-
#1100 by Rodal on 10 Apr, 2016 12:03
-
...
Dr. Rodal,
A clean DC driven power supply for the magnetron will provide a good narrow band width RF source.
Shell
Did Iulian Berca use "a clean DC driven power supply for the magnetron" you describe instead of AC power as in a home's microwave oven ? I thought Iulian Berca used AC power in his experiment at his home. If so, then one needs to consider the spectrum from 2.42 GHz to 2.48 GHz to figure out what mode shape was excited.
Same question for Dave Distler and Prof. Tajmar's experiments...
Also, concerning the frequency of AC power supplied for home use, power is supplied at 60 Hz in North America but at 50 Hz in Europe and most of Asia. Is there a difference in the frequency spectrum of the magnetron experiments of Iulian Berca and Tajmar (@50 Hz AC) from the frequency spectrum of the magnetron experiment of Dave Distler (@60Hz AC)? -
#1101 by Monomorphic on 10 Apr, 2016 12:45
-
I don't recall whether Tajmar used a magnetron, but given the fact that the experimental wiki lists 2.44 GHz, it seems probable that Tajmar also used a magnetron, somebody should check
Tajmar definitely used a magnetron, circled red in first image. And it looks like I see his high voltage transformer in the second image.
Dave also used the high voltage transformer in the microwave for his tests.
-
#1102 by Monomorphic on 10 Apr, 2016 13:08
-
If so, then one needs to consider the spectrum from 2.42 GHz to 2.48 GHz to figure out what mode shape was excited.
This is a time domain analysis of the bandwidth between 2.42Ghz and 2.48Ghz in Berca's frustum. I didn't see TM212 anywhere in the sweep either. I will probably need to broaden the sweep to find TM212.
EDIT: Berca uses the exact same dimensions as NASA. TM212 is at 1.9326Ghz using NASA dimensions. Why does Berca expect TM212 at 2.45Ghz using those dims?
-
#1103 by Rodal on 10 Apr, 2016 13:22
-
If so, then one needs to consider the spectrum from 2.42 GHz to 2.48 GHz to figure out what mode shape was excited.
This is a time domain analysis of the bandwidth between 2.42Ghz and 2.48Ghz in Berca's frustum. I didn't see TM212 anywhere in the sweep either. I will probably need to broaden the sweep to find TM212.
EDIT: Berca uses the exact same dimensions as NASA. TM212 is at 1.9326Ghz using NASA dimensions. Why does Berca expect TM212 at 2.45Ghz using those dims?
1) NASA's EM Drive resonated with mode shape TM212 at 1.9367 Ghz when tested in air, with a dielectric HDPE insert asymmetrically placed in the cavity. The dielectric polymer lowers the natural frequency of the cavity due to its higher value of relative electric permittivity compared to the relative electric permittivity of air, since the natural frequency in the dielectric is a function of the inverse square root of the relative electric permittivity.
2) Iulian Berca did not place any dielectric insert inside his EM Drive. There was only air inside his EM Drive.
Frank Davies at NASA calculated that the mode shape TM212 for NASA's EM Drive dimensions without a dielectric insert resonates at 2.4575 GHz. Therefore Iulian Berca was justified in expecting his EM Drive to resonate with mode shape TM212 when excited by a magnetron at 2.45 GHz +/-0.03 GHz.
NASA's dimensions and material used by Frank Davies in his COMSOL FEA analysis:
Height: 9.00 inch (228.6 mm)
Top diam.: 6.25 inch (0.1588 mm)
Bottom diam.: 11.01 inch (279.7 mm)
Material: 101 Copper Alloy
3) I verified NASA's COMSOL FEA solution of TM212 for NASA's EM Drive dimensions without a dielectric insert using the exact solution for the problem, which gave a natural frequency in TM212 towards the lower end of the magnetron spectrum range (this is to be expected because numerical solutions like COMSOL or FEKO converge from above, they are stiffer than the actual cavity, because the mesh is finite instead of having an infinite number of nodes).
ATTACHMENT: NASA's calculation for TM212 at 2.4575 GHz
-
#1104 by X_RaY on 10 Apr, 2016 13:25
-
Monomorphic I working on a similar design/calculation with waveguide coupling. (Not the same dimension like Tajmar!)
But at lower frequency I was able to get an result like you for the Tajmar cone. The actual resonant frequency for TE010 is a little higher. Since they had a tuning mechanism it is possible that the length was slightly smaller than the value in the paper. So try higher frequency or make the frustum shorter to find this resonance pattern.
The lower pic is a .gif so you may be one have to click on it to see the animation (at least in my browser it works this way).
-
#1105 by Monomorphic on 10 Apr, 2016 14:22
-
Berca uses the exact same dimensions as NASA. TM212 is at 1.9326Ghz using NASA dimensions. Why does Berca expect TM212 at 2.45Ghz using those dims?
I was wrong. Well, not wrong, the emdrive wiki page has the incorrect dims for Berca. They are slightly different. 3mm difference on the big end-plate diameter. That 3mm difference was enough to greatly weaken the TM212 mode pattern.
-
#1106 by Rodal on 10 Apr, 2016 14:38
-
Yes, we have to be careful with respect to what mode shapes are actually excited in an experiment, because a difference of 3 mm can make a difference, as you just found out.Berca uses the exact same dimensions as NASA. TM212 is at 1.9326Ghz using NASA dimensions. Why does Berca expect TM212 at 2.45Ghz using those dims?
I was wrong. Well, not wrong, the emdrive wiki page has the incorrect dims for Berca. They are slightly different. 3mm difference on the big end-plate diameter. That 3mm difference was enough to greatly weaken the TM212 mode.
This is the case for Dave Distler's experiment in particular: he used a perforated metal for the side walls that was very compliant: one can see from the videos that the shape was not perfectly conical. This will make a difference in what mode shape was actually excited (if any single mode was actually excited in the experiment: also remember that when many modes are nearby one may get participation from different modes).
Ditto for other experiments: particularly when the frustums are made at home by DoItYourself and/or with thin metal.
Actual dimensions during the experiment matter. Actual distortion matters.
Previously TheTraveller had challenged others over small differences in calculated natural frequency with different methods: I thought that such discussion is moot when the actual mode shapes are sensitive to actual geometry. This truth was also pointed out by the NSF poster from China that had a collaboration with Prof. Yang: discussion about small differences in natural frequency are moot without knowing precisely the actual measured internal geometry.
The actual measured geometry for Shawyer's experiments was never reported for all his dimensions in any of his reports.
TheTraveller also justified the difference between results in air and results in vacuum chambers based on the extremely small difference of the relative electrical permittivity between air and vacuum. This is very questionable because the difference in relative permittivity between air at ambient pressure and air at the pressure in the vacuum chamber experiments is very small and translates into differences in natural frequency that are much smaller than the uncertainty in geometrical measurements of what is being tested.
What is required in an experiment is for the experimenter to independently measure the mode shape. Experimenters measure, experimenters should not assume anything !
The only experimenter that has actually measured and reported an experimentally verified mode shape is NASA: TM212 which was verified by actual infrared thermal scan.
This is one of several reasons why NASA's experiments are far and above the highest quality experiments conducted on the EM Drive.
Shawyer and Yang have never provided a reported mode-shape-measurement of what mode shapes they actually excited in their claimed experimental reports. -
#1107 by Monomorphic on 10 Apr, 2016 14:44
-
The only experimenter that has actually measured and reported a mode shape is NASA: TM212 which was verified by actual infrared thermal scan.
Hopefully my next build update will include an infrared scan to confirm TE311. I have an infrared camera, I just need to solder my fustum together and paint the exterior with high temp paint (and then mount the magnetron). The copper frustum seems to just reflect IR like a mirror now. Perhaps there is a different setting...
-
#1108 by Monomorphic on 10 Apr, 2016 17:24
-
But at lower frequency I was able to get an result like you for the Tajmar cone. The actual resonant frequency for TE010 is a little higher. Since they had a tuning mechanism it is possible that the length was slightly smaller than the value in the paper. So try higher frequency or make the frustum shorter to find this resonance pattern.
I'm using Tajmar's dimensions, at least the ones listed on the emdrive wiki page. I haven't cross checked his paper yet. Can confirm TE010 at a higher frequency. Not sure if tuning will help in this case, and it doesn't look like tuning was modeled in COMSOL.
-
#1109 by X_RaY on 10 Apr, 2016 17:33
-
I would expect the opposite, first modeling in Comsol and after that build the frustum with the possibility to tune around the calculated length.But at lower frequency I was able to get an result like you for the Tajmar cone. The actual resonant frequency for TE010 is a little higher. Since they had a tuning mechanism it is possible that the length was slightly smaller than the value in the paper. So try higher frequency or make the frustum shorter to find this resonance pattern.
I'm using Tajmar's dimensions, at least the ones listed on the emdrive wiki page. I haven't cross checked his paper yet. Can confirm TE010 at a higher frequency. Not sure if tuning will help in this case, and it doesn't look like tuning was modeled in COMSOL.
...snip
BTW they had spherical end plates. (Does he really?)
Since Tajmar used Diameter instead of Radius in the original version of this paper I am careful to believe in any of the numbers in this work.
In the past Dr. Rodal put a lot of work in the exploration of the height of this frustum.
https://forum.nasaspaceflight.com/index.php?topic=39214.msg1477307#msg1477307 -
#1110 by Rodal on 10 Apr, 2016 18:02
-
But at lower frequency I was able to get an result like you for the Tajmar cone. The actual resonant frequency for TE010 is a little higher. Since they had a tuning mechanism it is possible that the length was slightly smaller than the value in the paper. So try higher frequency or make the frustum shorter to find this resonance pattern.
I'm using Tajmar's dimensions, at least the ones listed on the emdrive wiki page. I haven't cross checked his paper yet. Can confirm TE010 at a higher frequency. Not sure if tuning will help in this case, and it doesn't look like tuning was modeled in COMSOL.
...
These are the dimensions Tajmar lists in page 3 of his corrected paper:Quote from: TajmarWe started by designing a model optimized for a frequency of 2.45 GHz using COMSOL in order to be able to
(Bold added for emphasis)
use commercial magnetrons used in standard microwave ovens. We iterated our design several times by consulting
with R. Shawyer to be as representative as possible. Our final tapered cavity design had an internal top radius of
38.5 mm, a bottom radius of 54.1 mm and a height of 68.6 mm as well as a side entrance for the microwaves as
shown in Fig. 2. The cavity was made out of three copper pieces with a wall thickness of 3 mm where the lower and middle part as well as the side flange were hard soldered using silver and the top part was able to adapt its position in order to optimize for a high Q factor. A standard WR340 waveguide was then used to connect the magnetron to the EMDrive.
http://bit.ly/1XVP8yf
Contrast that with the dimensions in the wiki, which lists a height of 0.1777 m:
http://emdrive.wiki/Experimental_Results
the WR-340 waveguide has standard internal dimensions of 86.36 mm x 43.18 mm (3.400 inches x 1.700 inches) according to
http://www.microwaves101.com/encyclopedias/rectangular-waveguide-dimensions -
#1111 by eischei on 10 Apr, 2016 18:15
-
@ Monomorphic:
No, there is no different setting for the infrared cam. Your problem is the copper itself. If you take a look at the flight thruster it is painted black, and i'm quite sure it is not solely to prevent corrosion.
It would be best, to take a cheap matt spray paint to make your measurements perfect!
PS: you may find a setting that accounts for different emission coefficients (epsilon) but if you use any paint it will be close to 1 or 0.95.
Regards,
Sepp -
#1112 by Monomorphic on 10 Apr, 2016 18:34
-
BTW they had spherical end plates.
Do we know this? Nothing I see in any of the drawing indicates they had spherical end-plates. I know it is listed on the wiki, but so are a lot that I know had flat end-plates. -
#1113 by Monomorphic on 10 Apr, 2016 18:42
-
the WR-340 waveguide has standard internal dimensions of 86.36 mm x 43.18 mm (3.400 inches x 1.700 inches) according to
Those dims for the waveguide can't be right. The 8.636 cm is much larger than the frustum height of 6.86. And even with the angle figured in... something is wrong.
-
#1114 by X_RaY on 10 Apr, 2016 18:51
-
Great question!BTW they had spherical end plates.
Do we know this? Nothing I see in any of the drawing indicates they had spherical end-plates. I know it is listed on the wiki, but so are a lot that I know had flat end-plates.
I think it was the first pic below and the suggestion of someone (don't know how exactly).
While looking on the lower pic it seems flat.
https://forum.nasaspaceflight.com/index.php?topic=38203.msg1416868#msg1416868
https://forum.nasaspaceflight.com/index.php?topic=38203.msg1416872#msg1416872
https://forum.nasaspaceflight.com/index.php?topic=37642.msg1411785#msg1411785
-
#1115 by Rodal on 10 Apr, 2016 19:48
-
Yes, we came to the same conclusion, maybe that's why the wiki dimensions are larger.the WR-340 waveguide has standard internal dimensions of 86.36 mm x 43.18 mm (3.400 inches x 1.700 inches) according to
Those dims for the waveguide can't be right. The 8.636 cm is much larger than the frustum height of 6.86. And even with the angle figured in... something is wrong.
Again, Tajmar's et.al. reported numbers are a real mess. -
#1116 by FattyLumpkin on 10 Apr, 2016 20:16
-
Any takers for this downsized NASA frustum at TE012 After reviewing previous posts (other threads) re scaling up and down this should "ring" in this mode.
-
#1117 by X_RaY on 10 Apr, 2016 20:34
-
Any takers for this downsized NASA frustum at TE012 After reviewing previous posts (other threads) re scaling up and down this should "ring" in this mode.
Looks good.for details see here----->https://forum.nasaspaceflight.com/index.php?topic=39772.msg1516895#msg1516895
Using your dimensions I get 2,402521134GHz for TE012
-
#1118 by masterharper1082 on 10 Apr, 2016 20:55
-
@ Monomorphic:
Can confirm this from industry experience - matte black paint will work much better. It is almost impossible to get a good IR photo from bare metal - I've tried with copper, aluminum, and stainless.
No, there is no different setting for the infrared cam. Your problem is the copper itself. If you take a look at the flight thruster it is painted black, and i'm quite sure it is not solely to prevent corrosion.
It would be best, to take a cheap matt spray paint to make your measurements perfect!
PS: you may find a setting that accounts for different emission coefficients (epsilon) but if you use any paint it will be close to 1 or 0.95.
Regards,
Sepp
Sent from my SM-T710 using Tapatalk
-
#1119 by FattyLumpkin on 10 Apr, 2016 20:57
-
Thanks X_Ray, ---am glad it rings at some frequency... am still in a quandary about this though: I clearly recall " if you reduce the dimensions by 10 then you increase the frequency by 10" So if we scaled down frustum X by a factor of ten, then we must increase the frequency by a factor of ten. Therefore a frustum that rings best at 2.45 GHz will ring best at 24.5 GHz if the frustum scaled down by a factor of ten. IS THIS NOT COORECT? OR?
