-
#860 by Rodal on 28 Dec, 2015 20:53
-
...
(my Mod hat is off)
This will be my last public posting on this issue, but you've made public comments regarding my Phase I observations and so will I:
I find your assessment a bit perplexing since you had been away for so long and had no commentary during my experiment as it was being conducted...only well after the fact and it was very subjective/declarative. I accept specific quantifiable critique, not "generalized". This, of course, is a big problem on other forums and not usually here.
"That's why you resorted to a statistical analysis."
This is another problem...I did not seek/resort to Glenn's assistance. He volunteered it when other's like yourself fell silent. In fact, several others helped when I was struggling with the Laser-spot analysis. Other people here stepped up THEN, not weeks later with hindsight commentary.
So, to summarize, I do not agree with your belated assessment, mainly because you have no specific critique, but generalized statements such as "there are times at which the magnetron is turned on and there is no such reaction force". What about the majority of the time when something was noted?
Glenn has the statistical analysis which I don't believe you can dismiss off hand. Rather than take up bandwidth here, please PM him to sort things out. I only ask one thing:
Whether accidently or by intent, you have given people the impression that my Phase I observational tests were null. I do not agree with that, both from what I personally witnessed with my own eyes (ala Shell) and more importantly, what Glenn relayed to me thru statistical analysis. Check out my FT2B video where the "aha" moment was and compare that to FT1, FT2 and FT2A where there was none.
Look, if we don't have the computing power to do your fluid dynamics analysis...so be it, don't even bring it up. I happen to believe that lift can be predicted and accounted for statistically and there is no phantom downwards "turbulence" that must be accounted for. Did I have an ideal experiment? No. Did I do things simply and within a small budget? Yes.
Got that off my chest, now back to mod-mode.QuoteI accept specific quantifiable critique, not "generalized". This, of course, is a big problem on other forums and not usually here.
QuoteI happen to believe that lift can be predicted and accounted for statistically and there is no phantom downwards "turbulence" that must be accounted for.
Contrary to what you state, my discussion of your data and conclusions that you state "can be accounted statistically" has not been generalized. It has been specific : that your sample population is way too small for you to conclude that your conclusion can be "accounted statistically". Glenfish accepted this fact in his post and furthermore commented on the fact that you prematurely dismantled your experiment so that his proposed further verification, and further statistical sampling could not be carried out.
My discussion of the natural convection effect on your experiment is based on my background in experiments and analysis in the Aeroelastics and Structures Research Laboratory at Aero&Astro in a respected university. My assessment is that your experimental results are inconclusive and cannot convincingly support your belief that you have measured an anomalous force. I don't claim that my assessment on the natural convection effect is conclusive by any means, I only say that I found your belief that your results are due to an anomalous force is unconvincing (to me).
Your comments regarding <<people here stepped up THEN, not weeks later with hindsight commentary>> weakens your conclusion, because it suggests that you suspect that your conclusions are very fragile and cannot withstand hindsight criticism. They also suggest that you have personal emotion invested in your conclusion. Since when is hindsight commentary on experiments not allowed?
If anything, I would have thought that you would be seeking such hindsight critical commentary instead of objecting to it. Is your point that you only accept critical comments on your experiments if they are made prior to your personal decision to dismantle your experiments?
PS1: And your comment << if we don't have the computing power to do your fluid dynamics analysis...so be it, don't even bring it up>> ?: by that standard a large number of the posts in these threads should not be brought up according to you, including any and all speculation on what could be responsible for anomalous forces in EM Drive, since most theories involving such anomalous forces are much more difficult to analyze.
PS2: Your comments (you are the moderator, whether writing that you are wearing the moderator's hat or not) regarding your objection to hindsight critical comments of your experiment acts as as strong dissuader from continuing to write in these threads. You are effectively replacing the thread introduction <<Objective skeptical inquiry is strongly welcome>> with "Hindsight critical comments of my experimental conclusions are NOT welcome". And by discouraging hindsight critical comments of your experimental conclusions, as the moderator of the EM Drive you are effectively discouraging skeptical comments about any EM Drive experiments. Quelling criticism leads to fragility and progressively engulfs a group into a false sense of reality. -
#861 by oliverio on 28 Dec, 2015 21:00
-
@rfmguy:
It should always be the case that any forces delivered by thermally lifted air will have a net upward component over the time of heating, although at times it could push in any other direction given a specific measurement (due to the chaos inherent in vortices etc.).
So one way to think of it would be as thus: if the magnetron-on effect you observed consistently pushed the weight of your frustrum pointed down (or resting position) above the resting weight, for greater than half the runtime, there is almost no way for it to be attributable to thermal activity.
