Author Topic: EM Drive Developments - related to space flight applications - Thread 2  (Read 2165401 times)

Offline Rodal

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

Did you say you had an exact analytic solution for the truncated cavity ?
Yes, and I am working on an exact solution including the dielectric as well.

I obtained a closed-form solution for the integral of the wavenumber kz.  The expression contains square root terms and ArcTan terms.  Because of this one cannot invert the expression for the eigenvalue problem directly in terms of frequency and one has to solve the eigenvalue problem using a root-finding equation like Newton-Raphson.  (Similarly to what one has to do to solve eigenvalue problems in vibrations of beams and shells).

The inclusion of the dielectric is easy because the dielectric properties are constant.

Cutoff frequency conditions do not need to be added.  The cutoff is evident because real value solutions are only possible for high enough number of "p" (the longitudinal direction quantum number).

The solution produces evanescent modes (for complex values of the solution of the eigenvalue problem) and conventional resonant modes (TE or TM) for real solutions of the eigenvalue problem.

I have yet to compare my solution with the ones obtained by the Chinese (see: http://forum.nasaspaceflight.com/index.php?action=dlattach;topic=36313.0;attach=634645  ) but I have compared it to the cylindrical cavity and have established that the GeometricMean of the diameters plugged into the cylndrical cavity gives frequencies that are less than 1.5% from the truncated cone, for cone half-angle less than 20 degrees. 

« Last Edit: 02/08/2015 10:06 PM by Rodal »

Offline Notsosureofit

  • Full Member
  • ****
  • Posts: 656
  • Liked: 704
  • Likes Given: 1367
@ RODAL

Can you put it in terms of a dispersion relation that can be evaluated at the ends ?

Offline Rodal

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

Can you put it in terms of a dispersion relation that can be evaluated at the ends ?

Concerning your question, the non-dimensional parameters in the solution are:


1) fmnp * Length / cmedium    (where cmedium is the speed of light in the medium, therefore this non-dimensional parameter is the ratio of the length to the wavelength)

2) either Xmn/Pi  or X' mn/Pi (for TM and TE modes respectively)

3) (Dbig - Dsmall)/Length (the tangent of the half angle of the cone)

4) Dsmall/Length (the ratio of the small diameter to the length)



My plan:

1) work out the expression including the dielectric
2) compute and compare to NASA experiment, post here the results and the expression
3) compare to Egan and to the Chinese

 
« Last Edit: 02/08/2015 10:41 PM by Rodal »

Offline Notsosureofit

  • Full Member
  • ****
  • Posts: 656
  • Liked: 704
  • Likes Given: 1367
@  RODAL

That should keep you busy !  (but worth it)

Offline Rodal

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

That should keep you busy !  (but worth it)

Notice that my solution is in terms of Xmn and X'mn.  Rigorously speaking the solution should be in terms of spherical Bessel functions of the third kind.  It should not make much difference for truncated cones that are close enough to a cylinder, such that the ends are practically flat.  Basically it comes down again as to whether the truncated cone has flat ends or spherical ends.


NASA and Shawyer (initially) used flat ends instead of spherical ends.  Shawyer is using spherical ends for his superconducting version
« Last Edit: 02/08/2015 10:13 PM by Rodal »

Offline Notsosureofit

  • Full Member
  • ****
  • Posts: 656
  • Liked: 704
  • Likes Given: 1367
@ RODAL

Do you have the dielectric constants (@ freq) and the dimensions ?


Offline Rodal

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

Do you have the dielectric constants (@ freq) and the dimensions ?

Here are the dimensions supplied by Paul March http://forum.nasaspaceflight.com/index.php?action=dlattach;topic=36313.0;attach=634621

Concerning the dielectric constant, thanks for reminding me to ask Paul, see my message below
« Last Edit: 02/08/2015 10:39 PM by Rodal »

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
...
Best, Paul M.

Paul : did NASA (or an outside lab) measure with a dielectrometer the dielectric constant and the tan delta of the High Density PE you used for the dielectric  ?

Otherwise, do you have the relative electrical permittivity and the tan delta of the HD PE at a frequency of ~ 2 GHz at the temperatures experienced during the experiment, from some other source ?

If not, do you have the molecular weight of the HD PE used for the dielectric or some form of further characterization from the dielectric supplier ?

Thanks
« Last Edit: 02/08/2015 10:38 PM by Rodal »

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1116
  • Liked: 779
  • Likes Given: 1030
Still going to have to test it on a ballistic pendulum. http://er.jsc.nasa.gov/seh/Sgoddard.htm
« Last Edit: 02/08/2015 11:46 PM by Mulletron »
Challenge your preconceptions, or they will challenge you. - Velik

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1116
  • Liked: 779
  • Likes Given: 1030
Just want to put this out there in case it helps. During all the research of literature concerning QV momentum, these two materials came up again and again. Cr2O3 and FeGaO3. Or if that is too high speed, PTFE or PVDF. To get over the 100uN hump, it might be time to dump the PE and try something else.

« Last Edit: 02/09/2015 01:30 AM by Mulletron »
Challenge your preconceptions, or they will challenge you. - Velik

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
Just want to put this out there in case it helps. During all the research of literature concerning QV momentum, these two materials came up again and again. Cr2O3 and FeGaO3. Or if that is too high speed, PTFE or PVDF. To get over the 100uN hump, it might be time to dump the PE and try something else.
Or if that doesn't work...
Why should there be a target to achieve a given thrust with a single EMDrive?
Why not just achieve the required thrust at Glenn by using n drives to add up to the required thrust?
For 100uN one needs two EM Drives each providing 50uN, or three EM Drives each providing 33uN and so on
« Last Edit: 02/09/2015 03:02 AM by Rodal »

Offline Star-Drive

  • Member
  • Full Member
  • ****
  • Posts: 829
  • TX/USA
  • Liked: 876
  • Likes Given: 9
...
Best, Paul M.

Paul : did NASA (or an outside lab) measure with a dielectrometer the dielectric constant and the tan delta of the High Density PE you used for the dielectric  ?

Otherwise, do you have the relative electrical permittivity and the tan delta of the HD PE at a frequency of ~ 2 GHz at the temperatures experienced during the experiment, from some other source ?

If not, do you have the molecular weight of the HD PE used for the dielectric or some form of further characterization from the dielectric supplier ?

Thanks

What I used for the HDPE permittivity and loss tangent data was hand book based since we don't have the capability of accurately measuring these parameters in our lab.  So find attached the source for the HDPE we used and the dielectric constants used in our analysis.

BTW, the mass of the HDPE discs was around 486 grams each, which yields a density of 0.947 grams/cc.
Star-Drive

Offline Star-Drive

  • Member
  • Full Member
  • ****
  • Posts: 829
  • TX/USA
  • Liked: 876
  • Likes Given: 9
Just want to put this out there in case it helps. During all the research of literature concerning QV momentum, these two materials came up again and again. Cr2O3 and FeGaO3. Or if that is too high speed, PTFE or PVDF. To get over the 100uN hump, it might be time to dump the PE and try something else.
Or if that doesn't work...
Why should there be a target to achieve a given thrust with a single EMDrive?
Why not just achieve the required thrust at Glenn by using n drives to add up to the required thrust?
For 100uN one needs two EM Drives each providing 50uN, or three EM Drives each providing 33uN and so on


"Why should there be a target to achieve a given thrust with a single EM-Drive? Why not just achieve the required thrust at Glenn by using n drives to add up to the required thrust?"

The month plus of hand labor total provided by three different home workshops required to make each one of these sheet metal based copper frustums.  And we don't have the funding to contract out the duplication of more copper test articles until we prove to our management that this approach is viable.

BTW, the copper frustum is vented, so its internal pressure matches the chamber pressure after a short time period at vacuum.

Best,  Paul M.
Star-Drive

Offline NovaSilisko

  • Full Member
  • ****
  • Posts: 1828
  • Liked: 1435
  • Likes Given: 1305
So... I haven't really done a good long read of this thread in a huge amount of time, since before thread 1 reached a few pages long, even. Would it be possible for someone to produce a 1-2 paragraph summary of what's been going on here over the past few months? Are we... building our own EmDrive, now?
« Last Edit: 02/09/2015 04:00 AM by NovaSilisko »

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1116
  • Liked: 779
  • Likes Given: 1030
Dang! Shawyer didn't provide you a sample!? +10 more respect points for that build. Fantastic job.
Challenge your preconceptions, or they will challenge you. - Velik

Offline sanman

  • Senior Member
  • *****
  • Posts: 3892
  • Liked: 489
  • Likes Given: 7
Is copper the ideal material for the frustum? Or are there even better materials, theoretically?

Offline ThinkerX

  • Full Member
  • ***
  • Posts: 311
  • Alaska
  • Liked: 120
  • Likes Given: 59
Quote
So... I haven't really done a good long read of this thread in a huge amount of time, since before thread 1 reached a few pages long, even. Would it be possible for someone to produce a 1-2 paragraph summary of what's been going on here over the past few months? Are we... building our own EmDrive, now?

This is Doctor Rodal's summary as of page 20 or so of this thread.


Quote
1) We are exploring the validity of the claims of the different experimenters (NASA in the USA, Shawyer in the UK and Juan Yang in China) by systematically analyzing their experiments.

2) Some of the people in this forum are also designing and working to make their own prototypes and experiments.

3) One of the people in this forum (@NotSoSureOfIt), has made an outstanding contribution by independently deriving an equation that is not far from the claimed experimental results.

4) There are a number of possible physical reasons for the experimental results to be valid for space propulsion as well as for the results to be an experimental artifact that may not produce any propulsion in space.

5) There have been no further reports from NASA on the experiments that were supposed to be replicated at their other centers (JPL and Glenn) or at John Hopkins University.  Actually @wembley, who is an aerospace technology reporter, reports that it is his opinion that NASA has a "news blackout" on this matter, and China's Juan Yang is not saying much either.  The Chinese seem to be much further along than NASA, as they reported much greater thrust and they have conducted more thorough experiments (including being the first to numerically report the effects of temperature and temperature gradients with thermocouples embedded in the metal).  Shawyer in the UK seems to be much further ahead than NASA as he claims he is exploring a superconducting design.  Shawyer (UK) made a presentation in Canada late last year where he showed his latest design (using superconductivity) which he claims will result in much greater thrust/PowerInput (see image below).

The principle development since then is the return of 'Star Drive' who actually works with the Eaglework's team and was able to confirm this device does produce thrust in a hard vacuum.  (albeit at a reduced level).  Trying to figure out why this is so is more or less the main topic of discussion at the moment.

Offline francesco nicoli

  • Full Member
  • ****
  • Posts: 493
  • Amsterdam
    • About Crises
  • Liked: 225
  • Likes Given: 331
I forgot to thanks Rodal and the others for having taken the time to explain it to profanes like myself.

So:

Thanks! :)

Offline DIYFAN

  • Member
  • Posts: 48
  • Liked: 28
  • Likes Given: 149
Is copper the ideal material for the frustum? Or are there even better materials, theoretically?

I think maybe it was Mulletron that previously pointed out this paper:

https://hal.archives-ouvertes.fr/hal-01074608/document

The conjecture in that paper is that building the larger diameter plate (R1) out of metglas 2714A, a significant amplification of the effect would result. Perhaps there is a courageous person or group with some funds to put this to the test.  It would be interesting to see an experiment with the partial metglas construction to rule the theory in or out.

Given that metglas 2714A is a room-temperature material, it would be considerably easier to achieve an amplification that way than lining the interior of the test article with superconducting film and cooling to liquid nitrogen temperatures. 

« Last Edit: 02/10/2015 03:14 AM by DIYFAN »

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1116
  • Liked: 779
  • Likes Given: 1030
Is copper the ideal material for the frustum? Or are there even better materials, theoretically?

I think maybe it was Mulletron that previously pointed out this paper:

https://hal.archives-ouvertes.fr/hal-01074608/document

The conjecture in that paper is that building the larger diameter plate (R1) out of metglas 2714A, a significant amplification of the effect would result. Perhaps there is a courageous person or group with some funds to put this to the test.  It would be interesting to see an experiment with the partial metglas construction to rule the theory in or out.

Given that metglas 2714A is a room-temperature material, it would be considerably easier to achieve an amplification that way than lining the interior of the test article with superconducting film and cooling to liquid nitrogen temperatures.

As far as building the cavity itself, copper is good enough. It is easy to get and work with and has great electrical properties. Anything better would be precious metals which isn't worth the money. And superconductors aren't exactly "home workshop" compatible for most.

As far as something in place of the polyethylene dielectric, I'd suggest something that exhibits a higher relative permittivity or even a material that exhibits magnetochiral dichroism. http://www.academia.edu/1084905/Probing_magnetochirality
« Last Edit: 02/10/2015 03:54 AM by Mulletron »
Challenge your preconceptions, or they will challenge you. - Velik

Tags: