-
#2220 by Chrochne on 29 Sep, 2015 07:10
-
Alright, I always wanted to ask this. Do you think NASA Eagleworks were able to go above 100 micronewtons and keep it constant? and pass the testing to the Glenn Research Center?
I do not want to speculate. Just curious
.
My guess is they might have...but of course no paper yet to confirm or dismiss, but after the testing by rfmwguy I am more confident. On the other hand the "thrust" is still quite small so far.. -
#2221 by Flyby on 29 Sep, 2015 11:57
-
I don't think you'll find any insider or knowledgeable person willing to comment about that on this high profile public forum. Those who did in the past got slapped on the head by management, so unless there is an official NASA press release, all you'll hear are the crickets....
-
#2222 by SeeShells on 29 Sep, 2015 12:16
-
Alright, I always wanted to ask this. Do you think NASA Eagleworks were able to go above 100 micronewtons and keep it constant? and pass the testing to the Glenn Research Center?
I was asked the other day if Micro-newtons were just a little cookie.
I do not want to speculate. Just curious.
My guess is they might have...but of course no paper yet to confirm or dismiss, but after the testing by rfmwguy I am more confident. On the other hand the "thrust" is still quite small so far.. -
#2223 by graybeardsyseng on 29 Sep, 2015 12:19
-
Thanks for the welcome and the most appreciated comments!I would like to ask the members of the forum for some input/opinions.
Yay! Welcome to the DIY world! Fear not, as I took a lot of the naysayer hits when I started posting a few months ago, so in a sense I've absorbed a lot of the pressure for those who follow
Some of you know me from my posts on various RF and microwave issues. I am just now starting a DIY emdrive test effort I am finally getting my workshop back on line following a move and I want to try to build on the outstanding work of SeeShell, rfmwguy, and others to hopefully make some small contribution of my own.
Anyway to the request.
There are a multitude of test environment design factors of course and this is just a start but in order to begin limiting the possible test space I have narrowed down overall test configurations to the following:
1. Rotary Table configuration unit under test (UUT) thrusting either posigrade or retrograde to rotation
2. Linear (slide) configuration air or other low friction surface. UUT thrusting along the access of the slide.
3. Balance Beam (ala rfmwguy and seeshells) UUT thrusting either up or down
In my thinking all have things going for them as well as potential problems, but I would like to ask for your thoughts on each configurations. In particular these are the things I have been considering:
Overall Pros and Cons
Challenges to getting usable data output
Potential systemic and situational error sources
Data which should/can be measured for either signal or noise/error determinations
ANYTHING else you can think of
Thanks in advance. If you would prefer to reply by private message rather than on the forum that is fine with me however please indicate any information which you do not want disseminated further.
Herman
PS for those who dont know me and to provide some credibility background - I have worked in both professionally and DIY settings on RF from LF to 40+ Ghz, high power RF and RF power supplies, vacuum chambers and systems, nuclear power plants and various and sundry other technical fields including aerospace and defense for the past 39 years. Well acquainted with safety procedures - wrote some for my last company. This will be my first big retirement project and I cant wait to get started. 
I've thought long and hard about many of your questions. Rotary tables need an air source that will stir up all ambient air around the DUT. Same for a linear table. While I think horizontal measuring is best for ambient air measuring, the background artifacts can be a challenge.
Regarding vertical measurements, lift is an enemy, much more so that I would have thought, even with a wire mesh frustum. Extracting data out of the natural lift is difficult as you can see by the fine work done by data analysists here. I still think this is the way to go as it limits other mechanical and electrical variables.
Datalogging: suggest you go with a fast computer that can handle screen recording with ease. Datalogging is usually serial and even an old PC like mine handled it easily; not so with screen record. Quad core processor is a must. DAQ can be anything, but try to go with a 12 bit as a minumum. Locate several feet from frustum and power supplies.
Laser Displacement Sensors - highly recommend this for vertical measurements. Try to select a 40mm +/- 10mm range rather than a 100mm+ sensor. The closer, the better resolution...up to 7 digits. Try Omron or equivalent.
Setup in an area with no vents and cover windows for drafts. You can see the deflections as I simply approached the setup.
Have nothing else on AC lines feeding your gear that could draw a load (pretty basic advice).
Other than that, make sure you HAVE FUN. Thats really the bottom line Graybeard. We are here to help and support your efforts.
Re the rotary tables needing air sources and stirring up air around DUT yes this is definitely a concern with that approach. However I am researching other rotary table approaches without using air bearings in particular because I think that the rotary table approach may be one of the easier systems to put in a vacuum chamber more on that in a bit. But I am definitely in the study mode no decisions yet. And I appreciate any and all comments on this.
Re: data logging absolutely I have a spare quad i7 which I can go with and 14 bit or 16 bit DAQ is my desire depending on what is affordable. I plan on using NI Labview for a lot of data acquisition design and control. Mainly it seems, in a parallel to real estate, the key is measure, measure, measure.
I very much like the laser displacement sensors you used on NSF-1701 and I plan to use them where ever appropriate. Also good comments on A/C lines and minimizing air flow. In fact I hope to feed everything through an isolation transformer and/or inverter system even the measurement systems. That may be a bit of a stretch but thats how I set up my hamshack and it seems to work well. Several Kw rated iso transformers are a bit dear even on eBay but they last forever.
My initial build will likely use a faraday cage similar to Shells which will also double as a wind break for air movements. But eventually I hope to go to a vacuum chamber approach. I have a left over vacuum system which can get down to 10-4 torr or even 10-5 torr on a good day if all the connections are really well sealed. This was from some work maybe a decade ago on a Farnsworth fusor fun little project but have to be VERY careful about neutrons and shielding or you may glow in the dark. Even more than the emdrive, anyone playing with one should be fully trained on radiation and reactor safety!
Since the pumps are the expensive part, I got to noodling over building a DIY vacuum chamber to enclose a emdrive test fixture several interesting designs on the net - and its painstaking work and moderately costly but not out of the realm of possibility. This will NOT be soon definitely well into 2016.
Overall my approach is going to be to plan and design inward. What I mean by that is there is a great deal of discussion on frustum design and options and there will be much more in the next few months I expect so I am going to first concentrate on design and construction of a test facility which is reusable, reconfigurable and hopefully will provide data that can feed into our statistical and theoretical wizards here you guys just amaze me.
As I mentioned before I am developing a formal test plan analysis for the whole thing using - IEEE 829 as a very loose guide I have worked with it before and it has a lot of good formalism but is really both overkill in some areas and not detailed enough in others for this effort. As soon as I have to some coherent state I will post here on the forum.
Herman
-
#2224 by Stormbringer on 29 Sep, 2015 12:19
-
my guess using the force is: they got their s/n goal and reliable thrust. not necessarily steady state thrust. They were almost there last we heard before they were told to clam up so it is reasonable to assume they got there. But even then it is uncertain they sent it to Glen.
Even though none of the amateur/semi-pro replication got that magnitude of signal yet; the varied approaches both of test articles and of testing rigs probably were/are probably helpful in fertilizing EW's own set ups. -
#2225 by graybeardsyseng on 29 Sep, 2015 12:20
-
I **fig**ured that would come up sooner or later.Alright, I always wanted to ask this. Do you think NASA Eagleworks were able to go above 100 micronewtons and keep it constant? and pass the testing to the Glenn Research Center?
I was asked the other day if Micro-newtons were just a little cookie.
I do not want to speculate. Just curious.
My guess is they might have...but of course no paper yet to confirm or dismiss, but after the testing by rfmwguy I am more confident. On the other hand the "thrust" is still quite small so far..
H. -
#2226 by graybeardsyseng on 29 Sep, 2015 12:39
-
Thank you !I would like to ask the members of the forum for some input/opinions.
Some of you know me from my posts on various RF and microwave issues. I am just now starting a DIY emdrive test effort I am finally getting my workshop back on line following a move and I want to try to build on the outstanding work of SeeShell, rfmwguy, and others to hopefully make some small contribution of my own.
Anyway to the request.
There are a multitude of test environment design factors of course and this is just a start but in order to begin limiting the possible test space I have narrowed down overall test configurations to the following:
1. Rotary Table configuration unit under test (UUT) thrusting either posigrade or retrograde to rotation
2. Linear (slide) configuration air or other low friction surface. UUT thrusting along the access of the slide.
3. Balance Beam (ala rfmwguy and seeshells) UUT thrusting either up or down
In my thinking all have things going for them as well as potential problems, but I would like to ask for your thoughts on each configurations. In particular these are the things I have been considering:
Overall Pros and Cons
Challenges to getting usable data output
Potential systemic and situational error sources
Data which should/can be measured for either signal or noise/error determinations
ANYTHING else you can think of
Thanks in advance. If you would prefer to reply by private message rather than on the forum that is fine with me however please indicate any information which you do not want disseminated further.
Herman
PS for those who dont know me and to provide some credibility background - I have worked in both professionally and DIY settings on RF from LF to 40+ Ghz, high power RF and RF power supplies, vacuum chambers and systems, nuclear power plants and various and sundry other technical fields including aerospace and defense for the past 39 years. Well acquainted with safety procedures - wrote some for my last company. This will be my first big retirement project and I cant wait to get started.
A rotary rig will allow unlimited contious accelleration (well, depending on friction in your rig) and therefore most likely a higher signal to noise ratio. It would also be easy to do control experiments like rotating the frustum 180 deg. for reverse thrust, or 90 degrees (pointing towards or away from the rotational axis) for zero thrust.
On the other hand, it will arguably be the most complicated setup to build.
Whichever configuration you choose, be very careful when you design your experiment. What is your hypothesis? How can you falsify it? What control experiments do you need? The most accurate measurements in the world won't save you if you don't know what it is you're measuring.
As I have mentioned earlier, in my opinion, the most important control experiment to do is one where you are knowingly injecting EM at a non-resonating frequency. That should be your "negative control" experiment. Naturally, this means you must know the resonant freq. of your cavity and so on.
It is of course also important that you do at least 3 independent repetitions of each experiment, the more the better.
Good luck!
Outstanding thoughts.
First - yes the rotary table approach is likely the most complex. But it does seem to have some real advantages in flexibility and, as you mention, signal to noise factors. In particular the ability to change the axis of (potentially real) thrust is one of the major advantages I want to use. I may even consider 'dueling frustums' i.e. one posigrade and one retrograde to rotation on opposite sides of the table.
The specification of hypothesis and goals of each test is one of the reasons I will be formally documenting a test plan. I have been following the hypothesis discussion hear closely and I will say more on that later but there MUST be control experiments and null drive configurations. In particular I am very interested in what happens off design point with frequencies far from resonance. Likewise with heating the frustum in a manner which will be null for thrust but can heat it up etc.
I also completely agree with 3x or better tests. Likewise when test configuration or conditions are changed only one is changed at a time. This takes time but results in hopefully more useful data.
Herman
-
#2227 by glennfish on 29 Sep, 2015 12:49
-
BTW I just tried max-min and sd, the results were surpising. Now why is that.... Gotta think a bit before sharing.
Because both max-min and sd are measurements of the variability of a time series, and the periods when the magnetron is on are naturally going to be more noisy than when the magnetron is off, ceterus paribus?
Why do we know that "the periods when the magnetron is on are naturally going to be more noisy"?
In my mind's eye, we might have a step function being injected. I don't see a rigorous apriore justification for why the "noise" is greater for the "on" versus "off" period. We can start to make assumptions about 60Hz "steps" (or impulses) during the "on" period, but the resolution of the data is so low I don't see how such 60Hz "steps" could possibly be resolved. Instead, I would have thought the mechanical system acting as a crude low pass filter would be a more likely scenario... which leads back to the original question: why should we expect to see more noise during "on" than "off"?
If we can show statistically that the data *IS* noisier while the magnetron is on, then I think that would be very valuable piece of information. For example, if there is more "noise" while the magnetron is on, a follow-on question would be: what does the "noise" look like?
So in summary, I'm quite interested in hearing what @Glennfish noticed while performing simple min/max and sd calculations.
on noise.
Noise is typically calculated on the basis of a known signal, and a measurement of how that signal degrades. We don't really have that in this data.
A surrogate to noise is the standard deviation or variance.
Comparing two variances is generally verboten, but there are a some ways.
I chose, to answer your question, something called the Coefficient of Variation. I'm not sure it's a good method with this data. It's basically, the standard deviation / mean
It doesn't take the sample size into account, but it does sorta what you were asking about.
Spreadsheet attached ("macro's not included")
raw data in sheet raw2.php
summary in sheet Sheet1
-
#2228 by glennfish on 29 Sep, 2015 13:18
-
Anyway to the request.
...
Data which should/can be measured for either signal or noise/error determinations
As one of the data geeks here, one thing that's become apparant is that the data should be as complete and sharable as possible. The limited analysis I've done has depended on multiple other people massaging the data before I touch it. I'd like to propose some sort of data interchange standard, but I don't know squat about what we should be measuring. I don't know how fast you should sample, but as a rule of thumb, sample as fast as you can for as long as you can.
In general, include for each sample
1. time stamp (maximum possible accuracy... microsecond resolution would be cool)
2. state of the device (on/off)
3. other states, i.e. forward, backward, as many states as you can collect
...
4. measurement 1
5. measurement 2 as many simultaneous measurements as you can grab
... etc
CSV format is ideal, it works with almost everything
xls format is ok, most of us can work with that -
#2229 by graybeardsyseng on 29 Sep, 2015 13:41
-
Anyway to the request.
...
Data which should/can be measured for either signal or noise/error determinations
As one of the data geeks here, one thing that's become apparant is that the data should be as complete and sharable as possible. The limited analysis I've done has depended on multiple other people massaging the data before I touch it. I'd like to propose some sort of data interchange standard, but I don't know squat about what we should be measuring. I don't know how fast you should sample, but as a rule of thumb, sample as fast as you can for as long as you can.
In general, include for each sample
1. time stamp (maximum possible accuracy... microsecond resolution would be cool)
2. state of the device (on/off)
3. other states, i.e. forward, backward, as many states as you can collect
...
4. measurement 1
5. measurement 2 as many simultaneous measurements as you can grab
... etc
CSV format is ideal, it works with almost everything
xls format is ok, most of us can work with that
Outstanding idea. Likewise I would also suggest to naming data files or other data artifacts in a unique and unambiguous way and record somewhere a detailed description of your test configuration with the same name or description.
In my former life we tended to use something like "PROJ-TEST-TESTNUMBER-DATE/TIME" Where PROJ was a unique project name, TEST was a unique name for the overall test configuration being used, TESTNUMBER was a sequential number of tests in the TEST configuration or better yet a unique number in the whole PROJ area, and DATE/TIME were just that - a unique date and time stamp of the start or stop of the test - whatever your datalogger will support. It seems very pedantic and like a lot of extra work but 3 months later ( or 3 days sometimes) you will be able to unambiguously know which test generated each data set.
-
#2230 by rfmwguy on 29 Sep, 2015 14:00
-
Wow, impressive spreadsheet, especially sheet 2 graphs (my visual brain speaking). Can you conclude your first glimpse into ft 2d shows a statistical variance between on/off beyond the probability threshold?BTW I just tried max-min and sd, the results were surpising. Now why is that.... Gotta think a bit before sharing.
Because both max-min and sd are measurements of the variability of a time series, and the periods when the magnetron is on are naturally going to be more noisy than when the magnetron is off, ceterus paribus?
Why do we know that "the periods when the magnetron is on are naturally going to be more noisy"?
In my mind's eye, we might have a step function being injected. I don't see a rigorous apriore justification for why the "noise" is greater for the "on" versus "off" period. We can start to make assumptions about 60Hz "steps" (or impulses) during the "on" period, but the resolution of the data is so low I don't see how such 60Hz "steps" could possibly be resolved. Instead, I would have thought the mechanical system acting as a crude low pass filter would be a more likely scenario... which leads back to the original question: why should we expect to see more noise during "on" than "off"?
If we can show statistically that the data *IS* noisier while the magnetron is on, then I think that would be very valuable piece of information. For example, if there is more "noise" while the magnetron is on, a follow-on question would be: what does the "noise" look like?
So in summary, I'm quite interested in hearing what @Glennfish noticed while performing simple min/max and sd calculations.
on noise.
Noise is typically calculated on the basis of a known signal, and a measurement of how that signal degrades. We don't really have that in this data.
A surrogate to noise is the standard deviation or variance.
Comparing two variances is generally verboten, but there are a some ways.
I chose, to answer your question, something called the Coefficient of Variation. I'm not sure it's a good method with this data. It's basically, the standard deviation / mean
It doesn't take the sample size into account, but it does sorta what you were asking about.
