calculate Q?
Again, at the moment I don't believe in this Q values based on calculations with HOBF. I get freaky inconclusive results when using it.As you can see in the first two results the situation differs from calculation to calculate. The reason is slightly different mesh size and coupling factors.
Don't forget to include the spherical end-plate frustum with Q of 111,454. That is a pretty significant increase.
EDIT
This is the same frustum as used for the Q compare but using HOBF and fine mesh. Instead of natural possible QL~36000, i get 207000 loaded Q!
If yes can you check the results with EmPro?
FEKO SE calculation is above, EMPro (FEM Eigenresonance solver) below.
Dimensions are the same, material is copper in both cases. EMPro results should show Q0.
-
#159 by WarpTech on 13 Jan, 2017 17:54
-
The quote attached is a copy of my post in the EMDrive main thread #9.
The simulations are related to the discussion of the {Q}uality -factor when the small end of a truncated conical cayity resonator is below, at or above the cutoff diameter of a cylindrically waveguide.
To use the related value is frequently stated by TheTraveller(TT) (he says he quotes Shawyer in this regard) and have suggested to take it as a rule, i.e. to make the small diameter equal to the cutoff diameter.
Several explanations where given by TT but nothing conclusive.
For example, that there is no reflection at all if the small end plate diameter is below this cutoff rule. This was debunked as nonsense by Dr.Rodal and others.
As one of lastest "explanations" TT stated that the Q for such a cavity is much smaller ( or even: "...below anything usefull..") than for a cavity that fits the so called cutoff rule.
The results shown here debunk this to be nonsense also.
Please note that this tells nothing about differences related to (possible) thrust generation.
https://forum.nasaspaceflight.com/index.php?topic=41732.msg1626572#msg1626572
Can EmPro calculate Q?
Again, at the moment I don't believe in this Q values based on calculations with HOBF. I get freaky inconclusive results when using it.As you can see in the first two results the situation differs from calculation to calculate. The reason is slightly different mesh size and coupling factors.
Don't forget to include the spherical end-plate frustum with Q of 111,454. That is a pretty significant increase.
EDIT
This is the same frustum as used for the Q compare but using HOBF and fine mesh. Instead of natural possible QL~36000, i get 207000 loaded Q!
If yes can you check the results with EmPro?
FEKO SE calculation is above, EMPro (FEM Eigenresonance solver) below.
Dimensions are the same, material is copper in both cases. EMPro results should show Q0.
Can you lower the frequency of the one on the right to 2.45 GHz, by expanding the Big diameter? I need to know what those dimensions would be. 2.6 GHz is too far outside the amplifier's range.
