-
#1860 by Rodal on 02 May, 2016 16:49
-
...
NASA's, Cannae and other resonant cavity tests are still alive. McCulloch thinks his theory is still viable.
Yes I admit you were right Zen-In from the begining of this thing. That makes a fool of me to believe in such device and advance of the humanity. Guess I need to chill for a while.
So much hope destroyed.. lets keep to exploding engines.. Just hope it will not cost us of too much lives of astronauts
Zen-In has a propellant-less patented design that is supported by conventional physics (no argument) rather than anything esoteric. Many other concepts are viable: solar sails, electromagnetic tethers, etc etc.
MIT, CalTech, Stanford, NASA, Boeing, US Air Force have many other viable projects. Cheer up
-
#1861 by SeeShells on 02 May, 2016 16:50
-
Yes Dr. Rodal and I'm including it in the test bed build but a step further than was suggested testing both teeter-totter and a torsional pendulum on the same fixture. That is very doable....I couldn't agree more Dave. Small shop testing has it's limits with testing, budgets and cannot be expected to do level testing in vacuum unless funded to be able to do so. All of the builders here have pushed their test beds and drive designs far past where we were even 6 months ago.
We also have Mr. Li (TellMeAgain) as a builder that proved that spending $7 on a piano wire, he put together a test using a torsional pendulum to address Lorentz forces. Much better design that a Teeter_Totter device, as has been known in Aerospace R&D for micro-thrusters over 50 years. And as has been known for hundreds of years ever since Cavendish tests to measure G.
Back to work for me.
Shell
Torsional Pendulum --> much better instrument to measure micro-thrust for spaceflight applications than a teeter-totter
BOTTOM LINE: the selection of a teeter-totter over a torsional pendulum was not governed by budget limitations
Shell
-
#1862 by rfmwguy on 02 May, 2016 17:24
-
Appears Mike McCullough addressed Baez's commentary today on the Plank constant: http://physicsfromtheedge.blogspot.com/2016/05/new-emdrive-result-unmissing-planck.html
-
#1863 by Tellmeagain on 02 May, 2016 17:31
-
Yes I admit you were right Zen-In from the begining of this thing. That makes a fool of me to believe in such device and advance of the humanity. Guess I need to chill for a while.
So much hope destroyed.. lets keep to exploding engines.. Just hope it will not cost us of too much lives of astronauts
I am sorry to hear that. But we still have hope, not for human beings to go to outer space, but for artificial intelligence to go to outer space for us. I feel that a great revolution is imminent, and the project to build a true AI is the very last project for human beings. Maybe you should be like me, to spend almost all spare time in that direction. -
#1864 by Rodal on 02 May, 2016 17:38
-
Appears Mike McCullough addressed Baez's commentary today on the Plank constant: http://physicsfromtheedge.blogspot.com/2016/05/new-emdrive-result-unmissing-planck.html
Thanks for making us aware of it.
Not technically addressed. It only briefly states at the end:QuoteAlso, I gather there has been a frantic search for Planck's constant in the MiHsC formula. Planck's constant is implied in the formula, but the physical fact that the photons are resonating (tuned to fit) within the cavity means that it is possible to re-express h using E, c and L (the length of the cavity).
Since McCulloch himself admits that McCulloch equation is only a rough approximation, and since nobody (to my knowledge) has tackled the mathematical solution to the cavity for Unruh radiation Rindler horizon, the problem raised by Baez remains technically open.
Other points raised by Baez are not addressed either (yet?, perhaps McCulloch will address them given more time?). I imagine that McCulloch must be busy and he only wanted to briefly address this with a physical argument for the time being. -
#1865 by zen-in on 02 May, 2016 17:48
-
It is to her credit that Yang has disclosed these new experiments. The 20 experimental results shown in the graph don't look to me like convincing argument for any force. I don't see a need for fitting a curve to these data. If all 20 force measurements are averaged the result is -74 microNewtons. If the outliers are thrown out first (any force measurement greater than .4 mN or less than -.4 mN) the average becomes -30 microNewtons. While these calculations are just estimates themselves they do show a trend approaching zero.
Yes I admit you were right Zen-In from the begining of this thing. That makes a fool of me to believe in such device and advance of the humanity. Guess I need to chill for a while.
So much hope destroyed.. lets keep to exploding engines.. Just hope it will not cost us of too much lives of astronauts
I think we are a long way from replacing rockets to escape Earth's gravity well and anyone who claimed the em-drive would do that was unrealistic. -
#1866 by rfmwguy on 02 May, 2016 17:49
-
I'd suggest reflecting on the fact that Yang reported a very high number in comparison to other examples and not in line with Prof McCullough's MiHsC theory. Her test stand had high uncertainty below 3mN. It incorporated a laser, mirror and ruler measurement system taken from a mid-platform point rather than far end*...far from ideal. She used a waveguide aperture, but it does not appear it was on the frustum cavity wall, but at a mid-point spot in the waveguide. I would speculate this distorts the resonance modes just like the over-coupled waveguide in the Dresden experiments. She also used 3 wires instead of 1 torsion wire.It was my first reaction as well, however once some of the papers content began to appear, I realized their experiment had set a very high goal...3 mN minimum reolution/accuracy and a little over 200 watts RF. This is far above other tests. My own test observations were about 0.1 mN at about 900 watts. Granted, it was with a low Q device, not tuned for perfect resonance.
No, not 3 mN, please read their report and posts from TellMeAgain, look at what they measured in 20 tests with batteries, 0.7mN (80%) @ 220Watts = 3.18 mN/kW (the force is 0.7 mN (80%)!!!! ):
That being said, 3 mN is a robust force to measure at 220W. Wish their torsion wire test stand could resolve 100 μN, they might have observed something. Note per the emdrive.wiki page, they were claiming 160 and 170 mN last time. Probably why they felt 3 mN was sufficient to validate.
(snip)
As all readers can see from the plot, she actually measured less than 0.7 mN.
You appear to me misinterpreting the meaning of her statistical analysis !!! .
You are confusing measured force in an experiment with confidence estimates given uncertainty.
It is to her credit that Yang has disclosed these new experiments. The 20 experimental results shown in the graph don't look to me like convincing argument for any force. I don't see a need for fitting a curve to these data. If all 20 force measurements are averaged the result is -74 microNewtons. If the outliers are thrown out first (any force measurement greater than .4 mN or less than -.4 mN) the average becomes -30 microNewtons. While these calculations are just estimates themselves they do show a trend approaching zero.
Yes I admit you were right Zen-In from the begining of this thing. That makes a fool of me to believe in such device and advance of the humanity. Guess I need to chill for a while.
So much hope destroyed.. lets keep to exploding engines.. Just hope it will not cost us of too much lives of astronauts
IMHO, these 2014 tests simply nullified the hundreds of mN results she claimed earlier. It would be disingenuous to infer this nullifies all emdrive projects...only the Chinese one. BTW, the more I learn about the test article and method, the more surprised I am that someone did not design it for greater resolution.
* A $400 laser displacement sensor could have reduced the uncertainty greatly.
p.s. Sometimes you can find a good deal like this on ebay: http://tinyurl.com/zmmcsad
Edit - Guess a drawing will be best to illustrate the shortcoming of the Yang test stand:
Angle A is formed by rotation of test platform on the torsion wire. Assume the adjacent side is the side of the platform. Yang's team put the laser pointer mirror about halfway (best I can tell from drawings posted here) along the length, so the displacement of the platform is far less than it could have been if they simply located it at the end, making actual displacement equal to the entire opposite side length. Remember, the angle remains fixed, that's rotational displacement, the opposite side of a right triangle varies greatly depending on where the measurement point along the adjacent side is taken. Sorry if this is a little basic...all the more surprising Yang's test did not select a further measurement point. -
#1867 by Rodal on 02 May, 2016 18:08
-
Dave,It was my first reaction as well, however once some of the papers content began to appear, I realized their experiment had set a very high goal...3 mN minimum reolution/accuracy and a little over 200 watts RF. This is far above other tests. My own test observations were about 0.1 mN at about 900 watts. Granted, it was with a low Q device, not tuned for perfect resonance.
No, not 3 mN, please read their report and posts from TellMeAgain, look at what they measured in 20 tests with batteries, 0.7mN (80%) @ 220Watts = 3.18 mN/kW (the force is 0.7 mN (80%)!!!! ):
That being said, 3 mN is a robust force to measure at 220W. Wish their torsion wire test stand could resolve 100 μN, they might have observed something. Note per the emdrive.wiki page, they were claiming 160 and 170 mN last time. Probably why they felt 3 mN was sufficient to validate.
....
As all readers can see from the plot, she actually measured less than 0.7 mN.
You appear to me misinterpreting the meaning of her statistical analysis !!! .
You are confusing measured force in an experiment with confidence estimates given uncertainty.
in your most recent message (http://forum.nasaspaceflight.com/index.php?topic=39772.msg1527511#msg1527511) you are persisting with this statement:QuoteHer test stand had high uncertainty below 3mN
which is questionable when comparing her results to other test stands. Including yours. I think that you are using different statistical methodology to compare both test stands (actually, how many separate tests did you perform in your stand with the EM Drive to characterize it ?). An independent separate statistical test.
Now, at the time that you insisted on placing in the Experimental Results wiki, your experimental results as being about 0.1 mN at about 900 watts, I disagreed with the statistical analysis you used to justify this. We agreed to disagree. I agreed to stop curating the EM Drive Experimental wiki because I did not see it worthwhile to argue about DIY builders being able to report whatever number they saw fit to report in the Experimental Wiki.
However, now that Prof. Yang has reported her latest data, it is important to include it in the Experimental Wiki.
Your statements about her not being able to measure less than 3 mN in her test stand don't make any statistical sense, particularly when using the methodology that you used to come up with 0.1 mN in your test results:
Before placing Prof. Yang's results in the wiki (which are much closer to zero, by any objective observation of the data, as correctly observed by Zen-In) (*) I wanted to give you an opportunity to do a correct analysis: use the same methodology that you used to come up with 0.1 mN for your tests, on the reported 20 tests by Yang, and please report what is your assessment using the same methodology you used (apples and apples). If you don't know how to perform such statistical analysis and you have to rely on somebody else, please kindly ask him to do that for Yang's data.
If you are not interested, or unable to use the same statistical analysis to analyze Yang's data as you used to analyze your own data, please let us know so that a) I will update the wiki with my own estimate and b) we know that when you are comparing 0.1 mN for your data to Yang's data you acknowledge that you are not using the same statistical methodology to compare your experiments to her experiments.
__________
(*) Fine with me if DIY builders want to use their own numbers and assessments to report their own numbers, but when it comes to reporting other's data as Prof. Yang's we better come up with a mutually agreed methodology. -
#1868 by rfmwguy on 02 May, 2016 18:23
-
Glenn will have to answer the statistical analysis questions and I'd bet he'd gladly run the analysis if he had Yang's data runs. He's patiently been waiting on the sidelines for something to do
Dave,It was my first reaction as well, however once some of the papers content began to appear, I realized their experiment had set a very high goal...3 mN minimum reolution/accuracy and a little over 200 watts RF. This is far above other tests. My own test observations were about 0.1 mN at about 900 watts. Granted, it was with a low Q device, not tuned for perfect resonance.
No, not 3 mN, please read their report and posts from TellMeAgain, look at what they measured in 20 tests with batteries, 0.7mN (80%) @ 220Watts = 3.18 mN/kW (the force is 0.7 mN (80%)!!!! ):
That being said, 3 mN is a robust force to measure at 220W. Wish their torsion wire test stand could resolve 100 μN, they might have observed something. Note per the emdrive.wiki page, they were claiming 160 and 170 mN last time. Probably why they felt 3 mN was sufficient to validate.
....
As all readers can see from the plot, she actually measured less than 0.7 mN.
You appear to me misinterpreting the meaning of her statistical analysis !!! .
You are confusing measured force in an experiment with confidence estimates given uncertainty.
in your most recent message (http://forum.nasaspaceflight.com/index.php?topic=39772.msg1527511#msg1527511) you are persisting with this statement:QuoteHer test stand had high uncertainty below 3mN
which is statistically incorrect when comparing her results to other test stands. Including yours.
Now, at the time that you insisted on placing in the Experimental Results wiki, your experimental results as being about 0.1 mN at about 900 watts, I disagreed with the statistical analysis you used to justify this. We agreed to disagree. I agreed to stop curating the EM Drive Experimental wiki because I did not see it worthwhile to argue about DIY builders being able to report whatever number they saw fit to report in the Experimental Wiki.
However, now that Prof. Yang has reported her latest data, it is important to include it in the Experimental Wiki.
Your statements about her not being able to measure less than 3 mN in her test stand don't make any statistical sense, particularly when using the methodology that you used to come up with 0.1 mN in your test results:
Before placing Prof. Yang's results in the wiki (which are much closer to zero, by any objective observation of the data, as correctly observed by Zen-In) (*) I wanted to give you an opportunity to do a correct analysis: use the same methodology that you used to come up with 0.1 mN for your tests, on the reported 20 tests by Yang, and please report what is your assessment using the same methodology you used (apples and apples). If you don't know how to perform such statistical analysis and you have to rely on somebody else, please kindly ask him to do that for Yang's data.
If you are not interested, or unable to use the same statistical analysis to analyze Yang's data as you used to analyze your own data, please let us know so that a) I will update the wiki with my own estimate and b) we know that when you are comparing 0.1 mN for your data to Yang's data you acknowledge that you are not using the same statistical methodology to compare your experiments to her experiments.
__________
(*) Fine with me if DIY builders want to use their own numbers and assessments to report their own numbers, but when it comes to reporting other's data as Prof. Yang's we better come up with a mutually agreed methodology.
-
#1869 by Rodal on 02 May, 2016 18:27
-
...
Don't understand why you write "if he had Yang's data runs".
Glenn will have to answer the statistical analysis questions and I'd bet he'd gladly run the analysis if he had Yang's data runs. He's patiently been waiting on the sidelines for something to do
What have you been basing your 3 mN on ?
Please tell Glenn to use Yang's data as given in the above chart and in the rest of her paper . Her data is given in her paper. That's what he has to work with! Is the kind of data and number of data different than yours?
Of course. Admitting this is part of getting clarity here. Again what have you been basing your characterization of Yang data on?
Since, again I don't agree with the way you are characterizing Yang's data. I'll wait to see what Glenn has to say.
I would appreciate a rough idea of how long will it take for Glenn to assess Yang's data (a week ?) -
#1870 by Rodal on 02 May, 2016 18:35
-
...
When you have a chance I would appreciate an answer to this:
Glenn will have to answer the statistical analysis questions and I'd bet he'd gladly run the analysis if he had Yang's data runs. He's patiently been waiting on the sidelines for something to do
How many separate independent tests did you perform in your stand with your EM Drive?
I recall very few NSF-1701 tests, is my memory incorrect?
I see very few tests here:
https://plus.google.com/+NSF1701ElectromagneticEngineExperiment-rfmwguy/videos
Is this the correct place where the tests are stored? -
#1871 by jmossman on 02 May, 2016 19:19
-
...
NASA's, Cannae and other resonant cavity tests are still alive. McCulloch thinks his theory is still viable.
Yes I admit you were right Zen-In from the begining of this thing. That makes a fool of me to believe in such device and advance of the humanity. Guess I need to chill for a while.
So much hope destroyed.. lets keep to exploding engines.. Just hope it will not cost us of too much lives of astronauts
Zen-In has a propellant-less patented design that is supported by conventional physics (no argument) rather than anything esoteric. Many other concepts are viable: solar sails, electromagnetic tethers, etc etc.
MIT, CalTech, Stanford, NASA, Boeing, US Air Force have many other viable projects. Cheer up
Wasn't there a discussion/debate a few posts back about whether swapping a noisy RF magnetron source to a stable single frequency RF source (without dielectric) might result in a null test?
http://forum.nasaspaceflight.com/index.php?topic=39772.msg1524378#msg1524378
In fact, I believe there was a rather prophetic quote worth repeating:
"Cleaning" the input to a nonlinear process may ensure that the outcome is null or much reduced.
So, one possible outcome of "filtering the RF souce" is ensuring that the experiment will result in lower force/PowerInput or not a significant level of force.
Seems like that debate might have been put to rest with actual experimental results.
Unfortunately, I don't think we know the frustum geometry, Q, or mode shape in Yang's latest experiment... not sure how much her latest experiment narrows the remaining search space for those intrepid DIYers still willing to chase the dream. -
#1872 by rfmwguy on 02 May, 2016 19:26
-
From Mr. Li's low uncertainty (14%) at +/- 3mN here: http://forum.nasaspaceflight.com/index.php?topic=39772.msg1527132#msg1527132...
Don't understand why you write "if he had Yang's data runs".
Glenn will have to answer the statistical analysis questions and I'd bet he'd gladly run the analysis if he had Yang's data runs. He's patiently been waiting on the sidelines for something to do
What have you been basing your 3 mN on ?
Please tell Glenn to use Yang's data as given in the above chart and in the rest of her paper . Her data is given in her paper. That's what he has to work with! Is the kind of data and number of data different than yours?
Of course. Admitting this is part of getting clarity here. Again what have you been basing your characterization of Yang data on?
Since, again I don't agree with the way you are characterizing Yang's data. I'll wait to see what Glenn has to say.
I would appreciate a rough idea of how long will it take for Glenn to assess Yang's data (a week ?) -
#1873 by Rodal on 02 May, 2016 19:32
-
...
From Mr. Li's low uncertainty (14%) at +/- 3mN here: http://forum.nasaspaceflight.com/index.php?topic=39772.msg1527132#msg1527132
I also posted a note in McCulloch's blog http://physicsfromtheedge.blogspot.com/2016/05/new-emdrive-result-unmissing-planck.html.
People should look at the actual data, as done by Zen-In, since Yang's statistical statement is not being properly interpreted.
Take a look at your data, for comparison: I am able to find very few independent experiments. Only very few independent NSF-1701 experiments with the EM Drive "thrusting".
NSF-1701 Flight tests: A, B, C and D ?
You only conducted 4 independent tests ? (and is it less than that? were all 4 tests run under the same nominal parameters ?)
I'm going to need your help with your tests, as I was not active in the thread at that time -
#1874 by rfmwguy on 02 May, 2016 20:04
-
No offense, but I won't have time to go over the old testing again other than to say Glenn has my spreadsheets, which include tests done without the DAC. These were with a laser pointer, low resolution and analyzed optically. My test report has more info here: http://forum.nasaspaceflight.com/index.php?action=dlattach;topic=38577.0;attach=1072184...
From Mr. Li's low uncertainty (14%) at +/- 3mN here: http://forum.nasaspaceflight.com/index.php?topic=39772.msg1527132#msg1527132
I also posted a note in McCulloch's blog http://physicsfromtheedge.blogspot.com/2016/05/new-emdrive-result-unmissing-planck.html.
People should look at the actual data, as done by Zen-In, since Yang's statistical statement is not being properly interpreted.
Take a look at your data, for comparison: I am able to find very few independent experiments. Only very few independent NSF-1701 experiments with the EM Drive "thrusting".
NSF-1701 Flight tests: A, B, C and D ?
You only conducted 4 independent tests ? (and is it less than that? where all 4 tests run under the same nominal parameters ?)
I'm going to need your help with your tests, as I was not active in the thread at that time
The videos are chronological and contain real-time information about the Flight Tests and the progression of the test stand. Tried to be as comprehensive as possible to avoid having to repeat myself
I recall 4 "flight test" sessions on the beam under power.
-
#1875 by Rodal on 02 May, 2016 20:22
-
No offense either
No offense, but I won't have time to go over the old testing again other than to say Glenn has my spreadsheets, which include tests done without the DAC. These were with a laser pointer, low resolution and analyzed optically. My test report has more info here: http://forum.nasaspaceflight.com/index.php?action=dlattach;topic=38577.0;attach=1072184...
From Mr. Li's low uncertainty (14%) at +/- 3mN here: http://forum.nasaspaceflight.com/index.php?topic=39772.msg1527132#msg1527132
I also posted a note in McCulloch's blog http://physicsfromtheedge.blogspot.com/2016/05/new-emdrive-result-unmissing-planck.html.
People should look at the actual data, as done by Zen-In, since Yang's statistical statement is not being properly interpreted.
Take a look at your data, for comparison: I am able to find very few independent experiments. Only very few independent NSF-1701 experiments with the EM Drive "thrusting".
NSF-1701 Flight tests: A, B, C and D ?
You only conducted 4 independent tests ? (and is it less than that? were all 4 tests run under the same nominal parameters ?)
I'm going to need your help with your tests, as I was not active in the thread at that time
The videos are chronological and contain real-time information about the Flight Tests and the progression of the test stand. Tried to be as comprehensive as possible to avoid having to repeat myself I recall 4 "flight test" sessions on the beam under power.
But since you confirm that you had at most 4 independent tests (and I have to check whether it is less than that, if you did not keep all parameters identical, the number is even less than that) you just confirmed my worst fears: having at most 4 independent test I can hardly see how can you compare your tests with Prof. Yang's when she had 20 tests keeping all parameters the same.
You are using different standards for your testing than for Prof. Yang (*)
It looks like you do not have statistical basis of comparison to make statements comparing your tests with Prof. Yang's tests.
Therefore when discussing Prof. Yang's tests you may address them for example as done by Zen-In, instead of using her methodology, because if you would use her methodology (given your extremely small number of tests) there is not much that can be gathered about what Yang's translation calls "uncertainty" from your small number of tests (unless you make some strong assumptions about what would have happened if you would have made 20 independent NSF-1701 flight tests keeping all parameters the same, as she did).
____________
(*) I have no problem with that, several people said that DIY builders should not be held to the same standards as Prof. Yang, but then fine, you should not compare your "results" with Prof. Yang's statistics, because it is like comparing apples and oranges, let's be clear (when you attempted to compare your 0.1 mN with her 3 mN )
-
#1876 by Rodal on 02 May, 2016 20:40
-
Going over Dave's data:Quote
Flight Test #1 provided a null result with the magnetron mounted on the small plate (beneath
the frustum) with its power feed positioned 1/4 wave distance from the edge. The test stand
configuration did not use the laser displacement sensor, but a simple laser pointer and mirror to provide
approximately 28 feet of laser beam travel to the target. The initial testing also used Galinstan liquid
metal in copper cups for power connection to the frustum.
Flight Test #2 moved the magnetron to the center of the large diameter plate, on top of the
frustum. No other changes to the test stand were made. The results were inconclusive.
Flight Test #2A was a longer duration with the same configuration as Test #2 in an attempt to
improve the laser targeting with the laser pointer. It was also inconclusive.*
*After Flight Test #2A, a decision was made to improve the test stand by:
1) Removing the Galinstan, which added several milligrams of “drag” against electrode
movement and unwanted vibrations transmitted along the support and electrode copper
wires when controller keypad entries were made.
2) Changing from the laser pointer/spot display to a laser displacement sensor and DAQ
system (inclusive of the LDS, capable of logging and displaying beam displacement).
First Release 10/4/15 Page 4
Flight Test #2B contained all aforementioned improvements and the Final Bill of Materials. At
the time, only a video of the computer monitor could be recorded, as additional data-logging software
had not yet been integrated and screen recording was not possible with the slower PC. It also became
apparent that any natural thermal lift caused by the heating of the magnetron must be filtered out of the
recorded data to show any potential thrust. The magnetron heat sink temperature typically increases to
between 170 and 200 degrees C. A non-null test was deemed likely by myself for the first time but I
wanted to conduct further flight tests before reaching a conclusion.
Flight Test #2C had the same test stand equipment, hardware and software. The magnetron ON
cycle duration was varied from 20, 30, 50 and 70% power on cycles in order to differentiate frustum
movement from thermal lift and natural mechanical oscillations. 50% power cycle was deemed best to
avoid natural beam oscillations and fast thermal lift. Another non-null test was deemed likely by myself
but I desired one additional flight test to complete Phase I testing.
Flight Test #2D was a 50% power cycle test run in two separate 10 minute increments with an
approximate 10 minute delay in between. New data-logging software was installed and the test
provided over 2,700 data points per channel at a rate of about 75 samples per minute. The video was
simply to show the computer time stamp and allow data synch with magnetron ON/OFF time via the
audio track. This permitted insertion of a data set denoting the magnetron power state. The LDS was on
channel 1, the other channels were open (unloaded) which permitted an analysis of system noise. The
collected data was analyzed by a professional data analyst* using advanced algorithms. It was his
conclusion that with a probability of greater than .95, there was an anomoly causing the data
(displacement) to be distinctly different during ON cycles versus OFF cycles8-14. This professionally
confirms the visual changes I witnessed, which included displacement opposite of thermal lift, holding
steady against lift, and the attenuation of thermal lift while the magnetron was in the ON cycle. This
was the most rigorous review of any of the other Flight Tests.
Tentative conclusion (again, I was not active in the threads at the time, and this is just from rapid reading the above)
From the above, it looks like you run at most only 2 (two) tests (NSF-1701 tests) keeping all parameters the same.
This is in contrast to Prof. Yang conducting 20 tests keeping all parameters the same.
I can see that (going from memory) that Glenn made a valiant effort at analyzing the final test for what was going on during power on and power off and power on again, but this was all during a single test. Seem like different testing methodology as to what Yang refers to as 20 separate tests ( where it is not just a question of examining magnetron going on and off as in your last test, after all, Yang used a magnetron as well, in each of her separate tests)
Perhaps Glenn will have another viewpoint (although I seem to recall that he also wanted to have more tests) or perhaps can find a better way to analyze Prof. Yang's?
-
#1877 by rfmwguy on 02 May, 2016 20:45
-
Going over Dave's data:
Wrong, the several thousand data points show a series of powerups per session. Each session had two primary on/off cycles. Don't recall the sub-cycles for power up. Depends on how you want to split hairs on terminology.QuoteFlight Test #1 provided a null result with the magnetron mounted on the small plate (beneath
the frustum) with its power feed positioned 1/4 wave distance from the edge. The test stand
configuration did not use the laser displacement sensor, but a simple laser pointer and mirror to provide
approximately 28 feet of laser beam travel to the target. The initial testing also used Galinstan liquid
metal in copper cups for power connection to the frustum.
Flight Test #2 moved the magnetron to the center of the large diameter plate, on top of the
frustum. No other changes to the test stand were made. The results were inconclusive.
Flight Test #2A was a longer duration with the same configuration as Test #2 in an attempt to
improve the laser targeting with the laser pointer. It was also inconclusive.*
*After Flight Test #2A, a decision was made to improve the test stand by:
1) Removing the Galinstan, which added several milligrams of “drag” against electrode
movement and unwanted vibrations transmitted along the support and electrode copper
wires when controller keypad entries were made.
2) Changing from the laser pointer/spot display to a laser displacement sensor and DAQ
system (inclusive of the LDS, capable of logging and displaying beam displacement).
First Release 10/4/15 Page 4
Flight Test #2B contained all aforementioned improvements and the Final Bill of Materials. At
the time, only a video of the computer monitor could be recorded, as additional data-logging software
had not yet been integrated and screen recording was not possible with the slower PC. It also became
apparent that any natural thermal lift caused by the heating of the magnetron must be filtered out of the
recorded data to show any potential thrust. The magnetron heat sink temperature typically increases to
between 170 and 200 degrees C. A non-null test was deemed likely by myself for the first time but I
wanted to conduct further flight tests before reaching a conclusion.
Flight Test #2C had the same test stand equipment, hardware and software. The magnetron ON
cycle duration was varied from 20, 30, 50 and 70% power on cycles in order to differentiate frustum
movement from thermal lift and natural mechanical oscillations. 50% power cycle was deemed best to
avoid natural beam oscillations and fast thermal lift. Another non-null test was deemed likely by myself
but I desired one additional flight test to complete Phase I testing.
Flight Test #2D was a 50% power cycle test run in two separate 10 minute increments with an
approximate 10 minute delay in between. New data-logging software was installed and the test
provided over 2,700 data points per channel at a rate of about 75 samples per minute. The video was
simply to show the computer time stamp and allow data synch with magnetron ON/OFF time via the
audio track. This permitted insertion of a data set denoting the magnetron power state. The LDS was on
channel 1, the other channels were open (unloaded) which permitted an analysis of system noise. The
collected data was analyzed by a professional data analyst* using advanced algorithms. It was his
conclusion that with a probability of greater than .95, there was an anomoly causing the data
(displacement) to be distinctly different during ON cycles versus OFF cycles8-14. This professionally
confirms the visual changes I witnessed, which included displacement opposite of thermal lift, holding
steady against lift, and the attenuation of thermal lift while the magnetron was in the ON cycle. This
was the most rigorous review of any of the other Flight Tests.
From the above, it looks like you run at most only 2 (two) tests (NSF-1701 tests) keeping all parameters the same.
This is in contrast to Prof. Yang conducting 20 tests keeping all parameters the same. -
#1878 by Rodal on 02 May, 2016 20:51
-
....
No, you cannot compare like that. Prof Yang also used a magnetron, and her magnetron might have been firing on and off during each of her 20 tests. You don't have access to her magnetron time history. You cannot compare thousands of times of your magnetron firing on and off during A SINGLE TEST of yours, with Yang's 20 separate tests.
Wrong, the several thousand data points show a series of powerups per session. Each session had two primary on/off cycles. Don't recall the sub-cycles for power up. Depends on how you want to split hairs on terminology.
It seems to me like Glenn might have assumed an ergodic process
https://en.wikipedia.org/wiki/Ergodicity
Fine go ahead assume that your process is ergodic, but your comparison of 0.1 mN for you with 3 mN of Yang is comparing different things. -
#1879 by SeeShells on 02 May, 2016 20:54
-
Dr. Rodal, Yang used a magnetron for her last 20 tests? I thought she used a solid state. Don't have time to dig so could you clarify for me?....
No, you cannot compare like that. Prof Yang also used a magnetron, and her magnetron might have been firing on and off during each of her 20 tests. You don't have access to her magnetron time history. You cannot compare thousands of times of your magnetron firing on and off during A SINGLE TEST of yours, with Yang's 20 separate tests.
Wrong, the several thousand data points show a series of powerups per session. Each session had two primary on/off cycles. Don't recall the sub-cycles for power up. Depends on how you want to split hairs on terminology.
It seems to me like Glenn might have assumed an ergodic process
Shell
I recall 4 "flight test" sessions on the beam under power.