Author Topic: EM Drive Developments - related to space flight applications - Thread 2  (Read 2105938 times)

Offline SleeperService

  • Member
  • Posts: 48
  • Liked: 15
  • Likes Given: 15
Cross-posted from an EP thread..
http://pepl.engin.umich.edu/pdf/2014_Florenz_Thesis.pdf
Wouldn't it be cool if Mr. March and others could have just half the toys/funding to play with...

Offline Notsosureofit

  • Full Member
  • ****
  • Posts: 656
  • Liked: 704
  • Likes Given: 1364
@ Star-Drive

Trying to put together a reference table.

Did they calculate Q values for all of the COMSOL modes shown ?

As RODAL asked, I'm assuming these are w/o dielectric ?

Also, can they calculate a dispersion relation for the cavity including the dielectric ?

Thanks

Offline lele

  • Member
  • Posts: 56
  • France
  • Liked: 29
  • Likes Given: 65
If I understood correctly some things in this thread, EM radiations heat the frustum, which changes its shape, which changes the resonant frequency of the cavity, so it's difficult to keep the EM source tuned.

Would cooling (with flowing water for example) the copper plates 1) reduce the variation of the resonant frequency, by reducing the variation of temperature, thus reducing the geometrical changes of the frustum 2) not interfere with the thrust produced ?

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1113
  • Liked: 776
  • Likes Given: 1013

Wasn't that how Shawyer got the idea, only from a satellite not a ship?

Background of emdrive.
http://emdrive.com/background.html

I was joking about the ship rocking more.
Challenge your preconceptions, or they will challenge you. - Velik

Online Rodal

  • Senior Member
  • *****
  • Posts: 5838
  • USA
  • Liked: 5919
  • Likes Given: 5261
If I understood correctly some things in this thread, EM radiations heat the frustum, which changes its shape, which changes the resonant frequency of the cavity, so it's difficult to keep the EM source tuned.

Would cooling (with flowing water for example) the copper plates 1) reduce the variation of the resonant frequency, by reducing the variation of temperature, thus reducing the geometrical changes of the frustum 2) not interfere with the thrust produced ?
Cooling from the outside may actually be counterproductive, as it may increase the thermal gradient through the thickness (it would be induction-heated from the inside and cooled from the outside hence a large thermal gradient) and deform the very thin NASA EM Drive shell due to the non-uniform aspect of the induction heating due to the non-uniform EM field.

One solution (already offered by Shawyer) is to use Invar, an old alloy that has very low thermal expansion, another solution is to increase the thickness of the EM Drive.  NASA was constrained to use very thin copper for the EM Drive because they wanted to minimize the mass on their torque pendulum.
« Last Edit: 02/18/2015 03:09 PM by Rodal »

Offline SleeperService

  • Member
  • Posts: 48
  • Liked: 15
  • Likes Given: 15
If I understood correctly some things in this thread, EM radiations heat the frustum, which changes its shape, which changes the resonant frequency of the cavity, so it's difficult to keep the EM source tuned.

Would cooling (with flowing water for example) the copper plates 1) reduce the variation of the resonant frequency, by reducing the variation of temperature, thus reducing the geometrical changes of the frustum 2) not interfere with the thrust produced ?
Water cooling would either have to be internal to the thrust measuring device or external.
If external, then you have the problem of the coupling of the pipework and the externally powered thrust of the coolant.
If internal then you get the added mass of coolant, cooling jackets, pumps and radiators. And you have an pump running in close proximity to the cavity.

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1113
  • Liked: 776
  • Likes Given: 1013
Challenge your preconceptions, or they will challenge you. - Velik

Offline Notsosureofit

  • Full Member
  • ****
  • Posts: 656
  • Liked: 704
  • Likes Given: 1364
Or a phase locked loop circuit.
http://en.m.wikipedia.org/wiki/Phase-locked_loop#Frequency_synthesis

Yep, very easy these days.  In fact if they have VCO already, they must have it on the to-do list.

We lock to a satellite signal here w/ a tiny card.
« Last Edit: 02/18/2015 02:54 PM by Notsosureofit »

Offline SleeperService

  • Member
  • Posts: 48
  • Liked: 15
  • Likes Given: 15
Or a phase locked loop circuit.
http://en.m.wikipedia.org/wiki/Phase-locked_loop#Frequency_synthesis
How would apply this to tracking the resonant freq of the cavity to the input freq of the loop antenna?

Offline SleeperService

  • Member
  • Posts: 48
  • Liked: 15
  • Likes Given: 15
"The function of the PLL is to compare the distributed clock to the incoming reference clock, and vary the phase and frequency of its output until the reference and feedback clocks are phase and frequency matched."
At which spatial point shall we measure the freq. to compare with the incoming freq.?
Please forgive me if I am wrong and/or naive about this...
« Last Edit: 02/18/2015 03:01 PM by SleeperService »

Online Rodal

  • Senior Member
  • *****
  • Posts: 5838
  • USA
  • Liked: 5919
  • Likes Given: 5261
Or a phase locked loop circuit.
http://en.m.wikipedia.org/wiki/Phase-locked_loop#Frequency_synthesis
How would apply this to tracking the resonant freq of the cavity to the input freq of the loop antenna?
I also like Mulletron's original idea of distributing the spectral power density of the input into a spectral band instead of a spike at a single frequency.  As Mulletron pointed out, magnetrons used by Juan Yang in China and Shawyer in the UK deserve attention.
« Last Edit: 02/18/2015 03:05 PM by Rodal »

Offline Giovanni DS

  • Regular
  • Full Member
  • *
  • Posts: 186
    • ChibiOS/RT Project
  • Liked: 44
  • Likes Given: 162
A feedback circuit would be required in order to keep the VCO on the resonant frequency. If such a feedback is possible then a simple MCU could be programmed to do that using an ADC as output, it would not be a PLL anyway but more similar to an FLL.

Offline SleeperService

  • Member
  • Posts: 48
  • Liked: 15
  • Likes Given: 15
Maybe that was behind Cannae's technique of putting notches in his cavity. It produces more thrust with a broad, 'dirty' spectrum from a magnetron than a 'pure' sinusoidal source..

Offline SleeperService

  • Member
  • Posts: 48
  • Liked: 15
  • Likes Given: 15
A feedback circuit would be required in order to keep the VCO on the resonant frequency. If such a feedback is possible then a simple MCU could be programmed to do that using an ADC as output, it would not be a PLL anyway but more similar to an FLL.
Where does the feedback signal come from?

Offline matthewpapa

  • Member
  • Posts: 9
  • Houston, TX
  • Liked: 15
  • Likes Given: 22
I have 2 high performance boxes at my house you may want to consider using for simulation purposes
2p Sandy Bridge (16 cores, 32 threads total @2.8 GHz) about 50 GB of DDR3 memory
2p Haswell (20 cores, 40 threads total @3.1 GHz) and 32 GB of DDR4 memory

PM me if that can help at some point

Offline Giovanni DS

  • Regular
  • Full Member
  • *
  • Posts: 186
    • ChibiOS/RT Project
  • Liked: 44
  • Likes Given: 162
A feedback circuit would be required in order to keep the VCO on the resonant frequency. If such a feedback is possible then a simple MCU could be programmed to do that using an ADC as output, it would not be a PLL anyway but more similar to an FLL.
Where does the feedback signal come from?

This is the problem :) I do understand controls and realtime systems but microwaves and resonant cavities are not my field. How do you know it is resonating?

In theory the thrust could be your feedback, you tune the VCO in order to maximize that.

Offline SleeperService

  • Member
  • Posts: 48
  • Liked: 15
  • Likes Given: 15
A feedback circuit would be required in order to keep the VCO on the resonant frequency. If such a feedback is possible then a simple MCU could be programmed to do that using an ADC as output, it would not be a PLL anyway but more similar to an FLL.
Where does the feedback signal come from?

This is the problem :) I do understand controls and realtime systems but microwaves and resonant cavities are not my field. How do you know it is resonating?

In theory the thrust could be your feedback, you tune the VCO in order to maximize that.
I agree, the only way you could do it is to lock the thrust to the freq. How you would do that though to avoid false maxima is beyond me.

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1113
  • Liked: 776
  • Likes Given: 1013
Resonant cavities are also filters. You monitor a sample port to ensure the cavity is resonant at the desired frequency.

Another control loop compensating for doppler drift is needed. Shawyer mentions this type of control loop over at emdrive.com.
« Last Edit: 02/18/2015 03:46 PM by Mulletron »
Challenge your preconceptions, or they will challenge you. - Velik

Offline SleeperService

  • Member
  • Posts: 48
  • Liked: 15
  • Likes Given: 15
Resonant cavities are also filters. You monitor a sample port to ensure the cavity is resonant at the desired frequency.

Another control loop compensating for doppler drift is needed. Shawyer mentions this type of control loop over at emdrive.com.
It seems to me that it matters where you put the sample port with a truncated frustum.
Am I wrong? Would a sample port to feedback the PLF work in any location?
I'll leave Doppler drift for another time... :)
« Last Edit: 02/18/2015 03:51 PM by SleeperService »

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1113
  • Liked: 776
  • Likes Given: 1013
I think we're getting on a tangent. But here's your answer.

http://www.radartutorial.eu/03.linetheory/tl11.en.html
Challenge your preconceptions, or they will challenge you. - Velik

Tags: