-
#1200
by
A_M_Swallow
on 27 Aug, 2015 02:48
-
Great work rfmwguy!
Here is a montage of how the region of interest looks throughout the whole movie (stretched in Y to make things clearer).
The microwave oven has a timer on it so it will be drawing mains power through a transformer even when the magnetron is not transmitting. Transformers give off a magnetic field. Was the oven powered up when the weight was used to calibrate the laser?
Alternative possibility. Static electricity can build up slowly.
-
#1201
by
cosmo
on 27 Aug, 2015 02:54
-
@rfmwguy - Newbie poster here. Followed all the threads and have a suggestion. From an experimental standpoint, it would be useful to collect multiple data sets (in your case, the 10 min video of the laser dot position). Put simply, run the test 3 separate times, with a delay (an hour or two) between runs. Then post the 3 videos. This would allow those evaluating the data to see if there are repeatable patterns. Once a particular configuration has been torn down and reconfigured, it's non-trivial to recreate it exactly for verification purposes.
Please keep up the excellent work!
Kurt
Thanks, my first view of the vid appears to show something happening that wasn't there yesterday. If so, you bet I will start tweaking the setup, such as a new mirror, camera, target and who knows what else besides a longer static recording time. If movement confirmed, I'll not modify NSF-1701 from its current configuration.
My thinking is that it would be very useful to compare 3 sets of the video data reduced as @Croppa did here:
http://forum.nasaspaceflight.com/index.php?topic=38203.msg1419750#msg1419750If they don't correlate well for the same experimental configuration, then something is suspect.
Kurt
-
#1202
by
deltaMass
on 27 Aug, 2015 03:04
-
After the oscillations from starting the tests settle down, the laser is pointing lower on the target. I guess that would be the new 'zero' for each test. From there, I don't see any movement.
Of course, my technique of placing sticky notes on my monitor might not be the most accurate method. 
It's time for the data analysis group to zoom in and examine it frame by frame.
Indeed, also my observation. It's as if the floobiestick is permanently bent after a given trial. The obvious response to this is to call for
more detailed calibration. There seems to be excess "stickiness" in the system.
Deflection direction is the same as in test #1 (downward laser, upward frustum, therefore big-end-forward). There also seems to be a greater deflection with this new magnetron mount.
-
#1203
by
cee
on 27 Aug, 2015 03:27
-
Please include temps on big and little end walls as well as side walls, see if you can can a noticeable temp differential along the side and end walls.
-
#1204
by
rfmwguy
on 27 Aug, 2015 03:32
-
After the oscillations from starting the tests settle down, the laser is pointing lower on the target. I guess that would be the new 'zero' for each test. From there, I don't see any movement.
Of course, my technique of placing sticky notes on my monitor might not be the most accurate method. :)
It's time for the data analysis group to zoom in and examine it frame by frame.
Indeed, also my observation. It's as if the floobiestick is permanently bent after a given trial. The obvious response to this is to call for more detailed calibration. There seems to be excess "stickiness" in the system.
Deflection direction is the same as in test #1 (downward laser, upward frustum, therefore big-end-forward). There also seems to be a greater deflection with this new magnetron mount.
Once I see the video from croppa and others, I'll feel more confident in this configuration showing more movement. Only a guess, but the slow tracking downward is thermal lift even with mesh. If it tracks the same as yesterday, we have an answer. However, I did notice more movement overall this time.
-
#1205
by
rfmwguy
on 27 Aug, 2015 03:47
-
Great work rfmwguy!
Here is a montage of how the region of interest looks throughout the whole movie (stretched in Y to make things clearer).
The microwave oven has a timer on it so it will be drawing mains power through a transformer even when the magnetron is not transmitting. Transformers give off a magnetic field. Was the oven powered up when the weight was used to calibrate the laser?
Alternative possibility. Static electricity can build up slowly.
Fields are small within the closed oven for the display, etc. Would be surprised it it impacted the small amount of magnetic material on the test stand. Static is possible, but we are over 50% RH so not likely to have an impact...all imo only.
-
#1206
by
rfmwguy
on 27 Aug, 2015 03:49
-
Please include temps on big and little end walls as well as side walls, see if you can can a noticeable temp differential along the side and end walls.
Only temp over ambient was on board where mag was mounted. Sides and bottom were near ambient in static thermal tests.
-
#1207
by
WarpTech
on 27 Aug, 2015 04:02
-
After the oscillations from starting the tests settle down, the laser is pointing lower on the target. I guess that would be the new 'zero' for each test. From there, I don't see any movement.
Of course, my technique of placing sticky notes on my monitor might not be the most accurate method.
It's time for the data analysis group to zoom in and examine it frame by frame.
Indeed, also my observation. It's as if the floobiestick is permanently bent after a given trial. The obvious response to this is to call for more detailed calibration. There seems to be excess "stickiness" in the system.
Deflection direction is the same as in test #1 (downward laser, upward frustum, therefore big-end-forward). There also seems to be a greater deflection with this new magnetron mount.
I just had a thought. In the past, I have weighed a HV capacitor charged and discharged, on a balance beam much smaller than this one. Regardless of polarity, when the capacitor is charged, it is heavier and the balance beam would tip.
Here, when the microwave is turned on, the frustum is charging with energy, like a capacitor. It's weight will increase due to the energy input to the system, M = E/c^2. When the frustum is charged, it's heavier. This is only a concern when using a balance beam. When using a rotary test setup, this is not an issue. However, like buoyancy, it can be subtracted off by turning it upside down but a mesh isn't going to prevent it.
Todd
-
#1208
by
TheTraveller
on 27 Aug, 2015 04:35
-
Yet, each measure Q differently...see my point?
As far as I know all EMDrive experimental data Q quotes are based on measured S11 rtn loss max dB then 3dB down bandwidth. While Prof Yang did theory work with a Q equation, she measured the frustum Q as above.
-
#1209
by
Croppa
on 27 Aug, 2015 05:12
-
Great job again rfmwguy
I was actually still messing around with the file from yesterday. I took the audio from that movie and made a crude waveform with some free online software. I've highlighted the runs in green and you can see the periodical on/off of the magnetron within those runs.
Also here is the new run, treated exactly the same as before (0.25s/px). It's not likely I'll have chance to put landmarks in that one anytime soon but you can get a fair idea of time points from the laser gaps.
-
#1210
by
deltaMass
on 27 Aug, 2015 05:28
-
I just had a thought. In the past, I have weighed a HV capacitor charged and discharged, on a balance beam much smaller than this one. Regardless of polarity, when the capacitor is charged, it is heavier and the balance beam would tip.
Here, when the microwave is turned on, the frustum is charging with energy, like a capacitor. It's weight will increase due to the energy input to the system, M = E/c^2. When the frustum is charged, it's heavier. This is only a concern when using a balance beam. When using a rotary test setup, this is not an issue. However, like buoyancy, it can be subtracted off by turning it upside down but a mesh isn't going to prevent it.
Todd
Sorry, but I'm calling nonsense on that lot, as a cursory calculation shows. Take a 2 mF capacitor charged to 10 kV. That's 100 kJ or a mass equivalent of
1 nanogram.. The fact that you did not have that precision but think you did illustrates how difficult is experimental physics, and how easily fooled even smart people can be.
-
#1211
by
ThinkerX
on 27 Aug, 2015 05:46
-
I wonder...
...didn't the Romanian report a similar effect in one of his tests? A very slight 'bump' in an otherwise null result?
So...is this a bug? Or a clue?
-
#1212
by
kwertyops
on 27 Aug, 2015 05:57
-
Attached is my take on flight #2, audio aligned.
This is a 1 slice / second.
Some interesting oscillations are visible. The bump downwards was not during a run of the magnetron.
Not much going on during the low-power runs.
The most interesting thing I see is at the very beginning of the 100% test, there appears to be a slight initial bump upwards (the frustum moving downwards) that then slowly returns to baseline.
[Edit] Stretched the y-axis by 400%
-
#1213
by
WarpTech
on 27 Aug, 2015 06:17
-
I just had a thought. In the past, I have weighed a HV capacitor charged and discharged, on a balance beam much smaller than this one. Regardless of polarity, when the capacitor is charged, it is heavier and the balance beam would tip.
Here, when the microwave is turned on, the frustum is charging with energy, like a capacitor. It's weight will increase due to the energy input to the system, M = E/c^2. When the frustum is charged, it's heavier. This is only a concern when using a balance beam. When using a rotary test setup, this is not an issue. However, like buoyancy, it can be subtracted off by turning it upside down but a mesh isn't going to prevent it.
Todd
Sorry, but I'm calling nonsense on that lot, as a cursory calculation shows. Take a 2 mF capacitor charged to 10 kV. That's 100 kJ or a mass equivalent of 1 nanogram.. The fact that you did not have that precision but think you did illustrates how difficult is experimental physics, and how easily fooled even smart people can be.
This was a ~2.5 mil piece of plexiglass, covered in aluminum foil, both sides, about 24" on a side. I had it balanced on a stick of balsa wood, connected to a rectified 12kV neon sign transformer. The result was repeatable. When on, the balance tipped toward the capacitor. I didn't actually measure a weight, just a deflection. It was tiny, but visible to the naked eye, and I wasn't using a laser and mirror to magnify the deflection. I just looked at the height above the table, with a ruler at the other end of the stick. The balsa wasn't more than 4' long.
Todd
-
#1214
by
deltaMass
on 27 Aug, 2015 06:37
-
I just had a thought. In the past, I have weighed a HV capacitor charged and discharged, on a balance beam much smaller than this one. Regardless of polarity, when the capacitor is charged, it is heavier and the balance beam would tip.
Here, when the microwave is turned on, the frustum is charging with energy, like a capacitor. It's weight will increase due to the energy input to the system, M = E/c^2. When the frustum is charged, it's heavier. This is only a concern when using a balance beam. When using a rotary test setup, this is not an issue. However, like buoyancy, it can be subtracted off by turning it upside down but a mesh isn't going to prevent it.
Todd
Sorry, but I'm calling nonsense on that lot, as a cursory calculation shows. Take a 2 mF capacitor charged to 10 kV. That's 100 kJ or a mass equivalent of 1 nanogram.. The fact that you did not have that precision but think you did illustrates how difficult is experimental physics, and how easily fooled even smart people can be.
This was a ~2.5 mil piece of plexiglass, covered in aluminum foil, both sides, about 24" on a side. I had it balanced on a stick of balsa wood, connected to a rectified 12kV neon sign transformer. The result was repeatable. When on, the balance tipped toward the capacitor. I didn't actually measure a weight, just a deflection. It was tiny, but visible to the naked eye, and I wasn't using a laser and mirror to magnify the deflection. I just looked at the height above the table, with a ruler at the other end of the stick. The balsa wasn't more than 4' long.
Todd
What is wrong with you? You just made it about a hundred thousand times worse!
a) You have all these dimensions but can't be bothered to calculate the capacitance or the energy or the mass equivalent (don't worry, I did it for you and you are 10
5 down on my energy example)
b) You make no reference to the calculation I provided you before, nor to the back-of-the-envelope estimate that produced as a reference (as if I wrote nothing)
c) You don't use metric (yes, I know you're a USAn but that's a poor excuse).
d) You continue to insist that you measured a weight change even though it would have been one hundred thousandth of a nanogram!!!
e) It does not occur to you that you are measuring an artifact due most likely to static electricity (despite my warnings).
Really
-
#1215
by
SteveD
on 27 Aug, 2015 07:10
-
Looks like I'm a bit late to the party. Please leave on the light so that we can see the rig next time. For some reason you're camera picked up less ambient light this time (changing its exposure settings). I object to drawing any conclusions to a light moving in a dark room where I can't see what is going on.
I see others have done the analysis better than I. The final image is way too speculative. Enough has changed in the setup that I lack confidence that the target and camera are in the same location. That said, it looks like you ended almost exactly at the same point you did last time. It might be a thermal effect but I don't think its lift from heated air.
-
#1216
by
deltaMass
on 27 Aug, 2015 08:27
-
Early days for drawing any conclusions. Of course, you try the big and obvious things first. Repositioning the magnetron was one such trial. I'd suggest another big and obvious one would be to flip the frustum over so now it's small side up. Does the thrust reverse? That is what Iulian found.
-
#1217
by
Mr. Peter
on 27 Aug, 2015 09:32
-
Early days for drawing any conclusions. Of course, you try the big and obvious things first. Repositioning the magnetron was one such trial. I'd suggest another big and obvious one would be to flip the frustum over so now it's small side up. Does the thrust reverse? That is what Iulian found.
I definitely second that!
-
#1218
by
SlightPace
on 27 Aug, 2015 09:58
-
Not to be too critical, but this thread is really strange sometimes. People put in so much time, effort, and money to build test rigs, but don't bother with any kind of proper scientific conduct when testing them.
@rfmwguy: you have a nice rig, but where is your control experiment? Your hypothesis is that a resonating truncated conical cavity produces thrust, but that is not what you are testing with your experiments so far. Controls should be:
A) Non-resonating cavity, e.g. put something in the cavity which interferes with the resonance, or put the magnetron outside of the cavity (with relevant safety measures of course)
B) Non-conical cavity
To start changing other variables before having a control experiment makes no sense at all. Since you don't even know if your cavity is resonating, right now you are just testing what happens to a balance beam when you fire off a magnetron nearby.
The control experiments is the absolute first things you should do.
Again, sorry if it comes across as too critical, but it is frustrating to see a such a nice experimental setup go to waste (like Iulian).
-
#1219
by
bunjatec
on 27 Aug, 2015 11:58
-
