-
#600 by X_RaY on 22 Dec, 2015 19:36
-
I think thats because it changes the complex impedance of the cavity(and therefore the Z_in), this leads to over- or under- coupling. It could be helpful to change the placement of the antenna. If you have luck it fix the impedance / return loss problem.
I tried this and you are correct, it shifts higher, but at a terrible sacrifice of return loss...took it to about 10dB. My little experiment was simply to see if brass material behaved itself around RF and whether I could obtain a nice RL...mission accomplished...going w/unfinished brass.Not me Doc, the tuba (baritone) simply showed me that brass has a great resonance and return loss (30+dB) albeit too low in frequency for me to use. Sooo...I am contracting a brassmith to build a frustum cone of identical dimensions to NSF-1701. No compound curves...no bell shape, although it might be cool to play around with it some day. Wackiness is not afoot...yet
If you shorten the end with the large diameter of this instrument, the resonances will shift to higher frequencies.
I'll test for resonance, brassmith will make adjustments afterwards based on what I need. Then another RL test. Then polish it up...then perhaps silver plate. I live near Conn-Selmer instruments and lots of folks around here who can make brass sparkle with a variety of plating.
I figure, if NSF-1701A flops, I'll make it into a Tuba and serenade Doc. -
#601 by rfmwguy on 22 Dec, 2015 19:59
-
You seem to understand canned filter software, which is all current, but my experience started when a TO-8 replaced a much larger LC box. In the early 80's it was state of the art. Now, ceramics are used and no tuning involved, its more of a foundry operation. You simply weren't around early on and cannot appreciate the lack of canned packages, just the math...not to mention no powerful PCs. You have it easy by comparison, come on.The equations you are using for that kind of design are approximations and won't work when utilized outside of the assumptions that made the approximations possible. These days we use more complete set of equations that are complicated, but computers are up to the task. In a way it's vile how close the current stuff comes to actual on board components inside an enclosure when only a decade ago agonizing over Smith charts was still useful.
You've made my point, Anatol I. Zverev's math failed when matched up against the real world senarios. Technicians and production engineers added tweaks to these base formulae and practical software emerged. Thanks for helping me make my point. Math is a beginning, not an end.
Anyway there seems to be a large misconception about how graduate school works and the advisor student relationship. I've never seen it so follow the leader as you often insinuate here.
I've not insinuated nor commented on a PhD advisor before. It was simply what I would tell a student. Do something different, take a risk, push the envelope, etc.,
Surely you're not making the case that Risk Aversion is not a serious problem in the workplace, are you? If you are, check out hbr: https://hbr.org/2009/01/trapped-in-a-riskaverse-workpl/ plus many other articles. Been there, lived that.
No, we don't use Zverev. We use more math, not less, and nobody tweaks anything. If you have to tweak, you're dead. You can't make money. It has to come out of the math working, or else. Because it is possible, and if you don't do it, someone else will.
Everything is so incredibly integrated now all physics for any design must be considered. Multi-physics now rule, because you can't over engineer bandwidth to take care of temperature variation, there just isn't enough space/cost overhead. On top of that parts of the circuit are captured through different ways, like getting inductance from flex pcb, etching things directly to the inside of packages, etc. There is no room to fit TO-8s into cellphones, come on.
People in the industry still use those shake n bake books, but that's for supporting legacy military systems where there is no pressure to reduce size and cost at all.
As for risk aversion, nobody is risk averse in university labs. There are no resources to risk. You don't get paid well at all, the people that do it are there just to do that kind of thing. They like it. And there is no agreed upon dogma. Students argue with their advisor over things more often than married couples who are heading towards divorce.
"Nobody is risk averse in university labs"
I smell BS about risk aversion in university labs. Instrument time can be precious and a requestor might just get flak for the purpose of the experiment from other students and from the lab manager. Go try and request telescope time to look for cities on the moon and see how much priority you will receive. -
#602 by SeeShells on 22 Dec, 2015 20:02
-
Everyone is out looking at cars so I'm in for just a bit. Your right, I saw something don't know really what and right now I can't claim much of anything just a personal observation. It is a good starting point.
We've had several hand-waving "no thrust" at both our observation phases. You were there. Trust your personal observations, then data record/rinse & repeat.(...)
To answer your question in as few words to eliminate error. This is what was seen.
At least that's where the observed data seems to be taking me.
Shell
Added: the reason it is with the micrometer big up point up is it's easier to get to turn instead of up and down from the floor, getting older you know.
I find it interesting that despite injection point differences, the observations were the same...movement against lift towards the small end facing downwards. In your case, much higher thermal isolation of the signal source, and much less assumed thermal lift. Think you measured a cool frustum.
Translation: both your and my Observations were non-null and its time to roll up our sleeves and find out why.
Think you have a jump on me since you're already above me at 177 micros...I've still got some work to do.
I'm reminded of my dad's old mechanic that worked on his cars. My dad knew cars very well, but had a problem with his 55 Buick missing, he did everything you could think of to it, it still missed. He drove it in to this old mechanic and before my dad could get out of the car he said "Bill, your number 8 has a plug wire that's arcing, you should replace it, it will make you miss." I remember my dad looking at him like he had a chicken on his head.
That's a old school "feel" and where your art takes on a extra feel for what should work and not. In my power on test I felt the same way, that shouldn't be happening. Can't put a number to it, can't define it other than observation but it wasn't right to see the device produce what appeared to be a thrust. I've had my engineers who worked for me bring me a design and I'd scan it quickly and say nope that will not work and point to the area it would fail at, or be called out on the production floor because of some problem in a complicated piece of control electronics and an a few seconds say "it's the plug wire on number 8 that's arcing".
I've learned to see that my time after schooling was just another longer and harder school of hard knocks and instead of grading on a curve it becomes fail or not fail.
Shell -
#603 by Rodal on 22 Dec, 2015 20:04
-
Quote
I smell BS about risk aversion in university labs. Instrument time can be precious and a requestor might just get flak for the purpose of the experiment from other students and from the lab manager.
Gee, man, let's agree to give up this discussion about risk aversion in Universities.
Your experience runs completely contrary to my experience at MIT (and my lesser, much less thorough experiences at meetings at Stanford, Berkeley, Case Western, CalTech, Cambridge University (UK), and a number of German Universities).
Even if you have experience (at some unnamed University) concerning risk aversion, that is a subjective personal experience. What is the point of opening this line of discussion, about people's subjective experiences in universities, in the EM Drive thread?
You are going to get pushback from people having the complete opposite experience: students and advisors at universities seek the truth, push the risk envelope and encourage innovation, because I don't want my presence in this thread (and any silence on my part thereof, since I don't use a monicker) to be interpreted as sharing your opinion concerning universities.
Do we want participation in the EM Drive thread to be interpreted as a group of people that find universities to be risk-averse? -
#604 by Notsosureofit on 22 Dec, 2015 20:13
-
@Rodal
Working, so have to run back and forth....
So, 1/D^2 {as fn of x} = (1/Ds^2) - (x/L)((1/Ds^2)-(1/Db^2)) for x from 0 to L, to give linear dispersion (in this simple model) so as to be completely eliminated by the "gravitational" dispersion at some acceleration.
Make sense ?
-
#605 by A_M_Swallow on 22 Dec, 2015 20:28
-
Can the thermal forces be turned into torque by putting a hole on only one side of the EMDrive? Possibly strengthened by a single cooling fin. Since the trust from the microwaves is up/down this may permit separation of the two forces.
-
#606 by Notsosureofit on 22 Dec, 2015 20:54
-
@Rodal
Working, so have to run back and forth....
So, 1/D^2 {as fn of x} = (1/Ds^2) - (x/L)((1/Ds^2)-(1/Db^2)) for x from 0 to L, to give linear dispersion (in this simple model) so as to be completely eliminated by the "gravitational" dispersion at some acceleration.
Make sense ?
So in the final analysis, is the optimal shape a simple cone (linear with x?)
No, for linear dispersion D {as fn of x} = ( (1/Ds^2) - (x/L)((1/Ds^2)-(1/Db^2)) )^(-0.5)
Right ? (hard when brain gets this old) -
#607 by rfmwguy on 22 Dec, 2015 21:04
-
Glad you brought this up, because of attitudes like carroll, baez and perhaps some people right here, right now, not one university in the USA has announced any intention of investigating the emdrive. Risk adverse? I think the case is clear despite the relatively low cost of testing the device.QuoteI smell BS about risk aversion in university labs. Instrument time can be precious and a requestor might just get flak for the purpose of the experiment from other students and from the lab manager.
Gee, man, give up this campaign of talking about risk aversion in Universities.
Your experience runs completely contrary to my experience at MIT.
Even if you have experience (at some unnamed University) concerning risk aversion, that is a subjective personal experience. What is the point of opening this line of discussion in the EM Drive thread?
Of course you are going to continue getting pushback from people having the complete opposite experience: students and advisors at universities seek the truth, push the risk envelope and encourage innovation.
I don't understand your insistence on this line...
So the gauntlet has been thrown. Perhaps your alma mater, replete with their apparent non-risk aversion, will give it a go. I won't hold my breath. -
#608 by VAXHeadroom on 22 Dec, 2015 21:22
-
New ray-trace animation of SeeShells' frustum. MEEP data file by aero. Only change I have made is to output data every 0.02 of a wave cycle i.e. 51 frames per one wave. H fields only. POV-Ray with camera movement runs through the 51 frames 10x, and the video repeats this 3x. This generated 63GB of data. Currently working on freeing up more hard drive space as I'd like to do a finer grid in meep but the data files grow FAST when you do that. Enjoy. Comments / suggestions encouraged. I'm working to understand the magnetic fields in the device...
-
#609 by VAXHeadroom on 22 Dec, 2015 21:25
-
I also recorded this video on my phone - same data as previous post, this is Z slice of the H field data (x component only) at half-way through the device. Animation is from the H5 viewer. Note the numbers are 'MEEP VALUES' not a known value. Further investigation turned up the max was about 0.3, so > 2x higher than the values shown in this scale. Discussions with aero suggest the correct meep scale is 1/0.3 so 3.33x the numbers shown here, but I don't think that's authoritative (yet). And this is only the X component...
-
#610 by Flyby on 22 Dec, 2015 22:28
-
Paraboloid of revolution:
euh.. didn't I suggest that shape about 10 months ago and got spanked for that ?
http://forum.nasaspaceflight.com/index.php?topic=36313.msg1334659#msg1334659
But taking the opportunity here to discuss shapes and their consequences...
Would one of the readers here , that is skilled in programming, be able to create a program/routine that can calculate how many rays bounce in a forward direction (small end plate) and how many rays go backward towards the large end plate?
I might be wrong, but let's say for 10k bounces, I have a hunch that due to the shape of a frustum you'll get more forward bounces then backwards.
I get this idea from observing that the angle of indices of the frustum side walls, in the small plate direction, is more often steeper then in the other direction. But.. i could be wrong...hard to tell
Doing it 10000 times by hand doesn't seem like a good solution to verify this thesis... so.. any one care to help? -
#611 by ScottL on 22 Dec, 2015 23:15
-
Glad you brought this up, because of attitudes like carroll, baez and perhaps some people right here, right now, not one university in the USA has announced any intention of investigating the emdrive. Risk adverse? I think the case is clear despite the relatively low cost of testing the device.QuoteI smell BS about risk aversion in university labs. Instrument time can be precious and a requestor might just get flak for the purpose of the experiment from other students and from the lab manager.
Gee, man, give up this campaign of talking about risk aversion in Universities.
Your experience runs completely contrary to my experience at MIT.
Even if you have experience (at some unnamed University) concerning risk aversion, that is a subjective personal experience. What is the point of opening this line of discussion in the EM Drive thread?
Of course you are going to continue getting pushback from people having the complete opposite experience: students and advisors at universities seek the truth, push the risk envelope and encourage innovation.
I don't understand your insistence on this line...
So the gauntlet has been thrown. Perhaps your alma mater, replete with their apparent non-risk aversion, will give it a go. I won't hold my breath.
I think this is because Universities don't view this as a risk, but a waste of time. I'm certain you could probably talk an undergrad into doing it though with little effort. Most labs operate in a "whatever your interest extends to" model. Outside of the PI's project work, researchers are usually welcomed to do what they want providing they aren't damaging anything, that's been my experience at least. -
#612 by Notsosureofit on 22 Dec, 2015 23:34
-
@Rodal
Working, so have to run back and forth....
So, 1/D^2 {as fn of x} = (1/Ds^2) - (x/L)((1/Ds^2)-(1/Db^2)) for x from 0 to L, to give linear dispersion (in this simple model) so as to be completely eliminated by the "gravitational" dispersion at some acceleration.
Make sense ?
So in the final analysis, is the optimal shape a simple cone (linear with x?)
No, for linear dispersion D {as fn of x} = ( (1/Ds^2) - (x/L)((1/Ds^2)-(1/Db^2)) )^(-0.5)
Right ? (hard when brain gets this old)
##$^&& .. still can't send a picture !!
-
#613 by CuriousDreamer on 22 Dec, 2015 23:42
-
I have been following this forum for a short time now and find my curiosity getting the better of me. I have more of a conceptual understanding of physics than a in depth mathematical understanding and therefore only understand maybe 40% of what I have been reading here but even with that limited understanding it is exciting!
What I am currently curious about is something I saw in a video of Dr. Harrold White back in (I believe) April. Dr. White reported that during a EMDrive test they fired a laser through the test device and saw an increase in the travel distance of the laser. Dr. White suggested that a possible explanation of this was that the EMDrive somehow warped space. My question is, has this increase in travel distance been explained in a satisfactory, less science fictiony, way?
Personally I would be overjoyed if this device really was warping space.
-
#614 by zen-in on 22 Dec, 2015 23:45
-
Glad you brought this up, because of attitudes like carroll, baez and perhaps some people right here, right now, not one university in the USA has announced any intention of investigating the emdrive. Risk adverse? I think the case is clear despite the relatively low cost of testing the device.QuoteI smell BS about risk aversion in university labs. Instrument time can be precious and a requestor might just get flak for the purpose of the experiment from other students and from the lab manager.
Gee, man, give up this campaign of talking about risk aversion in Universities.
Your experience runs completely contrary to my experience at MIT.
Even if you have experience (at some unnamed University) concerning risk aversion, that is a subjective personal experience. What is the point of opening this line of discussion in the EM Drive thread?
Of course you are going to continue getting pushback from people having the complete opposite experience: students and advisors at universities seek the truth, push the risk envelope and encourage innovation.
I don't understand your insistence on this line...
So the gauntlet has been thrown. Perhaps your alma mater, replete with their apparent non-risk aversion, will give it a go. I won't hold my breath.
I wouldn't expect a research lab at MIT to spend any time investigating the em-drive. If an em-drive was being built at MIT it would likely be by some MechE students. There is a very strong maker culture there and most of what students do is not widely publicized. During my time at MIT I saw some interesting devices made by MechE undergrads, especially by a winner of the 2.01 design contest. However the theoretical conundrums of the em-drive would be obvious to any undergrad at MIT so I would expect that after the physics and experimental evidence were examined interest would drop off.
-
#615 by frobnicat on 22 Dec, 2015 23:50
-
I believe the argument is that constant power, creating constant thrust will -- with sufficient time -- result in the total kinetic energy in the system going over unity.
This is an incorrect characterisation of the overunity issue. I've heard others make similar statements, so I wonder if it's a common misunderstanding.
Maybe this stems directly from the mission profiles put forward by proponents (Shawyer, EW papers...) as the first shocking revelation to the vigilant reader is that the onboard generator total output doesn't quite amount (and by that I mean is less than a tenth) to the total energy that would be required to do such trajectory even if the solar system was a tarmac to drive on with a 100% efficient electric car.QuoteIt's not an issue of the total kinetic energy of the system exceeding the total energy consumed by the system. The issue is that above the critical velocity, the rate of increase of kinetic energy of the system exceeds the rate of consumption of energy (ie, power input).
Agree, the issue therefore appears "instantaneously", 1 second of "self powered frame agnostic* propellantless thrust" at above 3.33µN/kW is already too much not to cause apparent CoE issue in some (arbitrary) frame. And obviously other proven self powered frame agnostic propulsive schemes don't suffer this issue (because they are not propellantless), whatever the (arbitrary) frame used to account for changes of kinetic energy of both the craft and the expelled propellant.
* By frame agnostic I mean a device that with the same given power will have the same thrust (an accelerometer will record the same acceleration for the driven craft) whether or not it has previously been subject to some arbitrary deltaV by conventional mean.
Still, having "the total kinetic energy of the system exceeding the total energy consumed by the system" is a "sufficient issue" by itself to be relevant (requiring an explanation), even if it's a weaker argument (and possibly easier to dismiss by bad use of change of reference frame bogus counter examples with proven schemes)Quote.../...
the required densities [...] would be higher than dark matter, for which we already know there isn't quite enough even if it interacted with microwaves like crazy
Would this be the case if the dark matter was essentially non-rotating around the galaxy? So that the flow through the Earth was 200km/s +/- 30km/s. The V² in the Ek formula now drastically increases the amount of energy we can transfer from the dark matter, compared with a stationary (ie, low relative-v) "pool" around the Earth that the EMDrive is pushing against.
(I believe that previous testing has ruled out the implied directionality of this, I'm just wondering hypothetically.)
Do you mean, previous testing of Emdrive would have detected a preferred direction of thrust (relative to the stars), in the hypothesis that an interaction of powered Emdrive with fast dark matter wind is the source of measured force ? Probably if it is a passive "parachute" effect. Not compulsory if it is a powered jetting effect : consider the device as a propeller that swallows medium's mass around and is sending it back at much higher velocity than that of the incoming flow. This would show little directionality... But that would also imply a high ejection velocity, hence a low thrust/power (ejection speed>>200km/s => thrust/power<<0.01N/kW )
Anyway, this was discussed before (sorry rereading it that all not sound neat and tidy) :
http://forum.nasaspaceflight.com/index.php?topic=29276.msg1266616#msg1266616
The best case was for highest possible (known from observational bounds) densities at earth orbit, assuming they are higher than estimated galactic density. Even if it implies slower flow velocity (below sun escape velocity) this would allow so much higher densities (than galactic average) that it could more than compensate for the lower velocity. And while this best possible case does look like almost near enough, as far as momentum flow is concerned, this would still need a high interaction ratio of dark matter with the emdrive...
-
#616 by OnlyMe on 22 Dec, 2015 23:59
-
I have been following this forum for a short time now and find my curiosity getting the better of me. I have more of a conceptual understanding of physics than a in depth mathematical understanding and therefore only understand maybe 40% of what I have been reading here but even with that limited understanding it is exciting!
What I am currently curious about is something I saw in a video of Dr. Harrold White back in (I believe) April. Dr. White reported that during a EMDrive test they fired a laser through the test device and saw an increase in the travel distance of the laser. Dr. White suggested that a possible explanation of this was that the EMDrive somehow warped space. My question is, has this increase in travel distance been explained in a satisfactory, less science fictiony, way?
Personally I would be overjoyed if this device really was warping space.
SeeShells' design and plans I believe include an attempt to run a laser interferometer test, through a Quartz rod. -
#617 by Rodal on 23 Dec, 2015 01:32
-
@Rodal
Working, so have to run back and forth....
So, 1/D^2 {as fn of x} = (1/Ds^2) - (x/L)((1/Ds^2)-(1/Db^2)) for x from 0 to L, to give linear dispersion (in this simple model) so as to be completely eliminated by the "gravitational" dispersion at some acceleration.
Make sense ?
So in the final analysis, is the optimal shape a simple cone (linear with x?)
No, for linear dispersion D {as fn of x} = ( (1/Ds^2) - (x/L)((1/Ds^2)-(1/Db^2)) )^(-0.5)
Right ? (hard when brain gets this old)
##$^&& .. still can't send a picture !!
Does (1/2 of) the optimal cross-sectional shape (parallel to the axis of axisymmetry) for the NososureofitEMDrive look like this picture ?
(see attachment)
-
#618 by Notsosureofit on 23 Dec, 2015 01:56
-
Are those two sheets in the solar wind ?
-
#619 by SteveD on 23 Dec, 2015 02:05
-
Shell. Just saw your post on reddit. It strikes me that, while I'm not sure anyone is clear on the direction of movement in all these experiments, we seem to have a pattern emerging. Both plates solid, movement toward small end. Dielectric in small end, movement toward big end. Tuning screw in small end -- nothing. Tuning device in small end -- movement towards big end. Ends secured with loose clips -- nothing.
To answer your question in as few words to eliminate error. This is what was seen.
My working theory has been that very small forces are pernicious and end up working the tuning screw / clips off instead of driving the device forward. Seeing the direction of thrust reverse, similar to what seems to be reported for using a dielectric, makes me wonder about that. (Is your inside small base electrically conductive to the rest of the frustum? I know you have some form of seal on the inside so the thing isn't floating, but can this pass a current?)
Whatever the reason, I think you may have nulled the main force. Instead, you are seeing an opposite or retarding force against the large baseplate. Given the dimensions of the frustum (big base approximately equals length)this might be 1/4 of the force on the small base. Then again that NASA model somebody posted the other day showed much stronger fields on the small base than on the large one.
What I think this is showing is that the rf forces in the frustum are balanced. There's however a second force that is reacting to those rf forces on some kind of EM field strength per cm^2 of area basis. Fields are weaker at the large end, so more of that force is being produced at the small end. Null the smal; end and it moves towards the bigger end, though with less force. (Which makes me wonder how strong a none null main force would be).
At least that's where the observed data seems to be taking me.
Shell
Added: the reason it is with the micrometer big up point up is it's easier to get to turn instead of up and down from the floor, getting older you know.
One more thing, yes it's all electrically connected endplates to sidewalls.
I'm sorry if I somehow gave offense. So it seems that 1. much confusion reigns as to what direction the force is toward and 2. you measured force in the same direction as both NASA (presumably) at rfmwguy. Looks like the original reddit post got deleted. I do see why you would expect thrust in the other direction.
Might as well address some criticism I got for the original post. Let's say we had F = X in one direction and F = X/4 in the other direction. That actually makes some form of sense as this is, of course, a closed cavity in which we should expect F = X in both directions. Having F=0 in one direction would be unexpected (but then again so is X<1). If breaking electrical contact between one of the plates killed the force in that direction, then that would be relatively easy to test for. It would help explain the higher force on a curved large endplate (greater total area). It would also mean that we can stop trying to figure this out in terms of radiation pressure. Nicely this is also something testable and falsifiable (though perhaps not in Shells build). Put a removable insulator between the frustum and the baseplate (probably with a solid state build) and you can, relatively, quickly rule this in or out. Yes, I'm excited -- because some data is suggesting an experiment that would give us a better understanding of what's going on here.
