Author Topic: EM Drive Developments Thread 1  (Read 1472535 times)

Re: EM Drive Developments
« Reply #3180 on: 11/13/2014 06:02 pm »
Hi all, I came here from a reddit post regarding the EM drive: http://www.reddit.com/r/EmDrive/comments/2ih0mh/rapid_spread_of_emdrive_technology_by_the_diy/

I'm interested in funding a private sector test of this tech.  Does anyone here have the capabilities of assembling a team that can create a testable EMDrive?  If so, lets talk budget.

Thanks,

Jordan Greenhall

Offline JohnFornaro

  • Not an expert
  • Senior Member
  • *****
  • Posts: 10974
  • Delta-t is an important metric.
  • Planet Eaarth
    • Design / Program Associates
  • Liked: 1257
  • Likes Given: 724
Re: EM Drive Developments
« Reply #3181 on: 11/13/2014 06:05 pm »
Oh, I can assemble a team, all right.
Sometimes I just flat out don't get it.

Offline scienceguy

  • Regular
  • Full Member
  • ****
  • Posts: 836
  • Lethbridge, Alberta
  • Liked: 155
  • Likes Given: 279
Re: EM Drive Developments
« Reply #3182 on: 11/13/2014 06:09 pm »
They say this thing needs to be tested in space. Is that because it needs 0 g? What about testing it on an air track where there is almost no friction?
e^(pi*i) = -1

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1150
  • Liked: 837
  • Likes Given: 1071
Re: EM Drive Developments
« Reply #3183 on: 11/13/2014 06:15 pm »
Hi all, I came here from a reddit post regarding the EM drive: http://www.reddit.com/r/EmDrive/comments/2ih0mh/rapid_spread_of_emdrive_technology_by_the_diy/

I'm interested in funding a private sector test of this tech.  Does anyone here have the capabilities of assembling a team that can create a testable EMDrive?  If so, lets talk budget.

Thanks,

Jordan Greenhall

Hi Jordan, welcome to the group. We need intrepid individuals to help tackle this problem. Please invite your friends.
And I can feel the change in the wind right now - Rod Stewart

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1150
  • Liked: 837
  • Likes Given: 1071
Re: EM Drive Developments
« Reply #3184 on: 11/13/2014 06:19 pm »
They say this thing needs to be tested in space. Is that because it needs 0 g? What about testing it on an air track where there is almost no friction?

Well we certainly need to test this in vacuum, which can be done here on earth. Testing in space is certainly important too, that is if it can survive testing here on Earth. Welcome to the group. Please tell your friends too!
And I can feel the change in the wind right now - Rod Stewart

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5911
  • USA
  • Liked: 6124
  • Likes Given: 5564
Re: EM Drive Developments
« Reply #3185 on: 11/13/2014 06:26 pm »
I received another very interesting e-mail from Bob Ludwick, that I reproduce below:




From: Robert Ludwick
Sent: Wednesday, November 12, 2014 1:54 PM
To: Dr. J. Rodal
Subject: Testing the EmDrive

Hello Dr. Rodal

Although thrust without throwing something out the back is at least improbable (I am of course rooting for the improbable.), I think that the testing problem (to rule out heat artifacts) could be resolved by the test plan I proposed awhile back. 



i. e. 

A.  Establish the resonant frequency (s) and bandwidths of the thruster.

B.  Select a test frequency range that is at least double the bandwidth of the thruster, so that the start and stop frequencies are well outside the high Q region of the thruster.

C.  Select frequency steps so that you are guaranteed AT LEAST ten steps in the high Q region of the thruster.

D.  Set the test frequency to the start frequency, turn on the power amplifier, and wait 5 minutes or so for any thermal and current/magnetic field effects to stabilize.  Measure the residual thermal/magnetic/whatever ‘thrust’.

E.  Start the frequency sweep, with dwell times on each frequency long enough for the mechanical system to settle. Change NOTHING other than frequency.

F.  For each frequency step, record forward and reflected power from the thruster.

G.  After allowing for mechanical settling time, record the thrust.

H.  Go to the next frequency and repeat. 




If there is any ‘anomalous’ thrust related to thruster Q this procedure will detect it.  If the ‘thrust’ is due to thermal effects, it should remain constant throughout the test, as the power/current will be constant throughout the test

It DOES require a highly stable, computer controlled signal source rather than a VCO with a knob, but those, including those suitable for testing superconducting cavities, are available from any equipment rental place (such as ElectroRent) if the lab is too cheap to buy one.  Power meters, too.

I am completely baffled at the apparent disinterest of people and organizations who should be foaming at the mouth at the prospect of getting their hands on a relatively simple device that can convert microwave power into translational motion at efficiencies orders of magnitude better than simple photon rockets.  Apparently they have decided that it is prima facie impossible and therefore don’t want to waste any time or money in finding the problems in some fringe PhD’s test setup.

On the other hand, if I were controlling the budget for spaceships in any form and was aware that at least three disparate groups had detected thrust from EmDrive-like devices, I would want to confirm or refute this thing ASAP.  I’d have lab crews—more than one, at different labs, using different equipment--working overtime until I knew, one way or another, whether it was real or not.  And I would insist in more than one ‘fail’ before I called a halt.  Frankly, doing so should be cheap AND fast.  And the stakes are enormous.

Bob Ludwick
I completely agree with this except I would note that there is very serious challenge in the notion of changing frequency and NOTHING else.  Dr. Ludwick points this out, and I agree with him though I'm not sure how one would do what he suggests.

Bob Ludwick answered as follows:




From:   Robert Ludwick
Sent:   Wednesday, November 12, 2014 7:05 PM
To:   Dr. J. Rodal
Subject:   Re: Testing the EmDrive

As for the difficulty of changing the frequency and nothing else, I don’t see  it, but maybe I don’t understand the problem that is being referred to. 

To run the test, you turn on  the amplifier and all the test equipment, let it stabilize (Normally, in the labs I worked in, the sig gens et all remained powered up 24/7/365, so they didn’t require stabilization time.), and click ‘Run’ on the control computer.   The control program which you have written will ask for the start frequency,  the stop frequency, the frequency step size, and the desired output level for the signal generator driving the power amplifier.  Once those have been  entered, the computer will just execute the steps I have outlined.  Of course if you define ‘change nothing’ rigorously enough, it becomes difficult, but for the purposes of this exercise, changing the output frequency of the signal generator is done via software commands and occurs in microseconds, typically,  the drive is leveled to small fractions of a dB by the sig gen leveling circuits, and over narrow sweep ranges the variations in the current drawn by the amplifier are negligible, as are its variation in output level (which are monitored by the power meter and recorded).  So at least to a first approximation, ‘changing nothing' should be pretty simple. 

The program should run to completion, in a time depending on the number of steps and the settling time allowed for each step.  As the program runs, the computer plots a running graph of measured thrust vs frequency (while saving ALL data to a test file), so that the operator can see what’s happening in real time.

The thrust measurement in the torsional pendulum should occur under computer control, without any operator intervention, once the procedure is known. 

The only ‘moving part’ during the whole procedure is whatever mechanical movement occurs in the torsional pendulum as a response to the (hopefully) varying thrust as the frequency sweeps through thruster resonance.

The point is, with all the gory details coming with getting the thrust data back from the torsional pendulum, with my procedure everything should be steady state EXCEPT for the response of the thruster as the drive frequency sweeps through the high Q part of its frequency response.  There are no moving parts and no transients.  The current (and associated magnetic fields) to the amplifier should remain constant and the amplifier power into the thruster, and thus its heating effect, should remain constant within small fractions of a dB. 

You don’t have to worry about what material the end caps are made of, their temperature response curves, or any other material properties of the thruster, unless the measured thrust does in FACT vary with the Q at the drive frequency.  At that point, physicists need to figure out what is REALLY going on. 

On the other hand, if, after it is powered up and allowed to achieve its steady state temperature, it just hangs there doing nothing as the sig gen is stepped through the frequency of peak Q, we can all go home.
« Last Edit: 11/13/2014 06:29 pm by Rodal »

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5911
  • USA
  • Liked: 6124
  • Likes Given: 5564
Re: EM Drive Developments
« Reply #3186 on: 11/13/2014 06:41 pm »
Oh, I can assemble a team, all right.
What kind of a team can you assemble?

Who is on your team?

Offline Ron Stahl

  • Full Member
  • **
  • Posts: 210
  • Liked: 32
  • Likes Given: 2
Re: EM Drive Developments
« Reply #3187 on: 11/13/2014 07:27 pm »
No

1) They did not specify (in the "Anomalous ... " report) the Q before or after removing the dielectric

2) They did test (removing the dielectric) very early in the testing program

3) They did specify the frequency at which they performed this test (removing the dielectric) and it was a frequency much higher than for the other reported tests.  Therefore these tests (removing the dielectric) are highly questionable. 

4) Furthermore @Mulletron's concerns regarding resonance before and after are well thought out. 
No.  The dielectric slows the em through it, so the chamber is effectively a higher frequency chamber without it.  It is presuming the folks at Eagle are stupid in the extreme to think they would not note this most obvious issue.

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5911
  • USA
  • Liked: 6124
  • Likes Given: 5564
Re: EM Drive Developments
« Reply #3188 on: 11/13/2014 07:38 pm »
No

1) They did not specify (in the "Anomalous ... " report) the Q before or after removing the dielectric

2) They did test (removing the dielectric) very early in the testing program

3) They did specify the frequency at which they performed this test (removing the dielectric) and it was a frequency much higher than for the other reported tests.  Therefore these tests (removing the dielectric) are highly questionable. 

4) Furthermore @Mulletron's concerns regarding resonance before and after are well thought out. 
No.  The dielectric slows the em through it, so the chamber is effectively a higher frequency chamber without it.  It is presuming the folks at Eagle are stupid in the extreme to think they would not note this most obvious issue.

Why didn't "the folks at Eagleworks" report the Q with and without the dielectric in their report?

as you expected:

Quote from: Ron Stahl
I'm pretty sure they were specific that the Q was measured to be very high without the dielectric , which was I believe the first way they tested it.

You seem to have information from other sources that does not emanate from the "Anomalous..." report:

Where in the report are they so specific concerning the Q measured to be very high without the dielectric?

Please give the page in where that is stated

« Last Edit: 11/13/2014 07:54 pm by Rodal »

Offline Mulletron

  • Full Member
  • ****
  • Posts: 1150
  • Liked: 837
  • Likes Given: 1071
Re: EM Drive Developments
« Reply #3189 on: 11/13/2014 07:43 pm »
Posting this again.
https://www.fourmilab.ch/gravitation/foobar/

This is a DIY on how to build your own torsion balance at home. This setup is sensitive enough to measure the gravitation between chunks of lead. Good enough to measure "anomalous thrust" if you so desire to build your own test article.
And I can feel the change in the wind right now - Rod Stewart

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5911
  • USA
  • Liked: 6124
  • Likes Given: 5564
Re: EM Drive Developments
« Reply #3190 on: 11/13/2014 07:48 pm »
Posting this again.
https://www.fourmilab.ch/gravitation/foobar/

This is a DIY on how to build your own torsion balance at home. This setup is sensitive enough to measure the gravitation between chunks of lead. Good enough to measure "anomalous thrust" if you so desire to build your own test article.

Good info.  Much more similar to the classical Cavendish setup.

No Christmas-tree with hanging ornaments setup.
No 1.5 by 1.5 inch Faztek long cantilevered beam sticking out with the EM Drive at the end. 
No C-Flex Riverhawk with uncertain spring constant.
« Last Edit: 11/13/2014 07:59 pm by Rodal »

Offline Notsosureofit

  • Full Member
  • ****
  • Posts: 691
  • Liked: 747
  • Likes Given: 1729
Re: EM Drive Developments
« Reply #3191 on: 11/13/2014 07:56 pm »
Posting this again.
https://www.fourmilab.ch/gravitation/foobar/

This is a DIY on how to build your own torsion balance at home. This setup is sensitive enough to measure the gravitation between chunks of lead. Good enough to measure "anomalous thrust" if you so desire to build your own test article.

Good info.  The classical Cavendish setup.

No Christmas-tree with hanging ornaments setup.
No 1.5 by 1.5 inch Faztek long cantilevered beam sticking out with the EM Drive at the end. 
No C-Flex Riverhawk with uncertain spring constant.

Well, I certainly have vacuum chambers .......

Offline Ron Stahl

  • Full Member
  • **
  • Posts: 210
  • Liked: 32
  • Likes Given: 2
Re: EM Drive Developments
« Reply #3192 on: 11/13/2014 08:13 pm »
You seem to have information from other sources that does not emanate from the "Anomalous..." report:

Where in the report are they so specific concerning the Q measured to be very high without the dielectric?

Please give the page in where that is stated

I know Paul.  He was my mentor for two years.  We still pass notes from time to time.  If my memory was accessing stuff we discussed rather than the paper, that is my whoops.  Just saying, these are not the kinds of mistakes Paul nor anyone at Eagle would make.  If you have to presume they are stupid, your presumption will be wrong.

Sonny's model does not require anything in the chamber.  It is Paul who believes QVF and M-E theory are opposite sides of the same coin, who pressed to put a dielectric in the resonator.  He chose something with a fairly linear response, so he wouldn't have the difficulties of frequency dependence, etc.  In this he once mentioned he was looking at Teflon.  If I say more without rereading the report I might stumble upon something I was not meant to share, so lets leave it there, shall we?

it is worth noting just as Mulletron says, one can tune these sorts of cavities by putting in some dielectric.  If one wants to do a parametric study with changing frequency, the simplest way is to place varying amounts of dielectric thickness in, and then use the PPL circuit to find the resonance.  I know they're using PPL resonance tuning, so it would appear they intend to do this kind of study.  Lets remember, they only got the rig up and running a few days before this paper.  They should already have done much more since the paper than they did before it.  And note too, this method does not change only the frequency, so analysis of the generated data will be more difficult to interpret than what Dr. Ludwick was hoping for.

Offline frobnicat

  • Full Member
  • ****
  • Posts: 518
  • Liked: 500
  • Likes Given: 151
Re: EM Drive Developments
« Reply #3193 on: 11/13/2014 11:45 pm »
The sweeping methodology sounds nice, but a "simple" longer power on test could learn a lot too. What is the record of time with power on for any of the propellentless devices so far ? Why not try to let it run for longer than a few 10s of seconds ? 10 minutes, 1 hour ? 20W is not that much power to reach equilibrium and dissipate from inside a box of metal like a vacuum chamber...

Or is it the stability of the resonance that can't be kept that long ? And while I'm at it, what is the most convincing experimental proof so far that those devices really need to be operated at resonance ? I remember my search on fitting formulas did show a strong correlation with Q but it was only on 7 or 6 data points. And correlation with Q don't mean correlation with resonance : we can have a high Q cavity driven off resonance. Actually the higher the Q the higher the risk that it is driven off resonance.

I'm thinking of the results of Juan 2012 that did a parameter study on power input but that turns out to be a correlated parameter study on frequency since the power spectrum of their source varied with power output (output from source, input to cavity). They analyse the strong non linearity and even non monotonicity of their graph thrust vs power (fig 4 page 8 ) as the result of this complex frequency dependence on power of their source (fig 6 page 9). BTW, we see a number of data points on the fig 4 that's impressive when compared to what "anomalous thrust..." has to offer. On the other hand we are not gratified with a thrust vs time for pulses. Not clear, did they swept on power and recorded whopping hundreds milliNewtons all the way through ? For how long ? Need to read in detail again. Also the (a) and (b) curves of figure 4 are not in complete agreement, different sources ? Unfortunately the curves labels on fig 4 are not translated, multiple runs ?

Anyway, from what I understand it states that the experimental resonance is at 2.450GHz and experimental bandwidth is 2.4492 to 2.4508GHz (that is delta_f= 0.0016GHz, that would be Q=2.45/.0016=1531 only, not specially "high Q", have I missed something ???). And yet we see (hardly) from fig 6 page 9 that most of the source power (experimental) is outside this narrow window. Measured thrusts vs raw power do roughly follow the amount of power effectively in the resonance window (dependant on raw power) as discussed page 10.

Albeit I have a hard time at interpreting the vertical scales of fig. 6. For instance fig. 6 (b) (top right) shows what's going on at 300W raw power (again if I understand anything at all) which gives supposedly the higher "yield" in resonance bandwidth (little insert box). The text page 10 tells that the "actual maximum output power of the microwave source" (I interpret that as being the part of power in resonance) of 120W. But from the overall graph of fig 6 (b) we can't really see one third of power in the window, more like 2 or 3% (10W, not 120W) Most of the spectrum content is above 2.454, way above the bandwidth... The overall given power (not just on the resonance window) is also varying with raw power when looking at the vertical scales (that don't make sense for a spectrum : should be Watts per Hz, not Watts). Is a magnetron exhibiting such nonlinearities in overall RF power vs input power ?

I'm getting tired of those broken reports, or is it me ? Anyone willing to either confirm those inconsistencies or explain what is going on with Q and bandwidth and resonance on this report ? That Juan report was source of 2 of the 6 (or 7 with the "statistical oulier Brady b") that was used to get a grip on the parameters mutual dependencies with Mac Culloch's formula and with exhaustive searches...

If I'm interpreting the graphs of Juan et al 2012 at face value, we have either an effect that don't care that much on seeing most of its power being at some resonance or not (and that don't care having a Q of 1530). Or we have an effect that uses just the resonance part (the gross fraction that is off resonance is just wasted heat) and what I see is on the order of 10Watts (with a clean efficient source modestly tuned to Q 1530) to get 300mN, a staggering 30 N/kW. Round trip to Saturn in two weeks anyone ? 8)

ps. they do appear to use an inverted pendulum, ahem, that is with centre of gravity clearly above pivot + a flex bearing with stiffness to compensate the unstability. If I get it, they try to have a "metastable" equilibrium point to increase sensitivity ? Why would they need/want such sensitivity to measure 10s of millinewton ? (many gram-force !)
« Last Edit: 11/13/2014 11:54 pm by frobnicat »

Offline ThinkerX

  • Full Member
  • ***
  • Posts: 341
  • Alaska
  • Liked: 126
  • Likes Given: 63
Re: EM Drive Developments
« Reply #3194 on: 11/14/2014 12:40 am »
Quote
Hi all, I came here from a reddit post regarding the EM drive: http://www.reddit.com/r/EmDrive/comments/2ih0mh/rapid_spread_of_emdrive_technology_by_the_diy/

I'm interested in funding a private sector test of this tech.  Does anyone here have the capabilities of assembling a team that can create a testable EMDrive?  If so, lets talk budget.

Quote

They say this thing needs to be tested in space. Is that because it needs 0 g? What about testing it on an air track where there is almost no friction?

While there have been a couple of DIY types who dropped into this thread, most of the people posting are concerned with trying to figure out how and why the device works in the first place, and whether or not the reported results are the result of a 'false positive' or experimental artifact.

It must be pointed out that the explanations put forth by the creators of these devices run directly contrary to major, well established scientific laws, notably 'conservation of momentum.'  That said, the reported results, from different
groups in different countries using devices differing somewhat from each other did produce what APPEARS to be positive results.  The problem is reconciling or explaining those results in a manner consistent with known science.

Several options have been investigated to a greater or lesser extent over the past 150 pages or so of this thread.  These include:

1) The EM Drive is pushing against 'Dark Matter,' which is just barely workable if there is a lot of Dark Matter in the area.

2)  The EM Drive is a sort of Biefield - Brown device, essentially a high voltage all electric aircraft that fly's by ionizing the air underneath it.  These devices are legitimate; hobbyist of various sorts have been making the things for decades.  You can find videos of them in action on You-Tube under 'anti-gravity.'  This possibility was rejected because the EM Drive devices are low voltage mechanisms - simply not enough power.  Also, a Biefield-Brown device won't function in a vacuum. 

3) The EM Drive...attracts...'Unruh Radiation,' a theoretical 'force' behind Inertia.  This explanation gets into known cosmological problems involving expanding space-time.  One effect: despite gravitational attraction, galaxies are being 'pushed' away from each other at a constant rate roughly equal to 1 kilometer per second.  A Doctor McCulloch, noted physicist and occasional poster in this thread has published papers using Unruh Radiation as a means to explain tiny anomalies in the velocities of several spacecraft.  However, this effect, while fascinating, is both unconfirmed and probably confined to 'deep space' - at least by the posters here.

4) One or two posters here have recently begun looking again into the possibility the EM Drive may, after all, really be tapping into the Quantum Vacuum.  However, there remain severe problems with this.

5) The explanation most closely looked at now is that the effects produced by the EM Drive are a thermal artifact compounded by a flawed measuring device.  According to this hypothesis, the devices should produce miniscule amounts of thrust in an atmosphere, and no thrust at all in space, making it effectively worthless as a space propulsion system. 

It is also worth pointing out that the experimenters have been withholding key data about their devices, making independent verification difficult.  A EM Drive device probably could be built on a modest budget; the question is whether it would work.

















Offline JohnFornaro

  • Not an expert
  • Senior Member
  • *****
  • Posts: 10974
  • Delta-t is an important metric.
  • Planet Eaarth
    • Design / Program Associates
  • Liked: 1257
  • Likes Given: 724
Re: EM Drive Developments
« Reply #3195 on: 11/14/2014 01:55 am »
Oh, I can assemble a team, all right.
1) What kind of a team can you assemble?

2) Who is on your team?

1)  The best.

2) Top men.
Sometimes I just flat out don't get it.

Offline JohnFornaro

  • Not an expert
  • Senior Member
  • *****
  • Posts: 10974
  • Delta-t is an important metric.
  • Planet Eaarth
    • Design / Program Associates
  • Liked: 1257
  • Likes Given: 724
Re: EM Drive Developments
« Reply #3196 on: 11/14/2014 02:46 am »
This is the consequence of all the forces of physics being gauge invariant: absolute values don't exist--only differences matter.

Correct me if memory fails; this fellow wrote AutoCAD, which I've used to gracxe these pages here and there.

This is an example of why I continue to insist on the paramount importance of stating principles in English before devolving to math. Everything that needs to be said, and that has to be saod. should be reduced to English.

Sorry Frob.  French is fine in France, but dis be Amerika.

Given the philosophical implications of Newton's theory, it's interesting to speculate what might have happened had Archimedes discovered the universal nature of gravitation.

Other brains besides mine have speculated on what I call the "Spartacus Conjecture".  Mankind was already smart enough to attempt off planet travel and colonizxation as early as 0AD/BC.

Bottom line?  Take your math and shove it!

With the deepest respect for the utility and necessity of math.

Would such a discovery in Archimedes' time have had an impact comparable to Newton's or, occurring in a very different social and intellectual milieu, would it have been regarded as no more than a curiosity? How might human history have played out had the Enlightenment begun 1900 years before Newton?

All it takes is three hundred years to start colonizing the solar system.  Why haven't we?  We're being kept on planet.

But I digress.
Sometimes I just flat out don't get it.

Offline JohnFornaro

  • Not an expert
  • Senior Member
  • *****
  • Posts: 10974
  • Delta-t is an important metric.
  • Planet Eaarth
    • Design / Program Associates
  • Liked: 1257
  • Likes Given: 724
Re: EM Drive Developments
« Reply #3197 on: 11/14/2014 02:49 am »
The sweeping methodology sounds nice, but a "simple" longer power on test could learn a lot too. What is the record of time with power on for any of the propellentless devices so far ?

Me thank.  Me not get pragmatic utility of "xx" ms or whatever of power input.  Me think must have device on for many moons before get anywhere interesting.

That being if even can detect force period.
Sometimes I just flat out don't get it.

Offline ChrisWilson68

  • Senior Member
  • *****
  • Posts: 5266
  • Sunnyvale, CA
  • Liked: 4992
  • Likes Given: 6459
Re: EM Drive Developments
« Reply #3198 on: 11/14/2014 03:01 am »
Oh, I can assemble a team, all right.
What kind of a team can you assemble?

Who is on your team?

This is an example of why I continue to insist on the paramount importance of stating principles in English before devolving to math. Everything that needs to be said, and that has to be saod. should be reduced to English.

All it takes is three hundred years to start colonizing the solar system.  Why haven't we?  We're being kept on planet.

Rodal, I think this should be enough to answer your questions about John.

Offline zen-in

  • Full Member
  • ****
  • Posts: 541
  • California
  • Liked: 483
  • Likes Given: 371
Re: EM Drive Developments
« Reply #3199 on: 11/14/2014 04:30 am »
I received another very interesting e-mail from Bob Ludwick, that I reproduce below:




From: Robert Ludwick
Sent: Wednesday, November 12, 2014 1:54 PM
To: Dr. J. Rodal
Subject: Testing the EmDrive

Hello Dr. Rodal

Although thrust without throwing something out the back is at least improbable (I am of course rooting for the improbable.), I think that the testing problem (to rule out heat artifacts) could be resolved by the test plan I proposed awhile back. 



i. e. 

A.  Establish the resonant frequency (s) and bandwidths of the thruster.

B.  Select a test frequency range that is at least double the bandwidth of the thruster, so that the start and stop frequencies are well outside the high Q region of the thruster.

C.  Select frequency steps so that you are guaranteed AT LEAST ten steps in the high Q region of the thruster.

D.  Set the test frequency to the start frequency, turn on the power amplifier, and wait 5 minutes or so for any thermal and current/magnetic field effects to stabilize.  Measure the residual thermal/magnetic/whatever ‘thrust’.

E.  Start the frequency sweep, with dwell times on each frequency long enough for the mechanical system to settle. Change NOTHING other than frequency.

F.  For each frequency step, record forward and reflected power from the thruster.

G.  After allowing for mechanical settling time, record the thrust.

H.  Go to the next frequency and repeat. 




If there is any ‘anomalous’ thrust related to thruster Q this procedure will detect it.  If the ‘thrust’ is due to thermal effects, it should remain constant throughout the test, as the power/current will be constant throughout the test

It DOES require a highly stable, computer controlled signal source rather than a VCO with a knob, but those, including those suitable for testing superconducting cavities, are available from any equipment rental place (such as ElectroRent) if the lab is too cheap to buy one.  Power meters, too.

I am completely baffled at the apparent disinterest of people and organizations who should be foaming at the mouth at the prospect of getting their hands on a relatively simple device that can convert microwave power into translational motion at efficiencies orders of magnitude better than simple photon rockets.  Apparently they have decided that it is prima facie impossible and therefore don’t want to waste any time or money in finding the problems in some fringe PhD’s test setup.

On the other hand, if I were controlling the budget for spaceships in any form and was aware that at least three disparate groups had detected thrust from EmDrive-like devices, I would want to confirm or refute this thing ASAP.  I’d have lab crews—more than one, at different labs, using different equipment--working overtime until I knew, one way or another, whether it was real or not.  And I would insist in more than one ‘fail’ before I called a halt.  Frankly, doing so should be cheap AND fast.  And the stakes are enormous.

Bob Ludwick
I completely agree with this except I would note that there is very serious challenge in the notion of changing frequency and NOTHING else.  Dr. Ludwick points this out, and I agree with him though I'm not sure how one would do what he suggests.

Bob Ludwick answered as follows:




From:   Robert Ludwick
Sent:   Wednesday, November 12, 2014 7:05 PM
To:   Dr. J. Rodal
Subject:   Re: Testing the EmDrive

As for the difficulty of changing the frequency and nothing else, I don’t see  it, but maybe I don’t understand the problem that is being referred to. 

To run the test, you turn on  the amplifier and all the test equipment, let it stabilize (Normally, in the labs I worked in, the sig gens et all remained powered up 24/7/365, so they didn’t require stabilization time.), and click ‘Run’ on the control computer.   The control program which you have written will ask for the start frequency,  the stop frequency, the frequency step size, and the desired output level for the signal generator driving the power amplifier.  Once those have been  entered, the computer will just execute the steps I have outlined.  Of course if you define ‘change nothing’ rigorously enough, it becomes difficult, but for the purposes of this exercise, changing the output frequency of the signal generator is done via software commands and occurs in microseconds, typically,  the drive is leveled to small fractions of a dB by the sig gen leveling circuits, and over narrow sweep ranges the variations in the current drawn by the amplifier are negligible, as are its variation in output level (which are monitored by the power meter and recorded).  So at least to a first approximation, ‘changing nothing' should be pretty simple. 

The program should run to completion, in a time depending on the number of steps and the settling time allowed for each step.  As the program runs, the computer plots a running graph of measured thrust vs frequency (while saving ALL data to a test file), so that the operator can see what’s happening in real time.

The thrust measurement in the torsional pendulum should occur under computer control, without any operator intervention, once the procedure is known. 

The only ‘moving part’ during the whole procedure is whatever mechanical movement occurs in the torsional pendulum as a response to the (hopefully) varying thrust as the frequency sweeps through thruster resonance.

The point is, with all the gory details coming with getting the thrust data back from the torsional pendulum, with my procedure everything should be steady state EXCEPT for the response of the thruster as the drive frequency sweeps through the high Q part of its frequency response.  There are no moving parts and no transients.  The current (and associated magnetic fields) to the amplifier should remain constant and the amplifier power into the thruster, and thus its heating effect, should remain constant within small fractions of a dB. 

You don’t have to worry about what material the end caps are made of, their temperature response curves, or any other material properties of the thruster, unless the measured thrust does in FACT vary with the Q at the drive frequency.  At that point, physicists need to figure out what is REALLY going on. 

On the other hand, if, after it is powered up and allowed to achieve its steady state temperature, it just hangs there doing nothing as the sig gen is stepped through the frequency of peak Q, we can all go home.

I agree with most of the above.   However there is one problem with stepping the frequency.  The chamber has a very high Q so any RF power that is outside the very narrow passband will be reflected.   Maybe this is the reason why the em-drive didn't produce any "anomalous thrust" when the dielectric was removed:  The passband shifted.    I still think the best control for this experiment would be a 50 Watt resistor inside the chamber, fastened to the same location as the RF feed.    Apply a DC voltage to this resistor so that the same power is dissipated inside the chamber.   Is a thrust measured under this set of conditions?

Tags:
 

Advertisement NovaTech
Advertisement Northrop Grumman
Advertisement
Advertisement Margaritaville Beach Resort South Padre Island
Advertisement Brady Kenniston
Advertisement NextSpaceflight
Advertisement Nathan Barker Photography
0