-
#3800 by SeeShells on 12 Jul, 2016 21:08
-
I dissagree.
I disagree.Particularly just before it went below 2.45 GHz ? (long portion of flat response)
Just to be clear, the flat portions represent no movement of the beam. Time is horizontal, displacement is vertical. When the graph line drops, the frustum is moving in reverse direction (large end leading), when the graph line goes up, it is moving in forward direction (small end leading). So it would seem the the portions where the beam moved down quickly are the most interesting.
The figure of merit in all discussions about the EM Drive has been force/InputPower.
With constant input power, you should get a constant force on the torsional pendulum: it should settle down to a flat line. Constant input power = constant rotation of the torsional pendulum. Flat.
As we discussed several times, even this, if larger than thrust/inputPower of a photon rocket leads to a problem with conservation of energy.
Having increasing force with constant input power can trivially be shown to be a free-energy machine.
I doubt that the strongest advocate for the EM Drive as a source of space propulsion would seriously advocate for the EM Drive to result in increasing force at constant power. That would be a runaway free-energy machine.
Hence increasing force at constant power is clearly a transient or an artifact.
NASA Eagleworks plots were showing fairly good looking pulses with the force settling down at constant power. Not increasing force at constant power. That is a no-no.
It must be related to the fact that you are doing frequency sweeps and therefore the mode shapes are changing.
It must be due to the fact that they are transients and not steady-state.
If you want to understand what is going on, you better go for the simpler things: the steady-state.
Transients are always more complicated.
Transients for mechanical oscillations are more complicated than the steady-state solution: for example a mechanical system excited with a mechanical force oscillating harmonically will display a transient with a damped frequency different from the excitation frequency, and the shape of the transient will very much depend on the amount of damping and the shape of the transient excitation including initial conditions.
Therefore you should be looking for flat portions of the curve at constant power at constant frequency.
The worst thing to have, which was shown to be an experimental artifact is something like this, with the measured force increasing, apparently without bounds, at on/off pulsing of constant power in each on segment:QuoteThe figure of merit in all discussions about the EM Drive has been force/InputPower.
With constant input power, you should get a constant force on the torsional pendulum: it should settle down to a flat line. Constant input power = constant rotation of the torsional pendulum. Flat.
Not in monomorphic's magnetron driven drive.
https://www.youtube.com/watch?v=sKsV3qTStyE?t=168
If you draw a vertical line a little wider than the scope lines right where his resonance is for the mode he is needing to operating and watch the magnetron's output you'll see that the output of the magnetron doesn't fall into the resonance area all the time.
He isn't 100% effectively exciting the frequency window for resonance and to draw any serious conclusions other than the beam measures some kind of deflection during power on is all you can say.
Expect it to look more like EagleWork's data which still contains a noise component even when he goes solid state.
Shell
-
#3801 by jmossman on 12 Jul, 2016 21:26
-
I've superimposed the rate of change in beam displacement.
I haven't seen this mentioned yet: there is possible misalignment in the time axis of the "RF spectrum analyzer" versus the "laser displacement" plots.
My untrained eye would suggest that the the "RF spectrum analyzer" events would better fit the "laser displacement" profile if the "RF" events were shifted to the right (i.e. later in time) than the current combined plot shows.
It's very common in systems to have different delays from types of sensor outputs. In this case, we have a "RF spectrum analyzer" with an unknown time delay between when the signal gets set down the probing antenna/cable and when the information has been processed into something visible on an output port or computer screen (plus the time to establish an "initial lock" versus a "detected frequency drift" may also be different). Similarly, we have a "laser displacement" sensor which has an unknown time delay between when the torsion beam moves and when the laser time-of-flight has been computed and translated into a distance. There is a third unknown regarding the different delays between the A/D sampling channels.
While I completely agree with Dr. Rodal's and others observations about thermal responses and long decay time constants, Monomorphic's dataset is the first I've ever seen which seems to have a non-trivial correlation in the displacement force vs microwave frequency. Obviously this is very early data, and thermal differences between forward-vs-reflected microwave heating could indeed be the source of displacement delta vs frequency.
Alternatively, there are theories that claim to predict the amount of RF energy within the cavity will correlate with the total anomalous force generated; the RF energy within a resonant cavity will depend strongly upon the cavity's resonant frequency and bandwidth, the frequency broadcast by the antenna, and the impedance match between said antenna and said cavity.
I suspect many of us are eagerly awaiting more data with less dampening. Some more null runs with some calibrated force events would also be nice.
Keep up the excellent work! I've got plenty of popcorn ready; you've definitely got my attention.
Edit1: added "and bandwidth" to the resonant cavity description.
Edit2: replaced the word "discrepancy" with "misalignment". -
#3802 by Josave on 12 Jul, 2016 21:46
-
It was not an accident I chose these two modes. Not only are they closely adjacent, but the E-fields are concentrated on opposite ends of the frustum. The theory is TE311 produces reverse thrust, and then I tune to TE212, without making any other changes to the build, and get forward thrust.
Monomorphic, can you switch frequencies at a very high rate, in a sort of frequency modulation?
This can make the energy distribution of the radiation inside the frustum move rapidly from one end to another. In a kind of Woodward effect...
-
#3803 by rfmwguy on 12 Jul, 2016 22:06
-
Behold the cause of the magnetic displacement while the harness was aligned on the beam. Ferrite chokes...ferromagnetic...case closed.
-
#3804 by Monomorphic on 12 Jul, 2016 22:15
-
Powered test 11 completed. This time I optimized the laser displacement sensor (LDS). Highest resolution setting possible (3um), with 60ms response time, comes to a sample rate of 16.6/sec. Much higher quality data - pushing the LDS to its limits here.
-
#3805 by Rodal on 12 Jul, 2016 22:24
-
...
Of course he is not operating at resonance. I hope you are not constructing a strawman
I dissagree.QuoteThe figure of merit in all discussions about the EM Drive has been force/InputPower.
With constant input power, you should get a constant force on the torsional pendulum: it should settle down to a flat line. Constant input power = constant rotation of the torsional pendulum. Flat.
Not in monomorphic's magnetron driven drive.
https://www.youtube.com/watch?v=sKsV3qTStyE?t=168
If you draw a vertical line a little wider than the scope lines right where his resonance is for the mode he is needing to operating and watch the magnetron's output you'll see that the output of the magnetron doesn't fall into the resonance area all the time.
He isn't 100% effectively exciting the frequency window for resonance and to draw any serious conclusions other than the beam measures some kind of deflection during power on is all you can say.
Expect it to look more like EagleWork's data which still contains a noise component even when he goes solid state.
Shell
on an incorrect assumption : that I didn't consider that Monomorphic was changing frequency.
I explicitly said that he was changing the natural frequency. Monomorphic had indicated that in his graph. I even asked him what modes shapes and natural frequencies where in the region of the spectrum that he traversed. !
But, as he traverses the spectrum he will be hitting resonances if the natural frequencies are there. So he will briefly be in resonance. For how long? I don't know, it depends on what mode shapes are nearby and mode participation and how high is the Q. If the Q in Monomorphic's is very high I'll grant you that it will quickly go out of resonance. However if the Q is not too high then it can stay in resonance for a larger bandwidth.
Do we know high is the Q in his test for the mode shapes that he traversed?
And the point remains that for Monomorphic's or any such experiment, that if the force is not constant for constant power input, then it is either a transient or an artifact.
Otherwise to propose that the EM Drive is going to generate increasing thrust with time in space at constant power would produce a runaway free energy machine, that I question whether anyone seriously would propose such a machine.
In this case it is a transient, due to the changing frequency.
And of course it is very difficult to stay at constant natural frequency because the natural frequency changes as the frustum heats up and deforms due to thermal expansion, as well as any change in constitutive coefficients.
Jaydurst read my comment correctly and had an interesting view (difficult to comment on this because again, the behavior is due to a transient since Monomorphic did not stay at constant frequency):
http://forum.nasaspaceflight.com/index.php?topic=39772.msg1559247#msg1559247
I certainly agree with you that we need to wait for Monomorphic to run at a known mode shape at steady frequency to see what the response will look like...
-
#3806 by Monomorphic on 12 Jul, 2016 22:57
-
Do we know high is the Q in his test for the mode shapes that he traversed?
I calculated Q factor based on the Return Loss sweep at about 1,000. -
#3807 by Rodal on 12 Jul, 2016 23:00
-
Good to knowDo we know high is the Q in his test for the mode shapes that he traversed?
I calculated Q factor based on the Return Loss sweep at about 1,000.
OK that gives you 50 times larger bandwidth to stay at resonance than if it would be 50,000
10 times larger bandwidth to stay at resonance than if it would be 10,000
so an order of magnitude larger bandwidth to stay at resonance -
#3808 by Monomorphic on 12 Jul, 2016 23:07
-
I found the Return Loss scan I did with HDPE tuned to where it is now. I've marked it in the image. Dr. Rodal, notice the same 'flat' regions? One at RF on and one at RL peak. Damn interesting!
-
#3809 by Rodal on 12 Jul, 2016 23:08
-
Behold the cause of the magnetic displacement while the harness was aligned on the beam. Ferrite chokes...ferromagnetic...case closed.
Well, first time I see you were using ferrite chokes. What I had seen was this exchange that mentioned the cable:Quote[–]chongma 2 points 2 days ago
Did the cable turn the beam into a giant electromagnetic compass?
[–]rfmwguy-EMDrive Builder 2 points 2 days ago
Yes, I believe it was displaced towards north. The results were too repeatable to be thermal related especially after I secured the harness tighter
and this, mentioning the copper wires:Quote from: rfmwguy-EMDrive Builderwas actually a slow dissipation of a magnetic charge, not of the frustum cavity, but of the wires themselves
So, if I understand this correctly now, you are not saying that the cable turned the beam into a giant electromagnetic compass, but that it was the ferrite chokes?.
But if so, the way to "close the case" would be by re-running your experiment with the cable on the beam, without the ferrite chokes.
If re-running the experiment without the ferrite chokes eliminates the previous effect you have a reasonable case to close the case.
Otherwise, it remains just one among other explanations and the case is dormant.
As you eliminated the cable in its entirety from the beam, those other explanations remain alive. As, if the problem is eliminated or much ameliorated it will not be known whether thermal expansion of the cable, the ferrites, or SeeShells explanation or other explanations were the culprit, as they were all related to the cable on the beam. -
#3810 by SeeShells on 12 Jul, 2016 23:29
-
I would never do a straw-man construct with you although I wanted to clear up a possible misconception that could be seen in how you wrote your comment....
Of course he is not operating at resonance. I hope you are not constructing a strawman
I dissagree.QuoteThe figure of merit in all discussions about the EM Drive has been force/InputPower.
With constant input power, you should get a constant force on the torsional pendulum: it should settle down to a flat line. Constant input power = constant rotation of the torsional pendulum. Flat.
Shell on an incorrect assumption : that I didn't consider that Monomorphic was changing frequency.
I certainly agree with you that we need to wait for Monomorphic to run at a known mode shape at steady frequency to see what the response will look like...
Things are getting very interesting right now and being clear is very important.
Shell
You go Monomorphic! -
#3811 by Rodal on 12 Jul, 2016 23:49
-
I found the Return Loss scan I did with HDPE tuned to where it is now. I've marked it in the image. Dr. Rodal, notice the same 'flat' regions? One at RF on and one at RL peak. Damn interesting!
It is very interesting because one sees that the pendulum dynamics is past the trough (the bottom peak) and something, just at the vertical bar you indicated, is suddenly stopping the dynamics of the pendulum, and making it stay flat. And the effect is instantaneous, as you would expect from electromagnetism
After the RF is off, the pendulum dynamics (inertial, damping and torsional spring forces) make it oscillate. The effect after the RF is off is also almost instantaneous.
What made it go flat? It takes a force to stop the pendulum dynamics and make it stay flat
Put somebody on a swing and try to stop the swing, stone cold and you will see
-
#3812 by SeeShells on 13 Jul, 2016 01:11
-
I was researching electromagnetic braking a month or so ago and when I saw the "flats" I thought of this video.I found the Return Loss scan I did with HDPE tuned to where it is now. I've marked it in the image. Dr. Rodal, notice the same 'flat' regions? One at RF on and one at RL peak. Damn interesting!
It is very interesting because one sees that the pendulum dynamics is past the trough (the bottom peak) and something, just at the vertical bar you indicated, is suddenly stopping the dynamics of the pendulum, and making it stay flat. And the effect is instantaneous, as you would expect from electromagnetism
After the RF is off, the pendulum dynamics (inertial, damping and torsional spring forces) make it oscillate. The effect after the RF is off is also almost instantaneous.