Spreadsheet attached ("macro's not included")
raw data in sheet raw2.php
summary in sheet Sheet1 -
#2231 by glennfish on 29 Sep, 2015 14:16
-
Wow, impressive spreadsheet, especially sheet 2 graphs (my visual brain speaking). Can you conclude your first glimpse into ft 2d shows a statistical variance between on/off beyond the probability threshold?BTW I just tried max-min and sd, the results were surpising. Now why is that.... Gotta think a bit before sharing.
Because both max-min and sd are measurements of the variability of a time series, and the periods when the magnetron is on are naturally going to be more noisy than when the magnetron is off, ceterus paribus?
Why do we know that "the periods when the magnetron is on are naturally going to be more noisy"?
In my mind's eye, we might have a step function being injected. I don't see a rigorous apriore justification for why the "noise" is greater for the "on" versus "off" period. We can start to make assumptions about 60Hz "steps" (or impulses) during the "on" period, but the resolution of the data is so low I don't see how such 60Hz "steps" could possibly be resolved. Instead, I would have thought the mechanical system acting as a crude low pass filter would be a more likely scenario... which leads back to the original question: why should we expect to see more noise during "on" than "off"?
If we can show statistically that the data *IS* noisier while the magnetron is on, then I think that would be very valuable piece of information. For example, if there is more "noise" while the magnetron is on, a follow-on question would be: what does the "noise" look like?
So in summary, I'm quite interested in hearing what @Glennfish noticed while performing simple min/max and sd calculations.
on noise.
Noise is typically calculated on the basis of a known signal, and a measurement of how that signal degrades. We don't really have that in this data.
A surrogate to noise is the standard deviation or variance.
Comparing two variances is generally verboten, but there are a some ways.
I chose, to answer your question, something called the Coefficient of Variation. I'm not sure it's a good method with this data. It's basically, the standard deviation / mean
It doesn't take the sample size into account, but it does sorta what you were asking about.
Spreadsheet attached ("macro's not included")
raw data in sheet raw2.php
summary in sheet Sheet1
The sheet reflects my scatterbrained approach to figuring out what should be looked at.
There is no question that the on CV is greater than the off CV. p > .99
BUT, I'm not sure if that's real or an artifact of the different data counts during a downward trend.
Gotta think some more.
-
#2232 by SteveD on 29 Sep, 2015 14:16
-
The specification of hypothesis and goals of each test is one of the reasons I will be formally documenting a test plan. I have been following the hypothesis discussion hear closely and I will say more on that later but there MUST be control experiments and null drive configurations. In particular I am very interested in what happens off design point with frequencies far from resonance. Likewise with heating the frustum in a manner which will be null for thrust but can heat it up etc.
I also completely agree with 3x or better tests. Likewise when test configuration or conditions are changed only one is changed at a time. This takes time but results in hopefully more useful data.
Herman
I'm not sure a null test is possible in a frustum unless it's unpowered. Anything else will generate heat, and you're now testing theories about the importance of frequency and Q. -
#2233 by SteveD on 29 Sep, 2015 14:26
-
A while back, somebody noted an experiment that showed thrust by bouncing a laser back and forth. Was this a photonic laser rocket (cool tech, well understood) or did you mean it showed thrust by bouncing a laser off several mirrors and measuring the thrust at the last mirror. If so, can you post a link to the paper or report?
-
#2234 by rfmwguy on 29 Sep, 2015 14:46
-
Your thinking will pay off for future experimenters. The data overlay approach with mag ON/OFF is important. While OFF (lift) is moderately well behaved, ON has additional variables in the time domain such as mag frequency and power levels across a 40 MHz spectrum. Wish it were a square wave, instant ON with no level, thermal or frequency variations, but that is not the nature of the beast. RF power level in a mag is a delayed buildup akin to a cap charge best I can determine by reading and personal observation. Its frequencies slip across a natural frustum resonance; which itself changes with temperature.
Wow, impressive spreadsheet, especially sheet 2 graphs (my visual brain speaking). Can you conclude your first glimpse into ft 2d shows a statistical variance between on/off beyond the probability threshold?BTW I just tried max-min and sd, the results were surpising. Now why is that.... Gotta think a bit before sharing.
Because both max-min and sd are measurements of the variability of a time series, and the periods when the magnetron is on are naturally going to be more noisy than when the magnetron is off, ceterus paribus?
Why do we know that "the periods when the magnetron is on are naturally going to be more noisy"?
In my mind's eye, we might have a step function being injected. I don't see a rigorous apriore justification for why the "noise" is greater for the "on" versus "off" period. We can start to make assumptions about 60Hz "steps" (or impulses) during the "on" period, but the resolution of the data is so low I don't see how such 60Hz "steps" could possibly be resolved. Instead, I would have thought the mechanical system acting as a crude low pass filter would be a more likely scenario... which leads back to the original question: why should we expect to see more noise during "on" than "off"?
If we can show statistically that the data *IS* noisier while the magnetron is on, then I think that would be very valuable piece of information. For example, if there is more "noise" while the magnetron is on, a follow-on question would be: what does the "noise" look like?
So in summary, I'm quite interested in hearing what @Glennfish noticed while performing simple min/max and sd calculations.
on noise.
Noise is typically calculated on the basis of a known signal, and a measurement of how that signal degrades. We don't really have that in this data.
A surrogate to noise is the standard deviation or variance.
Comparing two variances is generally verboten, but there are a some ways.
I chose, to answer your question, something called the Coefficient of Variation. I'm not sure it's a good method with this data. It's basically, the standard deviation / mean
It doesn't take the sample size into account, but it does sorta what you were asking about.
Spreadsheet attached ("macro's not included")
raw data in sheet raw2.php
summary in sheet Sheet1
The sheet reflects my scatterbrained approach to figuring out what should be looked at.
There is no question that the on CV is greater than the off CV. p > .99
BUT, I'm not sure if that's real or an artifact of the different data counts during a downward trend.
Gotta think some more.
This is where a perfect experiment might be with a clean, single frequency injection and a frustum capable of "synching" or tuning resonance to track that signal. This would get us closer to a "square wave" ON condition I believe, maximizing the effect in the time domain.
This is probably old news to SPR, EW and yang, but worth the DIY community thinking about...unless you perfect the ability to separate the wheat from the chaff
-
#2235 by graybeardsyseng on 29 Sep, 2015 14:57
-
The specification of hypothesis and goals of each test is one of the reasons I will be formally documenting a test plan. I have been following the hypothesis discussion hear closely and I will say more on that later but there MUST be control experiments and null drive configurations. In particular I am very interested in what happens off design point with frequencies far from resonance. Likewise with heating the frustum in a manner which will be null for thrust but can heat it up etc.
I also completely agree with 3x or better tests. Likewise when test configuration or conditions are changed only one is changed at a time. This takes time but results in hopefully more useful data.
Herman
I'm not sure a null test is possible in a frustum unless it's unpowered. Anything else will generate heat, and you're now testing theories about the importance of frequency and Q.
Likely you are correct. But I think I didn't say what I was trying to say very well. Back a ways on the forum - maybe even back on Thread 3 - there as discussions of running a test in which there was heating but no RF energy actually coupled into the cavity. Discussion - IIRC - even included ideas for ohmic DC heating of the cavity and waveguides. The idea at that time was to attempt to duplicate the thermal input that was occurring during the "live" test but without the possibility for the potentially existing em generated thrust and thereby attempt to quantify the "lift" or "thrust" due to the thermal heating . At the time I was wondering about the potential fidelity of these sorts of tests, but I did think that if it could be done in am manner to duplicated the "flight" configuration but without coupling RF it might provide some useful control data. It might also be a bit of pursuing the wild goose.
H -
#2236 by rfmwguy on 29 Sep, 2015 15:32
-
Scientific proof that I have my old car and the garage back. Note the Floobie Stick safely hanging on the wall
-
#2237 by rfmwguy on 29 Sep, 2015 16:42
-
@glennfish, attached is the original ft 2d spreadsheet with data channels 2-4 (columns b-d) unhidden. Its in the original file as well, just hidden.
Note on these hidden data channels: The data inputs are unloaded, meaning nothing is attached to them, unlike channel 1 (column a on the spreadsheet), the LDS voltage input. This might be useful in looking at system noise in mag ON/OFF conditions.
While I cannot quantify the randomness, it will be the most variable (very sensitive) since there is no load (resistor) on the inputs. If mag ON imparts system noise, these 3 channels and over 8,100 data points will be the most sensitive to noise and will show it.
Another note. The balanced impedance of channel 1 is 470 ohms. Data on channels 2-4 (columns b-c) are open (infinite impedance by comparison) and are not directly equivalent to channel 1. However general deviation comparisons to mag ON/OFF should give you a relative indication of any system noise attributable to EMI, if present.
-
#2238 by SteveD on 29 Sep, 2015 16:45
-
The specification of hypothesis and goals of each test is one of the reasons I will be formally documenting a test plan. I have been following the hypothesis discussion hear closely and I will say more on that later but there MUST be control experiments and null drive configurations. In particular I am very interested in what happens off design point with frequencies far from resonance. Likewise with heating the frustum in a manner which will be null for thrust but can heat it up etc.
I also completely agree with 3x or better tests. Likewise when test configuration or conditions are changed only one is changed at a time. This takes time but results in hopefully more useful data.
Herman
I'm not sure a null test is possible in a frustum unless it's unpowered. Anything else will generate heat, and you're now testing theories about the importance of frequency and Q.
Likely you are correct. But I think I didn't say what I was trying to say very well. Back a ways on the forum - maybe even back on Thread 3 - there as discussions of running a test in which there was heating but no RF energy actually coupled into the cavity. Discussion - IIRC - even included ideas for ohmic DC heating of the cavity and waveguides. The idea at that time was to attempt to duplicate the thermal input that was occurring during the "live" test but without the possibility for the potentially existing em generated thrust and thereby attempt to quantify the "lift" or "thrust" due to the thermal heating . At the time I was wondering about the potential fidelity of these sorts of tests, but I did think that if it could be done in am manner to duplicated the "flight" configuration but without coupling RF it might provide some useful control data. It might also be a bit of pursuing the wild goose.
H
Well here's the thing. The only successful null tests I know of right now involve either a symmetrical resonance cavity or measurements with everything off. If you put a heating element in, it will produce photons of infrared radiation. I'm not at all sure that any form of electromagnetic radiation isn't going to produce a thrust effect at some level.
This effect appears so robust that the answer, when its found, has to be dirt simple. We're talking teach it in high school level simple, not "the logic gates in my silicon wafer are having difficulty because I've made the circuits so small that quantum tunneling is an issue."
So we have thrust from energy. What is the simplest explanation of thrust from energy: a photon rocket. We are getting more thrust than a photon rocket. How to you amplify the energy of a photon rocket: with a mirror. Simplest explanation is that reflection is somehow amplifying the photon rocket effect. Noether say's a closed system can't move on its own (though I get the feeling what she was really trying to say is that math can be used to deduce properties any physical system must have, so if you're looking at something new you have some idea what to look for). Since it can't be a closed system and go, it must not be a closed system. So either an outside force is working on it or something is escaping.
Given all of that then, if the outside force is not invoked, you can have a null test. If the answer is something is escaping, anything that emits photons is likely to produce a result. You can only get to null by changing the manner in which that thing escapes.
Which is why I'm looking for those laser studies, to see if something similar has been noted in tests that are bouncing photons around. -
#2239 by aero on 29 Sep, 2015 17:12
-
Regarding the null test. Aren't we pretty sure that a cylindrical resonator will not thrust? It will heat and do all the other things, but if something is escaping, that something should escape symmetrically, no thrust. If nothing is escaping then still no thrust due to symmetry.
Of course getting the heating rates right may be a little tricky when it comes to comparing to a real EM Drive configuration.
I do think the idea posted much earlier by (I forget who posted it), the idea of two identical frustums, one at each end of a balance beam, would go a long way toward eliminating thermal lift from the data. The idea was to mount both frustums upward, run to steady state then turn one off. Or mount one up, one down and run synchronously, doubling the thrust effect while significantly reducing the lift effect. This one would allow data collection from initial power on of the cold system and still reduce the effect of thermal lift.
Only problem is that it takes two frustums, but rfmwguy can tell us how much difficulty and cost would be involved in building the second identical frustum. And he could likely scope out the required modifications to the test rig as well.
.
