Author Topic: EM Drive Developments - related to space flight applications - Thread 3  (Read 1879516 times)

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
...OK, you've got mine at 11.01 x 6.25 x 9.91"L as discussed with you and Traveller.

OK, those dimensions are actually closer to NASA's and Iulian Berca's truncated cone geometry, and significantly different from Prof. Yang's dimensions (when taking into consideration the cone angle and the distance to the vertex of the cone).  The big and the small diameters are same as NASA's, the length is closest to Yang's.


(NASA has the following internal copper surface dimensions.Large OD : 11.00 " (0.2794m), Small OD: 6.25" (0.1588 m) & Length : 9.00 " (0.2286m) )




rfmwguy Dimensions

axial length = 0.252 meters = 9.91 inches
big diameter = 0.280 meters = 11.01 inches   [same as NASA]
small diameter = 0.159 meters =  6.25 inches  [same as NASA]

gives

r1= 0.340 meters

r2=0.599 meters

half cone angle = 13.5 degrees

See the dimensions of the other EM Drives here:  http://emdrive.wiki/Experimental_Results
Re. NASA dimensions. I am confused. You state that the measurements are internal dimensions then in the same breath say that they are OD. OD normally stands for Outer Diameter. I know that the dielectric disks used are 6.25 inches, OD, which makes the small diameter of the cavity >= 6.25 inches, ID. The others, I'm not sure as I'm not sure where you are coming up with the numbers, but I think it is the origonal Brady cavity. The dimensions of the NASA copper kettle cavity are a little different, given in the attached, as provided by Paul March.

The quote is direct from @Star-Drive (March, NASA Eagleworks) in the following message:

http://forum.nasaspaceflight.com/index.php?topic=36313.msg1326997#msg1326997

where he wrote:

Quote from: Star-Drive
The copper frustum we built and now are using has the following internal copper surface dimensions.
Large OD: 11.00" (0.2794m), Small OD: 6.25" (0.1588m) & Length: 9.00" (0.2286m) 

Notice that he writes <<internal copper surface dimensions>> as well as <<OD>>.

This may have been a mistake, or this may be telling us that the differences between ID and OD and the differences between 11.00" and 11.01" are entirely within the roundness and concentricity tolerances of their cavity because it was homemade in his living room, and not accurately machined.

That's why I don't agree with seeking super accuracy in frequency determination, providing numbers with many digits beyond geometrical tolerances.

I previously copied and pasted the number of output digits in my file, I'll go back and change them to 3 digits to reflect this.
« Last Edit: 06/20/2015 08:56 PM by Rodal »

Offline WarpTech

  • Full Member
  • ****
  • Posts: 1313
  • Do it!
  • Vista, CA
  • Liked: 1351
  • Likes Given: 1818
<snip>
The amplification could be due to a nonlinear effect that Todd has not taken into account.  Masers (and Lasers) involve amplification.  A solid state room-temperature Maser was invented during the last two decades but it is one meter long and it involves a special material to help amplification through emmission.

I have not calculated the combined effects of TE011, TE012 and TE111, and/or the effects of coupling with a DC component.

I can't calculate any thrust without mass/energy escaping the frustum somehow.  That's what bothers me the most.  I can't see how can the center of mass of an object be accelerated purely from the inside without expelling mass/energy to produce the thrust.

One thing I did find that "could" serve as an unexpected amplifier. Inside the frustum, the guide frequency is given by;

ωg = ω0*√(1 - (ωc / ω0)2)

Similarly for kg

if energy stored inside is ~ωg and if  ωc / ω0 ~ 1,

Then when that energy is released to the outside environment, it will receive a "boost" from;

1/√(1 - (ωc / ω0)2), as ωg -> ω0.

which can be substantial. It means, we are underestimating the Q inside at the guide frequency.

On the other hand, if all the energy is trapped inside and nothing gets out, this does nothing to boost 0-thrust.
Todd

Offline kitsuac

  • Member
  • Posts: 10
  • NW Ohio
  • Liked: 9
  • Likes Given: 4
Isn't group "velocity" sort of a misnomer, like waving a laser beam across the surface of the moon and saying that it's physically moving at a superluminal velocity? Just seems like a level of abstraction which is only meaningful in particular constrained situations, basing calculations on it seems very prone to yielding illegitimate conclusions.
« Last Edit: 06/20/2015 08:51 PM by kitsuac »

Online sghill

  • Full Member
  • ****
  • Posts: 1312
  • United States
  • Liked: 1440
  • Likes Given: 2019
Seems to me that if the thicknesses of the copper walls are varied, but the other dimensions remain the same, and thrust also differs between the models, then we know the drive is interacting with the outside environment somehow. If thrust does not vary between models of different wall thickness, then the thrust is occurring in-house- so to speak.
Bring the thunder Elon!

Offline deltaMass

  • Full Member
  • ****
  • Posts: 955
  • A Brit in California
  • Liked: 671
  • Likes Given: 275
Seems to me that if the thicknesses of the copper walls are varied, but the other dimensions remain the same, and thrust also differs between the models, then we know the drive is interacting with the outside environment somehow. If thrust does not vary between models of different wall thickness, then the thrust is occurring in-house- so to speak.
Cart before the horse? - if you will excuse me saying so.

To date we do not even have at least two independent laboratories using identical devices producing identical thrust to within experimental tolerances.

That for me is a prerequisite for declaring that there is any thrust at all.

Offline X_RaY

  • Full Member
  • ****
  • Posts: 778
  • Germany
  • Liked: 1001
  • Likes Given: 2195
Would it be useful to generate a geometrie like this in the pictures to multiply the effects of the force? I know its difficult without the definitiv correct math for the trust but i think it can help to maximize the effects of the frustrated cone... 
In the 24GHz case one can make easily  :)
« Last Edit: 06/20/2015 10:16 PM by X_RaY »

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
Seems to me that if the thicknesses of the copper walls are varied, but the other dimensions remain the same, and thrust also differs between the models, then we know the drive is interacting with the outside environment somehow. If thrust does not vary between models of different wall thickness, then the thrust is occurring in-house- so to speak.
Cart before the horse? - if you will excuse me saying so.

To date we do not even have at least two independent laboratories using identical devices producing identical thrust to within experimental tolerances.

That for me is a prerequisite for declaring that there is any thrust at all.

The geometrical dimensions of these two independent devices are identical.
Identical mode shape.
The force/PowerInput not rigorously the same (extreme rigor leads to "rigor mortis") but these two are very comparable [considering that reported Force/InputPower ranges from 0 to 1000 for other researcher tests]:

Description       Mode           Pressure    Length (m) Db (m)      Ds(m)    mN/kW

NASA Eagleworks  TM212      Ambient   0.2286   0.2794   0.15875   3.00

Iulian Berca          TM212       Ambient   0.2286   0.2794   0.1588    2.80


Berca Tests 3 & 3.1 (averaged w/up/down directional effects subtracted as per you, deltaMass -so you can't object to those numbers :)  )

Iulian's Berca Lab is very independent from NASA.

NASA Brady, White, March, Lawrence, and Davies, b

Pretty good agreement between these two (certainly within experimental tolerance).

Iulian Berca did not make noise, but he quietly humbly went about his business, the first independent researcher to produce results and his results (in force/PowerInput) are very close to NASA's.
« Last Edit: 06/20/2015 11:04 PM by Rodal »

Offline aero

  • Senior Member
  • *****
  • Posts: 2786
  • 92129
  • Liked: 724
  • Likes Given: 249
In my Meep runs with Q in the millions, I was making a fundemental error. I was chasing resonance at a frequency near 2.45GHz, and ended up at (set! fsi 2.46316014E+009). In doing this I continually narrowed the noise bandwidth, getting as low as 73.5 MHz and finally settling on a bandwidth of 246 MHz and a length of 0.232 meters (not .24 meters; To keep resonance close to 2.45 GHz). Of course I forgot to mention this in my post. So this is not really Yang's model although I don't know what the bandwidth of Yang's magnetron source should be.

I've made some more runs and with the bandwidth set somewhere between 500 MHz and 540 MHz, the Q value should be what we expect. (5,000, isn't it?)

But that's not what I want to post! Something interesting happened when I decided to quit fooling with resonance and look at the field patterns! The attached image is generated using the new copper model that we developed here on NSF in the last few days. There is energy outside of the frustum! Now we need to be very careful and certain the the copper model is correct for 2.4 GHz frequency.

If the copper model is correct, this image becomes a significant piece of data.
Retired, working interesting problems

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
In my Meep runs with Q in the millions, I was making a fundemental error. I was chasing resonance at a frequency near 2.45GHz, and ended up at (set! fsi 2.46316014E+009). In doing this I continually narrowed the noise bandwidth, getting as low as 73.5 MHz and finally settling on a bandwidth of 246 MHz and a length of 0.232 meters (not .24 meters; To keep resonance close to 2.45 GHz). Of course I forgot to mention this in my post. So this is not really Yang's model although I don't know what the bandwidth of Yang's magnetron source should be.

I've made some more runs and with the bandwidth set somewhere between 500 MHz and 540 MHz, the Q value should be what we expect. (5,000, isn't it?)

But that's not what I want to post! Something interesting happened when I decided to quit fooling with resonance and look at the field patterns! The attached image is generated using the new copper model that we developed here on NSF in the last few days. There is energy outside of the frustum! Now we need to be very careful and certain the the copper model is correct for 2.4 GHz frequency.

If the copper model is correct, this image becomes a significant piece of data.
Yes, Q=5000 is a reasonable value

Thanks for the update.

Please tell us more info as to how you find that there is energy outside the frustum.  First of all, do you have a Finite Difference mesh outside the frustum? (I thought you didn't)

If you don't have a mesh outside the frustum, how can you tell there is energy outside the frustum?

How does the energy leak to the outside?

Is your model still using HarmInv (basically only solving for frequencies) or is it a Time-Marching solution that enables you to monitor the time variation of the fields?
« Last Edit: 06/20/2015 11:42 PM by Rodal »

Offline aero

  • Senior Member
  • *****
  • Posts: 2786
  • 92129
  • Liked: 724
  • Likes Given: 249
Its a Time-Marching solution. The image shows the center slice of computational lattice at the end of the run, 16 peroids of the drive frequency, (2.45 GHz). The outer boundary of the .jpg image is the outer boundary of the computational lattice, or mesh, which is a cuboid. You can see the frustum boundaries within the mesh. The color represents the energy at that point within the mesh. It looks like there is color outside the frustum boundaries. I will make more runs and attempt to get some color discrimination within the frustum.
Retired, working interesting problems

Offline KittyMoo

  • Member
  • Posts: 23
  • Liked: 8
  • Likes Given: 33
Have lost hope.
IMHO what was promising has turned into dust.

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5895
  • USA
  • Liked: 6045
  • Likes Given: 5325
Have lost hope.
IMHO what was promising has turned into dust.


Quote from: Albert Einstein
Learn from yesterday, live for today, hope for tomorrow. The important thing is not to stop questioning.


Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2165
  • Liked: 2681
  • Likes Given: 1124
Have lost hope.
IMHO what was promising has turned into dust.

With so little test data, way too early for that opinion...lots of unfunded experiments going on...patience is paramount.

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2165
  • Liked: 2681
  • Likes Given: 1124
weight budget is 1.5 kg. amp below is not half that. exciter only 1.2 g.
« Last Edit: 06/21/2015 02:39 AM by rfmwguy »

Offline KittyMoo

  • Member
  • Posts: 23
  • Liked: 8
  • Likes Given: 33
You say there is little test data...
I say there is NONE past EW.
To this point we have EW...
Shaywer and the Chinese are mistaken. This is what Occam says and after much ado appears to be correct.
I see the hint from old timers here that I may be correct.

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2165
  • Liked: 2681
  • Likes Given: 1124
You say there is little test data...
I say there is NONE past EW.
To this point we have EW...
Shaywer and the Chinese are mistaken. This is what Occam says and after much ado appears to be correct.
I see the hint from old timers here that I may be correct.

You'd have to claim julian had null data. I'm not in a position to dismiss his results. Maybe u know something we do not. Regardless, my build goes on with an open mind, as others do. If mine are null, will have zero reservations in stating so.

We need data and math theories. Your contributions would be welcomed...the group here is apolitical for the most part.

Offline KittyMoo

  • Member
  • Posts: 23
  • Liked: 8
  • Likes Given: 33
What solid data did Julien garner? None.
Where is he now?
I disagree (agree?) we need more controlled experimental results, not more theory.

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2165
  • Liked: 2681
  • Likes Given: 1124
What solid data did Julien garner? None.
Where is he now?
I disagree (agree?) we need more controlled experimental results, not more theory.

Julian Berca on fb...new job in china. I also agree more experimental data needed. Was skeptical of julians data until downward momentum measured. Upwards could have been ballooning, downward is something very different. Did not appear to be falsified. May not have had lab standard accuracies...think relative accuracy shows something. What exactly...am not sure, so my curiosity remains.

Offline deltaMass

  • Full Member
  • ****
  • Posts: 955
  • A Brit in California
  • Liked: 671
  • Likes Given: 275
Not so much "ballooning" as a change in the contained mass of air via a temperature increase at constant volume.

Offline KittyMoo

  • Member
  • Posts: 23
  • Liked: 8
  • Likes Given: 33
Re New job in China
That right there should flash red lights. It is simple to end a fake FB account with that explanation.
I wonder how many other people will produce videos and then disappear?

EDIT: I include Shawyer in that
« Last Edit: 06/21/2015 03:50 AM by KittyMoo »

Tags: