-
#1300
by
zen-in
on 28 Aug, 2015 18:57
-
We are considering doing two experiments with an industrial magnetron transmitter (c/w waveguide launcher, isolator and waterload) and the University of Saskatoon; either a 100kW continuous into a copper frustrum 932Hz at TE012 (designing cooling system now..) or 5MW 10microsecond pulses. Looks like we would also try having the magnetron fire into the small base or the big base, maybe some with ports on the side. Have some ANSYS modelling going down now. We would also fire the unit towards the ground that buoyant effects are not in play (I don't know why anyone would aim to produce thrust in the up position right now since hot air rising / buoyant is probably the #1 explanation for positive tests). Objective would be to produce a non-null result an order of magnitude above background.
That should produce a more noticeable result. More power -> more heat and more ionization + 3rd order effects that haven't even been considered yet.
-
#1301
by
graybeardsyseng
on 28 Aug, 2015 19:18
-
I finally finished the loop antenna modeling exercise. Not sure if this is still useful but I will post here. Likely the patterns are the most useful part - as they should give some idea of what modes can be stimulated.
These models and results are made using a version of the NEC codes (specifically EZNEC). NEC antenna modeling is not perfect but if you stay within the constraints it can produce qualitatively useful results. As with all such models, quantitative results are highly dependent on many factors including geometry, signal quality, and many more. Hopefully these will be of use - particularly if the loops are in a spherical chicken in a vacuum.
I modeled three sizes of antenna 1/2, 1, and 2 wavelength circumference, at three different heights - 1/4, 1/2 and 1 wavelength. Also modeled each of these configurations for loops both parallel to and normal to their ground plane.
Sorry for how long this took - as some of you know I have been in the process of retiring and unfortunately this exercise mostly had to take place late in evening after all other boring but necessary stuff had been done. I am now free and hope to start detailed planning for my own DIY experiments soon.
Herman
-
#1302
by
rfmwguy
on 28 Aug, 2015 19:56
-
OK, ready to change test setup a bit after much thought and planning.
Three words: Laser Displacement Sensor
Basically mount the sensor under the opposite end to NSF-1701 and measure displacement of end of balance beam. This is a non-contact, triangulation-based micrometer measuring system typically used to measure surface mount component height in pick and place assemblies.
The data rates are overkill for what we're trying to measure, but it could be translated into a nice chart or data set.
Now, here's the bad news...too expensive for a unique-use home budget. Best I ask for a loaner or donation, but am convinced this is the way to go; avoiding long throw lasers, mirrors and targets.
-
#1303
by
SeeShells
on 28 Aug, 2015 20:08
-
I finally finished the loop antenna modeling exercise. Not sure if this is still useful but I will post here. Likely the patterns are the most useful part - as they should give some idea of what modes can be stimulated.
These models and results are made using a version of the NEC codes (specifically EZNEC). NEC antenna modeling is not perfect but if you stay within the constraints it can produce qualitatively useful results. As with all such models, quantitative results are highly dependent on many factors including geometry, signal quality, and many more. Hopefully these will be of use - particularly if the loops are in a spherical chicken in a vacuum.
I modeled three sizes of antenna 1/2, 1, and 2 wavelength circumference, at three different heights - 1/4, 1/2 and 1 wavelength. Also modeled each of these configurations for loops both parallel to and normal to their ground plane.
Sorry for how long this took - as some of you know I have been in the process of retiring and unfortunately this exercise mostly had to take place late in evening after all other boring but necessary stuff had been done. I am now free and hope to start detailed planning for my own DIY experiments soon.
Herman
Simply beautiful! This was on my bucket list! Love it.
I'm wanting to do a 1/10 loop to see what the patterns are.
-
#1304
by
Stormbringer
on 28 Aug, 2015 20:25
-
Interesting news:
http://arxiv.org/pdf/1506.08614v1.pdf
LHC confirms measurements of branching fraction ratio of tau-leptons at BaBar and Belle.
" This result, which is the first measurement of this quantity at a hadron collider, is 2.1 standard deviations larger than the value expected from lepton universality in the Standard Model."
Maybe there is really a "new, unknown kind of physics" in addition to the standard model 
This plus the recent paper about the sneaky way electrons hide part of their fields both are new to the standard model.
Speculation: if an electron hides part of it's fields that way it indicates there might be a way to bend the rules of physics with it in useful ways by hiding other stuff the way the electron and other leptons do.
-
#1305
by
zen-in
on 28 Aug, 2015 20:53
-
OK, ready to change test setup a bit after much thought and planning.
Three words: Laser Displacement Sensor
Basically mount the sensor under the opposite end to NSF-1701 and measure displacement of end of balance beam. This is a non-contact, triangulation-based micrometer measuring system typically used to measure surface mount component height in pick and place assemblies.
The data rates are overkill for what we're trying to measure, but it could be translated into a nice chart or data set.
Now, here's the bad news...too expensive for a unique-use home budget. Best I ask for a loaner or donation, but am convinced this is the way to go; avoiding long throw lasers, mirrors and targets.
Others have made some of these suggestions- Use a narror beam HeNe laser mounted against the back wall of your garage. Mount a flat first surface mirror on the far end of the balance (where the laser pointer was mounted) and another flat first surface mirror above and behind the HeNe laser. Now comes the hard part: adjust the mirror angles so there is a vertical line of laser dots on each mirror, bringing the dots as close together as possible. Since the laser beam is bouncing back and forth multiple times any movement of the balance arm will be amplified and readily observable by the seperation of the laser dots. A lot depends on the stability of the balance arm and how close it will return to the earlier rest position. If you can isolate one dot and position an image sensor or camera with no lens under it that will allow you to make displacement measurements of some accuracy. Maybe by measuring the PR of the camera to that dot and the others on either side you will be able to identify each dot in the series and so be able to return the balance beam to the same rest position before each test.
-
#1306
by
X_RaY
on 28 Aug, 2015 20:54
-
Which raises the question - if you're behind the Great Firewall of China, can you access NSF?
Or even (given the fact that Chinese are banned by NASA from visiting ISS) does NASA block China access to NSF?
I was able to access the Emdrive threads on NSF from Shanghai via cable modem connection during a 2 week stay with my in-laws. This was in June/July this year. Back to lurker mode for me.
Welcome to the forum cbuchner1 .
That's a useful information, so Yang is also able to follow us, if she like.
But there is no answer from her till now and that's the bad news hidden in your post,
although some people try to get contact
-
#1307
by
deltaMass
on 28 Aug, 2015 21:14
-
-
#1308
by
aceshigh
on 28 Aug, 2015 21:34
-
Which raises the question - if you're behind the Great Firewall of China, can you access NSF?
Or even (given the fact that Chinese are banned by NASA from visiting ISS) does NASA block China access to NSF?
I was able to access the Emdrive threads on NSF from Shanghai via cable modem connection during a 2 week stay with my in-laws. This was in June/July this year. Back to lurker mode for me.
Welcome to the forum cbuchner1 .
That's a useful information, so Yang is also able to follow us, if she like.
But there is no answer from her till now and that's the bad news hidden in your post,
although some people try to get contact
being able to follow us obviously doesnt mean she is allowed to post. Hell, even Paul March won´t post here anymore, nor NASA is making it easy to contact him or Dr. White after the whole press (a lot of it too sensationalist and the more scientific press too negative) surround ME and Warp months ago...
well, this thread is getting lots of pages per week in the last months, so I may have missed some Paul March post, but I guess Eagleworks people have been silent since then, right?
-
#1309
by
abuzuzu
on 28 Aug, 2015 21:54
-
I'm a RF microwave engineer who has been lurking here for a while.
I note RFguy found a laser displacement sensor that he feels is ideal for measuring DUT movement all but for the cost.
Has anybody considered using a capacitive displacement sensor? Analog Devices IIRC makes a capacity displacement ic sensor and Linear technology has published several capacitive displacement sensor circuits in there application notes, one IIRC authored by Jim Williams.
These will work in a vacuum if need be. The basic sensing technology is low power so ic heat management should be- well- manageable.
If anyone is interested in capacitive displacement sensing and unable to find useful information with google, you are welcome to contact me and I'll find the references I have in mind.
-
#1310
by
RERT
on 28 Aug, 2015 21:58
-
Two points:
Rfmwguy's first data set shows signs of patterns in the output of a similar frequency to the cycling of the magnetron (at least to my eye). Fourier analysis on the de-trended signal (postulating a thermal ramp) might show correlation with the magnetron power cycling. If I had some kind of time series of nominal spot centre positions as numbers, I could give this a go. (This is one reason why cycling the power isn't always necessarily a bad thing).
I have to disagree with Dr. Rodal's conclusion that the likely effect is small, at least from one respect.
Rfmwguy's experiment hasn't yet changed our state of knowledge: Q is completely unknown, as is the mode shape, nor do we yet have any estimate of the actual thrust which has been stated in this thread. Both the numerator and denominator of thrust/power in the test are largely unknown. The effect may indeed be small, but rfmwguy's tests haven't yet affected the assessment of the size of the effect.
R.
-
#1311
by
CraigPichach
on 28 Aug, 2015 22:27
-
Ironically there is a unit set up to do 5MW pulses that can adjust frequency within 10MHz of 930MHz ( frustum dimensions would not have to be exact that we could find resonance). At 10 microseconds the cooling of the frustum is managable without online cooling. Issue is would you see something in 10microseconds? Such a test hopefully could produce results that are an order of magnitude above background (into the ground no buoyant effects) or allow other phenomena (photon thruster leakage?) to be measurable. This will probably be experiment one as it is relatively cheap. If it's CoM I expect to see something on a digital scale within those 10us.
-
#1312
by
flux_capacitor
on 28 Aug, 2015 23:05
-
We are considering doing two experiments with an industrial magnetron transmitter (c/w waveguide launcher, isolator and waterload) and the University of Saskatoon; either a 100kW continuous into a copper frustrum 932Hz at TE012 (designing cooling system now..) or 5MW 10microsecond pulses. Looks like we would also try having the magnetron fire into the small base or the big base, maybe some with ports on the side. Have some ANSYS modelling going down now. We would also fire the unit towards the ground that buoyant effects are not in play (I don't know why anyone would aim to produce thrust in the up position right now since hot air rising / buoyant is probably the #1 explanation for positive tests). Objective would be to produce a non-null result an order of magnitude above background.
Good news! Will you use flat or spherical ends on this bigger frustum?
At such high power levels, I wonder if a hollow copper hull could be designed, the interstice being filled with liquid N
2 for active cooling, with longer stady-state duration and Q increase.
-
#1313
by
rfmwguy
on 28 Aug, 2015 23:33
-
-
#1314
by
Silversheep2011
on 29 Aug, 2015 00:11
-
Thought for today during coffee...
NSF1701 vents
-
#1315
by
Rodal
on 29 Aug, 2015 00:20
-
Thought for today during coffee...
NSF1701 vents
Rfmwguy has measured the temperature repeatedly and posted the results. What gets hottest by far is the magnetron itself, not the truncated cone.
-
#1316
by
RAID_RONIN
on 29 Aug, 2015 00:41
-
OMG moment at Phys.org.
New article advising of a new Theory on radiationless states of atoms and why orbiting electrons don't radiate.
http://phys.org/news/2015-08-theory-radiationless-revolution.htmlI think this might be a sleeper discovery with interesting applications.
Think Cherenkov radiation and either how to disperse it or hide it....
From the article;
Dr Miroshnichenko, in collaboration with colleagues from Germany and Singapore, successfully tested his new theory with a single silicon nanodiscs between 160 and 310 nanometres in diameter and 50 nanometres high, which he was able to make effectively invisible by cancelling the disc's scattering of visible light.
I could also be thinking way to much into this.
-
#1317
by
Rodal
on 29 Aug, 2015 00:51
-
Two points:
Rfmwguy's first data set shows signs of patterns in the output of a similar frequency to the cycling of the magnetron (at least to my eye). Fourier analysis on the de-trended signal (postulating a thermal ramp) might show correlation with the magnetron power cycling. If I had some kind of time series of nominal spot centre positions as numbers, I could give this a go. (This is one reason why cycling the power isn't always necessarily a bad thing).
I have to disagree with Dr. Rodal's conclusion that the likely effect is small, at least from one respect.
Rfmwguy's experiment hasn't yet changed our state of knowledge: Q is completely unknown, as is the mode shape, nor do we yet have any estimate of the actual thrust which has been stated in this thread. Both the numerator and denominator of thrust/power in the test are largely unknown. The effect may indeed be small, but rfmwguy's tests haven't yet affected the assessment of the size of the effect.
R.
1) Please don't state that you disagree with someone without providing an actual quotation of precisely what you disagree with. I made a "big picture" assessment referring to several tests, not just Rfmwguy's.
What I wrote was:
...
There are also experiments that have already been done, that had null results but have not yet been properly documented in the EM Drive wiki for experimental results, for example: Mulletron's experiment who deserves to be recognized as the true leader in these Do-It-Yourself experiments, had null results documented in early threads, also the experiments with the experimenter that had constant cross-section waveguide and was using dielectric inserts, gave null results.
I understand that null results in these experiments don't close the book, as any experiment is limited on what can be tested (type of force measurement, set-up, etc.), but as Shell, says, "there is no bad data" and we should not forget those null results as well, as they are part of the big picture.
The big picture that is emerging is that if there is any EM Drive force (yes, it may be zero), its force/InputPower magnitude is of the very small magnitude reported by NASA (~0.001 Newton/KiloWatt) and by Dresden University (Tajmar) (0.00003 to 0.00013 Newton/KiloWatt) but not at all the thousands of times greater magnitudes reported by Shawyer (0.4 Newton/KiloWatt) and Yang (1 Newton/KiloWatt) - who nobody has been able to reproduce-
The force/InputPower reported by Iulian Berca was similar to the one reported by NASA.
2) The statement I made was in reference to the big picture of several experiments. You have out of context, without providing a quotation, subsumed my statement as if it would be solely based on rfmwguy's experiment.
3) The statement that nobody up to this date has been able to reproduce the thrust/InputPower reported by Shawyer and by Yang is an undeniably correct statement, that stands on its own.
4) Rfmwguy didn't just blindly conduct the test without having any idea of calculated thrust. Actually discussions were made in the thread of the ability of the teeter totter measurement to measure given levels of thrust. Furthermore anyone can calculate the thrust based on existing formulas. Thrust (per a given value of Q) can be calculated using Shawyer, McCulloch (he has several formulas) and Notsosureofit's formulas for example. The geometry was provided by Rfmwguy.
5) It is incorrect that <<denominator of thrust/power in the test are largely unknown>> the denominator: the InputPower is known. Most assuredly RFMWGUY did not conduct a test blindly without knowledge of the input power. Rfmwguy even conducted careful tests at different power settings and measured the temperatures, much prior to this actual test.
6) I and others have (previous to your post) discussed that the experimental Q for this experiment is unknown. However, I have calculated the mode shape and the Q at 2.45 GHz for the designed geometry. Furthermore Q's for other tests have been posted by EM Drive researchers: by NASA and by Shawyer. If the Q were similar to those reported by NASA or Shawyer, or if the Q would be the one calculated for this mode shape and dimensions, and if the response were proportional to Q (as assumed by Shawyer, McCuloch and Notsosureofit) the response should be significantly larger than the measured response. One can also calculate how small the Q needs to be in order for the response (for the known power input) to be outside the limits of this measurement.
-
#1318
by
aero
on 29 Aug, 2015 01:01
-
Some new .csv files uploaded to Google drive. A few .png views also.
https://drive.google.com/folderview?id=0B1XizxEfB23tfmcxWjlORHBtazE5bW5zR0NHbXJzVU5QTHpHM1U1aXVoYzZkazRwUlpNaW8&usp=sharingThis is SeeShell's Crazy Eddie model 2, revision 6. The revisions are the result of changing the antenna configuration. We were assured that this cavity geometry would resonate, and it does resonate very well. It was just surprisingly difficult to find an antenna configuration that excited a resonance mode. The mode that did finally excite is hard to identify, perhaps Dr. Rodal or other experts here will be able to nail it down. It was hoped that the cavity would resonate at 2.47 GHz, and the final calculated resonant frequency was 2.50 GHz, well within the 3% error bounds on Harminv.
Some of you may recall that back in April we did considerable work to nail down the errors in Harminv resonant frequency calculation. Dr. Dominic, who is an experienced meep user helped (did most of the work) to determine this. I, on the other hand, had forgotten about that effort until recently.
-
#1319
by
jmossman
on 29 Aug, 2015 01:25
-
...
Rfmwguy's experiment hasn't yet changed our state of knowledge: Q is completely unknown, as is the mode shape, nor do we yet have any estimate of the actual thrust which has been stated in this thread. Both the numerator and denominator of thrust/power in the test are largely unknown. The effect may indeed be small, but rfmwguy's tests haven't yet affected the assessment of the size of the effect.
R.
...
4) It is incorrect that <<denominator of thrust/power in the test are largely unknown>> the denominator: the InputPower is known. Most assuredly RFMWGUY did not conduct a test blindly without knowledge of the input power. Rfmwguy even conducted careful tests at different power settings and measured the temperatures, much prior to this actual test.
...
Hi Dr. Rodal,
My interpretation of Rert's comments was that knowledge of raw magnetron InputPower is insufficient to define a denominator in the thrust/power equation. Unless I missed some critical information within this thread (quite possible), I don't think we know the impedance match between the magnetron and the frustum, and therefore don't know how much of the InputPower is actually getting coupled.
Do we currently have any way of determining reflected vs delivered power during rfmwguy's tests?
Admittedly I also don't know if reflected vs delivered power makes any difference in "thrust".... but given the Meep analysis so far being based upon delivered power to the frustum, I don't think I'm going to far out on a limb here with Rert.
If I recall, TT has been nudging rfmwguy to insert a small antenna of some kind to at least allow for some crude S11 analysis and potential real-time monitoring with a cheap VNA during a power-on test. If rfmwguy could follow SeeShell's lead with a GoFundMe URL, maybe the collective NSF lurkers (such as myself) could help send some pocket change his way to help fund things like a cheap USB VNA.
Thanks,
James