What made it go flat? It takes a force to stop the pendulum dynamics and make it stay flat
Put somebody on a swing and try to stop the swing, stone cold and you will see
Shell -
#3813 by Bob Woods on 13 Jul, 2016 04:22
-
Keep working folks. You are all doing real science here.
Openly. Sharing. Arguing. Learning. Improving. Testing.
Regardless of the outcome, this kind of open world-wide collaboration is, in and of itself, a great leap for science. -
#3814 by Star One on 13 Jul, 2016 06:29
-
I don't post my data here any longer, but I will address what is a misconception: As far as I know, EWs paper is not rejected but stalled. From here, I can only speculate and this is my best educated guess:Thanks. So that's the end of that then. Pretty downbeat ending to the whole enterprise.
Hard to accept...
but if so the Referees and Nasa must have had a reason for their decisions. I am quite sure by now that the concept is not working as we
expected it to work.
The Thing is: If an Experiment shows negative results (though scientifically constructive and in this sense positive) These results are never published nowadays.
So no News from Eagleworks means, that the EMDrive concept probably does not work, meaning that the thrust could not be measured in their improved Setup.
However, there are many more experimental ideas to test, so I am not losing my confidence, that at least the solar system will be colonized sometime.
Only one reviewer has not approved it.
The hang-up relates to the theory, not the test results.
This speculation is NOT based on anyone's info at EW, but I would consider it very trustworthy...and I didn't ask further questions.
The builders network is quite active off-forum and there are many private conversations we all choose not to make public for obvious reasons. So, drawing conclusions now is premature especially considering there is Summer recess for academia and reviewer(s) are likely out and about. So I wouldn't draw any conclusions, especially since many papers, especially at high levels of publishing, can take years to publish. While the wait is annoying, that's the way this thing is going.
Well you can hardly criticise people here as they can only draw their conclusions from the information that is publicly available. If there is other information that says different but is being kept secret then people are obviously not going to be able to factor this into their considerations. -
#3815 by frobnicat on 13 Jul, 2016 09:03
-
I found the Return Loss scan I did with HDPE tuned to where it is now. I've marked it in the image. Dr. Rodal, notice the same 'flat' regions? One at RF on and one at RL peak. Damn interesting!
It is very interesting because one sees that the pendulum dynamics is past the trough (the bottom peak) and something, just at the vertical bar you indicated, is suddenly stopping the dynamics of the pendulum, and making it stay flat. And the effect is instantaneous, as you would expect from electromagnetism
After the RF is off, the pendulum dynamics (inertial, damping and torsional spring forces) make it oscillate. The effect after the RF is off is also almost instantaneous.
.../...
Concerning this later statement : it takes almost half the time of the flat power-on period after power-off before the signal clearly deviates from the plateau. "Almost instantaneous" is a bit radical, it takes about 10s or a little more. Obviously the inertia of the mechanical system will tend to time shift any effect, "almost" needs to be related to known time constants, but then why would the causality of the flat power-on period after return loss peak is "instantaneous" and not "almost instantaneous" ?
Also it is hard to read a clear/clean "pendulum dynamics" with well behaved oscillations after power-off, looks to me more like a bit chaotic motion. Whatever mechanism B is causally responsible for the noisy output after the power-off is very likely to be already active by the end of the power-on run. So if the flat plateau just after return loss peak is to be considered significant and due to a causal mechanism A, the effect of A must cancel exactly the effect of B during 20s. This seems to me quite unlikely unless A=B, or at least both share "a lot".
I understand this is a bit imprecise (a lot, almost ...) I'm just trying to share my intuitive view of it, not to state a bullet proof conclusion : seems like all or a great part of what makes the flat plateau possible (if it is significant at all) is also acting long after the power-off, and hence can't be purely electromagnetic. Electromagnetism can play a central role though, but also in conjunction with long time constant effects (other than mechanics of the measuring system) likely thermals (again). For instance like jmossman puts it :.../...
Obviously this is very early data, and thermal differences between forward-vs-reflected microwave heating could indeed be the source of displacement delta vs frequency.
.../...