I haven't seen your test numbers specifically, but from the sound of it, I'd say that you either observed thrust or some magnetic artifact, especially given the fact that hot objects at rest tend to create a barrier of static air around their surface that deflect incoming thermal convection. (This is why a convection oven runs a fan, because moving air actually maximizes thermal change around a radiative body.)
In other words you could solve this by putting an emdrive in a very chaotic air environment and testing then, which is why I have advocated in the past for heating the entire test assembly to several hundred degrees passively before running experiments with the magnetron/antenna on. -
#862 by rfmwguy on 28 Dec, 2015 21:20
-
@rfmguy:
Thanks, these are good suggestions for next year. I had come to the same conclusion, it wasn't thermally induced for the exact reason you mentioned (think Glenn stated about 67% or thereabouts off the top of my head). In my test report, I never mentioned Thrust, only Something anomalous, statistically. If nothing was observed, I would have ended the experiment...and almost did until FT2B.
It should always be the case that any forces delivered by thermally lifted air will have a net upward component over the time of heating, although at times it could push in any other direction given a specific measurement (due to the chaos inherent in vortices etc.).
So one way to think of it would be as thus: if the magnetron-on effect you observed consistently pushed the weight of your frustrum pointed down (or resting position) above the resting weight, for greater than half the runtime, there is almost no way for it to be attributable to thermal activity.
I haven't seen your test numbers specifically, but from the sound of it, I'd say that you either observed thrust or some magnetic artifact, especially given the fact that hot objects at rest tend to create a barrier of static air around their surface that deflect incoming thermal convection. (This is why a convection oven runs a fan, because moving air actually maximizes thermal change around a radiative body.)
In other words you could solve this by putting an emdrive in a very chaotic air environment and testing then, which is why I have advocated in the past for heating the entire test assembly to several hundred degrees passively before running experiments with the magnetron/antenna on.
I'm sort of locked into the same configuration unless I totally redesign things; with the mag on top. I will plan to use a solid frustum next year plus try to develop a heat shroud around the magnetron; something that insulates as well as acts like a heatsink...it also needs to be light-weight. If anyone has any ideas on this, let me know...my first thought was copper shavings in a enclosed "shroud" around the mag. Have to be careful of thermal runaway, though. -
#863 by lmbfan on 28 Dec, 2015 21:28
-
Thanks so much for doing the model. I hope someone here provides you more info. Dimensional is all I need. Perhaps @Rodal can help since he is wanting this type of data. Here is the video that shows the magnetron ignited:<snip>
<snip>
So, a computational fluid dynamics model of rfmwguy's experiment natural convection interaction with his turning the magnetron on and off would be fun to see.
What CFD computer program do you have experience with, and are you willing to take your time to model the natural convection in these experiments?
Since this is read by a wide audience, this question is really addressed at large: it would be useful for somebody to model the low Reynolds number, natural convection in these experiments with a CFD model
Also, just on a side note, wouldn't this be rather hellish to model? It seems like having a fine mesh like rfmwguy's EMDrive would require an ultra fine grid to get near accurate results of the air interacting with the heated mesh, not to mention the difficulty of creating the model. I am sure there are tools that I am unaware of for modeling this kind of surface though.
What kind of models do you need? I made a simple IGES model of rfmwguy's magnetron housing located here:
http://forum.nasaspaceflight.com/index.php?topic=39004.msg1459224#msg1459224
I may have some free time in the next day or two to make more elaborate models, if given enough information.
...
See this message regarding Computational Fluid Dynamics (CD+FD) vis-a-vis rfmwguy's experiment:
https://forum.nasaspaceflight.com/index.php?topic=39004.msg1465399#msg1465399
There may be some confusion in rfmwguy's response (unless I am missing something):
1) The message deals only with a simple IGES model of rfmwguy's magnetron housing, not with a CFD analysis
2) The dimensions needed for rfmwguy's experiment must be given by rfmwguy since rfmwguy is the one that run the experiment. I don't understand on what basis rfmwguy writes << I hope someone here provides you more info. Dimensional is all I need.>>
3) The issues involved with doing a CFD analysis, as previously discussed, have to do with computational resources. Since even Meep models are presently run to only 0.01 microseconds because of lack of computational speed, and since as previously discussed a CFD transient 3D analysis with a full Navier-Stokes solution would make children's play of the Meep analysis. Compound that by the fact that a CFD analysis with the Navier Stokes equation would be nonlinear and hence much more demanding to formulate properly and to interpret properly.
My fault for not using precise terminology: The IGES files are strictly a GEOMETRIC model of the metal magnetron housing(s). I did not make a thermal/CFD model. DnA915's comment led me to believe that a geometric model was needed. I will thus amend my statement to:
I will gladly create any geometric models needed, as that is part of my professional skill set. Any analysis will have to be independently done as I don't have the necessary software or expertise. For best results, add as many dimensions as possible.
----
To add to all the parallel processing concerns raised above: I think one of the more pressing problems that have not yet been brought up directly is the matter of storage space. It would, in my opinion, be beneficial to extrapolate the storage requirements of running lengthy meep simulations. I would guesstimate multiple terabytes for the h5 files, if not petabytes. Also, to cut down on storage requirements, some discussion on what exactly needs to be saved for analysis would help. Along the lines of "20 time slices every 100 time slices" or some such, (which meep supports if I remember correctly), and a discussion of single precision vs double precision. -
#864 by ThereIWas3 on 28 Dec, 2015 21:45
-
Also, to cut down on storage requirements, some discussion on what exactly needs to be saved for analysis would help. Along the lines of "20 time slices every 100 time slices" or some such, (which meep supports if I remember correctly), and a discussion of single precision vs double precision.
Yes, MEEP lets you specify exactly how often data is to be written out, and which data that is, which field(s) out of six, along which axes out of three. Also, if all you are going to do is make pictures from the H5 data, you can have MEEP generate the PNG files directly and not write any H5 files at all. For initial runs to make sure what is going to be interesting, and to validate models, this can save huge amounts of time and storage. The PNG files are only a few tens of KB each. You imbed the h5topng commands directly in the code.
Internal computations are always done in double precision but you do have the option of having the H5 files written in single precision, whic I think is plenty for the purposes here. The PNG files are even less. -
#865 by frobnicat on 28 Dec, 2015 22:07
-
I showed this over multiple posts, using Mathematica to post-process the Meep data output by Aero (*). See for example http://forum.nasaspaceflight.com/index.php?topic=37642.msg1406309#msg1406309.../...
3) During the extremely short runs of Meep up to now, the Poynting vector and stresses were increasing at an exponential rate. What does the equilibrium balance between Poynting vector field rate and stress gradient look like vs time?
.../...
By "exponential rate" do you mean exp(-t/tau) as a first order charge at constant power against leaking power proportional to stored energy, asymptotically reaching a plateau (would make sense), or exp(t/tau) as ever increasing (at increasing rates) values, i.e. diverging (would not make sense, but your phrasing leaves ambiguity). Sorry for the nitpicking.
At hollidays with limited connectivity, happy celebrations everyone.
that shows an excellent fit of the net force (on the flat ends) data (from which the force has been computed and shown during the last two cycles ending at 0.013 microseconds from the time at which the Microwave feed was turned on) using the following fit:
A t +B (exp(t/tau) Sin [C t + D] + E
where the circles represent the data points output by Meep and the solid line represents the fitted model (with excellent R^2 = 0.999353)
I also wrote:Quotethe present Finite Difference model (from which the force has been computed at the last two cycles ending at 0.013 microseconds from the time at which the Microwave feed was turned on), would have to be marched forward for 1,000 times longer, to a total of 10 microseconds, for the force to be magnified by the calculated exponential growth to a value of 10 microNewtons (for an inputPower of 43 watts). Given the fact that the present Meep model takes an hour to run on a good PC modern computer, 1,000 hours of computer time represents over 41 days of computing time. Thus running the Meep model to steady state is impractical. Rather than using a supercomputer to perform such a computation, I suggest to use an implicit (unconditionally stable) Finite Difference model in time (rather than the explicit time difference model presently used that is subject to stability problems that limit the maximum finite difference time step). Such implicit finite difference models are well known (I developed a version of them in my PhD thesis 35 years ago) and can be run much faster than explicit FD models. There are also numerous alternative numerical schemes that are more accurate than Finite Differences.
The need to march forward the analysis to the order of tens of microseconds (instead of the extremely small time of 0.013 microseconds in the present Meep analysis) to ascertain what is going on agrees with a recent estimate by TheTraveller (where he proposed 30something microseconds was needed).
__________
(*) I have not seen anyone else similarly analyzing the Meep data in these threads, as they usually just output the electric or magnetic field values at one point in (extremely early) time (usually around 0.01 microseconds
-many times the Meep electric or magnetic field output are shown without indicating at what time, or without showing the audience that the output is increasing with time -), without analysis of the stresses or analysis of the time variation of the Meep model. Also, I have not seen the Meep model validated by checking it against the experimental results and COMSOL FEA analysis for TM 212 of NASA.
Oh yes, I see. Was less than attentive when this analysis work was carried on...
So, sorry, your phrase don't leave any ambiguity : the fitted curve does show amplitude "increasing at an exponential rate" (didn't thought it was possible) in the form e+t/τ1.
From your attached plot, in pN (10-12 Newton), t in time slice (meep units) :
Fxx=6.17×10-6 t + 1.53×10-4 e0.0222 t sin(1.23 t+0.938)+1.85×10-4
Which leaves me in a state of confusion as to why ? Doesn't make physical sense to me.
Surely, such an exponentially increasing value can't go on forever, this is diverging (and quick !). I understand we are looking at the first .01 µs or so of a phenomenon that reaches stability rather above 10µs, so we are at a thousandth of a complete view of the initial dynamic (assuming an instantaneous power on of the microwave generator). So is it an initial knee at the very first periods, that is to be followed by an inflection point and an overall slow down in the form max_amplitude×(1-e-t/τ2) where τ2>>τ1 ?
The motive of my questioning is that I would expect, overall (notwithstanding a specific shape of increasing rate around the few very first periods) something like a first order "charge" (starting at t=0 with E=0) :
E(t) stored electromagnetic energy in the cavity
Power_in(t)=P (constant)
Power_loss(t)=(ω/Q)E(t) (on average around a cycle)
dE/dt=Power_in-Power_loss
=> dE/dt=P-(ω/Q)E
=> E(t)=τ2×P×(1-e-t/τ2) with τ2=Q/ω
Here for Q using the "other common definition for Q is the ratio of the energy stored in the oscillating resonator to the energy dissipated per cycle by damping processes" (from wikipedia Q factor)
The amplitude you are fitting is that of a (not averaged) radiation_pressure(t), isn't this amplitude proportional to stored energy E(t) ?
Am I at lost ? Are we too early in the time marching simulation to hope see (fit) such a τ2 ?
And what about the linear "offset" 6.17×10-6t, it is also diverging (albeit more gently) so it has also to reach a plateau at some point ? Wasn't the simulations on which you based those analysis done at the time with wrong parameters (now corrected) that gave unnaturally high Q ?
-
#866 by Rodal on 28 Dec, 2015 22:18
-
I showed this over multiple posts, using Mathematica to post-process the Meep data output by Aero (*). See for example http://forum.nasaspaceflight.com/index.php?topic=37642.msg1406309#msg1406309.../...
3) During the extremely short runs of Meep up to now, the Poynting vector and stresses were increasing at an exponential rate. What does the equilibrium balance between Poynting vector field rate and stress gradient look like vs time?
.../...
By "exponential rate" do you mean exp(-t/tau) as a first order charge at constant power against leaking power proportional to stored energy, asymptotically reaching a plateau (would make sense), or exp(t/tau) as ever increasing (at increasing rates) values, i.e. diverging (would not make sense, but your phrasing leaves ambiguity). Sorry for the nitpicking.
At hollidays with limited connectivity, happy celebrations everyone.
that shows an excellent fit of the net force (on the flat ends) data (from which the force has been computed and shown during the last two cycles ending at 0.013 microseconds from the time at which the Microwave feed was turned on) using the following fit:
A t +B (exp(t/tau) Sin [C t + D] + E
where the circles represent the data points output by Meep and the solid line represents the fitted model (with excellent R^2 = 0.999353)
I also wrote:Quotethe present Finite Difference model (from which the force has been computed at the last two cycles ending at 0.013 microseconds from the time at which the Microwave feed was turned on), would have to be marched forward for 1,000 times longer, to a total of 10 microseconds, for the force to be magnified by the calculated exponential growth to a value of 10 microNewtons (for an inputPower of 43 watts). Given the fact that the present Meep model takes an hour to run on a good PC modern computer, 1,000 hours of computer time represents over 41 days of computing time. Thus running the Meep model to steady state is impractical. Rather than using a supercomputer to perform such a computation, I suggest to use an implicit (unconditionally stable) Finite Difference model in time (rather than the explicit time difference model presently used that is subject to stability problems that limit the maximum finite difference time step). Such implicit finite difference models are well known (I developed a version of them in my PhD thesis 35 years ago) and can be run much faster than explicit FD models. There are also numerous alternative numerical schemes that are more accurate than Finite Differences.
The need to march forward the analysis to the order of tens of microseconds (instead of the extremely small time of 0.013 microseconds in the present Meep analysis) to ascertain what is going on agrees with a recent estimate by TheTraveller (where he proposed 30something microseconds was needed).
__________
(*) I have not seen anyone else similarly analyzing the Meep data in these threads, as they usually just output the electric or magnetic field values at one point in (extremely early) time (usually around 0.01 microseconds
-many times the Meep electric or magnetic field output are shown without indicating at what time, or without showing the audience that the output is increasing with time -), without analysis of the stresses or analysis of the time variation of the Meep model. Also, I have not seen the Meep model validated by checking it against the experimental results and COMSOL FEA analysis for TM 212 of NASA.
Oh yes, I see. Was less than attentive when this analysis work was carried on...
So, sorry, your phrase don't leave any ambiguity : the fitted curve does show amplitude "increasing at an exponential rate" (didn't thought it was possible) in the form e+t/τ1.
From your attached plot, in pN (10-12 Newton), t in time slice (meep units) :
Fxx=6.17×10-6 t + 1.53×10-4 e0.0222 t sin(1.23 t+0.938)+1.85×10-4
Which leaves me in a state of confusion as to why ? Doesn't make physical sense to me.
Surely, such an exponentially increasing value can't go on forever, this is diverging (and quick !). I understand we are looking at the first .01 µs or so of a phenomenon that reaches stability rather above 10µs, so we are at a thousandth of a complete view of the initial dynamic (assuming an instantaneous power on of the microwave generator). So is it an initial knee at the very first periods, that is to be followed by an inflection point and an overall slow down in the form max_amplitude×(1-e-t/τ2) where τ2>>τ1 ?
The motive of my questioning is that I would expect, overall (notwithstanding a specific shape of increasing rate around the few very first periods) something like a first order "charge" (starting at t=0 with E=0) :
E(t) stored electromagnetic energy in the cavity
Power_in(t)=P (constant)
Power_loss(t)=(ω/Q)E(t) (on average around a cycle)
dE/dt=Power_in-Power_loss
=> dE/dt=P-(ω/Q)E
=> E(t)=τ2×P×(1-e-t/τ2) with τ2=Q/ω
Here for Q using the "other common definition for Q is the ratio of the energy stored in the oscillating resonator to the energy dissipated per cycle by damping processes" (from wikipedia Q factor)
The amplitude you are fitting is that of a (not averaged) radiation_pressure(t), isn't this amplitude proportional to stored energy E(t) ?
Am I at lost ? Are we too early in the time marching simulation to hope see (fit) such a τ2 ?
And what about the linear "offset" 6.17×10-6t, it is also diverging (albeit more gently) so it has also to reach a plateau at some point ? Wasn't the simulations on which you based those analysis done at the time with wrong parameters (now corrected) that gave unnaturally high Q ?
You can see why I was asking for verification of "Meep models" showing animations of the H and E fields at one very early point in time (0.01 microseconds)
and at least verification vis a vis NASA's experimental results for mode TM212 and COMSOL FEA
-
#867 by SeeShells on 28 Dec, 2015 22:26
-
I am having trouble reconciling all the dimensions in SeeShell's blueprint. The waveguide is shown as being 9.97cm high and protruding from the frustrum wall by 7.6 cm at the top. The end of the waveguide is given as 17.79cm from the central axis. This means that the point where the top of the waveguide meets the frustrum is 17.79 - 7.6 = 10.19 cm from the axis. The large end diameter is 29.5cm, so bigR = 14.75cm. Thus the slope of the wall is
atan((14.75 - 10.19) / 9.97) = atan( 4.56 / 9.97) = atan( .457 ) = 24.56 degrees.
But the frustrum itself is described as Height 24.79cm, smallD = 17cm, smallR = 8.5cm, for a slope ofatan( (14.75-8.5) / 24.79 ) = atan( 6.25 / 24.79 ) = atan( .252 ) = 14.15 degrees
Something is inconsistent. I notice the blueprint does not show the dimensions for height or small end diameter, so perhaps the numbers I am using for that, from an earlier post, are obsolete.
You'll find there are several different dimensions due to the confusion of the tuning chamber. But it's as simple as my small top plate is 165mm my large bottom plate is 295mm, I'm seeking TE modes and that determines the plate spacing along with the angle. That said my frustum isn't the exact solution from the calculations because I have a tuning chamber on top of the small section of the frustum.
So are you just trying to build what I've done, or model it, knowing will help me help you.
Shell -
#868 by rfmwguy on 28 Dec, 2015 22:44
-
#869 by RFPlumber on 29 Dec, 2015 02:07
-
W.r.t. building the cavity... Cutting the template out of sheet copper is easy, but then making the actual frustum out of it - not so much. In fact I am still trying to find out a way... The next attempt is going to be with a "jig" made from 2 wood boards with pre-routed paths for both the big and the small sides...
Or is it all because my copper is C110 and not the O2 free? Would the latter be much more pliable and hence easier to make into a frustum?
-
#870 by not_a_physicist on 29 Dec, 2015 02:25
-
W.r.t. building the cavity... Cutting the template out of sheet copper is easy, but then making the actual frustum out of it - not so much. In fact I am still trying to find out a way... The next attempt is going to be with a "jig" made from 2 wood boards with pre-routed paths for both the big and the small sides...
If your jig doesn't work out, I wonder if it would help to have 3 or so boards with successively smaller circles cut in them would help. You could start by roughly fitting the small end of the copper sheet into the biggest circle, and then kind of push it through into the next smallest hole. If the boards are held apart by threaded rods and wing nuts, then you could adjust them to fit the slope you want the frustum to be. I have no idea what I'm talking about, though... I've never done anything useful with copper sheets.
Or is it all because my copper is C110 and not the O2 free? Would the latter be much more pliable and hence easier to make into a frustum?
-
#871 by zen-in on 29 Dec, 2015 02:28
-
...
(my Mod hat is off)
This will be my last public posting on this issue, but you've made public comments regarding my Phase I observations and so will I:
I find your assessment a bit perplexing since you had been away for so long and had no commentary during my experiment as it was being conducted...only well after the fact and it was very subjective/declarative. I accept specific quantifiable critique, not "generalized". This, of course, is a big problem on other forums and not usually here.
"That's why you resorted to a statistical analysis."
This is another problem...I did not seek/resort to Glenn's assistance. He volunteered it when other's like yourself fell silent. In fact, several others helped when I was struggling with the Laser-spot analysis. Other people here stepped up THEN, not weeks later with hindsight commentary.
So, to summarize, I do not agree with your belated assessment, mainly because you have no specific critique, but generalized statements such as "there are times at which the magnetron is turned on and there is no such reaction force". What about the majority of the time when something was noted?
Glenn has the statistical analysis which I don't believe you can dismiss off hand. Rather than take up bandwidth here, please PM him to sort things out. I only ask one thing:
Whether accidently or by intent, you have given people the impression that my Phase I observational tests were null. I do not agree with that, both from what I personally witnessed with my own eyes (ala Shell) and more importantly, what Glenn relayed to me thru statistical analysis. Check out my FT2B video where the "aha" moment was and compare that to FT1, FT2 and FT2A where there was none.
Look, if we don't have the computing power to do your fluid dynamics analysis...so be it, don't even bring it up. I happen to believe that lift can be predicted and accounted for statistically and there is no phantom downwards "turbulence" that must be accounted for. Did I have an ideal experiment? No. Did I do things simply and within a small budget? Yes.
Got that off my chest, now back to mod-mode.
I have some knowledge of the difficulty of characterizing complex fluid mechanics behavior so can't offer any more guidance than Dr. Rodal has already offered. What I would suggest is to perform a series of experiments that will reveal how much of what you have observed is due to thermal effects. For example what happens when the fustrum is removed and a box made of Copper mesh is put in its place? The microwave energy will still be safely contained but there should be no em-drive thrust effect. The box can be made so it has the same weight as the fustrum. Another experiment would be to replace the magnetron with a weighted box (to equal the weight of the magnetron) covered with Aluminum heat sink fins and with a 200 Watt halogen bulb inside for heating. If we recall the experiments done to prove the existence of the ether; they instead showed the wave nature of light.
My overly simple analysis of rfmwguy's em-drive experiment is that the intense heating of the magnetron results in a lowering of the air pressure above it. This causes lift, just laike a plane's wing generates lift when air flows over it. However that can change because of air currents so it's possible to see a downward thrust as well. I have done my share of odd experiments. I have never been convinced the effect I am hoping to see is there if it is not always observable each and every time I throw the switch. If there is a natural pendulum motion can I stop that movement by throwing the switch at the right time? Can I increase the oscillation by throwing the switch at the opposite side of the cycle? And most of the time even after this has been observed there are other reasons for doubting a positive conclusion. Quoting Richard Feynman: "Science is a way of trying not to fool yourself. The first principle is that you must not fool yourself, and you are the easiest person to fool." -
#872 by rfmwguy on 29 Dec, 2015 02:31
-
Thanks for the pics! Shell did quite a nice job with her build and probably has better inputs but an internal frame would be nice. For my mesh, I simply took an old lampshade wire frame, cut it to size using aluminum tape to join the seams below:
-
#873 by SeeShells on 29 Dec, 2015 02:43
-
W.r.t. building the cavity... Cutting the template out of sheet copper is easy, but then making the actual frustum out of it - not so much. In fact I am still trying to find out a way... The next attempt is going to be with a "jig" made from 2 wood boards with pre-routed paths for both the big and the small sides...
Mine was O2 Free Copper .032" thick and very hard to work and struggled with it for couple days until I was sure I had it close.
Or is it all because my copper is C110 and not the O2 free? Would the latter be much more pliable and hence easier to make into a frustum?
Cut a circles out of plywood use it to "slide down the frustum" while curving it around. Do the same for the top smaller portion as well as the center.
I did a butt lap joint to cover the two edges and held it into place with "C" clamps
http://s1039.photobucket.com/user/shells2bells2002/library/CE%20Electromagnetic%20Reaction%20Thruster?sort=3&page=2
After I had the cone shape I scribed a line around the bottom plate to fit the frustum sides walls to.
http://s1039.photobucket.com/user/shells2bells2002/media/CE%20Electromagnetic%20Reaction%20Thruster/49image002_zpskidqc371.jpg.html
Working my way around the bottom I tacked the two together and ended up with somewhere around 12-14 solder tacks with the wall aligned to my scribe line. I filled with a silver epoxy and then with a heavier bead of PC7.
http://s1039.photobucket.com/user/shells2bells2002/library/CE%20Electromagnetic%20Reaction%20Thruster?sort=3&page=1
Take your time and make sure your surfaces are kept clean, I wore gloves to keep the oils and acids from my hands off the copper.
-
#874 by rfmwguy on 29 Dec, 2015 03:00
-
Good inputs zen, thanks. I think lift is low pressure for sure. Tests were in a large space without vents or fans. Air stillness was measured by the settling of the balance beam. Me just walking by the thing threw it into serious oscillation, so i started tests when still....
(my Mod hat is off)
This will be my last public posting on this issue, but you've made public comments regarding my Phase I observations and so will I:
I find your assessment a bit perplexing since you had been away for so long and had no commentary during my experiment as it was being conducted...only well after the fact and it was very subjective/declarative. I accept specific quantifiable critique, not "generalized". This, of course, is a big problem on other forums and not usually here.
"That's why you resorted to a statistical analysis."
This is another problem...I did not seek/resort to Glenn's assistance. He volunteered it when other's like yourself fell silent. In fact, several others helped when I was struggling with the Laser-spot analysis. Other people here stepped up THEN, not weeks later with hindsight commentary.
So, to summarize, I do not agree with your belated assessment, mainly because you have no specific critique, but generalized statements such as "there are times at which the magnetron is turned on and there is no such reaction force". What about the majority of the time when something was noted?
Glenn has the statistical analysis which I don't believe you can dismiss off hand. Rather than take up bandwidth here, please PM him to sort things out. I only ask one thing:
Whether accidently or by intent, you have given people the impression that my Phase I observational tests were null. I do not agree with that, both from what I personally witnessed with my own eyes (ala Shell) and more importantly, what Glenn relayed to me thru statistical analysis. Check out my FT2B video where the "aha" moment was and compare that to FT1, FT2 and FT2A where there was none.
Look, if we don't have the computing power to do your fluid dynamics analysis...so be it, don't even bring it up. I happen to believe that lift can be predicted and accounted for statistically and there is no phantom downwards "turbulence" that must be accounted for. Did I have an ideal experiment? No. Did I do things simply and within a small budget? Yes.
Got that off my chest, now back to mod-mode.
I have some knowledge of the difficulty of characterizing complex fluid mechanics behavior so can't offer any more guidance than Dr. Rodal has already offered. What I would suggest is to perform a series of experiments that will reveal how much of what you have observed is due to thermal effects. For example what happens when the fustrum is removed and a box made of Copper mesh is put in its place? The microwave energy will still be safely contained but there should be no em-drive thrust effect. The box can be made so it has the same weight as the fustrum. Another experiment would be to replace the magnetron with a weighted box (to equal the weight of the magnetron) covered with Aluminum heat sink fins and with a 200 Watt halogen bulb inside for heating. If we recall the experiments done to prove the existence of the ether; they instead showed the wave nature of light.
My overly simple analysis of rfmwguy's em-drive experiment is that the intense heating of the magnetron results in a lowering of the air pressure above it. This causes lift, just laike a plane's wing generates lift when air flows over it. However that can change because of air currents so it's possible to see a downward thrust as well. I have done my share of odd experiments. I have never been convinced the effect I am hoping to see is there if it is not always observable each and every time I throw the switch. If there is a natural pendulum motion can I stop that movement by throwing the switch at the right time? Can I increase the oscillation by throwing the switch at the opposite side of the cycle? And most of the time even after this has been observed there are other reasons for doubting a positive conclusion. Quoting Richard Feynman: "Science is a way of trying not to fool yourself. The first principle is that you must not fool yourself, and you are the easiest person to fool."
