EM Drive Developments - related to space flight applications - Thread 7

[Flyby]

Please review Rogers advise:

The route to high Q is to achieve very high precision in the machining of cavity components and their alignment, together with mirror finish on the conducting surface (copper, silver or gold) of at least 10X skin depth. Maintaining this quality of finish also requires a clean dry environment. This is typical flight standard for space qualified microwave equipment, and is therefore expensive to achieve.
 
If you aim for something like 75% of theoretical Q you will still get viable levels of thrust within a reasonable budget for a small business.

I'm still not convinced that a higher Q will automatically lead towards a higher generated force as it has yet to be established that there is a relation between them.

I'm still having trouble with the concept of having almost no energetic losses (= high Q) on one side and an increased force, which means a transfer of energy, on the other side.
In my mind, these to are in an apparent conflict.. no?

The only possible way i can see a high Q result in a higher force, is when the force is not generated on the frustum walls, but is somehow created inside the frustum and causes it to move "somehow" (the gravity/gravity field EMtheory). The only result of a high Q is that you store more electromagnetic energy in the cavity.

If the force is generated on the frustum walls (either on front/back plates or the side walls) it must come at the expense of the Q of the electromagnetic waves. Any (yet to be determined) interaction between the walls and the electromagnetic waves must come at the expense of the Q.

So that's , after nearly 2 years of following this topic, i remain a skeptic of the extreme high Q pursuit and the floating car promises they hold.

[rq3]

Have to agree! As I said I did this only to get an idea of the dimension in general of the frequency shift using this linear approximation. 


Its much more than the few KHz TT stated about the 13µm differences of a single endplate. Does he multiply it by two for both plates or does he meant the total displacement or surface roughness?

Please review Rogers advise:

The route to high Q is to achieve very high precision in the machining of cavity components and their alignment, together with mirror finish on the conducting surface (copper, silver or gold) of at least 10X skin depth. Maintaining this quality of finish also requires a clean dry environment. This is typical flight standard for space qualified microwave equipment, and is therefore expensive to achieve.
 
If you aim for something like 75% of theoretical Q you will still get viable levels of thrust within a reasonable budget for a small business.

This implies a surface finish of 13 microns, not a dimensional tolerance of 13 microns. Correct? A 13 micron surface finish is actually pretty crappy. As a SWAG on my part, it looks like Dave got better than 2 microns on his polished end plates (based on observation of haze and geometric distortion).

What we really need is a proper mechanical drawing, with dimensional tolerances and required flatness and surface finish, as any drafting student is taught to do.

[spupeng7]

Please review Rogers advise:

The route to high Q is to achieve very high precision in the machining of cavity components and their alignment, together with mirror finish on the conducting surface (copper, silver or gold) of at least 10X skin depth. Maintaining this quality of finish also requires a clean dry environment. This is typical flight standard for space qualified microwave equipment, and is therefore expensive to achieve.
 
If you aim for something like 75% of theoretical Q you will still get viable levels of thrust within a reasonable budget for a small business.

I'm still not convinced that a higher Q will automatically lead towards a higher generated force as it has yet to be established that there is a relation between them.

I'm still having trouble with the concept of having almost no energetic losses (= high Q) on one side and an increased force, which means a transfer of energy, on the other side.
In my mind, these to are in an apparent conflict.. no?

The only possible way i can see a high Q result in a higher force, is when the force is not generated on the frustum walls, but is somehow created inside the frustum and causes it to move "somehow" (the gravity/gravity field EMtheory). The only result of a high Q is that you store more electromagnetic energy in the cavity.

If the force is generated on the frustum walls (either on front/back plates or the side walls) it must come at the expense of the Q of the electromagnetic waves. Any (yet to be determined) interaction between the walls and the electromagnetic waves must come at the expense of the Q.

So that's , after nearly 2 years of following this topic, i remain a skeptic of the extreme high Q pursuit and the floating car promises they hold.

Flyby,
any thrust produced will raise these same issues, no matter how small. It is clear, however, that the best results do seem to be from frustums with very flat interior surfaces. I think the Traveler's point is a good one.

[TheTraveller]

Have to agree! As I said I did this only to get an idea of the dimension in general of the frequency shift using this linear approximation. 


Its much more than the few KHz TT stated about the 13µm differences of a single endplate. Does he multiply it by two for both plates or does he meant the total displacement or surface roughness?

Please review Rogers advise:

The route to high Q is to achieve very high precision in the machining of cavity components and their alignment, together with mirror finish on the conducting surface (copper, silver or gold) of at least 10X skin depth. Maintaining this quality of finish also requires a clean dry environment. This is typical flight standard for space qualified microwave equipment, and is therefore expensive to achieve.
 
If you aim for something like 75% of theoretical Q you will still get viable levels of thrust within a reasonable budget for a small business.

This implies a surface finish of 13 microns, not a dimensional tolerance of 13 microns. Correct? A 13 micron surface finish is actually pretty crappy. As a SWAG on my part, it looks like Dave got better than 2 microns on his polished end plates (based on observation of haze and geometric distortion).

My understanding is the actual skin depth layer (outer boundary if you will) should not vary from a perfect frustum model by more than +- 10 skin depth.

Mirror finish, as an ex telescope mirror grinder & polisher is understood to be just that. Mirror finish as Dave achieved.

We need to understand scratches of skin depth or deeper will cause alteration in surface current formation as they will stop or reduce current flowing across them. To me scratch free means any scratch or pit to be much less than skin depth deep.

I do remember the long hours it took to polish a 250mm diameter mirror & the use of a low power microscope to check the mirror surface to see if the polishing had removed all the grind pits & scratches.

[TheTraveller]

If the force is generated on the frustum walls (either on front/back plates or the side walls) it must come at the expense of the Q of the electromagnetic waves. Any (yet to be determined) interaction between the walls and the electromagnetic waves must come at the expense of the Q.

Exactly correct. The conversion of internal cavity energy to external kinetic energy is at the expense of reduced internal cavity energy & Q as this conversions adds to cavity Q losses.

No free lunches.

It is all detailed in Rogers papers.

[rfmwguy]

Have to agree! As I said I did this only to get an idea of the dimension in general of the frequency shift using this linear approximation. 


Its much more than the few KHz TT stated about the 13µm differences of a single endplate. Does he multiply it by two for both plates or does he meant the total displacement or surface roughness?

Please review Rogers advise:

The route to high Q is to achieve very high precision in the machining of cavity components and their alignment, together with mirror finish on the conducting surface (copper, silver or gold) of at least 10X skin depth. Maintaining this quality of finish also requires a clean dry environment. This is typical flight standard for space qualified microwave equipment, and is therefore expensive to achieve.
 
If you aim for something like 75% of theoretical Q you will still get viable levels of thrust within a reasonable budget for a small business.

This implies a surface finish of 13 microns, not a dimensional tolerance of 13 microns. Correct? A 13 micron surface finish is actually pretty crappy. As a SWAG on my part, it looks like Dave got better than 2 microns on his polished end plates (based on observation of haze and geometric distortion).

What we really need is a proper mechanical drawing, with dimensional tolerances and required flatness and surface finish, as any drafting student is taught to do.

I hate to bring up too much in the way of mechanical speculation. I'm just trying to get something more than I witnessed last time, but...
I have researched a lot on surface finishing and know where to obtain 100 angstrom flatness, or 0.001 microns. This is a semiconductor processor in chicago. It is hundreds of dollars per piece. Perhaps down the road if scaling is successful, this level of endplate precision could be useful. Now, I just think its something to keep in the back of our minds for future reference.
Tomorrow, the big diameter endplate will take shape, then the sidewalls.

[dustinthewind]

If the force is generated on the frustum walls (either on front/back plates or the side walls) it must come at the expense of the Q of the electromagnetic waves. Any (yet to be determined) interaction between the walls and the electromagnetic waves must come at the expense of the Q.

Exactly correct. The conversion of internal cavity energy to external kinetic energy is at the expense of reduced internal cavity energy & Q as this conversions adds to cavity Q losses.

No free lunches.

It is all detailed in Rogers papers.

I am fairly sure that a force in itself won't use up energy until the cavity actually begins to free accelerate by F.dx = Energy (if it does accelerate purely via light) .  This should be similar to two mirrors accelerated away from each other and the light between them red-shifts when the mirrors actually start to accelerate.  If the mirrors are held stationary and are perfectly reflective no red-shifting of the light between the mirrors should happen but a force will still be present.  (While the two systems [cavity or mirrors] are different, I am assuming if there were a force purely by the pressure of light inside the cavity that we could parallel between the two systems.)

On the other hand if the light is accelerating something other than the cavity as a propellant that passes through the walls then there is a chance of seeing Q or energy content drop even when the cavity isn't accelerating because the light is accelerating something else via F.dx=E . 

If there is some anomalous force the difference between these two scenarios might be a way to narrowing down where it is coming from. 

[rfmwguy]

EM propulsion study for USAF written in 1989 contains a lot about some of the theories and speculations similar to emdrive. 169 pages...5 D math...I suspect theory people will find this old paper interesting:

https://drive.google.com/file/d/0B4Ez9NDUxpYLZThJTy15TUdPVnM/view

[CW]

At the moment and in general there are much bigger problems (thermal expansion,

How much will a ~20cm copper frustum expand due to heating? If it is less than 1/2 a cm or so, then it shouldn't be too much of a concern.

Best to have a tunable frustum IMHO. And from what I've learned today, it might be best to have the big end be the side that is tuned.

I would design it in a way, that the frustum reaches its final (resonance-optimized) dimensions, when the final average operational temperature of the cavity metal walls is reached. I do not know, whether the EM drive designers are doing this already. So I thought it can not hurt to point this out. It's better to have the frustum expand into its perfect dimensions and work WITH nature, instead of trying to compensate for expansion and work AGAINST nature. Also, if the sought after effect is reproducible and we can go to space with it, we better design this thing mechanically in a self-calibrating/adjusting way.

[TheTraveller]

If the force is generated on the frustum walls (either on front/back plates or the side walls) it must come at the expense of the Q of the electromagnetic waves. Any (yet to be determined) interaction between the walls and the electromagnetic waves must come at the expense of the Q.

Exactly correct. The conversion of internal cavity energy to external kinetic energy is at the expense of reduced internal cavity energy & Q as this conversions adds to cavity Q losses.

No free lunches.

It is all detailed in Rogers papers.

I am fairly sure that a force in itself won't use up energy until the cavity actually begins to free accelerate by F.dx = Energy (if it does accelerate purely via light) .  This should be similar to two mirrors accelerated away from each other and the light between them red-shifts when the mirrors actually start to accelerate.  If the mirrors are held stationary and are perfectly reflective no red-shifting of the light between the mirrors should happen but a force will still be present.  (While the two systems [cavity or mirrors] are different, I am assuming if there were a force purely by the pressure of light inside the cavity that we could parallel between the two systems.)

On the other hand if the light is accelerating something other than the cavity as a propellant that passes through the walls then there is a chance of seeing Q or energy content drop even when the cavity isn't accelerating because the light is accelerating something else via F.dx=E . 

If there is some anomalous force the difference between these two scenarios might be a way to narrowing down where it is coming from.

As per attached.

Note that the reaction is either the acceleration a, or a force equal to Ma, but not both.

EmDrive Force Measurement:
http://www.emdrive.com/EmDriveForceMeasurement.pdf

The EmDrive is not a rocket. It does not generate acceleration unless it is free to move. When constrained there is no external Force generated.

[VAXHeadroom]

If the force is generated on the frustum walls (either on front/back plates or the side walls) it must come at the expense of the Q of the electromagnetic waves. Any (yet to be determined) interaction between the walls and the electromagnetic waves must come at the expense of the Q.

Exactly correct. The conversion of internal cavity energy to external kinetic energy is at the expense of reduced internal cavity energy & Q as this conversions adds to cavity Q losses.

No free lunches.

It is all detailed in Rogers papers.

I am fairly sure that a force in itself won't use up energy until the cavity actually begins to free accelerate by F.dx = Energy (if it does accelerate purely via light) .  This should be similar to two mirrors accelerated away from each other and the light between them red-shifts when the mirrors actually start to accelerate.  If the mirrors are held stationary and are perfectly reflective no red-shifting of the light between the mirrors should happen but a force will still be present.  (While the two systems [cavity or mirrors] are different, I am assuming if there were a force purely by the pressure of light inside the cavity that we could parallel between the two systems.)

On the other hand if the light is accelerating something other than the cavity as a propellant that passes through the walls then there is a chance of seeing Q or energy content drop even when the cavity isn't accelerating because the light is accelerating something else via F.dx=E . 

If there is some anomalous force the difference between these two scenarios might be a way to narrowing down where it is coming from.

As per attached.

Note that the reaction is either the acceleration a, or a force equal to Ma, but not both.

EmDrive Force Measurement:
http://www.emdrive.com/EmDriveForceMeasurement.pdf

The EmDrive is not a rocket. It does not generate acceleration unless it is free to move. When constrained there is no external Force generated.

"Clearly, in a static situation, where T and R both exist as forces , they will cancel out."

Not clear, and not proven, and only if Fg1=Fg2, an assertion I do not believe is accurate.
I have not read the whole paper, but will do so soon and comment further.

[TheTraveller]

If the force is generated on the frustum walls (either on front/back plates or the side walls) it must come at the expense of the Q of the electromagnetic waves. Any (yet to be determined) interaction between the walls and the electromagnetic waves must come at the expense of the Q.

Exactly correct. The conversion of internal cavity energy to external kinetic energy is at the expense of reduced internal cavity energy & Q as this conversions adds to cavity Q losses.

No free lunches.

It is all detailed in Rogers papers.

I am fairly sure that a force in itself won't use up energy until the cavity actually begins to free accelerate by F.dx = Energy (if it does accelerate purely via light) .  This should be similar to two mirrors accelerated away from each other and the light between them red-shifts when the mirrors actually start to accelerate.  If the mirrors are held stationary and are perfectly reflective no red-shifting of the light between the mirrors should happen but a force will still be present.  (While the two systems [cavity or mirrors] are different, I am assuming if there were a force purely by the pressure of light inside the cavity that we could parallel between the two systems.)

On the other hand if the light is accelerating something other than the cavity as a propellant that passes through the walls then there is a chance of seeing Q or energy content drop even when the cavity isn't accelerating because the light is accelerating something else via F.dx=E . 

If there is some anomalous force the difference between these two scenarios might be a way to narrowing down where it is coming from.

As per attached.

Note that the reaction is either the acceleration a, or a force equal to Ma, but not both.

EmDrive Force Measurement:
http://www.emdrive.com/EmDriveForceMeasurement.pdf

The EmDrive is not a rocket. It does not generate acceleration unless it is free to move. When constrained there is no external Force generated.

"Clearly, in a static situation, where T and R both exist as forces , they will cancel out."

Not clear, and not proven, and only if Fg1=Fg2, an assertion I do not believe is accurate.
I have not read the whole paper, but will do so soon and comment further.

As Roger has stated in his papers, there needs to exist a directionally differential doppler shift of the resonant EM waves to trigger the EmDrive into either Motor or Generator mode. In testing, this external tigger is generated by vibration. As Roger states in the paper linked, when external forces (vibrations) were eliminated, the reaction force stopped being generated.

To say it again. The EmDrive is not a fancy fuelless rocket. Thinking that it is a rocket & should work / act like a rocket is a really bad way to think.

[RERT]

EM propulsion study for USAF written in 1989 contains a lot about some of the theories and speculations similar to emdrive. 169 pages...5 D math...I suspect theory people will find this old paper interesting:

https://drive.google.com/file/d/0B4Ez9NDUxpYLZThJTy15TUdPVnM/view

rfmwguy - thanks!! A breath of fresh air. Helps the thesis that I'm not mad to find one other person who thinks that we should be investigating the coupling between EM and gravitational fields. Or maybe just a cellmate...

Section 2.10 looks interesting, esp equation 271. Made me wonder whether FEKO could map E.Edot and E.CurlB across the frustrum.

I think equation 168 may speak to CW's thought on the variation of the rate of flow of time.

Interesting the remarks around equation 221!

R.

[meberbs]

...

Note that the reaction is either the acceleration a, or a force equal to Ma, but not both.

EmDrive Force Measurement:
http://www.emdrive.com/EmDriveForceMeasurement.pdf

The EmDrive is not a rocket. It does not generate acceleration unless it is free to move. When constrained there is no external Force generated.

That force diagram is drawn in a nonsensical way. For now, I will assume Fg1 is not equal to Fg2, and there is no external force on the side walls (which implies some interaction with something that is not contained by the cavity, or breaking conservation of momentum). I will also absorb the factor of Q into the F terms, because for a force balance the total force is what matters.

To do the force balance:
ΣF = M*a
Fg2 - Fg1 = M*a (note: Fg1 is taken to be positive value, acting in opposite direction)

The diagram shows the vector for the net force R = M*a. The diagram also shows T. I have no idea what T is, there is no explanation. It has no place in this force balance.

This internal force F is measured by an outside observer as the Thrust T, a force acting against the observer in the direction shown.

This statement is completely nonsensical. The outside observer would see the device accelerate with acceleration a. Why would the observer feel a force on them? You don't feel a force just from looking at an accelerating object.

T could be the equal and opposite force to R that is felt by the mystery field/particles that carry away the balancing momentum, or it could be the tension in a rope holding the drive still. Assuming the second case, the force balance becomes:

Fg2 - Fg1 - T = M*a
Fg2 - Fg1 - T = 0 (because the rope holds it still)
Fg2 - Fg1 = T

You could do something similar where a rope tied to the back of the drive is connected to a mechanism that is designed to apply a constant force T. In this case the M*a term sticks around and the device will accelerate at a = (Fg2 - Fg1 - T)/M. (neglecting the mass of the rope tied to the back)

The following quote is just a mangling of terminology, an object with a net applied force R accelerates at a=R/M. you call R the reaction, "a" the acceleration.

Note that the reaction is either the acceleration a, or a force equal to Ma, but not both.

One more quote:

In  free  space,  the  thruster  will  simply  accelerate  at a m/s/s,  and R will  not  be measurable.

Except that R is defined as M*a, you know M and a is easy to measure.

The rest of this paper is not worth discussing, because it is clear the author does not understand the most basic concepts of applying a force balance or is just deliberately being confusing to come to a result they know is wrong. I will assume the first, since the second is an accusation that shouldn't be made without strong evidence.

[Rodal]

At the moment and in general there are much bigger problems (thermal expansion,

How much will a ~20cm copper frustum expand due to heating? If it is less than 1/2 a cm or so, then it shouldn't be too much of a concern.

Best to have a tunable frustum IMHO. And from what I've learned today, it might be best to have the big end be the side that is tuned.

I would design it in a way, that the frustum reaches its final (resonance-optimized) dimensions, when the final average operational temperature of the cavity metal walls is reached. I do not know, whether the EM drive designers are doing this already. So I thought it can not hurt to point this out. It's better to have the frustum expand into its perfect dimensions and work WITH nature, instead of trying to compensate for expansion and work AGAINST nature. Also, if the sought after effect is reproducible and we can go to space with it, we better design this thing mechanically in a self-calibrating/adjusting way.

1) Concerning <<I do not know, whether the EM drive designers are doing this already>> they are certainly not (*) conducting their room temperature EM Drive experiments so that <<the frustum reaches its final (resonance-optimized) dimensions, when the final average operational temperature of the cavity metal walls is reached>>.

2) As to whether what you propose is feasible, and whether such a test would be representative of space propulsion, consider the following:

a) whether a "final steady state" will be reached during the testing time available, is a function of the amount of cooling provided (and a function of how much % time is the RF feed on).  As long as the RF feed is on you have a constant source of induction heat (all EM Drive experiments have been conducted with either the RF feed constantly on, or with the RF feed pulsating on and off) .

b) Cooling only takes place by thermal conduction, convection or radiation:

b1) thermal conduction: this is limited by the amount of metal "sink" available.  For constructions such as for example, those of rfmwguy who has only 1 mm thickness on the side walls, this is very limited.  EM Drive constructions with thick walls have substantially more heat sink available.  The time at which a given temperature is reached is inversely proportional to the square of the thickness.  Everything else being the same, a wall thickness that is twice as thick means that a given temperature will be reached in four times longer time.  This follows from the Fourier number:
where L is the thickness, t is time and alpha is the thermal diffusivity of the metal (copper has a thermal diffusivity 111 mm^2/s).

b2)  convection:  all room temperature (*) EM Drive experiments have been conducted without any forced air convection (and certainly without fluid convection like water, or any other fluids that have substantially more cooling capability).  They rely on natural convection.  It is known that the amount of convection provided by natural air convection is very limited due to the very small coefficient of heat transfer due to natural air convection.

Natural convection - air ~0.5 to 20 (W/(m^2K))
Forced Convection - air ~ 10 to 1000 (W/(m^2K))
Forced Convection - water and liquids up to 10000 (W/(m^2K))


Most importantly, natural air convection is not available in space and hence tests relying on natural convection would be very unrepresentative of what would happen in space.

b3) radiation: the amount of heat radiation is proportional to the temperature to the fourth power. 


where σ is the Stefan–Boltzmann constant (σ = 5.670367×10^−8 W m^−2 K^−4) and epsilon are the emissivities.   Hence it is very nonlinear.  It would take a very large temperature increase for radiation to be effective and to be able to equal the amount of heat produced by induction heating.  Hence if you are counting on radiation, the "final temperature" would be high, and it would take place at a late time, considerably longer than what EM Drive experiments have been conducted for.

______________


(*) An exception are Superconducting EM Drive tests.  Although Fetta (Cannae) and Shawyer (SPR) have discussed such tests, there is no reported information on temperature vs. time to ascertain the effectiveness of the cooling system, but certainly this is of paramount importance to maintain superconductivity, so that a working superconducting EM Drive (under present superconducting materials) entails a forced cooling design.

[Willem Staal]

 
I have a sort of a theory, maybe too novel, but still.

In the original setup of the Roger Shawyer EM drive they use a hefty Microwave transmitter.

The other day my smartphone smashed on the floor  in pieces. (grumble) So i  examine all the parts in this device. (i was always puzzled how they achieve to build a reception antenna into these little devices)

And i found out that  the antenna in fact is build as a fractal unit!

That gives me a another  idea; is it possible to translate the shape of the EM frustrum into a fractal design? Maybe this is enoegh to boost the overall effect, as i think a fractal antenna has a great effect on reception, so it must have effect on transmitting as well (if you do the math right)   

[CW]

(...)

1) Concerning <<I do not know, whether the EM drive designers are doing this already>> they are certainly not conducting their experiments so that <<the frustum reaches its final (resonance-optimized) dimensions, when the final average operational temperature of the cavity metal walls is reached>>.

2) As to whether what you propose is feasible, and whether such a test would be representative of space propulsion, consider the following:

a) whether a "final steady state" will be reached is a function of the amount of cooling provided.  As long as the RF feed is on you have a constant source of induction heat (all EM Drive experiments have been conducted with either the RF feed constantly on, or with the RF feed pulsating on and off) .

b) Cooling only takes place by:

b1) thermal conductivity: this is limited by the amount of metal "sink" available.  For constructions such as for example, those of rfmwguy who has only 1 mm thickness on the side walls, this is very limited.  EM Drive constructions with thick walls have substantially more heat sink available.  The time at which a given temperature is reached is inversely proportional to the square of the thickness.  This follows from the Fourier number:

b2)  convection:  all EM Drive experiments have been conducted without any forced air convection (and certainly without fluid convection like water, or any other fluids that have substantially more cooling capability).  They rely on natural convection.  It is known that the amount of convection provided by natural air convection is very limited due to the very small coefficient of heat transfer due to natural air convection.

Natural convection - air ~0.5 to 20 (W/(m^2K))
Forced Convection - air ~ 10 to 1000 (W/(m^2K))
Forced Convection - water and liquids up to 10000 (W/(m^2K))


Most importantly, natural air convection is not available in space and hence tests relying on natural convection would be very unrepresentative of what would happen in space.

b3) radiation: the amount of heat radiation is proportional to the temperature to the fourth power. 

Hence it is very nonlinear.  It would take a very large temperature increase for radiation to be effective and to be able to equal the amount of heat produced by induction heating.  Hence if you are counting on radiation, the "final temperature" would be high, and it would take place at a late time, considerably longer than what EM Drive experiments have been conducted for.

Thanks for your great reply, because it should remind all here of the fact that this 'propulsion system' is eventually supposed to work in free space with its absence of convective cooling. I don't think, that the sort of testing that's going on right now, has any real merit, because the test articles never reach e.g. their thermodynamical operating point, which is, by the way, not even defined yet. The test articles not being in thermodynamical equilibrium can certainly cause readings that seem to indicate force production. But is it so?

Let me just say, that at my old workplace, we had 300+k€ high-end oscilloscopes, which needed to reach their operational temperature within about an hour after switching on, just to be able to make reliable measurements. Defining and keeping points of operation is vital in any engineering, which seems to have happily been forgone. At least it seems to me like this, with the information accessible to me. If the hitherto undertaken experiments have not run long enough for the frustum to even closely reach their stable operating points, then that is a problem of the hitherto taken approaches, not of me stating the fact. I think that I would never dare making any assumptions about any experimental setup without it having reached its operating point.

Just think LIGO, e.g. does anyone seriously think that they did not heavily control environmental conditions to the point where they could be practically 100% confident, that the measurement apparatus had reached its stable operating point and could hopefully produce expected results? The EM drive contraptions are all in fact measuring apparatuses that hope to measure a hitherto elusive way of producing propulsive forces in the low µN range. Don't you think we should apply similarly rigid considerations like in LIGO (although much simpler to achieve in this case) for trying to find out, whether there is really something going on, or whether there is some curious transitional thermodynamical effect emulating what is looked for? Since the EM drive seeks noting less than to overthrow centuries of confirmed physical experimental results so far, it is the greatest duty of experimenters to rigidly make sure that not only confirmed physical knowledge so far is applied correctly, but that especially the physical implementation and measurement setups in the form of test articles is made with prudent and known to be correct engineering approaches.

BR,
CW

[Rodal]

...
Thanks for your great reply, because it should remind all here of the fact that this 'propulsion system' is eventually supposed to work in free space with its absence of convective cooling. I don't think, that the sort of testing that's going on right now, has any real merit, because the test articles never reach e.g. their thermodynamical operating point, which is, by the way, not even defined yet. The test articles not being in thermodynamical equilibrium can certainly cause readings that seem to indicate force production. But is it so?

Let me just say, that at my old workplace, we had 300+k€ high-end oscilloscopes, which needed to reach their operational temperature within about an hour after switching on, just to be able to make reliable measurements. Defining and keeping points of operation is vital in any engineering, which seems to have happily been forgone. At least it seems to me like this, with the information accessible to me. If the hitherto undertaken experiments have not run long enough for the frustum to even closely reach their stable operating points, then that is a problem of the hitherto taken approaches, not of me stating the fact. I think that I would never dare making any assumptions about any experimental setup without it having reached its operating point.

Just think LIGO, e.g. does anyone seriously think that they did not heavily control environmental conditions until the point where they could be practically 100% confidential, that the measurement apparatus had reached its stable operating point and could hopefully produce expected results? The EM drive contraptions are all in fact measuring apparatuses that hope to measure a hitherto elusive way of producing propulsive forces in the low µN range. Don't you think we should apply similarly rigid considerations like in LIGO (although much simpler to achieve in this case) for trying to find out, whether there is really something going on, or whether there is some curious transitional effect emulating what is looked for?

BR,
CW

Great points.

Although relying on natural convection is not representative of operation in space, artificially incorporating forced convection to control the temperature is certainly an option. 

Superconducting EM Drive designs certainly need such forced convection cooling in order to achieve and maintain superconductivity (presently only achievable at temperatures below room temperature).

[rfmwguy]

Oh joy...jigsaw cut large endplate...grinding then finishing...

[SeeShells]

(...)

1) Concerning <<I do not know, whether the EM drive designers are doing this already>> they are certainly not conducting their experiments so that <<the frustum reaches its final (resonance-optimized) dimensions, when the final average operational temperature of the cavity metal walls is reached>>.

2) As to whether what you propose is feasible, and whether such a test would be representative of space propulsion, consider the following:

a) whether a "final steady state" will be reached is a function of the amount of cooling provided.  As long as the RF feed is on you have a constant source of induction heat (all EM Drive experiments have been conducted with either the RF feed constantly on, or with the RF feed pulsating on and off) .

b) Cooling only takes place by:

b1) thermal conductivity: this is limited by the amount of metal "sink" available.  For constructions such as for example, those of rfmwguy who has only 1 mm thickness on the side walls, this is very limited.  EM Drive constructions with thick walls have substantially more heat sink available.  The time at which a given temperature is reached is inversely proportional to the square of the thickness.  This follows from the Fourier number:

b2)  convection:  all EM Drive experiments have been conducted without any forced air convection (and certainly without fluid convection like water, or any other fluids that have substantially more cooling capability).  They rely on natural convection.  It is known that the amount of convection provided by natural air convection is very limited due to the very small coefficient of heat transfer due to natural air convection.

Natural convection - air ~0.5 to 20 (W/(m^2K))
Forced Convection - air ~ 10 to 1000 (W/(m^2K))
Forced Convection - water and liquids up to 10000 (W/(m^2K))


Most importantly, natural air convection is not available in space and hence tests relying on natural convection would be very unrepresentative of what would happen in space.

b3) radiation: the amount of heat radiation is proportional to the temperature to the fourth power. 

Hence it is very nonlinear.  It would take a very large temperature increase for radiation to be effective and to be able to equal the amount of heat produced by induction heating.  Hence if you are counting on radiation, the "final temperature" would be high, and it would take place at a late time, considerably longer than what EM Drive experiments have been conducted for.

Thanks for your great reply, because it should remind all here of the fact that this 'propulsion system' is eventually supposed to work in free space with its absence of convective cooling. I don't think, that the sort of testing that's going on right now, has any real merit, because the test articles never reach e.g. their thermodynamical operating point, which is, by the way, not even defined yet. The test articles not being in thermodynamical equilibrium can certainly cause readings that seem to indicate force production. But is it so?

Let me just say, that at my old workplace, we had 300+k€ high-end oscilloscopes, which needed to reach their operational temperature within about an hour after switching on, just to be able to make reliable measurements. Defining and keeping points of operation is vital in any engineering, which seems to have happily been forgone. At least it seems to me like this, with the information accessible to me. If the hitherto undertaken experiments have not run long enough for the frustum to even closely reach their stable operating points, then that is a problem of the hitherto taken approaches, not of me stating the fact. I think that I would never dare making any assumptions about any experimental setup without it having reached its operating point.

Just think LIGO, e.g. does anyone seriously think that they did not heavily control environmental conditions to the point where they could be practically 100% confidential, that the measurement apparatus had reached its stable operating point and could hopefully produce expected results? The EM drive contraptions are all in fact measuring apparatuses that hope to measure a hitherto elusive way of producing propulsive forces in the low µN range. Don't you think we should apply similarly rigid considerations like in LIGO (although much simpler to achieve in this case) for trying to find out, whether there is really something going on, or whether there is some curious transitional thermodynamical effect emulating what is looked for? Since the EM drive seeks noting less than to overthrow centuries of confirmed physical experimental results so far, it is the greatest duty of experimenters to rigidly make sure that not only confirmed physical knowledge so far is applied correctly, but that especially the physical implementation and measurement setups in the form of test articles is made with prudent and known to be correct engineering approaches.

BR,
CW

These are points that are to be taken seriously in a build  considering the limits of a pocketbook funding.

Personally it's been one of the cornerstones of my build to remove as best I could the thermal issues from powering the frustum, as much as the pocketbook would allow and still stay within basic design considerations.

It is a fine line to walk in the design. Providing enough power into the frustum to assure you will push the device out of the noise and error bars is no mean feat, even when you have a deep pocket for funding. EagleWorks found this out as well.

How a builder does it is truly up to them, but they should at least consider these points.

Thickness of the walls equates to not only to a greater heat sink but also adds increased structural stability to deformation of the cavity. That said it still will not remove all thermal expansion issues of the aluminum or copper frustum and provisions should be made in either the cavity compensating for thermal changes or tracking of the resonate changing frequency due to shape changes.

Keeping the RF amp or magnetron away from the frustum if at all possible otherwise you will need to profile the extra thermals from the devices. If you don't It just adds complexity to the build IMHO. Not insurmountable although the error bars widen out dramatically.

To this point one thing I have found is trying to keep the thermal signature as uniform as possible and migrate hot spots from mode generation, also waveguides and antennas (which cause deformations end endplates) to the sidewalls, so the deformations of the frustum are a more uniform in nature. I was excited to see the current sims and how they showed deformation issues.
Cooling it?
This is a inexpensive way to do it and it's the main reason your cell phones don't melt in your hands.
http://www.digikey.com/en/product-highlight/p/panasonic/pyrolytic-graphite-sheets


Well, that's my 2 cents this morning. Great insights from great posts.


Shell