In brief, the flat region of interest is suspiciously too flat given the expected level of noise in said region (given what is to be seen after). Either the explanations are over-fitting the data (flatness is a fluke) or something else is at play like something as improbable as data acquisition perturbed by sudden RF interference when resonance locks, or some mechanical stickiness actually causing the lock in resonance and not the reverse causality, or ferrite chokes ?
Added : @Monomorphic have you made a spreadsheet or raw data file available somewhere ? I must admit I haven't followed your publications with the care they'd deserve, only catching-up here on NSF thread and no time to follow all the links. Motivation : small test in the vertical scale for data acquisition artefacts (too much flattish steps to my taste, I understand you operate the LDS near its limits). -
#3816 by Flyby on 13 Jul, 2016 09:05
-
I was researching electromagnetic braking a month or so ago and when I saw the "flats" I thought of this video.
Shell
Very interesting video, SeeShells...
However, it makes me wonder what would happen if it had horizontal cuts instead of drilled holes, as obviously, the vertical cuts fails to generate a counter-magnetic field.
The only difference I see between the drilled holes and the vertical cuts, is the lack of electromagnetic induction current flow (hence lack of counter magnetic field) in the last one. Am I'm right in assuming that? -
#3817 by Rodal on 13 Jul, 2016 12:20
-
I found the Return Loss scan I did with HDPE tuned to where it is now. I've marked it in the image. Dr. Rodal, notice the same 'flat' regions? One at RF on and one at RL peak. Damn interesting!
frobnicat wrote:...In brief, the flat region of interest is suspiciously too flat given the expected level of noise in said region (given what is to be seen after). Either the explanations are over-fitting the data (flatness is a fluke) or something else is at play like something as improbable as data acquisition perturbed by sudden RF interference when resonance locks, or some mechanical stickiness actually causing the lock in resonance and not the reverse causality, or ferrite chokes ?
...
1) Have you measured whether you have any RF leaking from the cavity?
2) Is there any mechanical stickiness in your setup? Do you have such flat regions of response if the pendulum is given an initial torque, by hand, (free response to an initial angular displacement) and let it come back and oscillate on its own, without power going to the EM Drive? -
#3818 by Monomorphic on 13 Jul, 2016 13:16
-
I found the Return Loss scan I did with HDPE tuned to where it is now. I've marked it in the image. Dr. Rodal, notice the same 'flat' regions? One at RF on and one at RL peak. Damn interesting!
frobnicat wrote:...In brief, the flat region of interest is suspiciously too flat given the expected level of noise in said region (given what is to be seen after). Either the explanations are over-fitting the data (flatness is a fluke) or something else is at play like something as improbable as data acquisition perturbed by sudden RF interference when resonance locks, or some mechanical stickiness actually causing the lock in resonance and not the reverse causality, or ferrite chokes ?
...
1) Have you measured whether you have any RF leaking from the cavity?
2) Is there any mechanical stickiness in your setup? Do you have such flat regions of response if the pendulum is given an initial torque, by hand, (free response to an initial angular displacement) and let it come back and oscillate on its own, without power going to the EM Drive?
I assume there is some RF leaking from the cavity, or perhaps around the magnetron, otherwise the spectrum analyser wouldn't detect anything - as it is mounted externally. I did recently add another RF gasket to the insertion point to help better seal the cavity.
As for stickiness, it's something I'm looking into. I will need to redo the 'tap tests' since I switched to high resolution (3um). The previous taps are about 10 times stronger than the anamalous force seen in the latest powered runs. So I have to come up with something that can tap the beam much more lightly.
-
#3819 by Monomorphic on 13 Jul, 2016 13:25
-
Added : @Monomorphic have you made a spreadsheet or raw data file available somewhere ? I must admit I haven't followed your publications with the care they'd deserve, only catching-up here on NSF thread and no time to follow all the links. Motivation : small test in the vertical scale for data acquisition artefacts (too much flattish steps to my taste, I understand you operate the LDS near its limits).
Here is the raw data for the last few powered runs. Run 10 was corrupted, so while I could see the data, I couldn't export it to excel. I started using the LDS at run 7. On run 11, I switched the LDS to highest resolution (3um). That gives a 60ms response time. That's 16.6667 samples per second. My sample rate is set to 16, so I'm operating just within the limits of the LDS.
on an incorrect assumption : that I didn't consider that Monomorphic was changing frequency.
