Shawyer Cavity Simulation Optimizations

I will be back up at the University next week after the holiday Monday but just for a few days as I won’t be around on Thursday or Friday next week.)

This week a cavity (see the attached picture) was simulated and optimized to generate a pure TE0,1 mode and it turns out the diameter of the small end of the cavity was very important as even a difference of 1mm would mean a different mode. The mode was less sensitive to the length of the cavity as a change of 1cm was enough to change mode. The good news is once the cavity is properly tuned for a TE0,1 mode, it has a significantly higher Q because of low field strengths along the walls.

It should be possible to tune the cavity with a large movable circular plate as the short at the narrow end. Because the mode is TE0,1, the field strength around the edges are small. Having looked over the square tuners found in the lab, it is clear when field strengths are low on a certain wall, that fairly large spaces can be used (see attached picture).

A simple circular cavity at 152mm in diameter was simulated with a loop instead of a probe to properly generate a TE0,1 mode (see attached pictures). Fabrication of a loop probe maybe a difficult task considering it has to terminate in an N connector and would have to be inserted from the inside after the cavity has been built.

Here is what I will do next week:

  • Finish getting both probes matched to the cavity.
  • Fill the interior with Teflon and check the Q
  • Double the size of the large end to see if the TE0,1 at the narrow end changes to a TE0,2 mode in the large end or some other mode.
  • See if any further optimization of the cavity length can be done.


FWD: EMDrive is Going Backward – is Fg1 really > than Fg2?

Email from Shawyer below stating the mode is TE0,1, however the attached patent application from 1999 states TM0,1??  Maybe the mode in the cavity of the patent application is different then the demonstration engine.  I will be up today.


——– Original Message ——–

Subject: Re: EMDrive is Going Backwards – Is Fg1 really > then Fg2?
Date: Mon, 25 Jun 2007 14:35:04 +0100
From: Roger Shawyer
To: Sean Donovan
The cavity is operating in the TE01 mode.
No duplication of the work has yet been reported.
Roger Shawyer


Cable Heating Was Causing Movement

We did two 90 degree tests on Monday and the cavity did not move as expected, which means our previous movement was from cable heating.  For our first tests, the cavity by chance, moved in the expected direction.

Here are the animated gifs showing the first and last frames for two tests, each at 90 degrees to each other:

Note the slight bend in the coax?  In both tests it moved in the same direction, to the top left of the frame.  The only thing rotated was the cavity while the coax and camera stayed in the same orientation.  If the cavity had worked, there should have been a notable movement first toward the left and then toward the bottom.  With at least 10 minutes between each test, the cable had time to cool down.

There was none of the dramatic movement noted during the previous 75 second test.

Please note that we also modified the test setup to straightened the coax and moved the microwave source assembly including the fan, outside the Faraday cage.

The next step is to build a better cavity but we are going to keep the coax.  The movement from coax heating is small, we know which direction it moves and as long as the cable has enough time to cool down, we can account for it.  We also expect the cavity movement to be readily apparent.

Shawyer Reproduction Test Results & Eliminating Causes

Below are tests I will do in order to eliminate the other variables

  • Heated air escaping the cavity in a jet
    • Measure the temperature of the cavity – Systematically measure the temperature on all sides of the cavity after each test.
      • I haven’t done this systematically yet, but for the test runs this week, the cavity barely gets above ambient as measured by the IR sensor and by touch.
    • Fill the cavity with water and if it doesn’t leak, no air holes – Take the cavity to the sink, take out the probe and fill it with water to see if it leaks.  Leave it overnight and video tape the result in the morning.
      • I have already partially done this test before because when cleaning out the extra flux from the inside and I know the bottom half holds water with no leaks.  I am not sure about the larger flat plate I brazed onto the large end yet, but that should be easy to test and we have a sink in the lab.
  • Ion Wind
    • Make sure the voltage measured on the outside of the cavity is zero – Video tape the following – first, show the voltmeter makes a sound when it measures a contact between it’s two leads.  Next show that the Faraday cage is grounded to the grounding pin of an wall socket.  Then show, by beeping the voltmeter, that the cavity is grounded to the Faraday cage.  With the cavity grounded there is no way a voltage difference could build up across the cavity, especially not the thousands of volts DC necessary to create an Ion wind.
  • Buoyancy
    • Even if the cavity was heating, there is not enough volume to lift the cavity.  Weighing the cavity on a scale while powered will also confirm buoyancy is not the cause because the buoyancy should pull the cavity up with the larger end on top.
  • Magnetic or Static field attraction/repulsion
    • The tests are done in a Faraday with the cavity at the end of pendulum.  The cavity, to be confirmed in the Ion Wind tests, is also grounded.  The closest magnets are in the electrical motor of the fan and the magnetron, some six feet away.
  • Vacuum Test
    • I have started looking for suppliers for the pump, acrylic lid, and PVC pipe to build this:
    • It looks like I can get the pump for about $450CND shipped (a nice 5CFM model), the 2″ thick, 24″ by 24″ cell cast acrylic lid from here ($250) and in an oil and gas province, finding a short piece of large diameter PVC pipe ($70) should be easy.  I also need a 1/8″ neoprene sheet for the seal between the acrylic lid and the pipe.  The 1/2″ aluminum plate at the bottom I can get from the Metal Supermarket and then I will have to water jet cut (or laser cut) the aluminum, acrylic plate and neoprene.  I still have to figure out how to get the cavity inside to test with power…

Any more tests I should do?  Comments?

Reply to Kevin’s Email

Ahh, the consummate skeptic and exactly what we need at this point.  The peer review committee for our eventual paper, if they are any good, will want these same questions answered!

There is a relation between cable heating vs time of power source applied as seen by the animated GIFs.

There are two ways to eliminate cable heating as the source of the movement:

  • Put the cavity on a scale (I will bring one this coming week)
    • First, put the cavity on it’s side on the scale to see if the EM fields inside the cavity have any effect on the scales measurement.  This is our baseline.  We should measure nothing.
    • Second, put the cavity facing downward, i.e. on it’s narrow end, then zero out the scale and we should see a weight gain.
    • Three, put the cavity facing upward, i.e. on it’s large end, zero the scale again and we should see a weight loss
  • Run four tests with the cavity rotated 90 degrees each time, but leave the cable the way it is.
    • For the 1st 90 degree test, if the cavity moves sideways, it is the cable heating, if the cavity moves forward, toward the narrow end, it is the cavity.
    • Repeat for the other 270 degrees and we should see movement toward the narrow end for each test.

Easier then trying to cool the room.

IF the movement was caused by the microwaves, the movement would be apparent in the 1st few seconds as the steady-state condition would have been met by then (Rambabu, please correct me if i’m wrong).

Two points here and they are related to a discussion I had with Rambabu about a graph that Shawyer published:

  1. The graph is from one of Shawyer’s earliest tests, before he could run the magnetron indefinitely.  It shows that the force he measured (blue and green lines) is not produced until 10 to 12 seconds after power (red) is applied.  i.e. the resonating EM waves take about 10 to 12 seconds to build up and because our cavity is about the same size or larger, we won’t see steady state until about the same time.
  2. In our case, whatever is causing the force, is causing a very small force and we can only compare the start and ending frame to see the movement.  It takes 75 seconds for the cavity to move just millimeters.

The movement will not be apparent in the first few seconds with the low force cavity.

Due to this, it is expected for the cavity to move back slowly rather than immediately after the power to the magnetron is switched off.

This is true, and we can test this with the first three tests we did, the 30 second, 45 second and 75 second tests. I haven’t look closely, but last frame on the 30 second test should match the first frame on the 45 second test and the last frame on the 45 second test should match the first frame on the 75 second, because we did them in that order and nothing was moved between those tests. The 90 second test is an exception because the cable was touched or the camera moved because the cavity is in a different location.

To prevent the loop of the cable, the current set-up can be modified to hang the cable (and cavity) straight directly from the waveguide to N-type transition with the orientation of the transition flipped (i.e. facing downwards). Slight arrangement for the items on the shelf may be necessary.

Yes, good idea.  Another idea is that we can put all the weight of the cavity on a wire, and take tension off the cable.

See you guys on Monday,

First and Last Frame Comparisons

Before we started powering up the cavity, Kevin and I were testing the IR remote through the screen and we capture about a minute of footage which I had deleted. I went back and compared the first frame with the last frame and created an animated gif here:

The light shifting is because I am inside the cage and it shows negligible movement by the cavity.

What the comparison above shows is that the cable straightening is not the cause of the movement because over the minute that we recorded, with no power, the cavity did not move noticeably.  There were transient tremors even with the power off because when the door was opened on the Faraday cage, it would shake the apparatus.  However, we were aware of the movement and let the cavity stop moving before we initiated any tests, especially the longer ones because we also let everything cool down.

Below are before and after frame shots for our other shorter tests – below is the result from the 75 second test:


Below is the result from our 45 second test:


And our 30 second test:


It moved a shorter distance for shorter tests.  Wow!!

For fun, here is a before and after shot for our 90 second test:


The starting frame is taken just after the power is turned on and the last frame is taken just before power is turned off.

Possible cable Straightening & Other Ideas?

Another possibility is that the cable is straightening.  The cable had only been unwound that morning, having been in a circular shape during shipping and with the cavity hanging from it, it could be slowly straightening (and moving forward?).  When we run the 5-min-off, 90-sec-on, 5-min-off test in a couple days, the results will be more definitive because, by then, the cable should be in equilibrium.

Any other things that could cause such a slow uni-directional movement that you can think of?

I have started a list:

  • Coax cable straightening
  • Air currents – easy fix, enclose the shelf.
  • Coax cable heating – more thoughts on this – if we start the experiment with the coax cold, and it still moves through out the entire duration, that suggests heating is less likely to be a cause.
  • ?

Test Improvements Ideas

The next test we should run is to take a video of the cavity without any power applied over a five minute duration.  We should then initiate a 90 second run and see if the cavity moves during that time only.  After the test, we will take video for another five minutes.  If the cavity only moves during the 90 second test run, we can be safe in saying it is because of the power applied.  We also need to put a cowling on the shelf containing the fan in order to eliminate air movement.

As for heating the cable, how low can we drop the temperature in the lab?  I know there is a temperature control at the entrance and can we cool the lab down to 5c and run the tests again?

Err, did it move?

Hmm,  I was editing the video for the Curvity wiki  and when I compared the start and end positions of the cavity, I noticed something strange – the cavity is moving over a very small distance and doing it sooo slowly that when we watched it live, the movement was not apparent.

  • Download this [removed, put in Wiki] which is a high resolution version of what I put on the wiki.  I highlighted the distance moved in this clip.
  • Download this [removed, put in Wiki] which is the entire 90 second test and if you click through the entire video in three clicks with the position slider in your media player, it will make it apparent that the cavity is moving with the same speed over the entire test.

(You will need the latest version of VLC to play those clip properly.)

The problem is that it doesn’t swing back which could either be from cable stiffness or the movement was caused by the cable heating (In the direction we expect??).  The results are encouraging to say the least.

Do I now owe Kevin and Gary $60?

Ready for Next Cavity and Burnt Probe

I will be at home today writing up the results from yesterday, i.e. null.  The good news is that the new cable, for test runs under 2 minutes, just gets warm and there was no arcing, smoke or twisting.    As we figured out in the lab, what smoked was not the dielectric but the injected plastic which held the probe in place:

After taking the probe assembly apart it was apparent that the probe was held in place by plastic injected through holes (not shown) in the side of the metal mount, through the dielectric and around the probe top.  The plastic was then allowed to harden.  For our latest test runs, we just flipped the dielectric around, attached the coax and made sure the probe was pushed in far enough to seat on the N-type connector and then bolted the probe assembly into the cavity.

With the cable and probe problems solved, we can now iterate through a bunch of cavity designs to find one that moves.

Below is a picture of the video camera hooked up to an external monitor.  The camera is connected to the monitor with a $75 mini HDMI to DVI cable although the feed resolution is low, only 720×480, even though the camera is HD(??)  With the camera’s IR remote, we can start and stop the recording from outside the Faraday cage, which means we can close the cage, watch until the pendulum stops moving and then start the experiment.  We don’t have to enter the cage, except for temperature measurements.

The money on the table was part of an “asymmetrical bet” I made with Kevin and Gary – if the cavity moved I would pay them $60, but if it did not, they owed me 50c.  I made 50 cents!