It's so cool to see people that are much smarter than me running serious experiments on things that seem to violate the law of conservation of momentum, it shows that people are willing to try new ideas.

Hello, I found this recent article regarding EM drive physics and I thought people interested in this project might find it interesting. https://www.technologyreview.com/s/601299/the-curious-link-between-the-fly-by-anomaly-and-the-impossible-emdrive-thruster/

Thanx. Interesting article.

Have you considered lining the cavity with bismuth?

Putting a PTFE or PE dielectric inside without changing too much to the frustrum is quite tricky indeed. . You can try it with a dissasembled ceramic fuse . 

You can lead the rf wire trough it and avoid thermal issues too.  Oh, and you can consider a small piece of High-Temperature Control and Instrumentation Cable for the end of the lead.

Hi, I have some 24 GHz modules here originally from intruder alarms.

Cool! It looks sleek!

How did you build the small cap? flat or with a inside coneshape? as Gary's Mod predicted it must be coneshaped to yield any effect..  see image 

I also are puzzled with the original build from Roger Shawyer . Whats hidden in this section? a magnet? a induction coil? or something else??

Is there more information about spherical endplates? I think I will need to make a new silver cavity, then I can also print the endplates.
I see a problem however on the antenna side. It´s easy to get the antenna directly from the PCB through the center. Else I will need a short RF rigid cable to get to the center. I am also not using any dielectric. Maybe this would help.

the fact is: Experiments in the UK and China were able to measure thrust without a dielectric insert  because they used much higher input power. In your setup more power will probably melt your frustrum.  The key is in the presentation of the waves. Try to make your gain with the antenna. Maybe a small printed circuit inside the frustrum were you can attach the RF onto. 

I want to put a PTFE or PE dielectric inside, but I really don´t know how it will affect the excitation and so which changes to make to the cavity.

Yo dude, V4 looks awesome, hopefully it will provide you with more accurate data in the new test environment. But in general it looks sweet.

Thanx. First test has been run, they look good. I will perform two more. If the results are repetitive, I will post the raw data as ASCII and a viewer software for easier browsing.

I have concerns over this device and its application.
specially its saftey in relation to this video, both are 'microwaves' at high frequencies.

Your microwave oven, your cellphone, your computer monitor, and your hairdryer all emit microwaves :)

This is very low power. It's like saying that your iPod is dangerous because it has electricity, and this video shows that electricity is unsafe

Depends how you use it. Low power microwaves oscillating in a closed cavity in best case don´t radiate...

As i think of the waves in a closed cavity i often think of letting brandy tumble in a snifter to enhance the flavour.. ;) 

Any video of results #2? Good work. Also, can I get confirmation, you said big end leading? As in if this was to be a rocket engine, the small end would be the one pointing out of the rocket and the big end would be pointing up to the main part of the rocket? Not questioning your results because they are well backed by the data. Just asking because this is a different result then previous EMDRIVE tests, and would thus make it unique. :) Sorta has me excited.

When comparing to a rocket, in the actual results the small end would be the one where the flames come out.
In some earlier tests, I got peaks that showed it the other way around, this confused me, too. But they didn´t occur later, so I assumed they´ve been from another source.
The actual data with modified bias voltage of the amplifier looks different. Now according to the data the big end is leading. But I want make more tests to make sure the actual peaks are not bullshit. But I am scepetical if I can trust this "thrust".

According the videos: The videos look all the same, two LEDs in the dark. The tests duration is about seven hours, the results are only visible in the graph afterwards. I can put them somewhere on a server for download, but don´t have the space yet.

That is what I thought. Okay, try to see if the "thrust" can continue to be repeated over each new experiment. If anything it is helpful in learning the true direction of motion for the next stage, and if it can be confirmed you might be able to claim a brand new patient as this drive would then be working in a way counter that of the original.

In Shawyers original video it is going in both directions. But his drive had a motorized endplate to change the geometry.
It will go into the endplate with higher group velocity. Depending on the overall geometry this can be also the small plate.

Good on you for doing this despite it supposedly going against the laws of physics. Keep trying and all the best.

Thanx. I wouldn´t have started it if the Chinese Polytechnical University and NASA had not published promising results. But it seems to work in their labs, and there are interesting theories why it doesn´t break laws of physics - so it was worth a try.

Cool work,good luck!!

Thanx. Tests are running.

Just thought I would compliment you on your objective attitude and attention to detail during your experiments, also you have great documentation and analytical results posted here. Keep up the great work.

It would be awesome to get a test model into 0g.

Also if you can detect thrust on a floating platform, you could totally add a rudder and remote control it for some fun :).

Thank you, I´m trying my best :)

If there is thrust, I´ll try to machine a superconducting cavity made of YBCO - this should be really capable of propelling the platform. The forces now are negligable, I just want to make sure there are any before I proceed.

Shouldn't be too hard, liquid nitrogen is pretty easy to get. plus you could magetically suspend it if you wanted.

The hard part will be to mill a pressform for YBCO and cooking the powder itself. I already got an oven which can do 1000°C and stuff for injecting an oxygen flow - still must do the electronics for temperature control.
Then the YBCO must be fabricated, grinded, pressed and oxygen-annealed...
Getting Nitrogen is easo though, i got a dewar tank already.

You could always use the prototype you already have as a reverse mold and make a ceramic press form from it. It should hold up to temperature, plus you wouldn't need to mill it. May or may not be feasible, just an idea.

Also, do you happen to have a gold-leaf electroscope? Was wondering if you could detect any charge change at either the large or small side if the vessel 

Unfortunately I don´t have one.

Great work! Keep it up.

Got a question, could a stirrer fan help with resonating? I thought about it since you were down for a bit.

It is how they dispersed the microwaves in a microwave oven.

I don´t think so. Made some calculations with Traveller´s spreadsheet and it turns out that even slight variations in the geometry brings the cavity out of the frequency span of the transmitter.
Imho it´s better to adjust the frequency electronically. That´s my plan for the V3 version. I am still waiting for the PCBs.
The new board will (hopefully) have the ability to measure the reflected power in magnitude and phase to be able to find resonance by itself.

http://uspas.fnal.gov/materials/09UNM/ResonantCavities.pdf
On page 8 you can see the phase behaviour of a resonant cavity. If you can measure that, you can tune into resonance.

Sounds good. :)  Again good to see you back up and running. Going to modify a microwave oven to see if I can make my own microwave thruster. If it works will update you.

Good to see you back up and running.

Some thoughts.

a) the explanation of supposed trust is called "pressure of light". Anything else does not apply, because it will violate newton's third law (something has to get the opposite reaction, and the only something is the RF stream the thruster expels, which is just a flux of photons).

<EDIT>: after following the link @cosmicvoyager posted, I realized that the explanation of the thrust being sought indeed violates Newton's Third Law. See:

https://en.wikipedia.org/wiki/RF_resonant_cavity_thruster

The thoughts that follow are based on "classic" pressure-of-light explanation...</EDIT>

b) looked up the formulas for pressure of light on Wikipedia. the momentum of photon is hv/c. hv is photon energy, so momentum = E/c (c is speed of light). The amount of photons expelled per second is dn/dt=P/E (where P is power), and the force is F=dp/dt=P/c. So simple!

c) in b), I got that the thrust does not depend on frequency. It is in direct relationship with output power, and NOTHING else. So, how about using a light source to generate thrust.. maybe an LED... or maybe an incandescent lamp is even better!

d) the thruster requires buttloads of power to generate any sensible thrust. Where on earth .. I meant .. in space .. are we going to get that much power?

EDIT:

e) For 0.5 watts of power, the thrust is going to be 1.67E-09 N, which is

0.17 microgram-force.

But even though I think the idea is quite hopeless, the amount of science and effort put into measuring the thrust is enormous, and that alone makes the project very interesting to me =) Keep up! Detecting the pressure of light is an big achievement on its own.

The frequency becomes an issue if you want to use the Q factor of the cavity to multiply your output. If you hit resonance, you can charge up the cavity with more photons. The force will multiply by the Q factor. The higher the Q, the lower the bandwidth becomes, for superconductive cavities this would be a window of less than 50 kHz range from the resonance frequency.

If thats so easy, we could attach a lightbulb on the back of a spacevessel, turn on a switch and off we go! 

Altough.. light is a very odd phenomena. Consider this: we are able to detect light eminating from the other end of the known universe! 

So wat repels those particles or waves so long and so fast? It must be something very powerfull! If i emit a sound, it goes not much further that like 20 miles using a very powerfull amplifier. If you switch on a battery powered laserpointer then you  are able to reach the moon with a button cell. amazing.

As a EM Drive is a radiowave generator (we can detect radio transmissions from neutron stars far out in space) it can be possible that the same power that drives light particles might repell the drive, and as the gravitational force is stronger on earth , it obvious we cant see much movement here. (even light cant escape from strong gravity sources, as everyone knows ) so maybe its a good idea to test force measurements on different locations: around the equator and around the polar circle.

"something has to get the opposite reaction, and the only something is
the RF stream the thruster expels, which is just a flux of photons"

It is thought that what is expelled is virtual particles.

https://en.wikipedia.org/wiki/Quantum_vacuum_thruster

I have no idea, what are these virtual particles... It feels like no one really knows what they are. OK, if we assume the virtual particles are somewhat real - my conclusions are wrong. =)

Do you have anything grounding the vessel? Like an electric ground? Also can you test the same vessel except using the covered big plate assembly from the adjustable vessel?

Ground is the battery ground, the platform itself is isolating.
But why should this be an issue? There is no electron/ion emission which could charge up the vessel.

My thought was to why there was a time limit to the force, as in why is the force decreasing over time. And I wasn't worried about an electron/ion emission as microwaves are non-ionizing, more that if given a medium that the displaced electrons would move away from the places where they can produce the most force over time and return to their place after the system is turned off.

I understand now. The system was switched off between several tests  for recharging batteries. Even a cold start on the next day could not reproduce A1-A2 graphs, but B1-B4 are no problem, they show the same behaviour instantly.

Reason I asked about the second big plate was that it was covered. My question is do you get a different force profile with a covered big plate?

Do you mean "covered" as not mechanically tuneable? I didn´t measure anything useful with the tuneable steel screw, so I put fixed plates at both ends. All new tests are made with the fixed walls.

Sorry I might be getting confused, I am referring to the picture above, wherein the big plate appears to be covered on the outside by plastic.

Well if my theory is correct it might be certain electrons on the inner surface bouncing in between the two surfaces, the inner surface and the outer surface, or to say it better, they are pushing themselves away from the inner surface and causing a ripple effect. Reason it shows like that is that it is the tipping point. As basically we would be creating what actually made the microwaves in the first place, but in reverse. Instead of high energy electrons spinning around a copper or silver cavity making microwaves. We have a vessel, which is just a big cavity of copper, that we are pouring microwaves into, and getting the microwaves back up to resonance inside the vessel. It wouldn't surprise me if by doing so, we are reversing the process. It would also explain why it is easier to make a vessel out of copper, as it would just be a natural limiter as it was a copper cavity on most magnitrons (sorry for the mispell) that makes the microwaves for the bigger vessels. Also sorry for wall of text.

If we give more power to this, then can we get more power? Can you tell me?

More power would help to overcome the losses, so yes, the force should get higher.