EM Drive Developments - related to space flight applications - Thread 8

[aero]

Interesting curves, X-RaY. The area under the curve of power vs. time is Energy. What is the area under the power curve when plotted vs. frequency?

[Peter Lauwer]

My first test cavity (cylinder), measured with a network analyzer (Agilent E8364B). Length = 130 mm, r = 49 mm. Coupling loop, diameter 15 mm, at 1/4 of length (32.5 mm), 2 mm silvered copper wire.
Aim was to see what Q can be reached with untreated (unpolished) copper pipe and plates and this coupling (and to get some experience with copper soldering). And whether I can identify the modes (not yet).

In the picture 2-3 GHz, 6 peaks can be seen:
2.013 GHz   Q= low
2.136 GHz   Q=~600
2.358 GHz   Q=~23k
2.381 GHz   Q=low
2.642 GHz   Q= not calculated (order 100-1000)
2.917 GHz   Q= not calculated (order 100-1000)

Cheers,
Peter

[zellerium]

My first test cavity (cylinder), measured with a network analyzer (Agilent E8364B). Length = 130 mm, r = 49 mm. Coupling loop, diameter 15 mm, at 1/4 of length (32.5 mm), 2 mm silvered copper wire.
Aim was to see what Q can be reached with untreated (unpolished) copper pipe and plates and this coupling (and to get some experience with copper soldering). And whether I can identify the modes (not yet).

In the picture 2-3 GHz, 6 peaks can be seen:
2.013 GHz   Q= low
2.136 GHz   Q=~600
2.358 GHz   Q=~23k
2.381 GHz   Q=low
2.642 GHz   Q= not calculated (order 100-1000)
2.917 GHz   Q= not calculated (order 100-1000)

Cheers,
Peter

I ran a quick HFSS eigenmode sim on it without the loop and the resonant frequency seems to correlate very well. I'm not sure what mode you had ~ 2GHz, the TE111 should be the first one. Maybe an effect of the antenna.

Hope this helps

Kurt

[edit: each field plot is normalized to the same scale]

[flux_capacitor]

Interesting curves, X-RaY. The area under the curve of power vs. time is Energy. What is the area under the power curve when plotted vs. frequency?

The integral of the function (the area below the curve) would be in J.s^-2 (joules per second squared) which would be some sort of "energy acceleration unit" (of no meaning).

[Rodal]

Interesting curves, X-RaY. The area under the curve of power vs. time is Energy. What is the area under the power curve when plotted vs. frequency?

The integral of the function (the area below the curve) would be in J.s^-2 (joules per second squared) which would be some sort of "energy acceleration unit" (of no meaning).

The integral of Power over a frequency range, has units of Power times frequency, or equivalently, Power per unit time (Watt/second).  Since Power is force times velocity, Power per unit time has units of Force*acceleration or, equivalently, velocity times force per unit time.

Force per unit time is called the yank



So, equivalently, it has units of yank times velocity or  units of force times acceleration, or mass times acceleration squared or units of mass times jerk times velocity. 

The involvement of the force times  the acceleration, or the yank times velocity in these equations makes sense because mechanical resonance is a dynamic phenomenon governed by inertia, stiffness and damping.
(or in an electromagnetic circuit, by capacitance, inductance and resistance).



Momentum equals mass times velocity!
Force equals mass times acceleration!
Yank equals mass times jerk!
Tug equals mass times snap!
Snatch equals mass times crackle!
Shake equals mass times pop!!

http://math.ucr.edu/home/baez/physics/General/jerk.html

[X_RaY]

My first test cavity (cylinder), measured with a network analyzer (Agilent E8364B). Length = 130 mm, r = 49 mm. Coupling loop, diameter 15 mm, at 1/4 of length (32.5 mm), 2 mm silvered copper wire.
Aim was to see what Q can be reached with untreated (unpolished) copper pipe and plates and this coupling (and to get some experience with copper soldering). And whether I can identify the modes (not yet).

In the picture 2-3 GHz, 6 peaks can be seen:
2.013 GHz   Q= low
2.136 GHz   Q=~600
2.358 GHz   Q=~23k
2.381 GHz   Q=low
2.642 GHz   Q= not calculated (order 100-1000)
2.917 GHz   Q= not calculated (order 100-1000)

Cheers,
Peter

I ran a quick HFSS eigenmode sim on it without the loop and the resonant frequency seems to correlate very well. I'm not sure what mode you had ~ 2GHz, the TE111 should be the first one. Maybe an effect of the antenna.

Hope this helps

Kurt

[edit: each field plot is normalized to the same scale]

There seems to be a false mode notation, presented in your file.
The upper is called TE112, the lower one is TE211
regarding standard notation: TXmnp - in cylindical cavities "p" is the number of half waves in axial direktion.

EDIT
However, nice work!

[Peter Lauwer]

My first test cavity (cylinder), measured with a network analyzer (Agilent E8364B). Length = 130 mm, r = 49 mm. Coupling loop, diameter 15 mm, at 1/4 of length (32.5 mm), 2 mm silvered copper wire.
Aim was to see what Q can be reached with untreated (unpolished) copper pipe and plates and this coupling (and to get some experience with copper soldering). And whether I can identify the modes (not yet).

In the picture 2-3 GHz, 6 peaks can be seen:
2.013 GHz   Q= low
2.136 GHz   Q=~600
2.358 GHz   Q=~23k
2.381 GHz   Q=low
2.642 GHz   Q= not calculated (order 100-1000)
2.917 GHz   Q= not calculated (order 100-1000)

Cheers,
Peter

I ran a quick HFSS eigenmode sim on it without the loop and the resonant frequency seems to correlate very well. I'm not sure what mode you had ~ 2GHz, the TE111 should be the first one. Maybe an effect of the antenna.

Hope this helps

Kurt

[edit: each field plot is normalized to the same scale]

It surely helps, Kurt. Thanks!
Indeed I also measured a resonance at 3.21 GHz (and at 3.24, 3.76 and 3.93 GHz).
You would expect the TE111 to be little excited, don't you? As the field of the coupling loop is perpendicular to that of this mode. TM011, for example, should be much stronger.

Peter

[zellerium]

There seems to be a false mode notation, presented in your file.
The upper is called TE112, the lower one is TE211
regarding standard notation: TXmnp - in cylindical cavities "p" is the number of half waves in axial direktion.

EDIT
However, nice work!

Ahh thank you. A bit hasty on the labeling...

And no problem Peter!
If you wanted the coupling loop to work better for the TE111 I think you'd want it at half of the cylinder height where the H field is max.
Are you going for any mode in particular?
Also what kind of power source are you planning to use for your torsion balance test? 
(sorry I may have missed some of your previous posts)

[Peter Lauwer]

If you wanted the coupling loop to work better for the TE111 I think you'd want it at half of the cylinder height where the H field is max.
Are you going for any mode in particular?
Also what kind of power source are you planning to use for your torsion balance test? 
(sorry I may have missed some of your previous posts)

For this test I was not going for any particular mode. I want to study the effect of the placement and orientation of the coupling loop first.

I plan to use a low power (4 - 10 W) amplifier because I want to feed it with a battery and do not want too much heat generation. At 4 W, I can expect only something like 5 micronewton, which should be measurable with my torsion balance (resolution ~0.2 ľN) but a little more (10-20 ľM) would be nicer.
Peter

[Peter Lauwer]

Btw, I posted a discussion of the leaked paper by White et al. on https://forum.nasaspaceflight.com/index.php?topic=39214.msg1608047#msg1608047

[birchoff]

aie aie aie....Brace for impact guys/girls..... it's going around the globe.. FAST :

all articles not even 24hr old

http://www.digitaltrends.com/cool-tech/nasa-paper-emdrive/

http://www.epochtimes.com/gb/16/11/8/n8471367.htm

http://www.cnbeta.com/articles/555961.htm

https://lenta.ru/articles/2016/11/09/emdrive/

http://www.rtve.es/noticias/20161108/filtracion-revela-nasa-desarrollado-motor-sin-combustible/1439122.shtml

http://www.techworm.net/2016/11/leaked-nasa-papers-show-impossible-em-drive-work.html

A severe storm cloud of half truth's and over-hyped articles is forming near the horizon...
Take shelter, be safe...

Not sure if you should be congrat, due to your "act" and regarding your place in future history Phil.

I am guessing that the result will be the eventual closing of this topic on nsf as a result of 2 suppositions on my part and my part only. Ew winding down and aiaa not proceeding with the publication by request.

Disclaimer...I have no special insight...only gut feel. I hope I am wrong.

I too worry about this unauthorized release of non-confirmed EagleWorks data by Phil. EagleWorks was under the gun when the press got wind of some information last year and went nuts and NASA told the EagleWorks group to go quiet. I understand that.

I suspect and only just a feeling because the papers and information was not released per NASA's requirements that we may not hear anything from the EagleWorks group or NASA on this mess. NASA apparently hates bad unsubstantiated press and truthfully I don't blame them. IMHO Phil used bad judgement in doing this when we were told that the a paper was to be released in December, jeez Phil cancer or no cancer it was less than a month away.

IMHO more damage will come from this than good and that saddens me for it now will make my work tougher to validate anything good from my testing. And please Phil don't even reply to this as IMHO you may have done more damage than good and I have lost all respect.

Off my Soap Box,
Shell

I have a hard time fathoming any government agency would behave that way if their major concern was bad press. Seems to be they would simply feed the bad press flames even more. first they submit and then they retract AFTER going through months of peer review and being accepted for submission? That said I am still hopeful the paper will be released officially as previously stated. After having read the leak (Sorry I am like a man in a desert eagerly awaiting any solid research on this line of technology), if there is anything missing it looks like they included raw data. where as the leak only has graphs. I can see how that would be valuable for lots of interesting analysis.

Lastly, while I don't condone the leak. I did expect it. This topic has built up way too much inertia. It's like new music from a popular artist. The only way to have avoided a leak would have been to have released the paper the minute it cleared peer review. Especially if it is true that Open access option was paid for.

[Stormbringer]

Interesting curves, X-RaY. The area under the curve of power vs. time is Energy. What is the area under the power curve when plotted vs. frequency?

The integral of the function (the area below the curve) would be in J.s^-2 (joules per second squared) which would be some sort of "energy acceleration unit" (of no meaning).

The integral of Power over a frequency range, has units of Power times frequency, or equivalently, Power per unit time (Watt/second).  Since Power is force times velocity, Power per unit time has units of Force*acceleration or, equivalently, velocity times force per unit time.

Force per unit time is called the yank



So, equivalently, it has units of yank times velocity or  units of force times acceleration, or mass times acceleration squared or units of mass times jerk times velocity. 

The involvement of the force times  the acceleration, or the yank times velocity in these equations makes sense because mechanical resonance is a dynamic phenomenon governed by inertia, stiffness and damping.
(or in an electromagnetic circuit, by capacitance, inductance and resistance).



Momentum equals mass times velocity!
Force equals mass times acceleration!
Yank equals mass times jerk!
Tug equals mass times snap!
Snatch equals mass times crackle!
Shake equals mass times pop!!

http://math.ucr.edu/home/baez/physics/General/jerk.html

can I just say i want to nominate  the "yank" as a standard scientific unit. ' cause that'd be cool...

the instruments read 400 kilo-yanks!

[D_Dom]

How many jerks in a Kyank?
Oh, I see! Depends on tug/snap!

[RotoSequence]

How many jerks in a Kyank?
Oh, I see! Depends on tug/snap!

There are currently six named derivatives of position vectors with respect to time. 

[Rodal]

How many jerks in a Kyank?
Oh, I see! Depends on tug/snap!

There are currently six named derivatives of position vectors with respect to time. 

snap, crackle and pop: nose art on a Boeing B-17 Flying Fortress (WWII)

<<The name "pop", along with "snap" (also referred to as jounce) and "crackle" are somewhat facetious terms for the fourth, fifth, and sixth derivatives of position, being a reference to Snap, Crackle, and Pop. >>

[Peter Lauwer]

My first test cavity (cylinder), measured with a network analyzer (Agilent E8364B). Length = 130 mm, r = 49 mm. Coupling loop, diameter 15 mm, at 1/4 of length (32.5 mm), 2 mm silvered copper wire.
Aim was to see what Q can be reached with untreated (unpolished) copper pipe and plates and this coupling (and to get some experience with copper soldering). And whether I can identify the modes (not yet).

In the picture 2-3 GHz, 6 peaks can be seen:
2.013 GHz   Q= low
2.136 GHz   Q=~600
2.358 GHz   Q=~23k
2.381 GHz   Q=low
2.642 GHz   Q= not calculated (order 100-1000)
2.917 GHz   Q= not calculated (order 100-1000)

Cheers,
Peter

I ran a quick HFSS eigenmode sim on it without the loop and the resonant frequency seems to correlate very well. I'm not sure what mode you had ~ 2GHz, the TE111 should be the first one. Maybe an effect of the antenna.

Hope this helps

Kurt

[edit: each field plot is normalized to the same scale]

Hi Kurt,

Can you please also run a simulation 3-4 GHz? It will be interesting to identify more modes. Is there a TE011?

3.249 GHz      TM012?   
3.381 GHz         
3.747 GHz      TM110?     Q=12k
3.938 GHz      TM111?     Q=20k

The TM modes seem to have the highest Q for this cavity. Does that make sense? The loop is having its field in the axial direction of the cavity.

Thanks,
Peter

[Monomorphic]

The TM modes seem to have the highest Q for this cavity. Does that make sense? The loop is having its field in the axial direction of the cavity.

Thanks,
Peter

That is because you're using side wall injection. You are going to have a very difficult time (just like NASA did) trying to excite the TE01x modes. For example, I was able to locate TE012 at ~4.407Ghz, but it is VERY weak with the antenna mounted to the side and is very close to a TM mode.

If you move the loop to the center axis (2/3rd the way up) things clear up greatly. Also, FEKO says your antenna wire diameter of 2mm is too much. The ratio of wire radius and length are not ideal. You may need to switch to a thinner wire.

[SeeShells]

The TM modes seem to have the highest Q for this cavity. Does that make sense? The loop is having its field in the axial direction of the cavity.

Thanks,
Peter

That is because you're using side wall injection. You are going to have a very difficult time (just like NASA did) trying to excite the TE01x modes. For example, I was able to locate TE012 at ~4.407Ghz, but it is VERY weak with the antenna mounted to the side and is very close to a TM mode.

If you move the loop to the center things clear up greatly. Also, FEKO says your antenna wire diameter of 2mm is too much. The ratio of wire radius and length are not ideal. You may need to switch to a thinner wire.

Monomorphic, you truly rock. Perfect advice, I concur.
Shell

[Peter Lauwer]

The TM modes seem to have the highest Q for this cavity. Does that make sense? The loop is having its field in the axial direction of the cavity.

Thanks,
Peter

That is because you're using side wall injection. You are going to have a very difficult time (just like NASA did) trying to excite the TE01x modes. For example, I was able to locate TE012 at ~4.407Ghz, but it is VERY weak with the antenna mounted to the side and is very close to a TM mode.

If you move the loop to the center things clear up greatly. Also, FEKO says your antenna wire diameter of 2mm is too much. The ratio of wire radius and length are not ideal. You may need to switch to a thinner wire.

Thanks a lot for the advice and the sim, Monomorphic. I will make a new cavity with 1 mm wire. But, uhh..., how about the supply wires if you move the loop to the center? In a sim it is easy to do, but you will have strong influence of the wires from the side or endplate (either coax or two copper wires).
How about placing the loop on one of the endplates, halfway the center and edge, and opening (B-field) facing the center? The loop is not placed at a maximum then, that's true. I will try it out.
Peter

[SeeShells]

The TM modes seem to have the highest Q for this cavity. Does that make sense? The loop is having its field in the axial direction of the cavity.

Thanks,
Peter

That is because you're using side wall injection. You are going to have a very difficult time (just like NASA did) trying to excite the TE01x modes. For example, I was able to locate TE012 at ~4.407Ghz, but it is VERY weak with the antenna mounted to the side and is very close to a TM mode.

If you move the loop to the center things clear up greatly. Also, FEKO says your antenna wire diameter of 2mm is too much. The ratio of wire radius and length are not ideal. You may need to switch to a thinner wire.

Thanks a lot for the advice and the sim, Monomorphic. I will make a new cavity with 1 mm wire. But, uhh..., how about the supply wires if you move the loop to the center? In a sim it is easy to do, but you will have strong influence of the wires from the side or endplate (either coax or two copper wires).
How about placing the loop on one of the endplates, halfway the center and edge, and opening (B-field) facing the center? The loop is not placed at a maximum then, that's true. I will try it out.
Peter

Peter,
It doesn't matter the small effect of the coax, you can adjust for that. What matters is the balanced symmetry of field interactions and "locking" in a TE mode.

Shell