Author Topic: EM Drive Developments - related to space flight applications - Thread 3  (Read 1876760 times)

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2335
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 2982
  • Likes Given: 2601
On another note, my mother who (85) has just went under the care of Hospice and may not have much longer for this world, she has been by my side all of my career and childhood, encouraging me, pushing me. She was the one who stood beside me on that cold night in October of 1957 and said "Look there it is, see that twinkling light." It was Sputnik. And as much as the men and women at NSF and some of my dear friends have helped I want to dedicate this work I'm doing to her and to "Make it so".

Offline TheTraveller

.....
Just got this looks sweet.

The SA0314 looks very interesting. Price is good.
http://www.rfinstruments.com/php/pdf/SA0314%20datasheet.pdf

You should be able to measure and record the output power bandwidth of your magnetron and if you barely insert the tip of a probe inside your frustum (good to put in some attenuation so you don't blow the input stage of the spectrum analyser), should see the acceptance bandwidth of your frustum and record. Can then compare the charts to work out which of the magnetron frequencies are being accepted by your frustum and which are being rejected.

Is that how you plan to use this?
That is one way and the other is to record the baseline activity inside and outside the faraday cages when it's off.

Good idea.
"As for me, I am tormented with an everlasting itch for things remote. I love to sail forbidden seas.”
Herman Melville, Moby Dick

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2335
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 2982
  • Likes Given: 2601
.....
Just got this looks sweet.

The SA0314 looks very interesting. Price is good.
http://www.rfinstruments.com/php/pdf/SA0314%20datasheet.pdf

You should be able to measure and record the output power bandwidth of your magnetron and if you barely insert the tip of a probe inside your frustum (good to put in some attenuation so you don't blow the input stage of the spectrum analyser), should see the acceptance bandwidth of your frustum and record. Can then compare the charts to work out which of the magnetron frequencies are being accepted by your frustum and which are being rejected.

Is that how you plan to use this?
That is one way and the other is to record the baseline activity inside and outside the faraday cages when it's off.

Good idea.
Thanks. The price was so good I had to get it and couldn't touch a full fledged stand alone for that price.

Offline TheTraveller

.....
Just got this looks sweet.

The SA0314 looks very interesting. Price is good.
http://www.rfinstruments.com/php/pdf/SA0314%20datasheet.pdf

You should be able to measure and record the output power bandwidth of your magnetron and if you barely insert the tip of a probe inside your frustum (good to put in some attenuation so you don't blow the input stage of the spectrum analyser), should see the acceptance bandwidth of your frustum and record. Can then compare the charts to work out which of the magnetron frequencies are being accepted by your frustum and which are being rejected.

Is that how you plan to use this?
That is one way and the other is to record the baseline activity inside and outside the faraday cages when it's off.

Good idea.
Thanks. The price was so good I had to get it and couldn't touch a full fledged stand alone for that price.

PC side software looks good. Will buy one as well.
"As for me, I am tormented with an everlasting itch for things remote. I love to sail forbidden seas.”
Herman Melville, Moby Dick

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
...

tidux can get the current meep model done in about 45 minutes with 8 threads out of his 12 threads. Considering it scales linearly, he could get a performance of 30 minutes.
So, <0.5 hour / 0.013 microseconds>. So it would take tidux's server just a little bit over 2 days to complete 1.3 microseconds, which is completely reasonable on my opinion. Do you believe it would be a good use for his server, Doctor Rodal? And what do you think, Mister tidux?

It's not quite half an hour at 12 threads, it's closer to 35 minutes, but that's close enough to linear scale for our purposes.

I'm willing if Dr. Rodal thinks it would provide useful data.  The system is located in a datacenter, not my home, so there aren't any issues with noise, heat, or power failures from running a two day job.
Thanks for the offer.

1) Yes, it would be useful to have a Meep run for a file extended to run over 100 times longer, over a microsecond total run time, if and only if it has been previously run, examined and certified before, to match the input and output of a previously examined file.

2) To be useful data the same input would have to be shown previously to match another csv file previously run.  Having different number of rows and columns is not acceptable.  The number of rows and columns corresponds to the number of finite difference nodes in the finite difference discretization mesh, and thus it is of crucial importance to the partial differential equation solution.

3) To post-process output data it is preferable to have csv file output than ods file output. 
« Last Edit: 07/20/2015 01:13 PM by Rodal »

Offline Eer

  • Full Member
  • **
  • Posts: 274
  • Liked: 140
  • Likes Given: 129

Thanks for the offer.

1) Yes, it would be useful to have a Meep run for a file extended to run over 100 times longer, over a microsecond total run time, if and only if it has been previously run, examined and certified before, to match the input and output of a previously examined file.

2) To be useful data the same input would have to be shown previously to match another csv file previously run.  Having different number of rows and columns is not acceptable.  The number of rows and columns corresponds to the number of finite difference nodes in the finite difference discretization mesh, and thus it is of crucial importance to the partial differential equation solution.

3) To post-process output data it is preferable to have csv file output than ods file output.

To be clear:

2) Agreed. New results needs to be able to be matched to previous runs.  The issue with different number of rows and columns was to document a disconnect in specifications of what version of file output I was attempting to compare my runs against - it seems the file I thought I was supposed to use was not, in fact, a comparable run, and as a result I spent three days trying to make my output match the previous one.  Thus my suggestion that control files used to create output files always be provided with those output files in the future.  That, at least, should allow follow-on efforts to re-run the control file and verify the output files associated with it.

3) Agreed.  However, the use of files other than csv formats may be necessary when collecting multiple output files into a single package file to support direct comparisons of cell values across runs.  It's clumsy, but better than using file references for links between multiple csv files.  So consider it an artifact of the post-processing analysis like any other tool you might use.  CSV is a simple, easy to review, widely supported standard data interchange format that we should use for sharing data among researchers.  The alternative is to use versions of MEEP/HDF5 which store binary data in canonical format that is not machine independent, and I think that's not worth while.

Ed

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
..

To be clear:

2) Agreed. New results needs to be able to be matched to previous runs.  The issue with different number of rows and columns was to document a disconnect in specifications of what version of file output I was attempting to compare my runs against - it seems the file I thought I was supposed to use was not, in fact, a comparable run, and as a result I spent three days trying to make my output match the previous one.  Thus my suggestion that control files used to create output files always be provided with those output files in the future.  That, at least, should allow follow-on efforts to re-run the control file and verify the output files associated with it.

3) Agreed.  However, the use of files other than csv formats may be necessary when collecting multiple output files into a single package file to support direct comparisons of cell values across runs.  It's clumsy, but better than using file references for links between multiple csv files.  So consider it an artifact of the post-processing analysis like any other tool you might use.  CSV is a simple, easy to review, widely supported standard data interchange format that we should use for sharing data among researchers.  The alternative is to use versions of MEEP/HDF5 which store binary data in canonical format that is not machine independent, and I think that's not worth while.

Ed
I agree that it would be helpful to have unadulterated MEEP INPUT control files used to create OUTPUT files  be provided with output files.

The Meep input control file controls the Meep analysis, thus it is all one needs to run the analysis, "it is all you need".  More or different is not better because it would be subject to interpretation.   Same as providing numerical data in engineering drawings: redundant information is not better.  Extra information should be provided as comments. The Meep input control files could explained, with comments, as much as necessary but they should  never be substituted by any other type of input description that may be subject to interpretation or translation issues.



Ditto for the MEEP OUTPUT information.  To post-process the data, the actual output information from Meep is needed: 

* the total Meep time (the computer run time is completely irrelevant to post-processing),
* the total number of Meep time slices
* the total number of Meep time steps

etc.



It would be helpful to have both the MEEP INPUT control file and the MEEP OUTPUT file information referred to above be provided as .txt files in the same Google Drive folder where the csv files are provided, for easy reference to understand what is the input and output associated with the csv files.

This is all part of formalizing a collaboration between multiple parties.
« Last Edit: 07/20/2015 03:22 PM by Rodal »

Offline TheTraveller

This is what happens when the cavity input bandwidth is not aligned with the magnetron output bandwidth.

The small rectangle on the lower left of each diagram is the cavity input bandwidth versus the magnetron.

If you plan on using / are using a magnetron and don't get this right, very little energy will entry the cavity and the rest will be rejected. End result will be very low Force generation.

Data from the 2012 NWPU paper attached

Magnetron pwr, Cavity Pwr, Force generated, Specific Force Mag Pwr, Specific Force Cavity Pwr

a) 200Wmag, 013Wcav, 170mN, 0.85N/kW, 13.1N/kW
b) 300Wmag, 120Wcav, 270mN, 0.90N/kW, 2.25N/kW
c) 400Wmag, 085Wcav, 225mN, 0.56N/kW, 2.65N/kW
d) 500Wmag, 065Wcav, 200mN, 0.40N/kW, 3.08N/kW
e) 600Wmag, 045Wcav, 180mN, 0.30N/kW, 4.00N/kW  <<< Maybe where Dr. White got his 4N/kW figure?
f) 700Wmag,  048Wcav, 210mN, 0.30N/kW, 4.38N/kW  <<<    "         "       "       "       "   "       "          "
« Last Edit: 07/20/2015 03:50 PM by TheTraveller »
"As for me, I am tormented with an everlasting itch for things remote. I love to sail forbidden seas.”
Herman Melville, Moby Dick

Offline Devilstower

  • Member
  • Posts: 23
  • Liked: 10
  • Likes Given: 6
I suspect everyone has seen some version of Escape Dynamic's microwave-powered shuttle at this point, but just in case, here's a short (and badly written) article that includes a nice bit of embedded animation. It's only vaguely related, but hey, it is a spacecraft proposal, and it is using microwaves, so... here you go.

http://www.engadget.com/2015/07/20/escape-dynamics-microwave-spacecraft/

As several people have pointed out, the energy losses in using the kind of microwave sources that ED is proposing would seem to be pretty daunting. Of course, much could be solved if they could instead smack their shuttle with Masers.

Side benefit: developing the requisite high Q-factor microwave cavities necessary for building all those big Masers might give an opportunity to test... some other theories.

Offline deltaMass

  • Full Member
  • ****
  • Posts: 955
  • A Brit in California
  • Liked: 671
  • Likes Given: 275

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
I thought I would chime in with questions for the AIAA conference. I am going to be attending the event, so I will be able to report back to you guys about what happens, other questions that were asked, etc.

Let me know of a couple of definite questions that anyone would want me to ask during the Q and A, and I'll be sure to ask them and report back the answers. :)

-I

DrBagelBytes, pleas see below the questions that I already formulated:

did a search and did not find this... hope it was not posted yet
Quote
AIAA Propulsion and Energy Forum and Exposition
Hilton Orlando, Orlando, Florida...
TUESDAY, JULY 28, 2015
NFF-04. Future Flight Propulsion Systems
...5:00 PM - 5:30 PM
Direct Thrust Measurements of an EMDrive and Evaluation of Possible Side-Effects
Martin Tajmar
I wonder if someone from this thread could attend the conference and if there is a following Q/A, even mention some of the experiments discussed here, ask questions, etc.

I am looking forward to this presentation.  Unfortunately, I won't be attending.  I have tried to find out, from several different second-hand sources what has been the nature of Martin Tajmar's experiments.  It is my personal understanding that his EM Drive experiments have shown  very low force/InputPower readings for an EM Drive in a partial vacuum: less than a few dozen or so times the force/InputPower of a perfectly collimated photon rocket, thus much lower thrust force/InputPower than Yang (who reported 300,000  times a photon rocket) and Shawyer (25,000 to 70,000 times) have reported.  I understand that the quality factor of resonance (Q) in the experiments is extremely low, much lower than any researcher has reported up to now. 

Regarding possible questions to ask if anybody attends, one suggestion (if this is what is reported) is to ask why is his experimental  Q so low (less than 100): how could the experiments have been conducted under resonance if the Q was so low?. Another question: what was responsible for such a low Q in the experiments, and whether Tajmar thinks that the discrepancy with other researchers has to do with the different Q reported from different researchers.

Another suggested question to Prof. Tajmar: given the very low force/InputPower readings for an EM Drive in a partial vacuum measured by Prof. Tajmar (less than a few dozen or so times the force/InputPower of a perfectly collimated photon rocket),  does Prof. Tajmar see his (and Georg Fiedler's) experiments at The Technische Universität Dresden as a scientific nullification of the claims made by Yang  and Shawyer, since Yang and Shawyer claim over 1,000 to 10,000 times greater force/InputPower than what Tajmar measured) ?

Does Prof. Tajmar think that the reason why Shawyer and Yang claimed much higher thrust is because Shawyer and Yang reported tests at ambient pressure (unlike Prof. Tajmar who has performed his tests in a vacuum), and Shawyer and Yang just reported thermal convection artifacts?

If, not a nullification due to Shawyer and Yang not performing tests in vacuum, what does Prof. Tajmar think that the huge difference (1,000 to 10,000 times) is due to ?

Another suggested question to ask Tajmar (apparently their experimental measurements at some point showed some 60% orientation dependence if my understanding is correct, not clear whether experimental noise, and whether he will present some updated data):

QUESTION: why did the experiments show approximately 60% different thrust force measurements when the EM Drive was physically rotated 180 degrees from the "forward" thrust tests to the "reverse" thrust tests?  Shouldn't the thrust be the same regardless of space orientation?  Is this orientation-dependence indicative of an experimental artifact or a dependence on an external field ?

(motivated by TT's post http://forum.nasaspaceflight.com/index.php?topic=37642.msg1406672#msg1406672  )

QUESTION TO TAJMAR:  What is the explanation for the very low Q of only 50 in your EM Drive experiments?.  Shawyer has reported Q=45,000 for his Demonstrator at  your same tested frequency 2.45 GHz. Is it because you did not use the usual waveguide isolator and 3-stub tuner between the magnetron and test article?

Has Prof. Tajmar's team grossly over-coupled the RF input to the EM Drive copper truncated cone?
Over-coupling is a matter of putting a larger resistive load on the resonant cavity by shunting more of the source load onto the cavity. Over-coupling can give  an "equivalent" bandwidth which includes more than one mode.
How wide was the high filament current magnetron bandwidth?.
Was Tajmar trying to match it w/ ~ 50 MHz  ?
« Last Edit: 07/20/2015 04:38 PM by Rodal »

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2165
  • Liked: 2681
  • Likes Given: 1124
Interesting, a Q of 50 would translate roughly into the total frequency spread of a typical magnetron; about 49 MHz. This would avoid the need for a tuner either on the frustum or the magnetron. Wonder if this is telling us something...

Offline TheTraveller

Interesting, a Q of 50 would translate roughly into the total frequency spread of a typical magnetron; about 49 MHz. This would avoid the need for a tuner either on the frustum or the magnetron. Wonder if this is telling us something...

Q=50 * P=1,000W wideband magnetron input = 50kW equivalent field strength inside the cavity

Q=50,000 * P=100W narrow band input = 5MW equivalent field strength inside the cavity (100x greater)

Using F = (2 P Df Q) / c, Force generation will be 100x greater at 100W narrow band with Q 50,000 than at 1,000W wideband with Q 50.
"As for me, I am tormented with an everlasting itch for things remote. I love to sail forbidden seas.”
Herman Melville, Moby Dick

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
Interesting, a Q of 50 would translate roughly into the total frequency spread of a typical magnetron; about 49 MHz. This would avoid the need for a tuner either on the frustum or the magnetron. Wonder if this is telling us something...

Q=50 * P=1,000W wideband magnetron input = 50kW equivalent field strength inside the cavity

Q=50,000 * P=100W narrow band input = 5MW equivalent field strength inside the cavity (100x greater)

Using F = (2 P Df Q) / c, Force generation will be 100x greater at 100W narrow band with Q 50,000 than at 1,000W wideband with Q 50.
Last I heard, he had measured less than 50 microNewtons (what you call mosquito force) for several hundred watts of power, with a Q=50, in a partial vacuum

Less than 1/1,000 of what Prof. Yang reported as Force/InputPower
« Last Edit: 07/20/2015 05:04 PM by Rodal »

Offline deltaMass

  • Full Member
  • ****
  • Posts: 955
  • A Brit in California
  • Liked: 671
  • Likes Given: 275
Some time back I sent EW a paper on maintaining active tune resonance using a microwave PLL. I don't imagine anyone here has the resources for that, though. Which is why using very high Q is going to be problematic.

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2165
  • Liked: 2681
  • Likes Given: 1124
Interesting, a Q of 50 would translate roughly into the total frequency spread of a typical magnetron; about 49 MHz. This would avoid the need for a tuner either on the frustum or the magnetron. Wonder if this is telling us something...

Q=50 * P=1,000W wideband magnetron input = 50kW equivalent field strength inside the cavity

Q=50,000 * P=100W narrow band input = 5MW equivalent field strength inside the cavity (100x greater)

Using F = (2 P Df Q) / c, Force generation will be 100x greater at 100W narrow band with Q 50,000 than at 1,000W wideband with Q 50.
Last I heard, he had measured less than 50 microNewtons (what you call mosquito force) for several hundred watts of power, with a Q=50, in a partial vacuum

0.005098581064889641 gram force at the expense of several hundred watts...hmmm

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
Interesting, a Q of 50 would translate roughly into the total frequency spread of a typical magnetron; about 49 MHz. This would avoid the need for a tuner either on the frustum or the magnetron. Wonder if this is telling us something...

Q=50 * P=1,000W wideband magnetron input = 50kW equivalent field strength inside the cavity

Q=50,000 * P=100W narrow band input = 5MW equivalent field strength inside the cavity (100x greater)

Using F = (2 P Df Q) / c, Force generation will be 100x greater at 100W narrow band with Q 50,000 than at 1,000W wideband with Q 50.
Last I heard, he had measured less than 50 microNewtons (what you call mosquito force) for several hundred watts of power, with a Q=50, in a partial vacuum

0.005098581064889641 gram force at the expense of several hundred watts...hmmm

annus veritas for the EM Drive
« Last Edit: 07/20/2015 05:06 PM by Rodal »

Offline TheTraveller

Are Photons Degraded after they:

1) bounce off a non moving end plate, EMDrive in Idle mode?

2) bounce off a moving end plate, EMDrive in Motor mode?

Assuming the Photon hitting the moving end plate is 1st absorbed by the end plate atoms and then re emitted, is the re emitted Photon's wavelength longer (Red Shifted) to account for the lost momentum at absorption and then again at remittance?

Is Compton scattering in action inside an accelerating EMDrive? https://en.wikipedia.org/wiki/Compton_scattering

Interesting discussion here:
http://www.thenakedscientists.com/forum/index.php?topic=41821.0
« Last Edit: 07/20/2015 05:15 PM by TheTraveller »
"As for me, I am tormented with an everlasting itch for things remote. I love to sail forbidden seas.”
Herman Melville, Moby Dick

Offline TheTraveller

Interesting, a Q of 50 would translate roughly into the total frequency spread of a typical magnetron; about 49 MHz. This would avoid the need for a tuner either on the frustum or the magnetron. Wonder if this is telling us something...

Q=50 * P=1,000W wideband magnetron input = 50kW equivalent field strength inside the cavity

Q=50,000 * P=100W narrow band input = 5MW equivalent field strength inside the cavity (100x greater)

Using F = (2 P Df Q) / c, Force generation will be 100x greater at 100W narrow band with Q 50,000 than at 1,000W wideband with Q 50.
Last I heard, he had measured less than 50 microNewtons (what you call mosquito force) for several hundred watts of power, with a Q=50, in a partial vacuum

0.005098581064889641 gram force at the expense of several hundred watts...hmmm

annus veritas for the EM Drive

Would be good to see the actual experimental data, what the build / frustum was like and how it was tested before making statements? There are many effects to be accounted for and just maybe this was not a good design.
"As for me, I am tormented with an everlasting itch for things remote. I love to sail forbidden seas.”
Herman Melville, Moby Dick

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
Annus Veritas means the year in which we will find out the truth:

1) You are the one that claimed "all doubts will be removed" this year (you referring to the re-publication of Shawyer's 2014 Conference paper)

2) With the great testing set-up of rfmwguy, SeeShells and others, I have great expectations that we will learn a lot from rfmwguy, SeeShells  and others.

3) You have claimed that you will be reporting on tests too

4) With Prof. Tajmar reporting tests in a vacuum we will now have two prestigious institutions (NASA and TU Dresden, Germany) reporting on EM Drive tests in vacuum (something that neither Shawyer or Yang ever reported: not a single test in vacuum)

5) Hopefully NASA will report later in the year as to the progress of their tests?

So I expect 2015 to be an important year to learn about the EM Drive, yes, "Annus Veritas"
« Last Edit: 07/20/2015 05:34 PM by Rodal »

Tags: