-
#1200 by Rodal on 13 Apr, 2016 23:31
-
Understood, am not exactly sure why, but there was some confusion to start with when X_Ray was being so kind as to help with sims. This is because I did not indicate "cm" on my schematic. Additionally, I'm not sure he was aware that I was downscaling the NASA frustum and had chosen 1.8804GHz TE012...(as indicted in results had the "best" out come: mN/kW of 21.3). Having reviewed this I hope X_Ray is up to running a few more sims to find the "best" frequency for this "Cubesat frustum" Ciao!
I think that X-Ray needs to know:
1) are you planning to include a HDPE dielectric insert inside your EM Drive, as done by NASA, and therefore do you want X-Ray to model the dielectric insert in his analysis?
or
2) are you planning to run your EM Drive without a dielectric insert placed asymmetrically inside it, like Shawyer as of late, and like Yang, Tajmar, Berca, Distler, etc.?
X-Ray needs to know this, because the HDPE dielectric insert lowers the natural frequency for the same mode shape
It is not clear to me whether you are planning to run with or without a dielectric -
#1201 by frobnicat on 14 Apr, 2016 00:33
-
.../...
Also Dr. McCulloch has recently tried to tackle the energy-conservation paradox (looking forward to comments from Prof. Frobnicat
here ) using information theory (*)
http://physicsfromtheedge.blogspot.com/2016/04/informational-mihsc.html
.../...
I like the fact that Dr. Mike McCulloch tries to tackle the energy paradox raised by the EM Drive, rather than try to escape from it, something that I have not seen Mr. Shawyer or Prof. Yang do. Dr. McCulloch admits that if the EM Drive works as claimed by experimenters, it should be a new potential energy source (perhaps an energy source to feed into another EM Drive for propulsion purposes).
Whenever you summon me dr Rodal...
One thing escapes me about all those attempts : when we try to explain (the origin and/or consequences of) some F term on the left-hand side from the a term on the right-hand side, it's troubling that we forget that for a rigid body Newton actually predicts (quite accurately for low velocities) ∑F=ma (bold for vectors), and that implies the possibility of at least another F arbitrary term (externally imposed by conventional means, spring, electric actuator, chemical rocket, whatever) as we wish on top of a given thrust vector to get any arbitrary a ! I quite don't follow Notsosureofit hypothesis nor McCulloch's (nor one or two others that seem to "gravitate" around similar requirements) but they all seem to interpret thrust as linked somehow to acceleration. This is not far from Shawyer's (as reported by the Traveller) claims that the Emdrive "needs" to be free to accelerate, or be accelerated first before it "engages" in so called motor mode. All right but we need to see that experimentally and also yet to be told theoretically (quantitatively) how much of acceleration or "freedom to move" is needed. Despite your (polite) insistance dr Rodal, the Traveller never quite addressed this quantitatively.
My question is, what comes of those theories when the device is not allowed to accelerate, will it be unable to thrust ? Then it is to be expected that it can't quite hold a spring compressed (linked against an inertial massive "ground") which runs contrary to, for instance, usual way of measuring a constant thrust by measuring a constant proportional displacement against a spring's stiffness or static restoring force (various scales schemes...). I mean : while a spacecraft is mainly concerned by a single force, that of the main thruster, where F can be "equaled" to a, a thruster under test in the lab, after short transients, is concerned by two equal but opposite forces, that of the thrust and that of the test stand, and a is 0 (on average, assuming constant thrust).
If some constant level of thrust is claimed on scale or torsion pendulum in the lab with a=0 for some length of time while the scale is kept deviated from its rest position by the effect, then a theory should explain how such force is possible without relying on a≠0
As far as energy conservation is concerned with thrust/power>1/c, should a theory engage in the risky business of explaining where some apparent excess of energy is coming from, as is obvious (provided there is no absolute velocity limit, i.e. the theory is relativistic) in both the cases of a freely accelerating spacecraft (where F can be "equaled" to a) and that of pushing a generator at constant velocity (where F and a are independent variables since a=0), then such a theory should either not rely on a≠0 as starting principle to explain the force, or else make precise predictions as to the absence of measurable "static thrust" in the later case.
Is Mihsc claimed to be relativistic ?
If no :
what is the local privileged rest frame wrt earth ground at a given longitude/latitude/sidereal time ? What orientation dependence relative to this "wind" is experimentally expected ?
If yes :
Does Mihsc predicts that EMdrive's "thrust" can hold pressure against a spring against a stand ?
If no :
That still might be consistent theoretically, but not quite so experimentally (EagleWorks' claims for instance)
If yes :
There appears to be a major confusion, Force (Thrust) can't be "equaled" to acceleration in general. And in the case of "static" thrust (ie. a=0), how could a Rindler horizon "know" how much of data it's supposed to loose in exchange of a mechanical work done or not done, depending on relative constant velocity of the thrust vector to some "third party" part ? For instance we can have two EMdrives directly linked thrusting in opposite directions, for both a=0, for the system work=0. Or we can have two EMdrives, thrusting in opposite directions and moving in opposite directions at constant velocity, linked together through a generator (with regulation such that relative velocity is maintained constant in spite of force) : for both a=0, for the system work≠0.
Notwithstanding my ignorance about what really is a Rindler horizon... -
#1202 by FattyLumpkin on 14 Apr, 2016 00:38
-
definitely with dielectric with HDPE which I surmise is high density polyethylene --as said: extruded.
Perhaps a sim with and without the HDPE "disks": just dielectric antenna top middle.
-
#1203 by Rodal on 14 Apr, 2016 00:43
-
definitely with dielectric with HDPE which I surmise is high density polyethylene --as said: extruded.
OK, just remember:
Perhaps a sim with and without the HDPE "disks": just dielectric antenna top middle.
1) NASA TE012 without a dielectric, natural frequency=2.168 GHz
2.168 GHz * 2.794 = 6.057 GHz <------- Can scale accurately as per Maxwell's equations since they are linear. Only variable is the geometry.
X-Ray calculation: 6.117 GHz
https://forum.nasaspaceflight.com/index.php?topic=39772.msg1516895#msg1516895
Difference: 6.117/6.057 - 1 = 0.99 %
the difference is only 1 per cent, which is what I would expect when using a numerical method because the numerical solution converges from above: it is stiffer than the actual exact solution.
The difference can also be due to the fact that the internal geometry of the EM Drive tested by NASA may not be exactly equal to what was assumed in the numerical modeling. Anyway, a 1% difference is quite acceptable, no doubt about it.
2) NASA TE012 with a HDPE dielectric asymmetrically placed in the cavity, natural frequency=1.8804 GHz
1.8804 GHz * 2.794 = 5.254 GHz <------- Scaling depends on how dielectric is modeled, and how dielectric is scaled -
#1204 by Rodal on 14 Apr, 2016 01:00
-
...
There appears to be a major confusion, Force (Thrust) can't be "equaled" to acceleration in general. And in the case of "static" thrust (ie. a=0), how could a Rindler horizon "know" how much of data it's supposed to loose in exchange of a mechanical work done or not done, depending on relative constant velocity of the thrust vector to some "third party" part ? For instance we can have two EMdrives directly linked thrusting in opposite directions, for both a=0, for the system work=0. Or we can have two EMdrives, thrusting in opposite directions and moving in opposite directions at constant velocity, linked together through a generator (with regulation such that relative velocity is maintained constant in spite of force) : for both a=0, for the system work≠0.
Notwithstanding my ignorance about what really is a Rindler horizon...
Great statement:
"how could a Rindler horizon "know" how much of data it's supposed to loose in exchange of a mechanical work done or not done, depending on relative constant velocity of the thrust vector to some "third party" part "
-
#1205 by FattyLumpkin on 14 Apr, 2016 03:25
-
X_Ray, do you have the ability to simulate dielectric RF? Thanks, Kevin
-
#1206 by tleach on 14 Apr, 2016 04:07
-
I believe those dimensions are much larger than any emdrive built to date. See comparison below between my frustum dimensions. Also ran a sim using the 2.424757282Ghz. Being quite large, this model has over 650 triangles. So I can only do a one or two of these, and couldn't realistically do a frequency sweep unless I let it run for many hours.
Soooo... I wonder if I might talk you into modeling a frustum design based on a scalar horn?QuoteExponential horn (e) A horn with curved sides, in which the separation of the sides increases as an exponential function of length. Also called a scalar horn, they can have pyramidal or conical cross sections. Exponential horns have minimum internal reflections, and almost constant impedance and other characteristics over a wide frequency range. They are used in applications requiring high performance, such as feed horns for communication satellite antennas and radio telescopes.
https://en.wikipedia.org/wiki/Horn_antenna#Types
I've been really curious about this shape ever since I started lurking on this forum.
Pretty please? -
#1207 by X_RaY on 14 Apr, 2016 05:11
-
HI X_Ray, thank you for you work, while I'm sure that the down scaled geometry is correct whether multiplying the original NASA dimension by .35791 or dividing by 2.794, the only remaining question I believe is frequency:
the frequency of 1.8804 for the original size frustum was selected because it produced the best result.
Irrespective of which way method is used: division or multiplication one comes up with only two number for an upscale frequency: 5.2538 GHz or 6.7301 GHz. However Dr. Rodal (as I recall) and others have stated that an up-scaled frequency by the numbers would most likely not be the frequency that would produce the best resonance. (My reference re Dr. Rodal here is based on my recollection, so Dr. please chime in here if I've got it wrong)! Indeed, several pages back I believe Monomorphic found resonance at 2.4 Gigs.
These dimensions of this downsized frustum are vital, because NASA might delay testing in micro-gravity, and said well resonating and thrust tested frustum will fit into a Cubesat and may thereby make for a real demonstration of the "EM effect". Please let me (us) know what your find. And, yes as aforementioned TE012 and TM212 modes are noted the NASA data, however TE012 was noted at 1.8804 GHz. FL
Thanks Dr.Rodal for the clarification. well done!
https://forum.nasaspaceflight.com/index.php?topic=39772.msg1517929#msg1517929
https://forum.nasaspaceflight.com/index.php?topic=39772.msg1518028#msg1518028
FattyLumpkin I used the frustum without dielectric, for dims and frequencies I used the paper of Frank Davies (NASA/EW).
For any future calculations, the communication error has to be fixed first
. All dimensions are needed as well as the material properties.
Again, I only have the LIGHT version of Feko available with contains strong limitations such as the number of tetraeders is limited to 500 volume elements and so on.
That means I must use a predefined stiff mesh. It does not converge(like it would be the case for mesh optimization algorithm) as needed for high precision calculations. The mesh on a dielectric insert and in edges or other inhomogeneities normally has to be much denser to get precise results.
Therefore I think it would be better to ask Monomorphic for calculations with dielectric insert, he use the full version of FEKO.
-
#1208 by FattyLumpkin on 14 Apr, 2016 05:52
-
tleach, indeed several pages back I indicated that I'd hope for a simulation on an "Exponential" or "Scalar" horn. As you can see in this list of images from wiki a several of these geometries had already been simulated by Monomorphic. Horn "e" being the one to which you are referring. It also should be noted that Monomorphic has performed sims on several of these including convex and concave features. Conceptually, I'm not sure the Scalar would do well, unless a certain % arc be employed at least in the Large mirror. So many people are in the throes if testing, I'm not sure when we might expect a sim, but fingers are crossed. : )
-
#1209 by rfmwguy on 14 Apr, 2016 12:17
-
Decent review and commentary on aachen baby emdrive
https://hackaday.com/2016/04/11/flying-the-infinite-improbability-drive/ -
#1210 by Monomorphic on 14 Apr, 2016 13:23
-
Decent review and commentary on aachen baby emdrive
As pointed out by the top posts, thermal radiation pressure and low earth orbit atmospheric drag will probably be greater than the thrust a 3 watt emdrive can produce.
-
#1211 by rfmwguy on 14 Apr, 2016 13:37
-
I would agree. LEO is no place for that low of a power. I commend them for out of the box thinking, but LEO isn't the place. Cubesat is just too small of a footprint for useful EMDrive research. Rough minimum size I would think would be 1 cubic meter given frustum size, power plant requirements and folded solar panels. And that is a VERY rough guess but probably a placeholder for anyone thinking of sending one up on a payload.Decent review and commentary on aachen baby emdrive
As pointed out by the top posts, thermal radiation pressure and low earth orbit atmospheric drag will probably be greater than the thrust a 3 watt emdrive can produce.
Back of napkin calc...Cubic meter of aluminum weight 2800 kg, melt EMDrive probe assembly down to solid block about 1/4 volume and weight, about 700 kg or close to 1500 lbs. Lightweight materials use, shave off 500 lbs, maybe...just maybe a cubic meter payload of an EMDrive probe would weigh in at 1000 lbs.
At $10,000 per pound going rate, around $10 Mil to get one up there...Obviously not accurate but does put things in perspective if going thru NASA. Private companies might save 1/2 cost, so EMDRive launch enthusiasts can probably expect a $5 Mil price tag just to try it out...thus...the absolute need for ground verification tests before fundraising efforts to launch an EMDrive probe. -
#1212 by Rodal on 14 Apr, 2016 13:55
-
I look forward to somebody (preferably from the Aachen team) providing a review on this Aachen team's experimental results. Very difficult to make sense out of their sketchy reports: graphs where the vertical and horizontal coordinates are not labeled, noisy outputs, etc.
I would agree. LEO is no place for that low of a power. I commend them for out of the box thinking, but LEO isn't the place. Cubesat is just too small of a footprint for useful EMDrive research. Rough minimum size I would think would be 1 cubic meter given frustum size, power plant requirements and folded solar panels. And that is a VERY rough guess but probably a placeholder for anyone thinking of sending one up on a payload.Decent review and commentary on aachen baby emdrive
As pointed out by the top posts, thermal radiation pressure and low earth orbit atmospheric drag will probably be greater than the thrust a 3 watt emdrive can produce.
Here it is shown that smaller-size EM Drive's as in the Baby EM Drive Aachen experiments are expected to result in significantly smaller force/powerInput:
http://forum.nasaspaceflight.com/index.php?topic=39214.msg1474347#msg1474347
this fact, coupled with the much smaller power input lead to much smaller anomalous force, hence noisy output is not surprising.
One of the latest communications seems to be asking people in the Internet to make sense out of the data:Quote from: HackadayHere is the link to the test data of the EMDrive V4.
Looks still a bit noisy - many peaks occur irregularly, so many of them canīt come from thrust.
There are some repetitive patterns though.
I included a viewer so you can browse in the graph.
Raw ASCII data is also available.
https://goo.gl/ylD5kY
The data is colon separated, the format description is included in the repository.
Have fun and please let me know if you spot some thrust.
I will make the new tests longer in hope that the noise can be filtered out better.
https://hackaday.io/project/5596-em-drive
The problem with noisy data like this is that there are not enough separate experiments to provide a statistical population to enable a (frequentist) statistical analysis, and there is no prior basis to do a Bayesian statistical analysis either. Hence there is no basis to do a robust statistical analysis of the data. And there is no computer modeling of the experiment or simplified analysis of what is responsible for the noise, so one is left with a confusing picture as to whether their results are null or whether there could be something swamped by experimental noise.
Since their experimental results on Earth surface experiments are inconclusive due to all this experimental noise, the same situation is expected in LEO: due to atmospheric drag, variations in gravity, thermal radiation, etc.
As one of the most famous statisticians stated:
I prefer NASA's approach: where the experiments are thoroughly analyzed, in order to understand the source of noise and to design experiments where the noise is understood and minimized or even better: cancelled out (for example in NASA's vacuum chamber experiments the convection effects from air are made insignificant).
-
#1213 by Notsosureofit on 14 Apr, 2016 14:10
-
.../...
Also Dr. McCulloch has recently tried to tackle the energy-conservation paradox (looking forward to comments from Prof. Frobnicat here ) using information theory (*)
http://physicsfromtheedge.blogspot.com/2016/04/informational-mihsc.html
.../...
I like the fact that Dr. Mike McCulloch tries to tackle the energy paradox raised by the EM Drive, rather than try to escape from it, something that I have not seen Mr. Shawyer or Prof. Yang do. Dr. McCulloch admits that if the EM Drive works as claimed by experimenters, it should be a new potential energy source (perhaps an energy source to feed into another EM Drive for propulsion purposes).
Whenever you summon me dr Rodal...
One thing escapes me about all those attempts : when we try to explain (the origin and/or consequences of) some F term on the left-hand side from the a term on the right-hand side, it's troubling that we forget that for a rigid body Newton actually predicts (quite accurately for low velocities) ∑F=ma (bold for vectors), and that implies the possibility of at least another F arbitrary term (externally imposed by conventional means, spring, electric actuator, chemical rocket, whatever) as we wish on top of a given thrust vector to get any arbitrary a ! I quite don't follow Notsosureofit hypothesis nor McCulloch's (nor one or two others that seem to "gravitate" around similar requirements) but they all seem to interpret thrust as linked somehow to acceleration. This is not far from Shawyer's (as reported by the Traveller) claims that the Emdrive "needs" to be free to accelerate, or be accelerated first before it "engages" in so called motor mode. All right but we need to see that experimentally and also yet to be told theoretically (quantitatively) how much of acceleration or "freedom to move" is needed. Despite your (polite) insistance dr Rodal, the Traveller never quite addressed this quantitatively.
My question is, what comes of those theories when the device is not allowed to accelerate, will it be unable to thrust ? Then it is to be expected that it can't quite hold a spring compressed (linked against an inertial massive "ground") which runs contrary to, for instance, usual way of measuring a constant thrust by measuring a constant proportional displacement against a spring's stiffness or static restoring force (various scales schemes...). I mean : while a spacecraft is mainly concerned by a single force, that of the main thruster, where F can be "equaled" to a, a thruster under test in the lab, after short transients, is concerned by two equal but opposite forces, that of the thrust and that of the test stand, and a is 0 (on average, assuming constant thrust).
If some constant level of thrust is claimed on scale or torsion pendulum in the lab with a=0 for some length of time while the scale is kept deviated from its rest position by the effect, then a theory should explain how such force is possible without relying on a≠0
As far as energy conservation is concerned with thrust/power>1/c, should a theory engage in the risky business of explaining where some apparent excess of energy is coming from, as is obvious (provided there is no absolute velocity limit, i.e. the theory is relativistic) in both the cases of a freely accelerating spacecraft (where F can be "equaled" to a) and that of pushing a generator at constant velocity (where F and a are independent variables since a=0), then such a theory should either not rely on a≠0 as starting principle to explain the force, or else make precise predictions as to the absence of measurable "static thrust" in the later case.
Is Mihsc claimed to be relativistic ?
If no :
what is the local privileged rest frame wrt earth ground at a given longitude/latitude/sidereal time ? What orientation dependence relative to this "wind" is experimentally expected ?
If yes :
Does Mihsc predicts that EMdrive's "thrust" can hold pressure against a spring against a stand ?
If no :
That still might be consistent theoretically, but not quite so experimentally (EagleWorks' claims for instance)
If yes :
There appears to be a major confusion, Force (Thrust) can't be "equaled" to acceleration in general. And in the case of "static" thrust (ie. a=0), how could a Rindler horizon "know" how much of data it's supposed to loose in exchange of a mechanical work done or not done, depending on relative constant velocity of the thrust vector to some "third party" part ? For instance we can have two EMdrives directly linked thrusting in opposite directions, for both a=0, for the system work=0. Or we can have two EMdrives, thrusting in opposite directions and moving in opposite directions at constant velocity, linked together through a generator (with regulation such that relative velocity is maintained constant in spite of force) : for both a=0, for the system work≠0.
Notwithstanding my ignorance about what really is a Rindler horizon...
Kindly note that the force calculated for the Notsosureofit Hypothesis is the "static" force. This force drops off with actual acceleration of the cavity as a simple arithmetic calculation will show.
-
#1214 by Monomorphic on 14 Apr, 2016 14:15
-
EMDRive launch enthusiasts can probably expect a $5 Mil price tag just to try it out...thus...the absolute need for ground verification tests before fundraising efforts to launch an EMDrive probe.
That's just the launch cost right? Nobody is going to launch a DIY satellite. You'll need at least several million more to build a flight-rated spacecraft that can orient itself by gimbaling the emdrive, has reliable communications, solar arrays and batteries, proper cameras, mission control, etc. -
#1215 by rfmwguy on 14 Apr, 2016 14:31
-
Continuing thoughts on ~$5 Mil EMDrive probe launch...it is a small drop in the bucket in comparison to $100M effort to start the Hawking, et al, Starshot project.
Starshot...7 Gigawatt ground based laser firing for 2 minutes at targets, inducing 20% light speed, producing 60,000 G force...stop and think. Which makes more sense? EMDrive or Starshot?
Besides the "death ray" capabilities, the power consumed, the unknown effects to the atmosphere, the random scatter pattern of the light-sail nanosats, I am shocked at the proposal and the backers. What military on earth would allow such a potential weapon to be created capable of extinguishing anything in its path, including aircraft and existing satellites. Does not make sense.
Hey, we know deep space exploration is extremely important to many...throwing money away on impractical ideas based on existing physics is, well, misguided IMHO. -
#1216 by rfmwguy on 14 Apr, 2016 14:44
-
Absolutely, although the launch control and even mission costs could be part of the $5 Mil.EMDRive launch enthusiasts can probably expect a $5 Mil price tag just to try it out...thus...the absolute need for ground verification tests before fundraising efforts to launch an EMDrive probe.
That's just the launch cost right? Nobody is going to launch a DIY satellite. You'll need at least several million more to build a flight-rated spacecraft that can orient itself by gimbaling the emdrive, has reliable communications, solar arrays and batteries, proper cameras, mission control, etc.
I would envision it being nothing but an Arrow...shot into deep space with only one intent...speed.
No cameras, no directional control...basically a ballistic, accelerating "bullet" with standard telemetry: velocity, acceleration, power consumption, etc. No ground based commands, it would be autonomous.
This would be an engine test only, no need for other scientific payloads.
Cheap, straight forward, no trajectory management, nothing but "the need for speed". -
#1217 by SteveD on 14 Apr, 2016 16:13
-
Besides the "death ray" capabilities, the power consumed, the unknown effects to the atmosphere, the random scatter pattern of the light-sail nanosats, I am shocked at the proposal and the backers. What military on earth would allow such a potential weapon to be created capable of extinguishing anything in its path, including aircraft and existing satellites. Does not make sense.
I would think almost all of them actually, just so long as they get control of the technology. That ground based system probably isn't going to be able to rotate enough to make a decent weapon. Honestly, it's not the laser that has me worried, its how you're going to be able to build anything that can withstand the acceleration. -
#1218 by VAXHeadroom on 14 Apr, 2016 16:35
-
I design satellites for a living and I outlined an EMDrive mission about...3 threads ago
Absolutely, although the launch control and even mission costs could be part of the $5 Mil.EMDRive launch enthusiasts can probably expect a $5 Mil price tag just to try it out...thus...the absolute need for ground verification tests before fundraising efforts to launch an EMDrive probe.
That's just the launch cost right? Nobody is going to launch a DIY satellite. You'll need at least several million more to build a flight-rated spacecraft that can orient itself by gimbaling the emdrive, has reliable communications, solar arrays and batteries, proper cameras, mission control, etc.
I would envision it being nothing but an Arrow...shot into deep space with only one intent...speed.
No cameras, no directional control...basically a ballistic, accelerating "bullet" with standard telemetry: velocity, acceleration, power consumption, etc. No ground based commands, it would be autonomous.
This would be an engine test only, no need for other scientific payloads.
Cheap, straight forward, no trajectory management, nothing but "the need for speed".
1KW with a moderate duty cycle, off-the-shelf (my shelf!) hardware & software, NASA will probably pay for the launch costs, but it could get up to $1M as a ride-share - this is all stock stuff, I design these kinds of missions every day. Doppler shift in the telemetry carrier can detect down to mm/s of velocity change, plus on-board accelerometers get you very precise data, so just point it away from the Earth and turn it on
Give me a working drive and flying it is actually the easy part!! -
#1219 by rfmwguy on 14 Apr, 2016 16:37
-
With special regards to Vaxheadroom, the real expert Here it is...
Final "stream of consciousness" proposal for EMDrive spaceflight validation testing:
1. Launch Vehicle: Delta II or equivalent.
2. Base Platform: Deep Space 1: https://en.wikipedia.org/wiki/Deep_Space_1 = 2.kW power budget.
3. Modify Platform: Removal of NSTAR, 92 mN Ion Engine, replace with EMDrive(s). Add RTG capable of 2.5 kW. Jettison Solar Array and Batteries at appropriate distance from Sol, RTG to take over.
4. Subsystems: All are useful for EMDrive probe, remove as needed (i.e. Autonav and Pepe) for weight and power management.
5. EMDrive Engine(s): 2 kW power consumption max (500 W budget for other onboard systems).
here ) using information theory (*)
. All dimensions are needed as well as the material properties. 