-
#2800 by Monomorphic on 16 Nov, 2016 19:00
-
Is that field strength close to anything real or is it an artifact? Seems enormous: 12,000 times air dielectric breakdown!
I'm using idealized geometry and an idealized dipole antenna so I suspect it would be very difficult to achieve this kind of field strength practically. -
#2801 by X_RaY on 16 Nov, 2016 19:05
-
Monomorphic, what was the material definition for the frustum, PEC or Cu?Speaking of TE012, I have been working on a 5.8Ghz TE012 design with spherical endplates. E-field strength is higher than anything I have seen before. 3.6x107 kV/m whoa!
In the bottom image, the pink frustum is 5.8Ghz TE013, while the teal frustum is 5.8Ghz TE012. Shown to scale. All dims in cm.
Is that field strength close to anything real or is it an artifact? Seems enormous: 12,000 times air dielectric breakdown!
How does E-field is supposed to work on thrust force in the various ongoing theories?
Try to run a single frequency sim at the same frequency as shown in the pic with a defined input power value instead of "No power scaling". -
#2802 by TheTraveller on 16 Nov, 2016 19:09
-
What was the material definition for the frustum, PEC or Cu?Speaking of TE012, I have been working on a 5.8Ghz TE012 design with spherical endplates. E-field strength is higher than anything I have seen before. 3.6x107 kV/m whoa!
In the bottom image, the pink frustum is 5.8Ghz TE013, while the teal frustum is 5.8Ghz TE012. Shown to scale. All dims in cm.
Is that field strength close to anything real or is it an artifact? Seems enormous: 12,000 times air dielectric breakdown!
How does E-field is supposed to work on thrust force in the various ongoing theories?
Try to run a single frequency sim at the same frequency as shown in the pic with a defined input power value instead of "No power scaling"
What is then the resulting fieldstrength?
The question was:
"How does E-field is supposed to work on thrust force in the various ongoing theories?"
Answer: It doesn't. It is related of (2 Qu Pwr Df) / c -
#2803 by X_RaY on 16 Nov, 2016 19:23
-
It doesn't? What is your definition of POWER of the EM-field in your equation? "magic happens inside" or something like P=U·I ? What is the result when U=0 (therefore E=0)?
What was the material definition for the frustum, PEC or Cu?Speaking of TE012, I have been working on a 5.8Ghz TE012 design with spherical endplates. E-field strength is higher than anything I have seen before. 3.6x107 kV/m whoa!
In the bottom image, the pink frustum is 5.8Ghz TE013, while the teal frustum is 5.8Ghz TE012. Shown to scale. All dims in cm.
Is that field strength close to anything real or is it an artifact? Seems enormous: 12,000 times air dielectric breakdown!
How does E-field is supposed to work on thrust force in the various ongoing theories?
Try to run a single frequency sim at the same frequency as shown in the pic with a defined input power value instead of "No power scaling"
What is then the resulting fieldstrength?
The question was:
"How does E-field is supposed to work on thrust force in the various ongoing theories?"
Answer: It doesn't. It is related of (2 Qu Pwr Df) / c -
#2804 by zellerium on 16 Nov, 2016 20:16
-
I would like to test Todd's theory, and here's how I propose to do so:
I don't have fancy facilities like a vacuum chamber, torsion pendulum etc and there's no way I could create a better setup than EW. But what about a high power test with a separated power source?
I could weld together a WR340 sized aluminum waveguide that is long enough to add in extra magnetrons at quarter wave spacing to scale up the power if the force is below resolution. Connect that to a circulator that I might be able to borrow from work, otherwise I'll have to build one. I could probably connect some threaded pipes to jerry rig a triple stub tuner before the horn which will probably be necessary.
After that a horn antenna will transfer power to the another horn that is connect to a tapered prism to excite the TE013. Sure the horns will be leaky and the waveguide bends will be lossy but thats what the extra maggies are for right?
Have everything mounted on a mg resolution digital scale and potentially incorporate a tuning rod in the prism if necessary. And of course seal everything really well so I don't microwave myself...
I'm thinking copper prism walls .02" thick with removable end plates to test the dissipation theory. Future iterations could use the same prism-end flange and a straight section to connect a movable short plunger to test dielectrics with the same mode.
Luckily I have access to HFSS so the design shouldn't take too long, except the circulator which I haven't given much thought to. But it sounds to me like more data is needed to advance any of these theories. Although carefully controlling every spurious effect would be ideal, my approach would attempt to boost the signal out of the noise via excessive power and hopefully using the right theoretical approach.
Might be simpler and easier to replicate what I did.
Basically the same thing but much more KISS.
I measured 8mN (0.8g) of force small to big on an electronic scale with a gravity stacked frustum excited in TE013 mode with 95Wrf forward.
No centre of CG shift nor Lorentz forces. Ok maybe a thermal chimney effect but with power on being like 5 - 10 sec, nothing gets warm.
Swap the stack arrangement and the force reverses direction in reference to gravity but maintains a small to big direction.
Unfortunately I can't afford an amp, I could definitely convert a microwave magnetron to coax...
But my objective is to sufficiently separate the power source from the test article. Sure I can quantify Lorentz forces and thermal expansion of a coax but even then there will be doubt. I honestly doubt that "nothing gets warm" in the first 5 to 10 seconds, 95 W flowing through a coax would definitely heat up above room temperature. Maybe you couldn't feel it through the jacket...
Did you publish a paper or something that details your experiment? I've seen a few pictures, sparse descriptions and cogent results here and there but perhaps I missed the document. Replication is pretty difficult without a formal paper...If the scale is electronic, I would be concerned that leakage from high-power microwaves would throw off its readings.
As far as the digital scale is concerned, sufficient shielding is definitely necessary. Microwave leakage is going to be significant and far from ideal, but that's the price you pay for completely separating the power source. -
#2805 by TheTraveller on 16 Nov, 2016 20:48
-
As far as the digital scale is concerned, sufficient shielding is definitely necessary. Microwave leakage is going to be significant and far from ideal, but that's the price you pay for completely separating the power source.
Hang in there.
Working on a 10W Rf amp, TE012, 2.45Ghz design that will cost, for the components from EBay and your local Cu sheet supplier around $500.
2mN should do a good job on a 0.01g (100uN) scale. As for EMI suppression, how much do you actually think exits out of a 10K Qu cavity? Have you actually measured what exits from a cavity hit with a single freq of 100Wrf at a Qu of 8k? There was no noticeable effect on the scale. The 4 strain gauges use a differential input to the amp, which helps to eliminate external induction effects plus the 125mm diameter scale platform is ~1mm thick nickel / chrome coated copper.
Please work out the surface temp increase for 95Wrf for 5 sec with a radiative surface area of 0.265m^2 coated with a high emissivity stove black coating. -
#2806 by WarpTech on 16 Nov, 2016 20:50
-
Speaking of TE012, I have been working on a 5.8Ghz TE012 design with spherical endplates. E-field strength is higher than anything I have seen before. 3.6x107 kV/m whoa!
In the bottom image, the pink frustum is 5.8Ghz TE013, while the teal frustum is 5.8Ghz TE012. Shown to scale. All dims in cm.
Is that field strength close to anything real or is it an artifact? Seems enormous: 12,000 times air dielectric breakdown!
How does E-field is supposed to work on thrust force in the various ongoing theories?
Since in a TE mode, the electric field E = -dA/dt, we can calculate the magnetic gauge potential as;
A = - E/2pi*f
This is the amount of "potential" Electromagnetic Momentum/Coulomb of charge.
Curl(A) = B
Div(A) = 0 but Div(A2) =/= 0, and this is where we can exploit it for thrust.
-
#2807 by TheTraveller on 16 Nov, 2016 20:54
-
Div(A) = 0 but Div(A2) =/= 0, and this is where we can exploit it for thrust.
Thrust (small to big) is the difference in the Radiation Pressure as per Force = (2 Qu Pwr Df) / c. Force direction depends on the setup and can be either be a static Thrust force being big to small or a accelerative dynamic Reaction force being big to small.
Would enjoy reviewing any EmDrive experimental data that suggests this equation is not correct. -
#2808 by WarpTech on 16 Nov, 2016 21:03
-
Div(A) = 0 but Div(A2) =/= 0, and this is where we can exploit it for thrust.
Thrust (small to big) is the difference in the Radiation Pressure as per Force = (2 Qu Pwr Df) / c. Force direction depends on the setup and can be either be a static Thrust force being big to small or a accelerative dynamic Reaction force being big to small.
Would enjoy reviewing any EmDrive experimental data that suggests this equation is not correct.
"That" equation is fine. It's the DF that's BS. -
#2809 by zellerium on 16 Nov, 2016 21:05
-
As far as the digital scale is concerned, sufficient shielding is definitely necessary. Microwave leakage is going to be significant and far from ideal, but that's the price you pay for completely separating the power source.
Hang in there.
Working on a 10W Rf amp, TE012, 2.45Ghz design that will cost, for the components from EBay and your local Cu sheet supplier around $500.
2mN should do a good job on a 0.01g (100uN) scale. As for EMI suppression, how much do you actually think exits out of a 10K Qu cavity? Have you actually measured what exits from a cavity hit with a single freq of 100Wrf at a Qu of 8k? There was no noticeable effect on the scale. The 4 strain gauges use a differential input to the amp, which helps to eliminate external induction effects plus the 125mm diameter scale platform is ~1mm thick nickel / chrome coated copper.
Please work out the surface temp increase for 95Wrf for 5 sec with a radiative surface area of 0.265m^2 coated with a high emissivity stove black coating.
Sorry I think there were some misunderstandings:
I was replying to Donosauro who mentioned that my proposed horn experiment would interfere with a digital scale because there would be significant leakage due to the horn antenna power transfer.
Also I said that your coax could've heated up, perhaps it thermally expanded causing a slight moment and pushing down on the scale. Quick coax loss calculator says only a watt max wouldv'e been dissipated, so probably unlikely... But I really have no idea because I don't know the majority of the details about your experiment. Again if you had some sort of report with pictures, collected analyses, schematics etc that would help...
I think the low power tests have been done and done well. I think its time to revisit the sledgehammer approach. Anyone have an extra klystron laying around?
-
#2810 by TheTraveller on 16 Nov, 2016 21:28
-
Also I said that your coax could've heated up, perhaps it thermally expanded causing a slight moment and pushing down on the scale. Quick coax loss calculator says only a watt max wouldv'e been dissipated, so probably unlikely... But I really have no idea because I don't know the majority of the details about your experiment. Again if you had some sort of report with pictures, collected analyses, schematics etc that would help...
The Rf amp connector was aligned horizontally with the frustum connector plus the ~2kg frustum and the EMI shield were gravity stacked on the scale. Weight loss / gain followed frustum reversal.
So how will this arrangement, with 5 sec of Rf energy applied to the frustum, generate a 0.8g or 8mN weight increase or decrease, depending on frustum vertical orientation on the scale when the Rf connectors were horizontally aligned?
I have no photos as an incident occurred shortly later and it resulted in the Rf amp no longer working.
Should shortly reverse that situation.
Do have a picture of the 3kg x 0.01g scale.
The Rf amp was returned to the manuf for repair and the frustum was sent to my frustum fabricators to have flanges fitted to both end of the frustum and machined to be highly parallel and at right angles to the frustum axis.
I have due in mid Dec:
2 x 100W Rf amps.
2 x 250W Rf amps.
2 x spherical end plate Al frustums with multilayer very high Qu coating.
2 x flat end plate plate Cu frustums.
1 x rebuilt Cu original frustum.
Let the fun begin.
-
#2811 by xyzzy on 16 Nov, 2016 21:48
-
Hi All
For one thing, there are some good EM-Drive concepts appearing regularly on this thread, but the thread gets long and things tend to quickly go out of view. If somebody knows of a practicable solution to archive the salient points as they get developed, I think that would do a great service to many.
For myself, I'd like to point out some ideas that appeared here recently and that (to me at least) were showing great promise, and that might be considered for future interest. Also there are some ideas from an engineering standpoint on which might advance the state of the art and on which I'd like to see the opinions of our resident experts.
One thing is sure. We don't yet have a theory that is credible beyond reasonable doubt. Ans so long as it stays that way and as we don't have provable predictive power from a theoretical standpoint, engineering has to lead the way for a while. At least as long as the effect is not either demonstrated to a very high level of assurance or disproven to that same level. With better empirical data it should be easier to attract more serious attention from a wider range of scientists and that's what this topic has been in great need of, almost forever.
A way to go there is to increase the efficiency of the test devices. We don't yet know how, but there are many clues, and I think that now is a good time to collect them from the various places and sum them up, and to try to integrate various points that have so far been considered separately.
- There is the cavity loading and the field strength resulting from RF Power and Q.
- There are geometric cavity properties that lead to specific field properties like a high variation of the resulting guide wavelength and of the energy density along the longitudinal axis. According to experiments, both seem to be needed in order to obtain a good effect.
- There is the RF source frequency and phase stability issue.
- There are effects of the cavity and antenna geometry on the mode selectivity of the system.
Particularly the last 2 effects have been a challenge where different people tried different approaches, and yet better ones seem to be needed. Eagleworks has done everyone a really great service with their frequency-locked driver design where the PLL was directly being locked to the cavity resonance via a sense antenna, yet they had difficulties to excite specific modes, and in the end they had to settle for a possibly non-optimal mode choice. Given one particular weak point of their particular driver design (according to the leaked paper), I'm not too surprised that the difficulties had arisen, even though the design is excellent in other qualities. On the other hand, others seem to have made better choices considering geometry and antenna placement that might have benefited from EW's driver design approach if such drivers were more easily obtainable.
Therefore, I'd like to bring these points together and offer the people here some thoughts to consider.
Rather than writing a very long post all in one piece, I'd prefer to end the philosophical part here, and continue with the technical details in the next post...
-
#2812 by zellerium on 16 Nov, 2016 22:20
-
I made the tapered prism with the small end equal to WR340 (86.4 by 43.2 mm) waveguide internal dimensions and big end equal to a WG6 (165 by 83 mm) and height ~213 mm for 2.45 GHz TE013.
Put a WR340 in the end and started playing with the aperture... looks like its somewhere in between green and blue on the smith chart. Slightly shifted down from 2.45 but a little less height would bring it back.
-
#2813 by xyzzy on 16 Nov, 2016 22:55
-
... Now to the technical, hopefully more interesting part!
Eagleworks has made a risky, but from my point of view, good, decision, to use a PLL-based driver and to reference the PLL directly from the cavity resonator by a sense antenna. Active frequency control has been proposed by many others here, but directly referencing the cavity resonance itself is, from what I've seen, a new approach. They don't make fuss about it, and in fact, it's only noticeable if you look at the block schematic of their driver more closely.
There is one thing that they really "nailed" with that design however: it's that the driver is self-locking to the cavity resonance and the resonance is basically self-tracking in real time to a very high precision. I'm not aware of anyone here attaining comparably stable locking by other means.
There is also one major and one minor disadvantage: Their system is free-running (major) and it provides no phase tracking (minor).
The major problem is that it will always automatically lock to the strongest mode in the vicinity of the initial "seed" (see paper for details on "seed") frequency. It's a basic, unavoidable modus operandi for their system. Given the way it is made, it cannot act otherwise. It's like a "greedy" optimization algorithm that will always continue searching along an upwards slope until it finds a local maximum (mode with highest field strength). If a stronger (e.g. TM) mode is nearby, the system can't stay stable driving a weaker (e.g. TE) mode, it will always self-lock to the strongest local maximum (mode). That's why, when they said that they "had difficulties" exciting a particular weak more, this would be perfectly understandable - their drive system is self-optimizing in real time along a rising gradient towards the strongest local mode and it has no way to turn this "feature" off. Basically they made a very efficiently self-locking driver and once that worked, in some respects they've fallen prey to that driver's good automatic optimization abilities.
The minor problem (rather unrelated) is that they implement a manual phase setting that is preset in advance and does not get the benefit of automatic tracking in real time. But this should be less difficult to fix (see below).
Some people here will probably make the point "So don't make the driver so autonomous!", but I'd rather for now explore the idea, how such an "autonomous" driver might be put to a better use - and I think there are ways for doing that.
IMO, the key to this is a cavity and antenna design that is strongly mode-selective. That is, rather than forcing a driver to excite a weak mode that is difficult to lock into, better make a resonant system that has a mode selectivity sufficient that the desired mode is indeed the strongest mode - then the "greedy" self-locking PLL driver will find it by necessity rather than by luck.
We've now had many cavity designs of "classical" shapes - cones with end plates. They all have strong "nearby" undesired modes in the vicinity of the ones that would be preferred. But there are other, more advanced possibilities, for example the parabolic, di-parabolic and hemispherical ones mentioned by Todd and Shell on page 129 (ff) of this thread, one example re-attached below for reference. I think that advanced resonator geometries should be worth consideration as they might provide a high mode selectivity among other things. Plus they may also provide high field strengths and some may be amenable to a high quality prototype production on a CNC lathe. I think that advanced resonator geometries should be in our future rather than in our past.
Also, there have only been single antenna systems used so far. Phased arrays can provide an orders of magnitude higher directivitiy than any single small antenna, and inside a closed resonator the antenna directivity will practically translate to a mode selectivity. By driving a resonator with a phased antenna array, it should be possible to greatly enhance the mode selectivity of the whole antenna-resonator system. Basically up to the point where the resonator accepts but a single mode, uniquely chosen by the geometry of the phased array. This should also be one of the things for future consideration.
And finally to that pesky minor problem of the missing phase control. As long as the frequency is self-locking, the phase angle can be adjusted rather slowly, the control loop does not need to have microsecond reaction times. It may be perfectly fine to measure the forward and the reflected power and slowly adjust the carrier phase to find the best match by an optimum operating point tracking algorithm. Since it does not need high speeds, a microcontroller could do that.
In fact, The Traveller (IIRC) has proposed to control the drive frequency to minimize the reflected power based on the reflected power meter readings. That's certainly a good idea, but it has its difficulties because it only controls frequency, not phase, and the frequency stepping must be very "delicate", especially for narrowband High-Q resonators, plus the control loop can be no faster than the reflected power meter.
However if the frequency is taken care of by a fast real-time PLL like in Eagleworks's design, the phase angle becomes a quantity like the power factor in mains electricity networks. It would be easier to control at modest speed and The Traveller's reflected power minimization strategy could just as well be applied to phase control instead of frequency. Eagleworks adjusted their phase manually and they reported seeing a "Phase Q", that they could maximize by tuning. That's a good start, but it does not track dynamic changes, and this tracking could be added by a relatively modest control loop running at manageable speeds.
As a summary, I'd like to include a proposed block schematic of a modified driver. It's strongly based on the Eagleworks design, but includes the phase control as detailed above (needs a continuous 360 degree phase modulator) and a little reminder that multi-antenna phased array drive might also be an idea to be considered instead of (or in addition to) an advanced geometry resonator to improve selectivity...
Obviously the entire thing with all bells and whistles would be spendy, particularly for a homemade build, but there are also some institutional EM-drive builders reading here, and who knows, someone may be interested at least in some parts and aspects as a way to improve his existing design. Plus, I think that it was time to sum these things up that individually affect the critical EM-Drive operating parameters, and mention them together in one place for easier reference.
Particularly regarding proposed superconducting resonators, where loading times are long, so must also be the phase coherence of the driver. Otherwise the RF source will start "unloading" the cavity as the phase angle drifts, before it even got fully loaded. To that end, one should consider both fast real-time frequency tracking and stable precise phase tracking in order to maintain a stable resonant field inside.
Regards
-
#2814 by otlski on 17 Nov, 2016 00:39
-
I would like to test Todd's theory, and here's how I propose to do so:
I don't have fancy facilities like a vacuum chamber, torsion pendulum etc and there's no way I could create a better setup than EW. But what about a high power test with a separated power source?
I could weld together a WR340 sized aluminum waveguide that is long enough to add in extra magnetrons at quarter wave spacing to scale up the power if the force is below resolution. Connect that to a circulator that I might be able to borrow from work, otherwise I'll have to build one. I could probably connect some threaded pipes to jerry rig a triple stub tuner before the horn which will probably be necessary.
After that a horn antenna will transfer power to the another horn that is connect to a tapered prism to excite the TE013. Sure the horns will be leaky and the waveguide bends will be lossy but thats what the extra maggies are for right?
Have everything mounted on a mg resolution digital scale and potentially incorporate a tuning rod in the prism if necessary. And of course seal everything really well so I don't microwave myself...
I'm thinking copper prism walls .02" thick with removable end plates to test the dissipation theory. Future iterations could use the same prism-end flange and a straight section to connect a movable short plunger to test dielectrics with the same mode.
Luckily I have access to HFSS so the design shouldn't take too long, except the circulator which I haven't given much thought to. But it sounds to me like more data is needed to advance any of these theories. Although carefully controlling every spurious effect would be ideal, my approach would attempt to boost the signal out of the noise via excessive power and hopefully using the right theoretical approach.
If the scale is electronic, I would be concerned that leakage from high-power microwaves would throw off its readings.
This is a possibility of reading corruption with a strain gauge load cell scale. And, it is almost a certainty when using a force restoration transducer scale. Some of the pictures of of EM Drives directly on scales look like they are Sartorius or Mettler scales which generally are the force restoration types. If so then the results are already suspect. -
#2815 by rq3 on 17 Nov, 2016 03:12
-
... Now to the technical, hopefully more interesting part!
EW established exactly what I recommended, what, almost 2 years ago, a phase locked loop. Look back many threads. You must tune the source to the fabricated cavity. What they missed is that no one knows whether these things actually provide any thrust whatsoever, so microwave tuning is somewhat pointless.
Again, what we really need to see is a force locked looped. The signal source should be frequency and phase locked, and tracking, the maximum detected thrust. -
#2816 by Bob Woods on 17 Nov, 2016 07:32
-
Watch Zellerium, a very bright person who has yet to be obstructed by the conceit that effects those of us who have attained age.... Now to the technical, hopefully more interesting part!
EW established exactly what I recommended, what, almost 2 years ago, a phase locked loop. Look back many threads. You must tune the source to the fabricated cavity. What they missed is that no one knows whether these things actually provide any thrust whatsoever, so microwave tuning is somewhat pointless.
Again, what we really need to see is a force locked looped. The signal source should be frequency and phase locked, and tracking, the maximum detected thrust.
The universe belongs to the young. -
#2817 by flux_capacitor on 17 Nov, 2016 08:38
-
The following is not my idea but I feel it interesting enough to repost here so RF specialists can give some clues. This is already used to check multi-cavity klystrons for resonance. Maybe it could also increase efficiency and thrust for a superconducting EmDrive.
White noise wideband injection
In case of superconducting YBCO cavities, the input power can be small (milliwatts) in order not to reach breakdown voltages due to high Q. Then what about white noise wideband injection? When white noise is fed into a cavity with a Q of 108, assuming 10 mW per mode in the frustum, for 100% conversion one would theoretically reach a power of 1 megawatt.
High efficiency multimode operation
Assuming 10 useful modes over a wideband of 700 MHz, one could try to fill each mode slot with 100 mW white noise in 10 kHz slots, each meaning a total of 1W from an overall noise envelope of kilowatts.
The unused white noise would be rejected automatically from the frustum to a high power load at the circulator, or used in other slightly detuned frustums, adding to the available thrust.
I do not know how many frustums one could feed with that reflected white noise, but assuming an array of 10 frustums would amount to a total of 100 modes, each adding to the total thrust. There wouldn't be any overheating in each frustum since unused noise is reflected out. Sure that would only work IF it works for superconducting cavities.
No need for lots of PLL oscillators. The existing resonant modes should get filled with broadband noise. It's like an electronic version of a multichannel Helmholtz oscillator. Or a pipe organ where all the pipes start vibrating at the same time. Yet the narrow band-pass filter will convert white noise into sine waves. They are low Q devices. But if the cavity Q gets to 108 things could go interesting. One would need an S-band noise source followed by S-band wideband amplifiers and wideband 700 MHz band-pass filter followed by high power wideband amplifier, to get some 700 watts white noise in order to finally input 10 mW of noise in each resonant mode. If it works it would be a simple solution equipment wise, but open to really multichannel operation.
One last remark for space applications: Let's say we have many frustums consuming 1W each, and so easily passively cooled by radiating into cold space. The remaining unused waste white noise could be got rid of by radiating via waveguide horn antennas into space.
I know SeeShells tried multimodes with simulations and maybe with her frustums by accident, and she thinks this does not add thrust, because under broadband injection different frequencies are accepted or not by the cavity and the phasing of those signals either add or subtract (in phase or out or phase) so finally one particular mode is excited and the others cancel out. Maybe this would require several different antennas. -
#2818 by TheTraveller on 17 Nov, 2016 08:47
-
Would seem Roger and Gilo have been busy on a new EmDrive patent application. Please note that both Roger and Gilo are the inventors:
https://patentscope.wipo.int/search/docservicepdf_pct/id00000035187289/PAMPH/WO2016162676.pdf
Plus more technical details than before.
It is going to be very interesting to see what Gilo Industries releases as their 1st EmDrive product. -
#2819 by flux_capacitor on 17 Nov, 2016 09:14
-
Would seem Roger and Gilo have been busy on a new EmDrive patent application. Please note that both Roger and Gilo are the inventors:
https://patentscope.wipo.int/search/docservicepdf_pct/id00000035187289/PAMPH/WO2016162676.pdf
Plus more technical details than before.
It is going to be very interesting to see what Gilo Industries releases as their 1st EmDrive product.
Worth noting it is the first EmDrive international patent application.
Gilo Cardozo shares the invention with Roger Shawyer. The applicant is still SPR Ltd. and not their new joint-venture Universal Propulsion Ltd. though.