Heat rise was not instaneous per thermal vids so an opposite/downward force was unexpected. oliviero brought up static charge possibilities which I think was useful as well. Bottom line is my observations and data collection was brief due to time constraints with my test space. I cannot claim thrust, which I did not, but I can say with confidence there's something there that compells me to keep going with more testing next year.
You are skeptical and I appreciate that...really. Think the difference is you avoid judgemental calls as I tried to do by not calling it thrust. Whatever it is, its elusive and has been for a while. Thanks for your kind words...gets a little rough for diyers at times.
-
#875 by TheTraveller on 29 Dec, 2015 03:05
-
W.r.t. building the cavity... Cutting the template out of sheet copper is easy, but then making the actual frustum out of it - not so much. In fact I am still trying to find out a way... The next attempt is going to be with a "jig" made from 2 wood boards with pre-routed paths for both the big and the small sides...
Or is it all because my copper is C110 and not the O2 free? Would the latter be much more pliable and hence easier to make into a frustum?
I bought a 900 x 900 x 0.5mm (0.020") sheet of C110 that was rated as "soft". There are 2 harder variants of C110 available.
http://www.georgewhite.com.au/products/product_listing.asp?categorycode=CS-S
Plan to use these hoops on the inside and outside of the cone while forming to keep things circular and in place. 1.5" to 18" dia available.
http://www.ebay.com.au/itm/METAL-HOOP-9-18-DREAMCATCHER-indian-feathers-mobiles-ring-pentacles-/271454444908?var=&hash=item3f33f21d6c:m:mbQ27hxQC7J-8aeRx2IzewA
Will leave 3 hoops epoxied on the outside, as attached, for additional rigidity as the cone is only 0.5mm (0.020") thick.
-
#876 by ThereIWas3 on 29 Dec, 2015 03:13
-
You'll find there are several different dimensions due to the confusion of the tuning chamber. But it's as simple as my small top plate is 165mm my large bottom plate is 295mm, I'm seeking TE modes and that determines the plate spacing along with the angle. That said my frustum isn't the exact solution from the calculations because I have a tuning chamber on top of the small section of the frustum.
So are you just trying to build what I've done, or model it, knowing will help me help you.
Now I see what you are getting at. It is not a pure frustrum, but a frustrum with a cylinder on top. So the slope of the frustrum walls matches the ratio given by the waveguide dimensions in your blueprint, but the actual overall height is adjustable by moving the circular plate inside the cylinder. (See picture below, with exagerated cylinder height)
That probably does interesting things to the impendance. Watch out for heating where the two shapes meet.
I am trying to replicate your design in MEEP.
-
#877 by TheTraveller on 29 Dec, 2015 03:48
-
...Take your time and make sure your surfaces are kept clean, I wore gloves to keep the oils and acids from my hands off the copper.
Shell,
Any chance to see your plans and photos of your magnetron to coax adapter build?
Phil -
#878 by SeeShells on 29 Dec, 2015 10:16
-
I'll see what I can do Phil. I have a very busy day, Doctor's appointment this morning and then rebuild my system, have one being delivered this morning to replace the one that just died....Take your time and make sure your surfaces are kept clean, I wore gloves to keep the oils and acids from my hands off the copper.
Shell,
Any chance to see your plans and photos of your magnetron to coax adapter build?
Phil
Shell -
#879 by SeeShells on 29 Dec, 2015 10:23
-
The tune cone on top can do 2 things. Fine tune the frustum and also allow a placement of a block of material, PTFE (polytetrafluoroethylene) I believe, replicating the EagleWorks tests. Sorry... long night, up reading papers and need to get to bed, Dr. Apt in the morn.You'll find there are several different dimensions due to the confusion of the tuning chamber. But it's as simple as my small top plate is 165mm my large bottom plate is 295mm, I'm seeking TE modes and that determines the plate spacing along with the angle. That said my frustum isn't the exact solution from the calculations because I have a tuning chamber on top of the small section of the frustum.
So are you just trying to build what I've done, or model it, knowing will help me help you.
Now I see what you are getting at. It is not a pure frustrum, but a frustrum with a cylinder on top. So the slope of the frustrum walls matches the ratio given by the waveguide dimensions in your blueprint, but the actual overall height is adjustable by moving the circular plate inside the cylinder. (See picture below, with exagerated cylinder height)
That probably does interesting things to the impendance. Watch out for heating where the two shapes meet.
I am trying to replicate your design in MEEP.
