-
#1920 by ThinkerX on 18 Jan, 2016 23:55
-
Quote
Its hard for me to judge, but IF a delta energy level change equates to a delta mass change per Einstein, Doc is right, C2 is a huge denominator to overcome UNLESS you have additive, rapid, almost chaotic high energy level changes from multiple sources of GHz photons (a spectral spray so to speak).
The thing I get out of this is a single, CW shot of power only changes mass once (as long as the CW signal is there). Seems like the delta E on/off forces the delta M changes. Like I've said many times, theoretical physics is not my strong suit, so I can only rely on the basic fundamentals.
This being the case, it is not violating physical laws as we know them today.
Maybe the magnetrons in the EM drive get 'close enough' to meeting these conditions, at least for a short period of time?
Would this produce 'thrust' anywhere near the levels reported? -
#1921 by Rodal on 19 Jan, 2016 00:39
-
I snagged this image from the reference:
Please take into account that the internal height is adjusted internally with a screw prior to testing, so the internal height at resonance can be much smaller than the actual exterior height of the cavity
https://tu-dresden.de/die_tu_dresden/fakultaeten/fakultaet_maschinenwesen/ilr/rfs/forschung/folder.2007-08-21.5231434330/ag_raumfahrtantriebe/JPC%20-%20Direct%20Thrust%20Measurements%20of%20an%20EM%20Drive%20and%20Evaluation%20of%20Possible%20Side-Effects.pdf
and counted pixels using a tool new to me, so the count was not very repeatable, but these are the numbers:
Small diameter = 134 px
big diameter = 184 - 187 px
slant height = 142 - 147 px
Using small diameter, Ds = 77 mm and calculating by ratio, gives big diameter ~ 105.7 mm to 107.5 mm and slant height ~ 81.6 mm to 84.5 mm
I don't know what that proves, but as a sanity check it doesn't seem all that close to anyone's numbers.
Yes, consider that it can be shorter. But it can't be longer which kind of rules out axialLength = 2*(0.0686 - 0.003) meter (since the wall thickness is 3 mm)
Which means that
The dimensions I calculated a long time ago here: https://forum.nasaspaceflight.com/index.php?topic=39004.msg1477474#msg1477474
Axial Length = 0.100842 m = 0.735*2*0.0686 m
are very much in play !
Welll... The next thing I did was to snag the COMSOL image from the reference, attached. It seems to me that this drawing should be to scale and should show the interior dimensions. Using the pixel measuring method, I find:
px px ref. 77 77
sd 244 241.9 77 77 ave
bd 341.1 339.2 107.6422131148 107.0426229508 107.3424180328
shi 211.4 206.8 66.712295082 65.2606557377 65.9864754098
shi 212.5 208.1 67.0594262295 65.6709016393 66.3651639344
66.1758196721
That is, with small diameter = 0.77 mm, by ratio the big diameter = 107.34 mm and the height = 66.176 mm. But at least that is close to one of the dimensions given previously.
OK, hold your horses
Your persistence in looking at these dimensions was a good motivator to see what could be going on.
When I looked at the height being 0.0686 m instead of 2*0.0686 I was looking at mode TE111 instead of TM010 (see https://forum.nasaspaceflight.com/index.php?topic=37642.msg1410336#msg1410336).
It turns out that TM010 is fairly insensitive to height (which makes sense because it is the analog of the mode for a cylinder that is constant in the height direction).
Bottom line: the natural frequency does not change much for TM010 whether the height is 2*0.0686m or 0.0686 m. It is around the magnetron frequency (2.45 GHz) even if the height is changed by a factor of 2!!
Which makes the Tajmar EM Drive all the more weird, because if this is the mode he excited, what was the point of the screw-driven mechanism then ?
Details coming maybe tomorrow.
Making slow progress.
See chart for frequency (GHz) for mode shape TM010 vs. height for Tajmar's EM Drive.
Basically I conclude that aero was right and that the adjusted height was most likely ~ 0.06 meters
but it could be anywhere in this range 0.04m<height<0.08m
bigRadius = 0.0541 meters
smallRadius = 0.0385 meters
height = 0.06 meters
More details to come...
-
#1922 by rfmwguy on 19 Jan, 2016 00:50
-
I am not sure. Hoping doc or someone might plug in some numbers but unless we can assume multiple carriers at wildly changing dynamics I'm afraid a few micronewtons are out of the question. I still cannot find any experimental science that is close to the conditions we are setting up...much conjecture but no measurements of imparted kinetic energy into a closed cavity using moderately high ghz photon radiation levels. Seems no one considered it worthy of investigation...until lately.QuoteIts hard for me to judge, but IF a delta energy level change equates to a delta mass change per Einstein, Doc is right, C2 is a huge denominator to overcome UNLESS you have additive, rapid, almost chaotic high energy level changes from multiple sources of GHz photons (a spectral spray so to speak).
The thing I get out of this is a single, CW shot of power only changes mass once (as long as the CW signal is there). Seems like the delta E on/off forces the delta M changes. Like I've said many times, theoretical physics is not my strong suit, so I can only rely on the basic fundamentals.
This being the case, it is not violating physical laws as we know them today.
Maybe the magnetrons in the EM drive get 'close enough' to meeting these conditions, at least for a short period of time?
Would this produce 'thrust' anywhere near the levels reported? -
#1923 by rfmwguy on 19 Jan, 2016 01:08
-
Just received the following communication from Australia. I believe he could use some emdrive design help. See attachment for his contact info:
I am attaching the formal paper written by John Newell from Australia. Thanks for the heads-up John!
"Dave,
just a note to thank you for your report dated 10/2015 describing your thruster experiments. I am attempting to produce the effect using a 60GHz diode resonator but lack the experience required to design the waveguide and signal insertion.
Please let me know if anyone is interested in recommending design ideas.
Meantime, good luck with further experiments. What a useful method of propulsion this could be.
Yours John Newell..
N.B: An attempt to explain the mechanism of action is attached"
-
#1924 by Vultur on 19 Jan, 2016 01:56
-
So the intention was not to detonate the nuclear bombs in the Earth's atmosphere.
Unfortunately, the EMP would still happen even in space. This is a much more serious problem (really though not politically) than the radioisotopes in the atmosphere - there would be some tiny increase in cancer risk (if one accepts the linear-no-threshold model) but it would be enormously outweighed by the moving of industries into space (thus less conventional pollution) that Orion-class ultra-heavy-lift capabilities would allow for. -
#1925 by RFPlumber on 19 Jan, 2016 01:57
-
Frustum test on 01/17/2016.
Frustum testing with 28-29W of RF power at TE012 mode (2331 MHz). Frustum dimensions have been modified since the first test (http://forum.nasaspaceflight.com/index.php?topic=39004.msg1472667#msg1472667) to address the “cut-off” condition at the lower end.
Instead of going formal about it and throwing pages of data on you, I decided to make it interactive…
(But unfortunately I cannot make NSF display decent sized pictures from external links, so the post will be pointing to attachments...)
This is what I would expect to see for any abnormal force from applied RF power (Attachment1):
Yet this is what the experiment is showing(Attachment2):
At this point I am pretty convinced that if those 29W of RF power are resulting in any non-zero “thrust” then that thrust is most certainly less than 50 uN (!). Why? Because of this (Attachment3):
So much for Shawyer theory predicting hundreds of uN. Anticipating concerns that thrust is not there because of poorly constructed cavity, low Q, RF leak, no bolts, extra scratches, etc. the answer is – yes, it is possible the thrust is not there for any one of these reasons. However, the guiding theory for this experiment does not require bolts, or demands no RF leaks. All it asks for is Q factor, Design factor and small end cut-off condition. These are all present. Yet there is no thrust. Sorry.
The rest is boring stuff.
Total of 14 runs have been performed. All data is available and shared in this folder:
https://drive.google.com/open?id=0B3jbXEyEMvU8RmZGNk9pVF9GRk0
There is a ReadMe there describing the test as well as the Excel spreadsheet with summary.
A number of qualification runs have been done this time, including 2 long idle tests (runs 0 and 5), 4 runs with the high voltage pulse overlapping on the RF pulse (3,4,8,9), 2 HV-only passes with increased distance (hence lower test force) between electrostatic plates (10,11) and 2 runs with the same reduced electrostatic test force overlapping on the RF pulse (12,13). From all these qualification runs I developed the following humble understanding for the expected resolution of this test system:
-Anything above 200 uN is nearly impossible to miss despite an occasional noisy run.
- A force between 100-200 uN will most likely be detected either directly from mid-point charts or certainly with basic statistics across a few runs.
-A force between 50-100 uN may be detected under “optimal conditions”.
-Anything under 50 uN is unlikely to be detected or noticed.
For example, the last 2 runs here were done with only 160 uN of test force (Attachment 4):
Issues
RF Leak. Discussed here: http://forum.nasaspaceflight.com/index.php?topic=39004.msg1479442#msg1479442
Degradation of coax coupler when changing frustum orientation from East to West. New S11 scan (Attachment 5):
With the original coupler one could get reflected power down to 0.3W, growing to about 1.7W. With coupler degraded the reflected power was always around 2.4W. The Q factor did not appear to change much.
Time to start looking for some HDPE disks?
-
#1926 by Rodal on 19 Jan, 2016 02:01
-
Frustum test on 01/17/2016.
Frustum testing with 28-29W of RF power at TE012 mode (2331 MHz). Frustum dimensions have been modified since the first test (http://forum.nasaspaceflight.com/index.php?topic=39004.msg1472667#msg1472667) to address the “cut-off” condition at the lower end.
Instead of going formal about it and throwing pages of data on you, I decided to make it interactive…
(But unfortunately I cannot make NSF display decent sized pictures from external links, so the post will be pointing to attachments...)
This is what I would expect to see for any abnormal force from applied RF power (Attachment1):
Yet this is what the experiment is showing(Attachment2):
At this point I am pretty convinced that if those 29W of RF power are resulting in any non-zero “thrust” then that thrust is most certainly less than 50 uN (!). Why? Because of this (Attachment3):
So much for Shawyer theory predicting hundreds of uN. Anticipating concerns that thrust is not there because of poorly constructed cavity, low Q, RF leak, no bolts, extra scratches, etc. the answer is – yes, it is possible the thrust is not there for any one of these reasons. However, the guiding theory for this experiment does not require bolts, or demands no RF leaks. All it asks for is Q factor, Design factor and small end cut-off condition. These are all present. Yet there is no thrust. Sorry.
The rest is boring stuff.
Total of 14 runs have been performed. All data is available and shared in this folder:
https://drive.google.com/open?id=0B3jbXEyEMvU8RmZGNk9pVF9GRk0
There is a ReadMe there describing the test as well as the Excel spreadsheet with summary.
A number of qualification runs have been done this time, including 2 long idle tests (runs 0 and 5), 4 runs with the high voltage pulse overlapping on the RF pulse (3,4,8,9), 2 HV-only passes with increased distance (hence lower test force) between electrostatic plates (10,11) and 2 runs with the same reduced electrostatic test force overlapping on the RF pulse (12,13). From all these qualification runs I developed the following humble understanding for the expected resolution of this test system:
-Anything above 200 uN is nearly impossible to miss despite an occasional noisy run.
- A force between 100-200 uN will most likely be detected either directly from mid-point charts or certainly with basic statistics across a few runs.
-A force between 50-100 uN may be detected under “optimal conditions”.
-Anything under 50 uN is unlikely to be detected or noticed.
For example, the last 2 runs here were done with only 160 uN of test force (Attachment 4):
Issues
RF Leak. Discussed here: http://forum.nasaspaceflight.com/index.php?topic=39004.msg1479442#msg1479442
Degradation of coax coupler when changing frustum orientation from East to West. New S11 scan (Attachment 5):
With the original coupler one could get reflected power down to 0.3W, growing to about 1.7W. With coupler degraded the reflected power was always around 2.4W. The Q factor did not appear to change much.
Time to start looking for some HDPE disks?
CONGRATULATIONS !!!!
A null result for Shawyer's strange hypothesis that a closed cavity like an EM Drive would have the same cut-off condition as an open waveguide, and a null result for Shawyer's "theory" predicting a larger force.
VERY IMPRESSIVE and scientifically conducted. !!
You should publish this work, or at least present it at the AIAA Propulsion conference.
The nullification of a Woodward-type MLT thruster paper by Brito, Marini and Galian was first presented at an AIAA conference and then later chosen for publication in an AIAA peer-reviewed journal.
45th AIAA/ASME/SAE/ASEE Joint Propulsion Conference & Exhibit
2 - 5 August 2009, Denver, Colorado
AIAA 2009-5070
Null Findings on Electromagnetic Inertia Thruster Experiments using a Torsion Pendulum
Hector H. Brito,* Ricardo Marini† and Eugenio S. 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 -
#1927 by RFPlumber on 19 Jan, 2016 02:30
-
Frustum test on 01/17/2016.
...
CONGRATULATIONS !!!!
A null result for Shawyer's strange hypothesis that a closed cavity like an EM Drive would have the same cut-off condition as an open waveguide, and a null result for Shawyer's "theory" predicting a larger force.
VERY IMPRESSIVE and scientifically conducted. !!
You should publish this work, or at least present it at the AIAA Propulsion conference.
...
Thank you. While it would have been awesome to witness some abnormal force, this was still a very satisfying project! I have no idea about how to publish anything though and whether it is worth the hassle. I can update the EmDrive wiki to point to these results
-
#1928 by Tellmeagain on 19 Jan, 2016 02:36
-
Thank you. While it would have been awesome to witness some abnormal force, this was still a very satisfying project! I have no idea about how to publish anything though and whether it is worth the hassle. I can update the EmDrive wiki to point to these results
Thank you!
I think you can further improve your error bar by doing the following:
1. Use 2x2 wood to make a frame that sits on a solid concrete floor to hang your pendulum. Hanging it on ceiling is a poor choice.
2. Use some thin ply board to surround the frame to remove air disturbance.
3. If oscillation still exist after 1 and 2, Use a oil damper so that you do not need to resort to middle points.
But anyway, this experiment is already good enough.
-
#1929 by Tellmeagain on 19 Jan, 2016 02:47
-
Thank you. While it would have been awesome to witness some abnormal force, this was still a very satisfying project! I have no idea about how to publish anything though and whether it is worth the hassle. I can update the EmDrive wiki to point to these results
Unfortunately "25 - 27 July 2016 | Salt Lake City, Utah 52nd AIAA/SAE/ASEE Joint Propulsion Conference" had just passed deadline to submit a paper (see http://www.aiaa.org/Research/). But if you want to publish it on arxiv I can help you on formatting. -
#1930 by RFPlumber on 19 Jan, 2016 03:06
-
Thank you. While it would have been awesome to witness some abnormal force, this was still a very satisfying project! I have no idea about how to publish anything though and whether it is worth the hassle. I can update the EmDrive wiki to point to these results
Unfortunately "25 - 27 July 2016 | Salt Lake City, Utah 52nd AIAA/SAE/ASEE Joint Propulsion Conference" had just passed deadline to submit a paper (see http://www.aiaa.org/Research/). But if you want to publish it on arxiv I can help you on formatting.
PM sent. Thank you. -
#1931 by TheTraveller on 19 Jan, 2016 03:31
-
Frustum test on 01/17/2016.
.....
Time to start looking for some HDPE disks?
Can you please get a S11 VNA scan done with and without your directional coupler in the circuit and over the range 2-2.5GHz.
Also a high res S11 scan done at the driven frequency that shows the peak rtn loss dB value, resonant freq and the bandwidth at -3dB from the peak rtn loss freq?
I believe I understand how you are currently making these measurements. Before claiming they are accurate, they need to be checked with a real VNA. Several forum members do have miniVNA tiny units you may be able to arrange a short term loan of.
How did you monitor & measure forward power? I trust you did not assume 30W output and subtracted the measured reflected to obtain forward power, which may not be accurate due to many reasons. Here again a real VNA S11 scan will give you the VSWR. But you still need to monitor the real time forward & reflected power to know what is happening to the frustum input power and impedance. That is why I invested in a Rf amp that provides no insertion loss, forward & reflected power monitoring output.
In my opinion having a real VNA to confirm resonance and measure VSWR so you can make alter coupler design and position ro obtain close to calculated Ql is vital before trying to power up a frustum. Just accepting a low Q resonance is really forming the right mode shape and not exciting multiple degenerate mode is maybe not the way to go. These cavities have been shown to be capable of forming high Q cavities (~50k loaded Q) but only if properly constructed and excited by a good coupler. Doing the work to get there may need the use of a VNA.
This is my pathway. I have no doubt it will be successful. -
#1932 by SteveD on 19 Jan, 2016 03:48
-
Frustum test on 01/17/2016.
Nicely done. Can you normalize the data between the two sets of runs?
and
Seem to be showing somewhat different behaviors.
Also, when you get time, can you give an official final statement as to measured Q and power into the frustum?
-edit can you clarify. Is rf on a measurement with rf and hv both on and hv on a measurement with hz on but no rf entering the frustum? (If that is the case, suggest also conducting a run in the "west" orientation.) -
#1933 by RFPlumber on 19 Jan, 2016 05:47
-
Questions and challenges are very welcome. However, please, understand that I am not able to follow up on every request "just because". Those requests ideally need to be accompanied by an explanation of why you think they would result in producing thrust (under the theory being tested)
Thank you!
I think you can further improve your error bar by doing the following:
1. Use 2x2 wood to make a frame that sits on a solid concrete floor to hang your pendulum. Hanging it on ceiling is a poor choice.
2. Use some thin ply board to surround the frame to remove air disturbance.
3. If oscillation still exist after 1 and 2, Use a oil damper so that you do not need to resort to middle points.
But anyway, this experiment is already good enough.
Sorry, not going to build it. It would take a lot of room in my garage (not to mention time and effort required), and I strongly doubt it will result in producing thrust. I even doubt it will reduce oscillations much as the frustum will pick up any minor excitation and will start oscillating again. It would indeed help with air movement so I could maybe be able to turn and breath during those runs Frustum test on 01/17/2016.
.....
Time to start looking for some HDPE disks?
Can you please get a S11 VNA scan done with and without your directional coupler in the circuit and over the range 2-2.5GHz.
Also a high res S11 scan done at the driven frequency that shows the peak rtn loss dB value, resonant freq and the bandwidth at -3dB from the peak rtn loss freq?
I believe I understand how you are currently making these measurements. Before claiming they are accurate, they need to be checked with a real VNA. Several forum members do have miniVNA tiny units you may be able to arrange a short term loan of.
How did you monitor & measure forward power? I trust you did not assume 30W output and subtracted the measured reflected to obtain forward power, which may not be accurate due to many reasons. Here again a real VNA S11 scan will give you the VSWR. But you still need to monitor the real time forward & reflected power to know what is happening to the frustum input power and impedance. That is why I invested in a Rf amp that provides no insertion loss, forward & reflected power monitoring output.
In my opinion having a real VNA to confirm resonance and measure VSWR so you can make alter coupler design and position ro obtain close to calculated Ql is vital before trying to power up a frustum. Just accepting a low Q resonance is really forming the right mode shape and not exciting multiple degenerate mode is maybe not the way to go. These cavities have been shown to be capable of forming high Q cavities (~50k loaded Q) but only if properly constructed and excited by a good coupler. Doing the work to get there may need the use of a VNA.
This is my pathway. I have no doubt it will be successful.
Hey TT! Lots of questions to go through...
Both narrow and wide range S11 scans have been previously posted for the first test:
http://forum.nasaspaceflight.com/index.php?topic=39004.msg1471219#msg1471219
Sorry, don't know how to make S11 scans without a directional coupler. Real VNAs most likely have an internal one already.
Forward power was measured with Boonton 4210-4A RF microwattmeter and various combinations of attenuators summing up to 40 dB (as the max input power for this meter is +10 dBm). It was measured both through a 50W 20 dB attenuator serving as dummy load as well as via the -20 dB directional coupler in forward configuration. All measurements agree on the number between 29-30W. It takes a long continues operation (10+ minutes without any extra heat sinks) of the amplifier to overheat it and to drop the output power to ~26W.
During the run there is both a LED indicator for the ON signal to the amplifier and a digital voltage meter to the main battery. With the amp starting to take 10 A a drop in battery voltage is immediately noticeable.
Reflected power is checked before and after a series of runs. There has not been a single run where reflected power at the end was more than 2.6W.
Cavity leaks RF, but only at its resonance mode (changing the freq by a mere 1 MHz results in no leaked RF). There was 1 (failed) run where I was planning to use a USB spectrum analyzer connected to the same computer as my DAQ system for visual monitoring of leaked RF as an indicator of resonance and power during the test. My DAQ refused to collect in this configuration for some reason, but control signals still worked, and the specrum analyzer promptly showed the same level of leaked RF as was observed during bench tests. This is a good indication that the resonance is there and the power to the frustum during the actual test is the same as during bench measurements.
W.r.t. exciting multiple degenerate modes, apparently this is very different with frustum shapes compared to cylinders. Take a look at the attached comsol screenshot where the nearest simulated degenerate mode is at least 5 MHz away.
Yes, the frustum Q will need to be improved before proceeding to HDPE disk tests, but this test was not intended as an all-exhaustive proof that it is not possible to get thrust from RF energy no matter what. It was a test for getting thrust under one specific theory.Frustum test on 01/17/2016.
Nicely done. Can you normalize the data between the two sets of runs?
...
Also, when you get time, can you give an official final statement as to measured Q and power into the frustum?
-edit can you clarify. Is rf on a measurement with rf and hv both on and hv on a measurement with hz on but no rf entering the frustum? (If that is the case, suggest also conducting a run in the "west" orientation.)
Thank you. How do you suggest to normalize it? Test force will be always different as it is never the same across sets of runs (those plates are on X-stage, and I am just eyeballing the distance between those before starting).
The runs themselves will never be the same because of all kinds of noises present (air movement, thermal forces, vibration, etc.). Take a look at the 2 idle runs - there is a lot of error sources in there.
The only runs ever having both RF and HV on at any one time (that is, partially overlapping as RF starts at 20s, HV - at 30 s and then both are turned off at 40s) are those which have the "HvOverRfAt30s" tag in their name. One can always just open up the particular csv and check data for Ch 2 (HV) vs Ch 3 (RF) to get their timing relationship.
Measured (loaded, at -3 dB S11) Q for this test is ~2000. Power is 27-28W.
EDIT: Attaching an example of one idle run (no RF no HV). Also, is the normalization request to just diff the displacement from its initial value at t=0? Right now all charts show absolute position of the pendulum platform (from some random base location).
-
#1934 by TheTraveller on 19 Jan, 2016 08:50
-
It was a test for getting thrust under one specific theory.
That "theory" is based on very high Q excitement of a single mode. Nothing you have shown says that requirement was achieved. What tests did you do to ensure only the desired mode was being excited and there were no degenerative modes excited, which would have reduced the Q? This is where S21 dual port tests helps.
Just as the models predict the resonant freq, they also predict the Q of the excited resonant mode. Failing to achieve the high Q may mean your frustum is not operating in the way required to generate thrust. You need to verify the excited mode is what you think it is and why the Q is so low. The low Q means something is not right. It needs to be addressed, fixed and then run your test. Assuming a low Q will work is no guarantee it will.
You also need to be able to directly measure forward power, along with reflected power as both are check and balance for each other.
Should also state that "theory" is what Roger used to build his EmDrives and his reported data is why you are here building an EmDrive.
-
#1935 by Rodal on 19 Jan, 2016 11:51
-
...This is my pathway. I have no doubt it will be successful.
You have no doubt, ahead of time, that your test will be successful ?
The scientific method of testing is that one doesn't know ahead of time what the outcome of a future test will be. -
#1936 by TheTraveller on 19 Jan, 2016 12:05
-
...This is my pathway. I have no doubt it will be successful.
You have no doubt, ahead of time, that your test will be successful ?
The scientific method of testing is that one doesn't know ahead of time what the outcome of a future test will be.
I not experimenting with the build. Have seen more than enough test data, some I can't yet disclose, to know what I'm building will work. That is as long as I do a really good job on the build or it becomes rubbish in, rubbish out. If I have thrust issues, I know where is error is and it will be in my build. So I need to do the build very well and do everything possible to excite only ONE high Q mode, which will extend to drilling a small hole in the centre of the small end plate and inserting a loop antenna to physically map out the excited mode.
What I don't know and will be experimenting to determine, is the relationships between power supply energy draw, Rf amp output energy, frustum input energy and kinetic energy gain of the accelerating rotary table.
For me that is the only unknown.
BTW the FEKO solver shows the single end plate and 2 x sidewall current loops I believed should be there for a TE012 excitation of Shell's frustum. I seems to be a very capable solver for frustum simulations.
-
#1937 by RERT on 19 Jan, 2016 12:16
-
RFPlumber - Congratulations on the getting this done!
I had a go at aggregating your data across multiple runs. If one uses the HV deflection as a thrust scale of 260 microNewtons, the answers I get for the RFon - BfRF are:
Runs 1267 58 micro-N +- 194 (2 sigma of highest aggregate midpoint sigma, which was 'Before RF')
Runs 3489 74 micro-N +- 54 (2 * same sigma)
[So 2000-3000 microN/kW.]
The first result is definitely in the noise, but not the second. However, of eight runs with the same HV, only one shows negative deflection on RF. Including the two low power HV runs, you have eight positive and two negative. If the odds were even, the chances of two or less in 10 are less than 6%.
I'm not enough of a statistician to tell at a glance how the strong oscillations you show in your raw data would affect the analysis, other than to guess that they probably make error bands even wider than calculated.
Nonetheless I agree you have ruled out thrust in the hundreds of micro-newton range. There is a hint of a signal of low thrust, but not enough data to confirm or refute it. The levels seen may be in the range of experimental Lorentz errors.
Let me know if you think anything here doesn't make sense for any reason.
Cheers, and congrats again!
R. -
#1938 by Rodal on 19 Jan, 2016 12:18
-
...
This shows precisely what I was talking about: rampant CONFUSION about what mode is being excited, with people not using the correct fields to determine the mode shape.
BTW the FEKO solver shows the single end plate and 2 x sidewall current loops I believed should be there for a TE012 excitation of Shell's frustum. I seems to be a very capable solver for frustum simulations.
IslandPlaya excellent work using FEKO
https://drive.google.com/folderview?id=0B6juR48k_XoTOWJMMHRHdm1fUmc&usp=sharing
shows mode shape TM112, a transverse magnetic mode, completely different mode shape than TE012, for SeeShell experiment, even when the rectangular waveguide enters the frustum from the Big Base, as shown iin the pictures below.
The mode shapes, as we learnt at school, are determined by the shape of the electric and magnetic fields, and not by the current loops !!!
The standard interpretation of mode shapes is that they are identified by the Electric and Magnetic field distributions. That is why mode shapes are known as TM (transverse magnetic field) and TE (transverse electric fields). They are not identified as TC (transverse currents)
And take a look at the surface current picture from FEKO (the last one attached below): it does NOT look like the surface current distribution for mode shape TE01, it looks like the surface current distribution for TM112
-
#1939 by Tellmeagain on 19 Jan, 2016 12:21
-
[wood frame]
Sorry, not going to build it. It would take a lot of room in my garage (not to mention time and effort required), and I strongly doubt it will result in producing thrust. I even doubt it will reduce oscillations much as the frustum will pick up any minor excitation and will start oscillating again. It would indeed help with air movement so I could maybe be able to turn and breath during those runs
It is fair not to build it. Here is just some explanation. The purpose is not to produce thrust, but to reduce error bar, so you can say "if there is any thrust, it would be below 1uN" instead of "it would be below 50uN".
Your ceiling may introduce much of the "minor excitation" you see, that's why I suggest a frame. It is easy to build anyway, just four wood sticks tied together to form the pyramid shape.
But since we are on the opposite side of EmDrive, and my purpose is improve "no thrust" claim and your purpose is to find under what condition there is thrust, it is fair for you not to take my suggestion.
