Author Topic: EM Drive Developments - related to space flight applications - Thread 2  (Read 3314773 times)

Offline frobnicat

  • Full Member
  • ****
  • Posts: 518
  • Liked: 500
  • Likes Given: 151
Ok, despite my best efforts (which, admittedly are rather feeble), I can no longer tell what 'frobnicat' is attempting to accomplish.  Is he arguing that all of the 'thrust' produced by this device is some sort of thermal artefact?   From what I can tell, the goalposts seem to have moved at least once.

Executive summary time, please.

First we were on recoil effects. Recoil effects depend on a mass accelerating and keeping a constant acceleration if they are to mimic thrust effects. There is not enough thermal expansion room (by far) to account for such constant acceleration that could explain a sustained "thrust" for 40s or so.

The disparity of values in the vertical scale (µm readings) from chart to chart, for identical 29.1µN calibration pulses, have made me to wonder if the whole balance was tilted, and that the variation of tilt and/or offbalance could account for that disparity. I asked to Star-Drive if the balance was tilted or not and he replied that it was, that it was necessary because a perfectly horizontal arm (perfectly vertical axis) wouldn't settle to stable rest position (before runs). Whether this tilt/offbalance does indeed explain (in whole or in part) the vertical variations (µm readings) from chart to chart is still unconfirmed.

What we know for sure is that there is a tilt and an offbalance, enough to play a role in the rest equilibrium of the arm. That means that gravity plays a role in the angular rest position and that the angular rest position (the position when there is no force and no thrust imposed to the arm) will react to changes in position of masses of the test article relative to fixation point on the arm. The orientation of this change is counter-intuitive : when looking the apparatus in front ( frustum's small end to the left, big end to the right) a shift of mass to the left will shift the rest equilibrium to a new position with the front arm leftward, thus increasing the measured distance (a rise in the charts). This is because the global centre of mass of the whole rotating assembly is behind the axis of rotation and the tilt is such that this backward part of the arm is below the front part. If all is inverted (meaning putting more weight in front, and lowering front relative to back) then the orientation would be also inverted.

Anyway, there is here a mechanism by which a slow displacement of mass from point A to point B (typically 1kg by a few µm, or 100g by a few 10s of µm) will change the readings from a distance to another distance. What would appear as a sustained "thrust" would in fact be a mass just sitting at another place when heated relative to when not heated. Is it certain that such mechanism could account for whole the signal ? No, this is not certain for me, at this point. It would take an accurate simulation of thermal deformations in function of time and a proper characterisation of the pendulum dynamic (tilt included) together to reconstruct the signal that "change of rest position effect" would give. I would say that it is at least likely messing with the thrust signal (if there is one).

To add confusion even more, we already know of another "change of rest position effect" that is not linked to thermal change of mass position (also this is not completely sure it isn't) but rather to heating of the flexure bearings either by IR at the back of electronic amplifiers (thread 1, mitigated since) and/or by return ground currents. This is the cause of the infamous "drifting baseline", effect that easily reaches in magnitude the magnitude of the signal (also with a different shape).

So yes the goalposts moved a little bit. And my leading hypothesis of thread 1 (warm air jet) and a number of other hypothesis are now nullified by Eagleworks working in vacuum. Great. This is advancing.

Offline ThinkerX

  • Full Member
  • ***
  • Posts: 341
  • Alaska
  • Liked: 126
  • Likes Given: 63
Just to keep things straight, a list of experimental artefacts that have been fully or partially accounted for might be in order.  Otherwise, I see a fair possibility of 're-inventing the wheel' so to speak.

Offline frobnicat

  • Full Member
  • ****
  • Posts: 518
  • Liked: 500
  • Likes Given: 151
....

I wanted to show that the measured "force" (LDS reading really) might not be a force at all, but a change in equilibrium rest position, as per the diagram showing how the equilibrium rest position is different depending on test article CoM position...
Sorry again for not recalling your prior posts and needing a brief summary.

QUESTION1: have you calculated the change in Center Of Mass position for these different cases?  (yes or no ?)

 if your answer to QUESTION1 is no, stop reading.
Er, yes, no, this is in progress but would need a simulation with professional level tools I haven't knowledge of to be assessed with any certainty. Needs both buckling and thickness gradient flexing deformations, with asymmetrical thin plate modelling (single sided PCB). Amongst others, Code Aster is a free simulation framework of more than 1000000 lines code that could do the job but require quite a training to be used correctly. Right now I'm trying to do order of magnitude estimates with small personal code ~500 lines of code. Is it in the ballpark or not ? Preliminary results say maybe.

I haven't "published" yet because, you know, people want to know the details ( rightly so  :) ) and it takes time. The problem with this line of reasoning is not the magnitude but rather the shapes : with low thermal evacuation (in vacuum) the temperatures tend to rise linearly for the whole run of 40s power-on, and then tend to stay at their new acquired temperature at power-off rather than falling back to equilibrium in a symmetrical manner. It gives linear ramp up followed by slowly decay plateau rather than (more or less) quick rise to plateau followed by (more or less) quick decay back to baseline, as observed. That's for the absolute temperatures. But some temperature differences (gradients) show a temporal shape more like the expected shape. See for instance this very rough and simplified model (FR4 heat capacity at 600, will try the value of 1200 you found later, my value came from another source that looked serious, don't have the link at hand).



A gradient across the thickness will make the PCB plate warp. How much ? I don't know : the PCB plate is attached on a rim... if it was a thin strip just floating in vacuum with a cylindrical deformation (rather than a stiffer plate with a spherical deformation) deflection at the belly = L²/(8*d) * ExpanCoef*(T2-T1) where T2 is hot side, T1 cold side, L the rest length (28cm), d the depth (1.6mm) and ExpanCoef is about 1.3e-5m/m/K (FR4 inplane, neglecting the copper for now). That is around 80µm for 1°C difference across the thickness. How much less for a real plate fixed to rim ? All I can say at this point is, we are not talking below µm thermal effects, thermal effects with the appropriate dynamic shape.

The "active part" of PCB (that can warp) of 28cm diameter should weigh around 200g. Very very roughly (order of magnitude) rest of test article is in the 2kg, 10 times more. So the displacement of CoM of test article is roughly a tenth of the displacement of CoM of PCB due to buckling/warping. To "explain" a 1µm deviation in rest position at the LDS could possibly need a bit less than 1µm change of CoM position of the test article. This last claim is unsubstantiated by equations so far, it is just an observation of the 3 bottom diagram above (to scale), and will depend on the relative stiffness of flexure bearings (why I insist so much on a clarification on this point). Let's say this is in the same ball park. Conclusion (fragile) : sustained 10µm displacement of mean PCB membrane position relative to rim could account for 1µm sustained plateau above baseline on the charts.

Quote
if your answer to QUESTION1 is yes, then,

QUESTION2: do your calculations for the change in center of mass position give a displacement that is close to the displacement vs. time measurements at NASA Eagleworks?

It could.
At least it needs to be properly assessed before it can be ruled out. Also it is important to understand that the magnitude of the effect don't depend linearly on the tilt, if it wasn't for the flexure bearings restoring torque, the same deltaLDS/deltaCoM would be observed whether the tilt is 0.1° or 1° or 10°. Since we have the flexure bearings restoring torques, the magnitude will depend on tilt and rotating assembly's CoM distance from Z and bearing's stiffness. And doing the calculations one will invariably fall on a visible inconsistency of the data : the apparent stiffness in µm/µN (whatever part is due to bearings vs g tilted pendulum position restoring torque) from the cal pulses is in contradiction with the reasonably well known moment of inertia of the rotating assembly + clearly visible ~4.5s pseudoperiod on underdamped charts. (+ in contradiction with the known .007lb-in/° stiffness of bearings + in contradiction with tilt needed to reach such equivalent stiffness that would be above 10°)
 
« Last Edit: 03/15/2015 11:55 pm by frobnicat »

Offline Giovanni DS

  • Regular
  • Full Member
  • **
  • Posts: 220
    • ChibiOS/RT Project
  • Liked: 67
  • Likes Given: 286
Would be possible to mount the thruster on the pendulum in a way that its main axis is aligned with force of gravity? this would put to rest this thermal expansion hypothesis or confirm it.

It could be mounted like on scales, with a passive mass on the other side.

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5911
  • USA
  • Liked: 6124
  • Likes Given: 5564
Observation #1) The explanation (e.g. the jet model from a poster in this thread, and alternative air current conjectures in several blogs in the Internet) that the thrust measurements were due to thermal convection in the air has been nullified by NASA Eagleworks tests in a vacuum at  6.6*10^(-9) standard atmosphere = 0.0000000066 standard atmosphere
If some kind of perfect Ion engine expels a gas with a density as thin as the one in the NASA Eagleworks vacuum chamber, through a one cm3 nozzle and at a relativistic speed (let say 200,000 km/sec) the thrust due to this very thin gas is in the order of several Newtons.

A far more imperfect Ion engine can certainly provide 100,000 less thrust by accident at this level of pseudo-vacuum.
Did you calculate the thrust that your "thought-experiment ion-engine" would produce with "air propellant expelled through a 1 cm^3 nozzle" stored at 1 standard atmosphere and heated in the microwave cavity?  (the same "imperfect Ion engine" that would produce the measured thrust force of 50 microNewtons thrust with "air propellant expelled through a 1 cm^3 nozzle" stored at 0.0000000066 standard atmosphere "by accident"?)

Is the thrust produced by your "thought-experiment ion-engine" consistent with both the NASA Eagleworks results at 1 standard atmosphere and at 0.0000000066 standard atmosphere?



In your "thought-experiment ion-engine", if the directional thrust force is due to an accidental  1 cm^3 opening acting as a nozzle  at one end of the truncated cone , how is it possible for the experimenters to reverse the direction of the thrust force at will?

« Last Edit: 03/17/2015 02:12 pm by Rodal »

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5911
  • USA
  • Liked: 6124
  • Likes Given: 5564
...
Er, yes, no, this is in progress but would need a simulation with professional level tools I haven't knowledge of to be assessed with any certainty...It could.
At least it needs to be properly assessed before it can be ruled out. ...
Suppose (for argument's sake) that your conjecture for shifting center of mass due to thermal expansion could explain the measured response in one direction, which will call the "forward" direction.



Then, how could it possibly explain the measured response in the opposite direction, with the experimenters able to put the EM Drive into "reverse" at will?

Wouldn't that necessitate a cooling effect or a thermal contraction in order for your conjecture to still hold when the experimenters put the EM Drive in reverse?

Due to the laws of thermodynamics, there can only be power dissipation which leads to heat generation, which leads to temperature increase , the microwave electromagnetic field cannot lead to a cooling effect, and hence not to a temperature decrease when put in reverse.

Also the temperature increase leads to a thermal expansion in these materials because they have a positive coefficient of thermal expansion, therefore it is not possible for the electromagnetic field to lead to a thermal contraction.

Hence it appears that your conjecture has already been nullified by NASA Eagleworks reverse experiment, insofar as explaining the total measurement as an artifact: it may only serve to explain a % portion of the measurement as a biasing artifact::

MeasurementFORWARD= EMDriveFORWARDThust + @frobnicatMechanism
MeasurementREVERSE= EMDriveREVERSEThust + @frobnicatMechanism

Assume that the absolute value of the EMDrive thrust is really the same in the forward and reverse directions such that:

+ EMDriveFORWARDThust = - EMDriveREVERSEThust , then:

EMDriveThrust = (MeasurementFORWARD - MeasurementREVERSE) / 2

frobnicatMechanism =  (MeasurementFORWARD + MeasurementREVERSE) / 2



Here is a numerical example, to quantify the above statements:

Consider the Reverse Experiment in relative vacuum where

MeasurementREVERSE = -9.9 microNewtons at 30 W input Power

Consider the Forward Experiment in relative vacuum, taking the initial onset force (see data below) such that:

MeasurementFORWARD = + 30 microNewtons at 30 W input Power

Then

EMDriveThrust = (MeasurementFORWARD - MeasurementREVERSE) / 2 = (+30 - (- 9.9))/2 = 19.95 microNewtons at 30 Watt = 19.95/30 = 67 % of the MeasurementFORWARD

@frobnicatMechanism  = (MeasurementFORWARD + MeasurementREVERSE) / 2 = (+30 + (- 9.9))/2 = 10.05 microNewtons at 30 Watt = 10.05/30 = 33 % of the MeasurementFORWARD

Therefore, @frobnicat's mechanism could be responsible for 1/3 of the forward measurement.

2/3 of the forward measurement could be due to a mechanism of another nature (a mechanism capable of reversing direction).







« Last Edit: 03/17/2015 09:58 pm by Rodal »

Offline Notsosureofit

  • Full Member
  • ****
  • Posts: 691
  • Liked: 747
  • Likes Given: 1729

Offline frobnicat

  • Full Member
  • ****
  • Posts: 518
  • Liked: 500
  • Likes Given: 151
...
Suppose (for argument's sake) that your conjecture for shifting center of mass due to thermal expansion could explain the measured response in one direction, which will call the "forward" direction.



Then, how could it possibly explain the measured response in the opposite direction, with the experimenters able to put the EM Drive into "reverse" at will?

Please define "at will". By the turn of a knob ? I vaguely recall claim that for some modes the thrust is not toward the small end but toward the big end, with Star-Drive concluding something in the line of "there is not one simple explanation fit all". I don't find back in what post there was such statement, maybe you have better luck at it. I certainly don't recall having seen a chart showing such microwave driven mode reversal, where the thrust is in the opposite direction, toward the big end. ( I collect piously all charts published ) and have no recall either of any explicit detailed statement specifying precise data for such a microwave driven mode reversal where the thrust is toward the big end (power, frequency, mode, thrust, duration, dielectric, pressure, conditions ?). If you are aware of such data, please link to, since it could be certainly a decisive information.

The chart you just posted is not a microwave mode driven reversal, it is not a reversal of thrust "at will" by the turn of a knob. It is simply that the test article is turned 180°. Naturally any local effect (barren interactions with nearby vacuum chamber wall) would also reverse, whether real thrust or thermal expansions effects. That the chart you are posting (Chasing Thrust Reversal...) corresponds to a situation where the test article is turned 180°, with the big end to the left and small end to the right, is my understanding of the post by Paul March that disclosed the chart :

Quote from: Star-Drive
...  a picture of the new heat shields for our torque pendulum's upper and lower torsion springs, (more belts and suspenders to mitigate thermal drifts in the TP baseline), the reversed test setup drawing and the best reversed thrust plot obtained just before or during when our second and last 120W max RF amplifier was dying from internal corona discharges around its RF output circulator.

I added bold for the same use of qualifier reversed concerning the second and third image here :

bigger

Here we see that the 01/16/2015 between 16:15 and 16:45 the test article is mounted with the small end toward the right.

bigger

This is qualified as "reversed", small end toward the right. Thrust toward the right. : "Thrust is Right or Down in Plots". Reverse don't qualifies a device that thrusts toward big end. The thrust is always toward the small end. Don't indulge ourselves in absurd controversies on things thrusting in a way, but accelerating the other since reversed and so on. Let's define the thrust by the direction relative to small end. The orientation to the left or to the right of said small end should then be irrelevant.

BTW : we see from comments on top right that Paul March is still (or was at the time of this comment) analysing the impact of thermal expansions on pendulum responses as recoils. Such recoils, while not strictly 0, will have a negligible transient contribution relative to the Z tilted pendulum CoMShift component, this latest greater and sustained effect being in the opposite direction of what is given by the (correct) comments about recoil.

bigger

This is qualified as "reversed", is there any possible doubt that this corresponds to the small end to the right again ? Date/time is 01/16/2015 17:53. About an hour or so after the previous pictures were taken, where the small end was to the right. And the door wide open...
So, the dip in the plot is a thrust toward the small end (again). My thermal effect will (or will not) show an effect toward the small end, whether the small end is left or right is irrelevant (should be irrelevant).


I would agree with the 98.7% of the rest of your comments about frobnicat's conjecture being 2/3 nullified if it wasn't for the "reversal of thrust" being a 180° turn of test article.

Obviously, and I hope you will make me the courtesy of not engaging me on lengthy quibble on the fact that a 180° turn of test article, that is defining Forward as "small end to the left" and Reverse as "small end to the right" will make not only
+ EMDriveFORWARDThrust = - EMDriveREVERSEThrust
but also
+ @frobnicatFORWARDMechanism = - @frobnicatREVERSEMechanism

So such 180° "reversal" will not discriminate between EMDriveThrust and frobnicatMechanism

But it could discriminate between an effect purely local to test article (be it EMDriveThrust or frobnicatMechanism) where we would expect the same magnitude, only in reverse, and a vicinity effect with the vacuum chamber walls, where we could expect vicinityFORWARD != - vicinityREVERSE (the position of test article is not symmetric in chamber in both situations). So if we want to take the single test "Chasing Thrust Reversal" chart seriously, it nullifies EMDriveThrust as much as frobnicatMechanism...

SO :

1) Do you agree that the only serious data we have about "reverse" is in fact just a 180° turn of test article ?
2) That for a 180° turn of the test article, it will expand in the opposite direction (for a same heating)
3) That a 180° turn of the test article will therefore reverse a thermal effect as much as it reverses a thrust ?
4) That we don't have precise data concerning a situation where some sustained level of apparent thrust would be toward the big end of test article ? I do recall some mention by Paul March to such thrust toward big end was made, would appreciate if you or anyone find that.

I do agree that precise and consistent data on point 4 would indeed nullify a lot of my conjectures.
« Last Edit: 03/17/2015 11:28 pm by frobnicat »

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5911
  • USA
  • Liked: 6124
  • Likes Given: 5564

1) Do you agree that the only serious data we have about "reverse" is in fact just a 180° turn of test article ?
2) That a 180° turn of the test article will reverse a thermal effect as much as it reverses a thrust ?


DEFINTION OF THE WORD REVERSE:

re·verse
rəˈvərs/
verb
1.
move backward.
"the truck reversed into the back of a bus"
synonyms:   back, back up, drive back/backward, move back/backward
"the car reversed into a lamppost"



1) Please refer to the image I provided: of a car's shift mechanism, showing a REVERSE shift. I certainly do NOT consider turning the EM Drive around by 180 degrees to equate "running in reverse", any more than turning a car engaged in "D"  (or in 1st shift) around by 180 degrees would make it run in reverse: the car would still run FORWARD (but in the opposite direction).  Reverse, as indicated by the graph I provided refers to going backwards: a car's reverse !  That's why I attached the huge image showing a car's reverse shift (the significance of the "reverse" R cannot be misconstrued )



2) Of course, "That a 180° turn of the test article will reverse turn a thermal effect as much as it reverses turns a thrust ?"
So what ?

That has nothing to do with running in reverse (from the definition of the word reverse), in so much as than turning a car around by 180 degrees still runs forward (in the opposite direction) when engaged in "D" or in "1st shift"



Your post above questions whether the EM Drive can be run in reverse.  You state:

Quote from: @frobnicat
..we don't have precise data concerning a situation where some sustained level of apparent thrust would be toward the big end of test article ? I do recall some mention by Paul March to such thrust toward big end was made, would appreciate if you or anyone find that.

@Mulletron [hat tip for being the first one in the thread to use the car's reverse shift image]  has a number of posts also interpreting that the EM Drive had been run in reverse, for example certain modes run the EM Drive forward and certain modes run the EM Drive in reverse.  One can, in principle, then run the EM Drive in reverse, at will, by "turning the (virtual) knob" of the exciting frequency such as to excite a natural frequency that has a mode shape such as producing an EM Drive's thrust in reverse

I will let Paul March:

1) describe how else the EM Drive can be run in reverse at will [using the literal definition of "reverse"].

2) determine whether my interpretation of the "reverse" measurement of -9.9 microNewtons is indeed a reverse thrust measurement (my interpretation was based on a literal interpretation of the word "reverse") or whether it is a measurement with the EM Drive turned around by 180 degrees (your interpretation).  The image for "reverse" shows the same frequency as for the "forward" images, so if the frequency is correct it means that it was run in the same mode shape both forward and backwards.   Either A) you are correct that the image is really not an EM Drive in reverse (but instead it is the EM Drive turned around and still running forward), or B) if it is indeed running in reverse it must have been accomplished by some other means, other than by changing frequency.

3) if my interpretation of the image showing  "reverse" measurement of -9.9 microNewtons was NOT a reverse thrust measurement, but instead was forward thrust with the EM Drive turned around 180 degrees, then could Paul March please point out to where is the data for the EM Drive run in Reverse. 




PS: No need to warn about "absurd controversies" or about "indulging ourselves."  Such warning (and your discussion about turning the EM Drive around by 180 degrees) was unnecessary if you take into account 1) the definition of the word reverse, and 2) the image I provided of the car's shift mechanism showing a Reverse shift.  The meaning of "R" reverse in a car cannot be misconstrued.

No need for us to get upset about this. It is important to settle this issue, as you recognize:

Quote from: @frobnicat
I do agree that precise and consistent data on point 4 [Reverse tests data] would indeed nullify a lot of my conjectures.
« Last Edit: 03/18/2015 01:10 am by Rodal »

Offline Stormbringer

  • Full Member
  • ****
  • Posts: 1340
  • Liked: 239
  • Likes Given: 92
When antigravity is outlawed only outlaws will have antigravity.

Offline aero

  • Senior Member
  • *****
  • Posts: 3628
  • 92129
  • Liked: 1145
  • Likes Given: 360
Regarding running these thrusters in reverse, I for one am not the least surprised by that result. See the attached graph of meep generated data showing force/power over a range of gap widths around the perimeter on the ends of the frustum. I think it does similar things for constant gap with frequency/mode changes but I don't have the data handy and probably didn't keep it. If someone cares to provide a set of frequencies I'd run them just to document what the curve looks like. Meep gives the the following frequencies as flux generators.

freq. Detected
for epsilon = 1, in Hz
1.5486E+009
1.7422E+009
1.9358E+009
2.1294E+009
2.3229E+009
That is, for the Copper Kettle thruster model with no dielectric.

I do have a question though. Meep measures both flux and force. Each carries it's own sign and is always measured in the positive coordinate direction. In calculating  the direction of the force should I use the signed values, or should I take the sign of the force and just the magnitude of the flux?
« Last Edit: 03/18/2015 12:54 am by aero »
Retired, working interesting problems

Offline zen-in

  • Full Member
  • ****
  • Posts: 541
  • California
  • Liked: 483
  • Likes Given: 371
The thrust waveforms have changed significantly in the past few months.  The first picture below is from April 2014.   19.9 Watts of RF power resulted in  a peak thrust of 105 microNewtons.   There is no mention what the error thrust due to magnetic interaction is however.   The waveform has a fast rise time and has an overshoot and other characteristics of an underdamped response.   It is very similar to the calibration drive and the error thrust.

The next thrust waveform, in a vacuum, is very different.   I think it is the result of  a long duration thermal effect and doesn't indicate any thrust at all.    With 50 Watts and a vacuum surrounding the cavity a thermal response would be expected.

The reverse test, also with 50 Watts, brings us back to the underdamped response, but at a very low level.

With no consistency in any of these data the only thing anyone knows for sure is:
1)  Heat is being dissipated in the cavity.
2)  DC current near a magnet generates torque.
« Last Edit: 03/18/2015 05:30 am by zen-in »

Offline Left Field

  • Member
  • Posts: 27
  • Liked: 20
  • Likes Given: 21
I disagree that "reverse" only means to move backwards, so may I suggest that a different description such as "flipped", "turned around" or "rotated" be used to better describe the experimental setup, especially given the potential for there to be a way to change the directionality electrically (and if that is true then perhaps we could follow the circuit symbol for a cell to help us communicate such directionality i.e. thrust towards the big end is positive; small end is negative)

http://www.thefreedictionary.com/reverse

adj.
1. Turned backward in position, direction, or order: the reverse side of the poster.
2. Moving, acting, or organized in a manner contrary to the usual: in reverse order.
3. Causing backward movement: a reverse gear.
4. Printing Printed in such a way that the normally colored part appears white against a colored or black background.
n.
1. The opposite or contrary: All along we thought Sue was older than Bill, but just the reverse was true.
2.
a. The back or rear part: the reverse of the flyer.
b. The side of a coin or medal that does not carry the principal design; the verso.
3. A change to an opposite position, condition, or direction.
4. A change in fortune from better to worse; a setback: suffered financial reverses.
5.
a. A mechanism, such as a gear in a motor vehicle, that is used to reverse movement.
b. The position or operating condition of such a mechanism.
c. Movement in an opposite direction.
« Last Edit: 03/18/2015 07:07 am by Left Field »

Offline frobnicat

  • Full Member
  • ****
  • Posts: 518
  • Liked: 500
  • Likes Given: 151
...
1) Please refer to the image I provided: of a car's shift mechanism, showing a REVERSE shift. I certainly do NOT consider turning the EM Drive around by 180 degrees to equate "running in reverse", any more than turning a car engaged in "D"  (or in 1st shift) around by 180 degrees would make it run in reverse: the car would still run FORWARD (but in the opposite direction). 

Very well put, this would also be my definition of reverse, but this is not how the qualifier "reversed" was used by Paul March on the chart "Chasing Thrust Reversal". Paul March called reversed a result that was just a 180° turn. Do you agree or not on that this is the (unusual) specific use of of the qualifier reverse by Paul March in regard to the chart you used as a proof of (true usual) reversed mode ? That therefore the chart "Chasing Thrust Reversal" is not reversed in the usual sense and brings next to nothing to the discussion (about thrust vs thermal expansions)

Quote
Reverse, as indicated by the graph I provided refers to going backwards: a car's reverse !  That's why I attached the huge image showing a car's reverse shift (the significance of the "reverse" R cannot be misconstrued )

No, the graph your provided (if we are speaking the one titled "chasing thrust reversal") does not use the qualifier reversal in this sense, it only means turned 180°, small end toward the right. The car's reverse shift analogy is irrelevant as far as this chart and accompanying data is concerned. The significance of reverse R on a car's reverse shift cannot be misconstrued. The significance of "reverse" as used (in an unusual way) on the chart "chasing thrust reversal" was misleading. Insofar, all we have is a car that was turned 180°.

Quote
2) Of course, "That a 180° turn of the test article will reverse turn a thermal effect as much as it reverses turns a thrust ?"
So what ?

Frankly, sometimes things that seem so obvious by one get misconstrued by another. I just wanted to make sure that we agree on that. Now we just have to agree (or disagree) on the fact that on published charts so far "reverse" was not reverse, it was just 180° turn, that we don't have data about "true" reverse modes.

Quote
...
Your post above questions whether the EM Drive can be run in reverse.  You state:

Quote from: @frobnicat
..we don't have precise data concerning a situation where some sustained level of apparent thrust would be toward the big end of test article ? I do recall some mention by Paul March to such thrust toward big end was made, would appreciate if you or anyone find that.

@Mulletron [hat tip for being the first one in the thread to use the car's reverse shift image]  has a number of posts also interpreting that the EM Drive had been run in reverse, for example certain modes run the EM Drive forward and certain modes run the EM Drive in reverse.  One can, in principle, then run the EM Drive in reverse, at will, by "turning the (virtual) knob" of the exciting frequency such as to excite a natural frequency that has a mode shape such as producing an EM Drive's thrust in reverse

Very well, I'm tired of reading the archives again and again. Now you want to make a point about the test article being driven in true reverse having such or such consequence (that I would agree), but you post a chart that only shows data that is not true reverse but only 180° turn, that is therefore irrelevant for your argument.

To make your point stand, please find the most accurate data about a true reverse situation,  certain modes run the EM Drive forward and certain modes run the EM Drive in reverse : what modes ? what frequencies ? what powers ? what magnitude ? what durations ? what shapes ?

Quote
I will let Paul March:

1) describe how else the EM Drive can be run in reverse at will [using the literal definition of "reverse"].

Actual charts and very clear statements about direction of thrust relative to geometry of frustum would be greatly appreciated. Clear as with "thrust toward the big end".

Quote
2) determine whether my interpretation of the "reverse" measurement of -9.9 microNewtons is indeed a reverse thrust measurement (my interpretation was based on a literal interpretation of the word "reverse") or whether it is a measurement with the EM Drive turned around by 180 degrees (your interpretation).  The image for "reverse" shows the same frequency as for the "forward" images, so if the frequency is correct it means that it was run in the same mode shape both forward and backwards.   Either A) you are correct that the image is really not an EM Drive in reverse (but instead it is the EM Drive turned around and still running forward), or B) if it is indeed running in reverse it must have been accomplished by some other means, other than by changing frequency.

I'm also waiting for Paul March confirming either A or B, but frankly do you seriously believe, with the elements I gave in my last post, that B stands correct ? If it takes hours and hours of discussion to settle such simple matter...

Quote
3) if my interpretation of the image showing  "reverse" measurement of -9.9 microNewtons was NOT a reverse thrust measurement, but instead was forward thrust with the EM Drive turned around 180 degrees, then could Paul March please point out to where is the data for the EM Drive run in Reverse. 

It is my understanding no such data (complete, detailed) were disclosed so far.

Quote


PS: No need to warn about "absurd controversies" or about "indulging ourselves."  Such warning (and your discussion about turning the EM Drive around by 180 degrees) was unnecessary if you take into account 1) the definition of the word reverse, and 2) the image I provided of the car's shift mechanism showing a Reverse shift.  The meaning of "R" reverse in a car cannot be misconstrued.

I'm warning that a leading author driving the discussion has a use of the word Reverse that is misleading. The meaning of "R" reverse on Eagleworks plots not only can but has been misconstrued.

Quote
No need for us to get upset about this. It is important to settle this issue, as you recognize:

Quote from: @frobnicat
I do agree that precise and consistent data on point 4 [Reverse tests data]
 would indeed nullify a lot of my conjectures.

I'm actually (a bit) upset indeed that we are not really requiring of the leading authors half of the consistency in wording that seems to be required from me. That ambiguities, when they occur, seem to be hastily interpreted in favour of EMdriveThrust hypothesis and in disfavour of competing schemes, instead of being recognized as ambiguities that need proper and neutral investigation.

Edit : My claims, while needing further justifications and clarifications, and maybe experimentally wrong in the end, are within proven frameworks of physics. The claims of real EMDrive thrust are the extraordinary claims, needing extraordinary clarity in result reports and consistency of data. For recall we have a 30 time factor unexplained at the core of the process of measuring pendulum displacement vs calibrations pulse, we have varying (by 1 to 5) vertical scale readings from chart to chart, we have a signal that should be symmetric when turning the test article 180° and that is not symmetric, we have a Z tilted axis (of weakly characterised tilt angle) that transforms stationary shifts in mass parts of test article into sustained displacement readings, we have undisclosed charts for the "0 thrusts" when operating without dielectric disc (or with neoprene), we have not seen the 20 charts of the 5 test at each power 30W 40W 50W 60W of the "forward thrust campaign" that would allow to average curves and make some statistics instead of arguing on noise plagued single cases, we have a disparity of thrust in vacuum relative to air... All those aspects must be investigated, not just the easy ones. I say that as I'm sure it must be itching to answer the easy ones and forget the hard ones or sweep them under the carpet.


To clarify things : I am not thinking that Paul March used the word "reverse" in a unusual way voluntarily to mislead the reader, I think it was very clear to him that he was speaking about a 180° turn, and that it would be equally clear to reader. Reading his post and attached picture it was indeed relatively clear and straightforward to me that, in this chart, "reverse" was just a 180° turn, even if a bit awkward formulation.
« Last Edit: 03/18/2015 11:32 am by frobnicat »

Offline frobnicat

  • Full Member
  • ****
  • Posts: 518
  • Liked: 500
  • Likes Given: 151
The thrust waveforms have changed significantly in the past few months.  The first picture below is from April 2014.   19.9 Watts of RF power resulted in  a peak thrust of 105 microNewtons.   There is no mention what the error thrust due to magnetic interaction is however.   The waveform has a fast rise time and has an overshoot and other characteristics of an underdamped response.   It is very similar to the calibration drive and the error thrust.

The next thrust waveform, in a vacuum, is very different.   I think it is the result of  a long duration thermal effect and doesn't indicate any thrust at all.    With 50 Watts and a vacuum surrounding the cavity a thermal response would be expected.

The reverse test, also with 50 Watts, brings us back to the underdamped response, but at a very low level.

With no consistency in any of these data the only thing anyone knows for sure is:
1)  Heat is being dissipated in the cavity.
2)  DC current near a magnet generates torque.

This is what we have (same color, same topic) :

Dr. Rodal:

Looking over some of our test results from last spring and summer just now, I realized that I had forgotten to provide this forum a few test runs at some of the other resonant modes we've looked at that used different dielectric discs and locations including polycarbonate which I hadn't mentioned before.  These examples are attached and you will note that these other copper frustum resonant modes (TM010 & TE012) have very different thermal torque pendulum responses than the TM212 (your TM221 I think) cases we've discussed of late.  That fact might be of importance when discussing whether these thrust signals are real or just artifacts...

PS: The magnitude of the torque pendulum's overshoot is directly related to what version of the magnetic damper that was used for the particular test in question and how thick the copper damper blade was at the time.  At the moment we are using an arrangement that is slightly over-damped in an attempt to smooth out some of the pesky low frequency seismic noise in the system.

DPS:  The magnitude of the RF amp's dc current induced thrust signal offset is directly related to how much leakage current is going through the torque pendulum's steel torsion bearings.  As I became aware of this problem over the last year, I've found various ground wiring tricks to mitigate this dc current induced negative going offset, but I've never found a way to get rid of all of it.

Best, Paul M

Offline frobnicat

  • Full Member
  • ****
  • Posts: 518
  • Liked: 500
  • Likes Given: 151
I disagree that "reverse" only means to move backwards, so may I suggest that a different description such as "flipped", "turned around" or "rotated" be used to better describe the experimental setup, especially given the potential for there to be a way to change the directionality electrically (and if that is true then perhaps we could follow the circuit symbol for a cell to help us communicate such directionality i.e. thrust towards the big end is positive; small end is negative)


Yes, reverse is polysemous. I agree for a clear convention. Using negative to mean "toward the small end" would have the advantage of being compatible with my (relatively arbitrary) direct OXYZ conventions on my drawings, with thrust measured relative to  axis Y, when the small end is to the left (as seen from front of experiment) which was the case for all the tests qualified as "forward" (most of the tests). It has the disadvantage of most of the time talking of negative thrusts, since all the published data so far was toward the small end. The only way to make it crystal clear is to state explicitly "thrust toward the small end" (all published data so far) or "thrust toward the big end" (data yet to be seen), and saying if small-end is mounted toward the left (most of the experiments) or to the right (the so called "reversed" experiment).
« Last Edit: 03/18/2015 09:36 am by frobnicat »

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5911
  • USA
  • Liked: 6124
  • Likes Given: 5564
...

Quote from:  Star-Drive
one can also reverse this thrust vector for this copper frustum by just changing which excited resonant mode is used





I will calmly wait for the person that conducted the experiments (Paul March "Star-Drive") to confirm whether the EM Drive can indeed be run at will (by changing the exciting frequency) with the thrust force directed towards the big base (instead of towards the small base) of the truncated cone, and if so, to point out the relevant data demonstrating that fact.

Mr. Frobnicat, with all due respect, since you did not perform these experiments, you cannot write with pontificating authority to resolve this matter.

If Paul March answers that the EM Drive cannot be run with the thrust force directed towards the big base (instead of towards the small base) of the truncated cone, I will be very pleased to have learned this fact and to have corrected my misunderstanding.

On the other hand if Paul March answers that the EM Drive can indeed be run with the thrust force directed towards the big base (instead of towards the small base) of the truncated cone, by changing the exciting frequency, your conjecture (if it pretends to explain the total measurement) will be shown to have been already nullified by NASA Eagleworks.  In that case, your conjecture might, at best, serve to explain a fraction of the measurement as a biasing artifact. 

« Last Edit: 03/18/2015 07:06 pm by Rodal »

Offline aceshigh

  • Full Member
  • ****
  • Posts: 792
  • Liked: 269
  • Likes Given: 22
regarding the earlier quote by Paul March saying he could retire (or be retired?) from NASA by the end of March...

what are the conditions for that happening? Not getting positive results for EM Drive?

is the whole Eagleworks Lab being shut down? Which would also mean the end of the Warp Drive and QM Vacuum Thruster Experiments and also the EM Cannae Drive experiments?




Offline Mulletron

  • Full Member
  • ****
  • Posts: 1150
  • Liked: 837
  • Likes Given: 1071
Before today, I've never used my set of tin snips for anything more than hacking difficult cables in two. I gotta say, there is definitely a skill set required on how to use tin snips properly if you want to make precision cuts.

There is a lot of bad information out there on the internet, like the red ones only cut left, or green ones only cut right. The real info is that red ones push waste material to the left, and green ones push waste material to the right. They can cut either way. Make sure the waste side is away from your work. That is all that matters.

So I watched a couple youtube videos and practiced on some extra sheet metal I had laying around and got good enough to cut out some circles from expensive copper sheet. It is pretty easy. I made some perfect circles no sweat. Lucky for me, I like to hoard tools, so those tin snips I got 2-3 years ago finally got used for something useful. So you potential replicators out there, make sure you're good with tin snips before you cut, it ain't as easy as scissors. Just widdle down so there is no more than 1/8" of excess around your work then, go in for detail work.

I had trouble seeing my score marks I made with the sheet metal dividers, even with the room well lit. I discovered that an LED head lamp set to red makes the score marks POP, so I could easily see them and follow the cut. White light just blinded me. I didn't want to use a sharpie. I wanted sub millimeter accuracy.

I opted against using the left over sheets from the sample pack I got to be used as end plates. Instead I got 2 12"x12" sheets of 16oz copper on Amazon to make the end plates. The whole copper kettle is 16oz copper.

For now (and possibly forever) the end plates are clamped on with numerous stainless steel paper clamps in. I know I'll probably get some crap over using the clamps, but I have to tell you that these things are crazy strong. I can't pull them apart with my finger tips. This isn't a flight ready thruster, it is a rapid prototype for research. All I care about is keeping the ends firmly against the frustum and that the entire assembly is electrically short and mechanically true. Drilling >9000 bolt holes doesn't seem appealing to me right now; especially since the clamps hold over a wide surface area and do a good job.

After putting the whole thing together, it is extremely sturdy. Thanks to 1952rickey on Ebay for building me the frustum. He did a really good job. He still hasn't answered me back about listing the microwave resonant cavity for research, yet. Feel free to send him a message if you want one too. I want to be clear, he is making resonant cavities, not thrusters or Emdrives.

Ok, so I was very careful to never touch the inside of the cavity without gloved hands. The outside has some fingerprints. Is there anything I could use to wipe down the inside and outside to stave off corrosion?.......while not messing up the RF characteristics? Copper loves to turn green. I need the Q to be as high as possible.

To do, I have all the dielectric sheets in except the UHMW PE. My to do list for tomorrow is to jig saw out 2, 6.25" HD PE circles and drill/mount them. I think this 1" thick sheet of plastic is going to do in my saw. This sheet is very dense and heavy. Wish I had a band saw.

I got a length of very lightweight extruded rectangular aluminum. It is 150cm long. I'm going to cut it down to 1 meter. This will serve as the platform of the balance. This thing is extremely light and doesn't bend.

I have a scale on the way which can/will accurately weigh all this stuff down the sub gram accuracy. It is important to know how much all this stuff weighs as accurately as possible.

Pics are here:
https://drive.google.com/folderview?id=0B4PCfHCM1KYoTXhSUTd5ZDN2WnM&usp=sharing



« Last Edit: 03/18/2015 07:18 pm by Mulletron »
And I can feel the change in the wind right now - Rod Stewart

Offline timbatman1960

  • Member
  • Posts: 2
  • Liked: 1
  • Likes Given: 1
All:

Sorry I didn't make the time to participate in this ME-Drive forum for the last 6-to-8 months up, but I will try to catch up with everyone else in due course.  That said lets try to answer the questions that popped up since my morning post.

1.  I was not the lead author for the Eagleworks' 2014 AIAA/JPC paper and in fact I only supplied pictures and data for same during that period because Dr. White thought that my time was best spent in the lab gathering data instead of report writing.  Thus some of the details that Dr. Rodal is looking for may have been lost or garbled in the report writing by the others on the author list.

2.0  The thrust vector for the four resonant modes examined in detail, (the cavity's fundamental TM010, TE012, TM211 & TM212 for our copper frustum is normally in the frustum's large OD to small OD direction for most, but not all the E&M resonant modes checked.  However, one can also reverse this thrust vector for this copper frustum by just changing which excited resonant mode is used and/or mounting the dielectric discs at the large OD end of the cavity instead of the small OD end, see attached resonant mode map.  Sorry, but a one size fits all solution to this EM-Drive thrust direction is not available in this venue because of the importance of the ExB phase relationship of the expressed Lorentz forces between the excited E&M fields and the possible dielectric and QV plasma flow phenomenon that may be at work in each resonant mode expressed.  That is why this type of E&M thruster is so hard to get a handle on, for there are far too many degrees of freedom in the system to track let alone directly control.

3. The Eagleworks vacuum chamber's main body is made from 304L stainless steel while its swing out door is made from aluminum.  Most of the nuts and bolts in the vacuum chamber are also made from 18-8, 304 or 316 stainless steel alloys. 

Now to try to answer Dr. Rodal's specific questions:

"1) In the NASA experiments the truncated cone's center of  mass moved towards the [  ? ] diameter end  (where ? stands for big or small)"

For the TE012 and TM212 excited resonant modes, our copper frustum's center of mass moved toward the small OD end of the frustum when RF power was applied to the copper frustum.

"2) In the NASA experiments, we at NASA Eagleworks define the thrust force direction to be in the  [? ] direction as the movement of the truncated cone's center of  mass  (where ? stands for same or opposite)"

For just the TE012 & TM212 excited resonant modes, the thrust force direction AKA thrust vector was observed to be in the same direction as the movement of the frustum's center of mass when RF power was applied to the frustum's magnetic loop antenna.

If I missed a question along the way keep asking, but I'll be in and out of the house for the rest of the day, so I may not get to answer them until late this evening or tomorrow afternoon USA based CST.

Best, Paul M.

I understand that we need to wait for Paul March to explicitly "re-verify" the manner in which thrust reversal was achieved, but it does appear to me that he has already  addressed this in the above quote from Feb 14.  Quoting Mr. March:  "one can also reverse this thrust vector for this copper frustum by just changing which excited resonant mode is used and/or mounting the dielectric discs at the large OD end of the cavity instead of the small OD end".   To my admittedly untrained eye,  this appears to be in agreement with Dr. Rodal's earlier statements, and in conflict with Mr. Frobnicat's contention that reversal was achieved by simply rotating the entire mechanism 180 degrees.

Tags:
 

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