Author Topic: EM Drive Developments - related to space flight applications - Thread 3  (Read 1804254 times)

{list type=lower-greek}l m n{/list}

That doesn't work with square brackets.
But I saw a post by A. N. Swallow (because of the god-awful search here I cannot locate his posts) which used Greek "mu".

So how do I shoehorn Greek into posts here?

Its a bit manual, but you could just copy and paste unicode symbols, like from here http://tinyurl.com/2vrnz4q

μ

Edit: The math versions of Greek do not seem to show up on all browsers by default, the regular Greek unicode can be pasted from here; [ http://en.wikipedia.org/wiki/Greek_and_Coptic ]
« Last Edit: 06/08/2015 07:06 PM by zurael »

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2165
  • Liked: 2681
  • Likes Given: 1124
Velocity = roughly constant from t=0 to t=2 sec, and it becomes roughly zero after 2 sec until power is turned off

All good points. Perhaps the system had velocity at T=0 and the power-on condition temporarily halted it. Too much chart snipping in the time domain IYAM. He left himself wide open for critique.

I would have done the test differently. Instead of putting all the weight onto a air bearing I would have suspended from the ceiling from the center point of the test jig. As even a small air bearing would have provided enough air bearing surface and balancing it out. Eliminate a lot of potential errors.

Great minds...etc.,: I had a similar thought in my suggestion here: http://forum.nasaspaceflight.com/index.php?topic=37642.msg1386392#msg1386392

Didn't have a cool drawing, though ;)


Offline deltaMass

  • Full Member
  • ****
  • Posts: 955
  • A Brit in California
  • Liked: 671
  • Likes Given: 275
My comment is that dv/dt = 0 for a photon

It's not a photon, it's a Dirac particle. dv/dt is the acceleration of the group velocity of the particle.
Be careful to formulate your variable mass piece correctly:
http://en.wikipedia.org/wiki/Momentum#Objects_of_variable_mass
Note the correct vs. incorrect forms. It's a matter of using the correct frames

Nevertheless, solving your equation at F=0 one obtains after integration
1/2 (𝛾 v/c)2 = ln(𝜆)

and no, I don't know the correct limits. It appears to blow up as v->0, doesn't it?
« Last Edit: 06/08/2015 07:03 PM by deltaMass »

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2165
  • Liked: 2681
  • Likes Given: 1124
EMDYIers, seems a lot of the RF injection into the frustum has been monopole antenna (magnatron) or coupling loop (EW). Been trying to think past this...I'll start out with a unity gain 1/2 wave monopole, then the loop, then a colinear array which provides about gain up to about 10dB with series 1/2 wave elements. Since this would exceed a linear dimension in the frustum, was thinking perhaps of a spiral arrangement.

Gain is important in my design, since I'm only planing on about +37dbm to start...FWIW: http://www.nodomainname.co.uk/Omnicolinear/2-4collinear.htm

Oops, forgot the helical pic...
« Last Edit: 06/08/2015 07:13 PM by rfmwguy »

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2326
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 2956
  • Likes Given: 2589
Velocity = roughly constant from t=0 to t=2 sec, and it becomes roughly zero after 2 sec until power is turned off

All good points. Perhaps the system had velocity at T=0 and the power-on condition temporarily halted it. Too much chart snipping in the time domain IYAM. He left himself wide open for critique.

I would have done the test differently. Instead of putting all the weight onto a air bearing I would have suspended from the ceiling from the center point of the test jig. As even a small air bearing would have provided enough air bearing surface and balancing it out. Eliminate a lot of potential errors.

Great minds...etc.,: I had a similar thought in my suggestion here: http://forum.nasaspaceflight.com/index.php?topic=37642.msg1386392#msg1386392

Didn't have a cool drawing, though ;)
I like the idea but don't use a white paper stock and a red laser as you'll get a fuzzy image from the fringing patterns from the laser. Use a darker paper. We used that same process to measure some new equipment.

Offline deltaMass

  • Full Member
  • ****
  • Posts: 955
  • A Brit in California
  • Liked: 671
  • Likes Given: 275
@SeeShells: FYI your quotes are usually messed up. Final /quote(s) is missing

Offline Prunesquallor

  • Full Member
  • *
  • Posts: 174
  • Currently, TeV Brane Resident
  • Liked: 157
  • Likes Given: 73
Thoughts for emdrive advanced experimenters: These threads have mentioned cubesat, which is a fixed form factor payload availability. Believe the Aachen group is planning to use it if success is achieved at 26 GHz.

An interesting article on a recent cubesat experiment via crowdfunding and The Planetary Society recently appeared. However, there are rumblings from contributors that its of a lot of money for a very limited duration:

"LightSail's days in orbit are now numbered. The cubesat launched to a relatively low orbit, and atmospheric drag will likely pull it back down to Earth in the next two to 10 days, Planetary Society representatives have said." -
http://www.space.com/29588-lightsail-spacecraft-deploys-solar-sail.html

FWIW, cubesats are low-earth orbit platforms with apparently short lives. Thought the Lightsail could overcome the low-orbit forces and get moving outward...apparently not. Could emdrive? Maybe...maybe not.

This illustrates why I'm not too enthusiastic about an early test flight.  There are ALL kinds of effects in low earth orbit that would have to be accounted for if possibly small effects of a little understood device were trying to be detected.  These include gravity gradients, radiation, atomic oxygen, atmospheric drag, orbital debris, thermal cycling, interference from spacecraft systems, etc. that cannot be shielded against.  I think there would be a far better chance of figuring out what is going on in the lab.
Retired, yet... not

Offline mwvp

  • Full Member
  • **
  • Posts: 254
  • Coincidence? I think Not!
  • Liked: 169
  • Likes Given: 30
EMDYIers, seems a lot of the RF injection into the frustum has been monopole antenna (magnatron) or coupling loop (EW). Been trying to think past this...I'll start out with a unity gain 1/2 wave monopole, then the loop, then a colinear array which provides about gain up to about 10dB with series 1/2 wave elements. Since this would exceed a linear dimension in the frustum, was thinking perhaps of a spiral arrangement.


Not that I have experience designing waveguide couplers, but AFAIK the idea is to couple energy by matching impedance from source to sink.

An antenna is used to match impedance to free space. An E-field probe (or slot) or B-field loop is used to couple to waveguides, and very similarly, cavities.

And with high-Q cavities and filters, you'll want to match a pretty low impedance (~50 ohms) to a very high-Q and impedance. So a very short probe or link, away from the high-field area would be in order.

I don't think putting the magnetron output stub axially, on an end of the frustrum is a good idea, because the end regions may be "hot" (high E-field). Locating it near a node, or low-field, will match better.

Most important, it won't load the cavity which would kill the Q.

It has been noted previously, if you use a loop, you can rotate it to reduce the coupling/impedance/loading.

Offline Prunesquallor

  • Full Member
  • *
  • Posts: 174
  • Currently, TeV Brane Resident
  • Liked: 157
  • Likes Given: 73
Re. the recent flyby reference to the Aachen group's Baby EmDrive and CubeSats, I'm reminded that their team leader has already flown a couple of amateur space missions with an outfit called PoqetQub (from memory).

This is a NASA forum, so presumably packed to the brim with orbital mechanics specialists!! So... what value of k (N/W) is needed to get EmDrive up from LEO, O Experts?

ETA: On reflection that's a dumb question  :-[
Any positive k value will do.

You would have to determine what constitutes an orbit change that is outside the natural decay forces.  Cubesats don't have much power, so they may not get much thrust.  Would a retardation of orbital decay convince anyone?  That is a tricky deal, because orbit decay is sensitive to upper atmosphere expansion/contraction, which is affected by solar activity, etc.

If the thrust was significantly greater than the decay forces, you can use something like the Edelbaum approximation to determine the altitude change you should see with constant, tangential acceleration.  If there is interest, I'll run some quick parametrics to see what that might be.
Retired, yet... not

Offline deltaMass

  • Full Member
  • ****
  • Posts: 955
  • A Brit in California
  • Liked: 671
  • Likes Given: 275
That would be good. Maybe assume 100 Kg all-up EmDrive weight and maybe 100 mN thrust?

p.s. Doctor or Irma?  ;)
« Last Edit: 06/08/2015 08:15 PM by deltaMass »

Offline mwvp

  • Full Member
  • **
  • Posts: 254
  • Coincidence? I think Not!
  • Liked: 169
  • Likes Given: 30
...The paper on quantum ground states by Hal Puthoff is available online and very easy to read, relatively speaking. He derives the equilibrium using Stochastic Electrodynamics
...
Puthoff has a very clear way of explaining things, so does Milonni. I believe this will help most of the people here to understand this, without getting too deep into the Standard Model physics.
...

I like PV and SED, I can kind of understand it and my skull doesn't swell as bad as other stuff.

It would be nice to have a soliton-like model of a particle, a simulation depicting energy oscillating between the trapped EM mode and a non-linear PV around it in rest, motion and acceleration. Then the idea of an EM drive wouldn't seem un-physical. A particle with point charge, wavelength and mass is un-physical and improbable.

Offline Prunesquallor

  • Full Member
  • *
  • Posts: 174
  • Currently, TeV Brane Resident
  • Liked: 157
  • Likes Given: 73
That would be good. Maybe assume 100 Kg all-up EmDrive weight and maybe 100 mN thrust?

Uhh, Cubesats mass about ONE kilogram, 10cmx10cmx10cm.

Edit: PocketQubes are even smaller: 180 grams, 5cmx5cmx5cm.

Dr. :)
« Last Edit: 06/08/2015 08:21 PM by Prunesquallor »
Retired, yet... not

Offline deltaMass

  • Full Member
  • ****
  • Posts: 955
  • A Brit in California
  • Liked: 671
  • Likes Given: 275
Oh duh. So a baby one. Say 1 Kg and k = 5*10-4 N/W ? So F for 10 W is 5 mN, and a ~5*10-3 m/s2
« Last Edit: 06/08/2015 08:28 PM by deltaMass »

Online aero

  • Senior Member
  • *****
  • Posts: 2744
  • 92129
  • Liked: 705
  • Likes Given: 239
@SeeShells: FYI your quotes are usually messed up. Final /quote(s) is missing

I've found that the final quotes are ususlly there, just scroll down to the end of the inserted quoted material. It is very easy to read the last sentence of the quoted post and overlook the fact that the author may have added some blank lines at the end, which are also quoted. So the closing quote comes after the quoted blank lines. Here, try to quote this post, I have added 4 blank lines.





Retired, working interesting problems

Offline Paul Novy

  • Member
  • Posts: 12
  • Poland
  • Liked: 11
  • Likes Given: 3
EMDYIers, seems a lot of the RF injection into the frustum has been monopole antenna (magnatron) or coupling loop (EW). Been trying to think past this...I'll start out with a unity gain 1/2 wave monopole, then the loop, then a colinear array which provides about gain up to about 10dB with series 1/2 wave elements. Since this would exceed a linear dimension in the frustum, was thinking perhaps of a spiral arrangement.

Gain is important in my design, since I'm only planing on about +37dbm to start...FWIW: http://www.nodomainname.co.uk/Omnicolinear/2-4collinear.htm

Oops, forgot the helical pic...


I think that the antena length might have some influence if we consider near field (as an evanescent waves) distribution inside the frustum.

After wikipedia:



http://en.wikipedia.org/wiki/Near_and_far_field


Offline hhexo

  • Member
  • Posts: 31
  • An Italian in the UK
  • Liked: 24
  • Likes Given: 1
Sorry if I interject into @WarpTech's and @deltaMass's conversation...

As a humble engineer I'm trying to make sense of this GR discussion as I tried to do so for when @StrongGR hypothesised and proved a gravitational effect within the EmDrive (but that one turned to be too small by about 25 orders of magnitude).
I can sort of follow the math this time, but I'm struggling a bit with the physical meaning of it all. Let's see if I can write down my thoughts in somewhat layman terms.

We make an assumption which I could agree with. Technically, there is no inertial frame of reference in our universe. Every point in the universe is subject to a combination of fields/forces/actions so that everywhere and everything is in an accelerated frame, even if the acceleration is tiny. This is just because the universe exists ("sum ergo accelero"? :) ). Inertial frames are just an approximation which may be valid when accelerations are close to zero, but they don't really exist. So we must use GR for everything. Fine by me.
GR also tells us of the Equivalence Principle, which is basically that any cause of acceleration is indistinguishable from gravity. That's fine too.

So let's consider an entity which is subject to several "causes of acceleration": a gravity well, some force exerted by spending energy / expelling mass, other stuff (maybe an EM field?). Every one of these "causes of acceleration" can be seen as gravity if we want to, it doesn't matter where they come from. They could be magic for all we care.

If I understand correctly, what your model is saying is that there is continually a flow of energy from the whole universe (represented by the combination of all "causes of acceleration") into the entity (increasing its inertial mass, affecting the De Broglie wavelength of its particles, and possibly affecting the macro properties of the entity, including its motion) and a flow of energy from the entity back into the universe (presumably the entity affecting everything else in the universe with its own "cause of acceleration" due to its mere existence and properties).

At some point these two flows will reach an equilibrium when they are identical. Once that equilibrium is reached, the entity is in a certain state: its inertial mass has been altered, its properties (including motion) have been altered. At that equilibrium, in "visible", "macro" terms you are still pumping energy into it (power > 0) but you're just fighting its inertial mass and all other forces that contribute to spend that power. For example, hovering.
It's a system of interconnected "buckets" of energy where the pressure in each "bucket" has been equalised but there is still a current from one "bucket" to the next, to the next, in a circle; if you then want to change the properties of the entity you have to pump more power in, which causes an imbalance in the energy "buckets" and they all change to reach a new equilibrium.

I can sort of see where you're going with this... you're basically stating a different form of CoE. The two differences with our (my?) usual way of thinking are:
- this is similar to a "thermodynamic" equilibrium as opposed to the "static" one that we are used to considering (where we think "sum of energy in each bucket equals constant, no flow"), and
- the only closed system in existence, and the only one you can consider, is "the whole universe".

So, the way I see it, this theory isn't saying anything about frames of reference, a preferred frame, or measuring velocity and getting confused. This theory states that an equilibrium will be reached (or approached) subject to certain conditions. Then, every observer will probably measure properties of the entity differently, and there may not be any way of reconciling these measurements, and some of them might look wrong; the exact way in which they differ is for the rest of GR to determine (if it can).
But in the context of "the whole universe" it won't matter, the equilibrium is reached anyway by the universe and we can just be baffled about it. Maybe that "limiting velocity" will not be exactly what we predict in our frame of reference and we won't see why, because we don't know all the interactions of our test entity with the rest of the universe and we don't know what the situation of our initial frame of reference was. But the entity will be in equilibrium somehow, and we can't get free energy from the equilibrium.

@WarpTech, am I interpreting what you are saying correctly? :)

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2165
  • Liked: 2681
  • Likes Given: 1124
EMDYIers, seems a lot of the RF injection into the frustum has been monopole antenna (magnatron) or coupling loop (EW). Been trying to think past this...I'll start out with a unity gain 1/2 wave monopole, then the loop, then a colinear array which provides about gain up to about 10dB with series 1/2 wave elements. Since this would exceed a linear dimension in the frustum, was thinking perhaps of a spiral arrangement.


Not that I have experience designing waveguide couplers, but AFAIK the idea is to couple energy by matching impedance from source to sink.

An antenna is used to match impedance to free space. An E-field probe (or slot) or B-field loop is used to couple to waveguides, and very similarly, cavities.

And with high-Q cavities and filters, you'll want to match a pretty low impedance (~50 ohms) to a very high-Q and impedance. So a very short probe or link, away from the high-field area would be in order.

I don't think putting the magnetron output stub axially, on an end of the frustrum is a good idea, because the end regions may be "hot" (high E-field). Locating it near a node, or low-field, will match better.

Most important, it won't load the cavity which would kill the Q.

It has been noted previously, if you use a loop, you can rotate it to reduce the coupling/impedance/loading.

Well said. I've been trying to visualize this. EW had their loop in the top of the frustum, off-center, but according to models, right on a "hot spot". Shawyer seems to have his on the side as well as Iulian. My first tests will be side injection per my paper, at 3 different points. If I go with a helical radiator, think I can avoid the 4 hotspots I believe Doc said were TM212 which highlights 4 E fields at the top and bottom sides. Thks for the inputs...

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2165
  • Liked: 2681
  • Likes Given: 1124
EMDYIers, seems a lot of the RF injection into the frustum has been monopole antenna (magnatron) or coupling loop (EW). Been trying to think past this...I'll start out with a unity gain 1/2 wave monopole, then the loop, then a colinear array which provides about gain up to about 10dB with series 1/2 wave elements. Since this would exceed a linear dimension in the frustum, was thinking perhaps of a spiral arrangement.

Gain is important in my design, since I'm only planing on about +37dbm to start...FWIW: http://www.nodomainname.co.uk/Omnicolinear/2-4collinear.htm

Oops, forgot the helical pic...


I think that the antena length might have some influence if we consider near field (as an evanescent waves) distribution inside the frustum.

After wikipedia:



http://en.wikipedia.org/wiki/Near_and_far_field

Thanks. Would you recommend 1/2 wavelengths or loops to avoid far-field interactions?

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2165
  • Liked: 2681
  • Likes Given: 1124
Re. the recent flyby reference to the Aachen group's Baby EmDrive and CubeSats, I'm reminded that their team leader has already flown a couple of amateur space missions with an outfit called PoqetQub (from memory).

This is a NASA forum, so presumably packed to the brim with orbital mechanics specialists!! So... what value of k (N/W) is needed to get EmDrive up from LEO, O Experts?

ETA: On reflection that's a dumb question  :-[
Any positive k value will do.

You would have to determine what constitutes an orbit change that is outside the natural decay forces.  Cubesats don't have much power, so they may not get much thrust.  Would a retardation of orbital decay convince anyone?  That is a tricky deal, because orbit decay is sensitive to upper atmosphere expansion/contraction, which is affected by solar activity, etc.

If the thrust was significantly greater than the decay forces, you can use something like the Edelbaum approximation to determine the altitude change you should see with constant, tangential acceleration.  If there is interest, I'll run some quick parametrics to see what that might be.

This may mean that cubesat V 2.0 might be in order, although I thought I read somewhere there was a lunar orbit insertion...gotta check that.

Even cubesat's 3U size is difficult to work with unless you "unfurl the sails" of a 2.4 GHz frustum somehow.

Offline WarpTech

  • Full Member
  • ****
  • Posts: 1229
  • Do it!
  • Vista, CA
  • Liked: 1296
  • Likes Given: 1753
My comment is that dv/dt = 0 for a photon

It's not a photon, it's a Dirac particle. dv/dt is the acceleration of the group velocity of the particle.
Be careful to formulate your variable mass piece correctly:
http://en.wikipedia.org/wiki/Momentum#Objects_of_variable_mass
Note the correct vs. incorrect forms. It's a matter of using the correct frames

Nevertheless, solving your equation at F=0 one obtains after integration
1/2 (𝛾 v/c)2 = ln(𝜆)

and no, I don't know the correct limits. It appears to blow up as v->0, doesn't it?

Thanks. It sort of makes sense to me if it's ln(𝜆/𝜆o).  Then it goes to 0 at v=0. I'll give it some thought after work, if my brain isn't too scrambled when i get home.  :o There's a lot rattling around in there at the moment.

Todd


 

Tags: