Author Topic: EM Drive Developments - related to space flight applications - Thread 3  (Read 1804257 times)

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2326
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 2956
  • Likes Given: 2589

////

QUESTION: Concerning your rotary test rig, are you planning to use an air bearing set-up, like Shawyer's ?
========================

As for the test rig, will use a DIY magnetic thrust bearing as the max load will be around 10kg as against SPR's 100kg test load. Will have 1,000s of output pulses per rev to enable easy accel calcs. All data logged. Will not have adjustable load as such but can vary the mass to get different accel as per A = F/M.

When SPR did this they found the KE transferred to the 100kg load mass was what which was drawn from the power supply, adjusting for losses.

As SPR have not yet published this data, my tests may be the 1st to show the KE gain by the load is sourced from the power supply, which eliminates ZPE as the energy source. Can't do this with scales.

I need to stop as phone battery is almost dead and the nurse will be around any minute. Will be good to be home again.
From one DIYer to another....
I'm glad to see you'll be going home and are getting things taken care of. Take it easy, don't overdo it, but I'm sure your wife will make sure of that. ;)
Shell


Online Rodal

  • Senior Member
  • *****
  • Posts: 5838
  • USA
  • Liked: 5919
  • Likes Given: 5263
As a means of comparison, this is an early movie (attached below with title "2.1-cone-out2.1" as a mp4 attachment - do NOT confuse with the YouTube movie shown as reference) that was posted by aero of NASA's EM Drive with a dielectric insert.  The Meep model in this movie is 2-D instead of 3-D.  Observe that it converges to a p=3 mode which was not observed in NASA' experiments (they only observed TM212 and TE012 both of which have p=2 insted of p=3).

The interesting thing that I would like to point out is that this movie shows:

1) The fields settle into standing waves, with a sinusoidal--in-time variation.  The behavior is as predicted by standard resonance in a cavity with standing waves fixed in space, instead of the always changing fluctuating steady state with time-asymmetry shown in the recent 3D modeling without a dielectric insert.

2) One can also clearly observe that although the EM Drive being modeled has flat ends, the wave field wants to settle into a spherical wave pattern in its interior, as it should be according to resonant standing waves.

So, it is clear that Meep can predict a standing wave field inside the EM Drive or a fluctuating, time-asymmetric field, depending on the Meep model.

Compare the attached movie mp4 movie using a 2-D model, to the behavior of Ex -y on this 3-D model on the YouTube here :

https://www.youtube.com/watch?t=26&v=Cm9Nl-x1hj4
« Last Edit: 07/01/2015 03:03 PM by Rodal »

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2165
  • Liked: 2681
  • Likes Given: 1124
Good job, Shell...this helix has high power capability and appears to be a double helix, just like DNA, so if it doesn't provide thrust, it might become a new life form ;)

Welcome back Mr. T. slow and steady as she goes.

Aero and all, very impressed with the NSF-1701 analysis. Need a summary: is 10.2L looking less promising than 9.0L? Initial images appear to be. I can still change the exoskelelton to 9.0L easily now, but the new copper support bars will arrive today and need to have a consensus from the braintrusts here: stick with 1701@10.2L or go to 1701A@9.0L? I have no issues either way.

Onward and upward...

Online aero

  • Senior Member
  • *****
  • Posts: 2744
  • 92129
  • Liked: 705
  • Likes Given: 239
Good job, Shell...this helix has high power capability and appears to be a double helix, just like DNA, so if it doesn't provide thrust, it might become a new life form ;)

Welcome back Mr. T. slow and steady as she goes.

Aero and all, very impressed with the NSF-1701 analysis. Need a summary: is 10.2L looking less promising than 9.0L? Initial images appear to be. I can still change the exoskelelton to 9.0L easily now, but the new copper support bars will arrive today and need to have a consensus from the braintrusts here: stick with 1701@10.2L or go to 1701A@9.0L? I have no issues either way.

Onward and upward...

You'd need to get Dr. Rodal's input but as far as I can tell, nothing significant has changed. 10.2 inches seems to be the choice. All that has changed is that some csv files have been generated but they don't seem to contain any definitive data, or if they do, it has yet to be extracted and understood. I doubt we will reach any sort of conclusion soon, maybe in a day or two, but not today.

I "vote" for 10.2 inches but it is not much more than a vote.

Added: Progress -------
I currently have generated the 6 EM field files for NSF-1701 (10.2 inches) at higher resolution both for the copper model and perfect metal, so those long runs are complete. I plan to create and upload csv files for both, then field images. This will help Dr. Rodal to understand just what he is seeing as the perfect metal images (from a quick look) appear to behave as expected regarding the cavity boundaries. Haven't looked at the copper yet except to note that it does show differently in the cavity image, without energy added.
« Last Edit: 07/01/2015 04:17 PM by aero »
Retired, working interesting problems

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2326
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 2956
  • Likes Given: 2589
Good job, Shell...this helix has high power capability and appears to be a double helix, just like DNA, so if it doesn't provide thrust, it might become a new life form ;)

Welcome back Mr. T. slow and steady as she goes.

Aero and all, very impressed with the NSF-1701 analysis. Need a summary: is 10.2L looking less promising than 9.0L? Initial images appear to be. I can still change the exoskelelton to 9.0L easily now, but the new copper support bars will arrive today and need to have a consensus from the braintrusts here: stick with 1701@10.2L or go to 1701A@9.0L? I have no issues either way.

Onward and upward...
I really hope the "new life form isn't DOA" hehehe

I like this helix design because it gives me a beautiful focused symmetrical pattern with some very nice gain.  I'm able to position it directly into the center of the small end and use it as a ground plane. And add a bandwidth that covers the 2.45 Ghz magnetron spectral mix for exciting various modes and I'm happy. The dipoles do ok but you can see from the meep animations the cavities anharmonic response during a phase reversal, I wanted to avoid that. The circular polarization is something I'm also looking for but still doing my pidgin maths (Thanks Elizabeth, I love that word, it truly describes how I feel sometimes).

I still am up in the air on the size too so I heck I'm testing both with one cavity, just change a plate. ;) the inner and outer dimensions of the hexagonal base plate cover both. Plus, I might have figured out to have the cavity do both TM and TE at the same time, and that is a work in progress to see if i can do it or need to. Gads what a EMBlender that would be!

Shell

Online Rodal

  • Senior Member
  • *****
  • Posts: 5838
  • USA
  • Liked: 5919
  • Likes Given: 5263
Since many times posters have inquired as to why we even bother to understand the reported claims behind the EM Drive, I have added the reported experimental results of a conventional electromagnetic thruster using propellant: Ad Astra's VASIMR VX-200 magnetoplasma engine with argon propellant http://www.adastrarocket.com/aarc/research-and-development to the experimental results in http://emdrive.wiki/Experimental_Results for comparison purposes.

This will help people to understand that NASA's present reported results for the (claimed to be propellant-less)  EM Drive force/inputPower are not too far off from the argon-propellant VASIMIR's results, and that Prof. Yang at Northwestern Polytechnical University, College of Aeronautics, in China reported results for her EM Drive have 38 times higher force/InputPower than VASIMIR's (and that, supposedly, is without the need for carrying propellant, which is needed for VASIMIR's case). 





Of course, there is the important question of scaling, as VASIMIR is reporting 5.7 Newtons with 200 kW InputPower, while Prof. Yang reported 0.27 Newtons with 0.3 kW InputPower.

For discussion of VASIMIR, NSF has threads dedicated to this topic, for example: "Why VASIMIR ?"  http://forum.nasaspaceflight.com/index.php?topic=18382.0
« Last Edit: 07/01/2015 05:12 PM by Rodal »

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2326
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 2956
  • Likes Given: 2589
Since many times posters have inquired as to why we even bother to understand the reported claims behind the EM Drive, I have added the reported experimental results of Ad Astra's VASIMR VX-200 magnetoplasma engine with argon propellant http://www.adastrarocket.com/aarc/research-and-development to the experimental results in http://emdrive.wiki/Experimental_Results for comparison purposes.

This will help people to understand that NASA's present reported results for the (claimed to be propellant-less)  EM Drive force/inputPower are not too far off from the argon-propellant VASIMIR's results, and that Prof. Yang at NorthWestern Polytechnic University in China reported results for her EM Drive have 38 times higher force/InputPower than VASIMIR's. 



Of course, there is the important question of scaling, as VASIMIR is reporting 5.7 Newtons with 200 kW InputPower, while Prof. Yang reported 0.27 Newtons with 0.3 kW InputPower.

For discussion of VASIMIR, NSF has a thread dedicated to this topic.
Big differance
1.2814 foot lbs for 200kW on the VASIMIR
vs
Yang's @ 200Kw would be 40.4656 foot lbs

I like cars so forgive me Foot Pounds works a little better for me.

Online Rodal

  • Senior Member
  • *****
  • Posts: 5838
  • USA
  • Liked: 5919
  • Likes Given: 5263

Big differance
1.2814 foot lbs for 200kW on the VASIMIR
vs
Yang's @ 200Kw would be 40.4656 foot lbs

I like cars so forgive me Foot Pounds works a little better for me.
Unfortunately for the EM Drive, Prof. Yang has shown that the force/PowerInput for the EM Drive does not scale linearly beyond that,



so to achieve those numbers at 200 kW would take 666 EM Drive engines of the Yang type (each EM Drive engine giving 0.27 Newtons with 0.3 kW InputPower, for 666 engines giving a total 180 Newtons at 200 kW )



Conversely, it would take twenty-one (21) EM Drive engines of the Yang type, to produce the same total thrust that VASIMIR but only consuming 6 kW instead of 200 kW

« Last Edit: 07/01/2015 05:47 PM by Rodal »

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2326
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 2956
  • Likes Given: 2589

Big differance
1.2814 foot lbs for 200kW on the VASIMIR
vs
Yang's @ 200Kw would be 40.4656 foot lbs

I like cars so forgive me Foot Pounds works a little better for me.
Unfortunately for the EM Drive, Prof. Yang has shown that the force/PowerInput for the EM Drive does not scale linearly beyond that,


I've suspected that they ran into thermal issues with that closed cavity that gave them a non-linear profile the greater the power but... could be wrong.

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2165
  • Liked: 2681
  • Likes Given: 1124
Since many times posters have inquired as to why we even bother to understand the reported claims behind the EM Drive, I have added the reported experimental results of Ad Astra's VASIMR VX-200 magnetoplasma engine with argon propellant http://www.adastrarocket.com/aarc/research-and-development to the experimental results in http://emdrive.wiki/Experimental_Results for comparison purposes.

This will help people to understand that NASA's present reported results for the (claimed to be propellant-less)  EM Drive force/inputPower are not too far off from the argon-propellant VASIMIR's results, and that Prof. Yang at NorthWestern Polytechnic University in China reported results for her EM Drive have 38 times higher force/InputPower than VASIMIR's. 

...

Of course, there is the important question of scaling, as VASIMIR is reporting 5.7 Newtons with 200 kW InputPower, while Prof. Yang reported 0.27 Newtons with 0.3 kW InputPower.

For discussion of VASIMIR, NSF has a thread dedicated to this topic.
Big differance
1.2814 foot lbs for 200kW on the VASIMIR
vs
Yang's @ 200Kw would be 40.4656 foot lbs

I like cars so forgive me Foot Pounds works a little better for me.
I've lived in technical marketing for so many years, that I cannot help but comment...IMO, the propulsion industry has failed to have enough breakthrough innovations that it has become relatively stagnant (buying russian rocket motors) but probably still very profitable. This is not uncommon in many industries, but what may be going on in our area of interest is better "mousetraps" trying to move past theory and into repeatable demonstration mode. This HAS to be a concern to big players, not involved in "speculative" R&D. Ion engines seem to have taken the early steps and hopefully, other technologies will catch up and surpass. After all, if one considers "gravity assist" as propulsion, then we have a long way to go to reach the stars. Perhaps we never can...I choose not to subscribe to that defeatest mentality.  8)

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2326
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 2956
  • Likes Given: 2589

Offline tchernik

  • Full Member
  • **
  • Posts: 247
  • Liked: 297
  • Likes Given: 577
http://captiongenerator.com/48295/Hitler-Reacts-to-current-EmDrive-Situation

That was both hilarious and poignant.

But when has science really at the limits of knowledge not been fraught with both joy and disappointment? And work, lots of sometimes grueling, ungrateful work.

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2326
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 2956
  • Likes Given: 2589
http://captiongenerator.com/48295/Hitler-Reacts-to-current-EmDrive-Situation

That was both hilarious and poignant.

But when has science really at the limits of knowledge not been fraught with both joy and disappointment? And work, lots of sometimes grueling, ungrateful work.
So right tchernik, a good laugh is a great reset and makes the effort a little easier.

Online Rodal

  • Senior Member
  • *****
  • Posts: 5838
  • USA
  • Liked: 5919
  • Likes Given: 5263
http://captiongenerator.com/48295/Hitler-Reacts-to-current-EmDrive-Situation
Hilarious,  rolling on the floor laughing :)

Whoever put this together, knows this NSF EM Drive thread very well...

The video says "YouTube" but I could not find it on YouTube.  I wonder why...
« Last Edit: 07/01/2015 06:17 PM by Rodal »

Online aero

  • Senior Member
  • *****
  • Posts: 2744
  • 92129
  • Liked: 705
  • Likes Given: 239
@Dr. Rodal
I think I have found what I was doing wrong, maybe not all of it, but one thing for sure. Here's what I did.
- - - Increased resolution by 2.5 times
- - - Replaced copper with perfect metal
- - - Looked at the csv file of the big end base. Same view as I have uploaded
- - - Made the same run using Copper model instead of the perfect metal.
- - - Looked at the corresponding csv file.

By noting that the Copper model csv file showed energies on the order of 10-30 (smaller than before due to increased resolution) while the perfect metal showed exact zero energies throughout, I determined that my x-slice was actually outside of the cavity. That is why you don't see the cavity in the uploaded data. All of the previously uploaded data is wrong and will be removed.

I have attached the perfect metal csv file, as .txt, with the x-slice moved to inside the cavity. (the forum won't attach files with .csv extensions. (same with .sh extensions))

I need to make 72 more runs of h5totxt and h5topng. Can someone help me by converting the attached list of commands into a BASH shell command file. Running those commands manually at the terminal is challenging, exhausting and very prone to errors and I haven't bothered to learn BASH since moving to Ubuntu from Windows a few months ago. Given one working command file, I can likely make the rest myself, and maybe learn to recognize a BASH command file in the process.

Retired, working interesting problems

Online Rodal

  • Senior Member
  • *****
  • Posts: 5838
  • USA
  • Liked: 5919
  • Likes Given: 5263
@Dr. Rodal
I think I have found what I was doing wrong, maybe not all of it, but one thing for sure. Here's what I did.
- - - Increased resolution by 2.5 times
- - - Replaced copper with perfect metal
- - - Looked at the csv file of the big end base. Same view as I have uploaded
- - - Made the same run using Copper model instead of the perfect metal.
- - - Looked at the corresponding csv file.

By noting that the Copper model csv file showed energies on the order of 10-30 (smaller than before due to increased resolution) while the perfect metal showed exact zero energies throughout, I determined that my x-slice was actually outside of the cavity. That is why you don't see the cavity in the uploaded data. All of the previously uploaded data is wrong and will be removed.

I have attached the perfect metal csv file, as .txt, with the x-slice moved to inside the cavity. (the forum won't attach files with .csv extensions. (same with .sh extensions))

I need to make 72 more runs of h5totxt and h5topng. Can someone help me by converting the attached list of commands into a BASH shell command file. Running those commands manually at the terminal is challenging, exhausting and very prone to errors and I haven't bothered to learn BASH since moving to Ubuntu from Windows a few months ago. Given one working command file, I can likely make the rest myself, and maybe learn to recognize a BASH command file in the process.

Thanks so much for doing this, like a true hero you have apparently solved the problem in a quiet way.   Warm congratulations, as being able to output those csv files would be a giant step forward for people to:

1) Do their own postprocessing (calculating fields at diferent angles: 45 degrees to the axis, etc., calculating Poynting Vector field, etc etc)

2) Plotting Vector Fields, plotting 3D views, etc etc

When you are finally able to output revised csv files, would appreciate also having some trapezium (x z and x y plane views).
« Last Edit: 07/01/2015 06:22 PM by Rodal »

Online Rodal

  • Senior Member
  • *****
  • Posts: 5838
  • USA
  • Liked: 5919
  • Likes Given: 5263
Yes, I plan to generate and upload the full set of 18 views and csv data files.

here it is !!!!!!!!!!!!
« Last Edit: 07/01/2015 06:44 PM by Rodal »

Online aero

  • Senior Member
  • *****
  • Posts: 2744
  • 92129
  • Liked: 705
  • Likes Given: 239
Yes, I plan to generate and upload the full set of 18 views and csv data files.

It sure would help though if I had the BASH shell command file with which to make the runs. I know that someone or several people reading this knows how to use BASH to make it so. My problem is of course, like almost all software problems, that I need it now. And if I get multiple responses, that won't bother me a bit. Thank all in advance.
Retired, working interesting problems

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2326
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 2956
  • Likes Given: 2589
Yes, I plan to generate and upload the full set of 18 views and csv data files.

here it is !!!!!!!!!!!!
Oh. My. God.

Offline deuteragenie

  • Member
  • Posts: 71
  • Germany
  • Liked: 22
  • Likes Given: 0
As a means of comparison, this is an early movie (attached below with title "2.1-cone-out2.1" as a mp4 attachment - do NOT confuse with the YouTube movie shown as reference) that was posted by aero of NASA's EM Drive with a dielectric insert.  The Meep model in this movie is 2-D instead of 3-D.  Observe that it converges to a p=3 mode which was not observed in NASA' experiments (they only observed TM212 and TE012 both of which have p=2 insted of p=3).

The interesting thing that I would like to point out is that this movie shows:

1) The fields settle into standing waves, with a sinusoidal--in-time variation.  The behavior is as predicted by standard resonance in a cavity with standing waves fixed in space, instead of the always changing fluctuating steady state with time-asymmetry shown in the recent 3D modeling without a dielectric insert.

2) One can also clearly observe that although the EM Drive being modeled has flat ends, the wave field wants to settle into a spherical wave pattern in its interior, as it should be according to resonant standing waves.

So, it is clear that Meep can predict a standing wave field inside the EM Drive or a fluctuating, time-asymmetric field, depending on the Meep model.

Compare the attached movie mp4 movie using a 2-D model, to the behavior of Ex -y on this 3-D model on the YouTube here :

https://www.youtube.com/watch?t=26&v=Cm9Nl-x1hj4

Congratulations to Stephen: this is a very well done video.
After a quadrillion of dependency downloads and re-compiles, I am close to having the latest Meep source building completely on Linux... In fact it does build successfully already but is not able to locate the .h5 library, for unknown reasons.

Tags: