Quote from: TheTraveller on 07/31/2015 05:12 pmQuote from: flux_capacitor on 07/31/2015 02:54 pmI asked Martin Tajmar directly by email about cavity dimensions that would be off by a factor 2, and he replied confirming the numbers were indeed internal radii instead of diameters.He added he already uploaded a revised manuscript altogether with some other typo corrections and some additional clarifications at the AIAA website, but revisions from the conference will appear only after 21st of August.For now, the updated paper is online on the UD-Dresden website.In the updated paper the height is still confirmed to be 68.6 mm:Quote from: Martin TajmarOur 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 mmSo:- internal big diameter = 0.1082 m- internal small diameter = 0.077 m- height = 0.00686 mTo @Rodal, @TheTraveller and others: can you try to find resonances and modes with your COMSOL and spreadsheets programs with those dimensions?No resonance found using those revised numbers.http://forum.nasaspaceflight.com/index.php?topic=37642.msg1410218#msg1410218
Quote from: flux_capacitor on 07/31/2015 02:54 pmI asked Martin Tajmar directly by email about cavity dimensions that would be off by a factor 2, and he replied confirming the numbers were indeed internal radii instead of diameters.He added he already uploaded a revised manuscript altogether with some other typo corrections and some additional clarifications at the AIAA website, but revisions from the conference will appear only after 21st of August.For now, the updated paper is online on the UD-Dresden website.In the updated paper the height is still confirmed to be 68.6 mm:Quote from: Martin TajmarOur 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 mmSo:- internal big diameter = 0.1082 m- internal small diameter = 0.077 m- height = 0.00686 mTo @Rodal, @TheTraveller and others: can you try to find resonances and modes with your COMSOL and spreadsheets programs with those dimensions?No resonance found using those revised numbers.
I asked Martin Tajmar directly by email about cavity dimensions that would be off by a factor 2, and he replied confirming the numbers were indeed internal radii instead of diameters.He added he already uploaded a revised manuscript altogether with some other typo corrections and some additional clarifications at the AIAA website, but revisions from the conference will appear only after 21st of August.For now, the updated paper is online on the UD-Dresden website.In the updated paper the height is still confirmed to be 68.6 mm:Quote from: Martin TajmarOur 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 mmSo:- internal big diameter = 0.1082 m- internal small diameter = 0.077 m- height = 0.00686 mTo @Rodal, @TheTraveller and others: can you try to find resonances and modes with your COMSOL and spreadsheets programs with those dimensions?
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
Quote from: TheTraveller on 07/31/2015 04:55 pmQuote from: rfmwguy on 07/31/2015 03:54 pmThink of it as horsepower in the cavity world. The bigger the better, right? Over the top pronouncements get attention. When anyone claims super high Qs, its all relative to they test methodology they are using in the real world. I thought we had settled how Q was measured.To be very clear, the Chinese, EW, Shawyer and myself are taking unloaded 1 port S11 -3db off the peak return loss dB bandwidths. That is the way the Q is measured for these cavities. It may not be how you would measure the loaded Q but it is the way Q is measured in EMDrives. Shawyers Force equation uses S11 1 port return loss dB driven unloaded Q.Attached is an example of a 1 port S11 return loss Q measurement Paul March posted on NSF. The cavity did not have a dielectric. Clearly Qs of 50k are possible with a plain hand made copper frustum with flat end plates. Curve the end plates and the Q will go higher. Machine the cavity to 0.05mm accuracy and the Q will go higher. Highly polish all the interior surfaces and the Q will go higher."Unloaded" is relative, the port (for S11) measurement have already a 50 Ohm impedance, its design to be almost free of reflections... But i don't know if there's a better way to discover the Q, actually no i think german file with explanations how to dohttp://www-elsa.physik.uni-bonn.de/Lehrveranstaltungen/FP-E106/E106-Erlaeuterungen.pdf
Quote from: rfmwguy on 07/31/2015 03:54 pmThink of it as horsepower in the cavity world. The bigger the better, right? Over the top pronouncements get attention. When anyone claims super high Qs, its all relative to they test methodology they are using in the real world. I thought we had settled how Q was measured.To be very clear, the Chinese, EW, Shawyer and myself are taking unloaded 1 port S11 -3db off the peak return loss dB bandwidths. That is the way the Q is measured for these cavities. It may not be how you would measure the loaded Q but it is the way Q is measured in EMDrives. Shawyers Force equation uses S11 1 port return loss dB driven unloaded Q.Attached is an example of a 1 port S11 return loss Q measurement Paul March posted on NSF. The cavity did not have a dielectric. Clearly Qs of 50k are possible with a plain hand made copper frustum with flat end plates. Curve the end plates and the Q will go higher. Machine the cavity to 0.05mm accuracy and the Q will go higher. Highly polish all the interior surfaces and the Q will go higher.
Think of it as horsepower in the cavity world. The bigger the better, right? Over the top pronouncements get attention. When anyone claims super high Qs, its all relative to they test methodology they are using in the real world.
Quote from: X_RaY on 07/31/2015 05:17 pmQuote from: TheTraveller on 07/31/2015 04:55 pmQuote from: rfmwguy on 07/31/2015 03:54 pmThink of it as horsepower in the cavity world. The bigger the better, right? Over the top pronouncements get attention. When anyone claims super high Qs, its all relative to they test methodology they are using in the real world. I thought we had settled how Q was measured.To be very clear, the Chinese, EW, Shawyer and myself are taking unloaded 1 port S11 -3db off the peak return loss dB bandwidths. That is the way the Q is measured for these cavities. It may not be how you would measure the loaded Q but it is the way Q is measured in EMDrives. Shawyers Force equation uses S11 1 port return loss dB driven unloaded Q.Attached is an example of a 1 port S11 return loss Q measurement Paul March posted on NSF. The cavity did not have a dielectric. Clearly Qs of 50k are possible with a plain hand made copper frustum with flat end plates. Curve the end plates and the Q will go higher. Machine the cavity to 0.05mm accuracy and the Q will go higher. Highly polish all the interior surfaces and the Q will go higher."Unloaded" is relative, the port (for S11) measurement have already a 50 Ohm impedance, its design to be almost free of reflections... But i don't know if there's a better way to discover the Q, actually no i think german file with explanations how to dohttp://www-elsa.physik.uni-bonn.de/Lehrveranstaltungen/FP-E106/E106-Erlaeuterungen.pdfThe defacto way to measure unloaded Q in the EMDrive world is via 1 port S11 -3db off the peak return loss dBs. Like it or not, it is the way the measurement is done.
Quote from: X_RaY on 07/31/2015 05:22 pmQuote from: TheTraveller on 07/31/2015 05:12 pmQuote from: flux_capacitor on 07/31/2015 02:54 pmI asked Martin Tajmar directly by email about cavity dimensions that would be off by a factor 2, and he replied confirming the numbers were indeed internal radii instead of diameters.He added he already uploaded a revised manuscript altogether with some other typo corrections and some additional clarifications at the AIAA website, but revisions from the conference will appear only after 21st of August.For now, the updated paper is online on the UD-Dresden website.In the updated paper the height is still confirmed to be 68.6 mm:Quote from: Martin TajmarOur 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 mmSo:- internal big diameter = 0.1082 m- internal small diameter = 0.077 m- height = 0.00686 mTo @Rodal, @TheTraveller and others: can you try to find resonances and modes with your COMSOL and spreadsheets programs with those dimensions?No resonance found using those revised numbers.http://forum.nasaspaceflight.com/index.php?topic=37642.msg1410218#msg1410218 The length/height I used was twice what was quoted. The cavity at 68.6mm is way too short to resonant at 2.45GHz.I suggest Tajmar needs some help that knows what they are doing.
I am with you at this point like i've said"But i don't know if there's a better way to discover the Q, actually no i think "
The quality factor of this resonator under no load can be calculated by the following equation:Qu=∫|H|2dv/h/2∫|nxH|2ds+tgd∫|H|2dv = (14)Where tg is the electric loss within the cavity, n is the normal vector of the wall, s is the cavity surface area, v is the volume of the cavity.
Quote from: X_RaY on 07/31/2015 05:34 pmI am with you at this point like i've said"But i don't know if there's a better way to discover the Q, actually no i think "Prof Yang has an equation to calc unloaded Q but I have never been able to get it working. It is in the 2010 paper attached.QuoteThe quality factor of this resonator under no load can be calculated by the following equation:Qu=∫|H|2dv/h/2∫|nxH|2ds+tgd∫|H|2dv = (14)Where tg is the electric loss within the cavity, n is the normal vector of the wall, s is the cavity surface area, v is the volume of the cavity.If anyone else can get it to work, please share how you worked it out.
Quote from: TheTraveller on 07/31/2015 05:42 pmQuote from: X_RaY on 07/31/2015 05:34 pmI am with you at this point like i've said"But i don't know if there's a better way to discover the Q, actually no i think "Prof Yang has an equation to calc unloaded Q but I have never been able to get it working. It is in the 2010 paper attached.QuoteThe quality factor of this resonator under no load can be calculated by the following equation:Qu=∫|H|2dv/h/2∫|nxH|2ds+tgd∫|H|2dv = (14)Where tg is the electric loss within the cavity, n is the normal vector of the wall, s is the cavity surface area, v is the volume of the cavity.If anyone else can get it to work, please share how you worked it out.what is "tgd"? couldn't find a definition in the paper
Quote from: X_RaY on 07/31/2015 05:49 pmQuote from: TheTraveller on 07/31/2015 05:42 pmQuote from: X_RaY on 07/31/2015 05:34 pmI am with you at this point like i've said"But i don't know if there's a better way to discover the Q, actually no i think "Prof Yang has an equation to calc unloaded Q but I have never been able to get it working. It is in the 2010 paper attached.QuoteThe quality factor of this resonator under no load can be calculated by the following equation:Qu=∫|H|2dv/h/2∫|nxH|2ds+tgd∫|H|2dv = (14)Where tg is the electric loss within the cavity, n is the normal vector of the wall, s is the cavity surface area, v is the volume of the cavity.If anyone else can get it to work, please share how you worked it out.what is "tgd"? couldn't find a definition in the papertg is defined. Equation attached. Would be really good to get this working as then we can see the Q change as the length and end plate diameters change. So can tune the cavity dimensions for both high Df and high Q. Magic.
The idea that someone might put their hand on Yang's device and "feel it push" put me in mind of a story from many years ago—the Dean Drive.In the 1950s, well-known science fiction author and editor John W. Campbell claimed to have encountered a "reactionless drive" created by a man named Dean."In particular, Campbell said he had seen this thing sit on a bathroom scale; it weighed, say, nine pounds when it was at rest. When it was turned on -- by plugging in an ordinary quarter-inch electric drill that was incorporated into the gadget -- gears turned, weights whirled, and the scale indicated a weight of perhaps 8.5 pounds! This is an impressive thrust; if you could apply continuously that much thrust (1/18 g!) in free fall you would be able to colonize the solar system. The Dean Drive would be the key to space travel even if it never produced enough thrust to actually lift itself."The device was later shown to aerospace engineer and author, G. Harry Stine, who swore he felt it push against his hand.However, Dean was very secretive about his device. He purposely fudged diagrams given to others, even taking out patents that didn't resemble the device he demonstrated. After he died, the device was never found. I've read accounts from Campbell, Stine and Pournelle, and much of their experience with Dean comes to mind when I look at Shawyer.But I often find myself remembering what Stine said. "I felt it push."http://www.jerrypournelle.com/science/dean.html
Quote from: TheTraveller on 07/31/2015 05:55 pmQuote from: X_RaY on 07/31/2015 05:49 pmQuote from: TheTraveller on 07/31/2015 05:42 pmQuote from: X_RaY on 07/31/2015 05:34 pmI am with you at this point like i've said"But i don't know if there's a better way to discover the Q, actually no i think "Prof Yang has an equation to calc unloaded Q but I have never been able to get it working. It is in the 2010 paper attached.QuoteThe quality factor of this resonator under no load can be calculated by the following equation:Qu=∫|H|2dv/h/2∫|nxH|2ds+tgd∫|H|2dv = (14)Where tg is the electric loss within the cavity, n is the normal vector of the wall, s is the cavity surface area, v is the volume of the cavity.If anyone else can get it to work, please share how you worked it out.what is "tgd"? couldn't find a definition in the papertg is defined. Equation attached. Would be really good to get this working as then we can see the Q change as the length and end plate diameters change. So can tune the cavity dimensions for both high Df and high Q. Magic.The tg term (tg is the "electric loss within the cavity") is not included in the classic electromagnetic theory of resonant cavities since for standing waves the energy is such that when B is max, E is zero, and vice versa, the energy goes from B to E to B for standing waves. For standing waves all you need to consider is B and the surface losses (eddy currents) due to B. The inclusion of the term tg is due to her implication that the fields inside the cavity are not standing waves. She doesn't clarify why except to ask readers to consider ions inside the cavity. She NEVER gives an explicit equation for tg, she just gives the words "electric loss within the cavity" And what is the "d" after tg ? the other d's are differentials. It cannot be a differential here otherwise it would negate the integral that follows it. Is that "d" a typo?The term including tg doesn't make sense to me. If tg is due to electric losses it should belong in the denominator, with the losses, it should not be added !Which means that she is probably missing a parenthesis, and the tg term should be added to the denominator of the previous term:Qu=∫(|H|^2) dv/(2 h∫(|nxH|^2) ds+tgd ∫(|H|^2) dv ) This shows a very sloppy paper and very poor "peer review" Then this becomes the standard equation to calculate Q, the equation present in Wikipedia:It is what I used to calculate Q as well.
...May be the change of the tg if the dimensions are changed so not tgd but dtg?Don't know if this make much more sense jet...Would be the loss over the surface or volume integral
Quote from: TheTraveller on 07/31/2015 05:30 pmQuote from: X_RaY on 07/31/2015 05:17 pmQuote from: TheTraveller on 07/31/2015 04:55 pmQuote from: rfmwguy on 07/31/2015 03:54 pmThink of it as horsepower in the cavity world. The bigger the better, right? Over the top pronouncements get attention. When anyone claims super high Qs, its all relative to they test methodology they are using in the real world. I thought we had settled how Q was measured.To be very clear, the Chinese, EW, Shawyer and myself are taking unloaded 1 port S11 -3db off the peak return loss dB bandwidths. That is the way the Q is measured for these cavities. It may not be how you would measure the loaded Q but it is the way Q is measured in EMDrives. Shawyers Force equation uses S11 1 port return loss dB driven unloaded Q.Attached is an example of a 1 port S11 return loss Q measurement Paul March posted on NSF. The cavity did not have a dielectric. Clearly Qs of 50k are possible with a plain hand made copper frustum with flat end plates. Curve the end plates and the Q will go higher. Machine the cavity to 0.05mm accuracy and the Q will go higher. Highly polish all the interior surfaces and the Q will go higher."Unloaded" is relative, the port (for S11) measurement have already a 50 Ohm impedance, its design to be almost free of reflections... But i don't know if there's a better way to discover the Q, actually no i think german file with explanations how to dohttp://www-elsa.physik.uni-bonn.de/Lehrveranstaltungen/FP-E106/E106-Erlaeuterungen.pdfThe defacto way to measure unloaded Q in the EMDrive world is via 1 port S11 -3db off the peak return loss dBs. Like it or not, it is the way the measurement is done.Its an invented technique that any RF engineer will say is nonconformal. An EM drive is an RF device and should be tested with integrity as opposed to unconventional methodology. IOW this is specsmanship. In legitimate RF systems, Qs of 100K+ will be viewed as nonsensical to the engineering community.Therefore, it is settled, Shawyer and Yang are likely using unconventional methodology to define Q and should consult with the British Standards Institute or other reputable body to resolve their specification claims.
Quote from: TheTraveller on 07/31/2015 05:30 pmQuote from: X_RaY on 07/31/2015 05:17 pmQuote from: TheTraveller on 07/31/2015 04:55 pmQuote from: rfmwguy on 07/31/2015 03:54 pmThink of it as horsepower in the cavity world. The bigger the better, right? Over the top pronouncements get attention. When anyone claims super high Qs, its all relative to they test methodology they are using in the real world. I thought we had settled how Q was measured.To be very clear, the Chinese, EW, Shawyer and myself are taking unloaded 1 port S11 -3db off the peak return loss dB bandwidths. That is the way the Q is measured for these cavities. It may not be how you would measure the loaded Q but it is the way Q is measured in EMDrives. Shawyers Force equation uses S11 1 port return loss dB driven unloaded Q.Attached is an example of a 1 port S11 return loss Q measurement Paul March posted on NSF. The cavity did not have a dielectric. Clearly Qs of 50k are possible with a plain hand made copper frustum with flat end plates. Curve the end plates and the Q will go higher. Machine the cavity to 0.05mm accuracy and the Q will go higher. Highly polish all the interior surfaces and the Q will go higher."Unloaded" is relative, the port (for S11) measurement have already a 50 Ohm impedance, its design to be almost free of reflections... But i don't know if there's a better way to discover the Q, actually no i think german file with explanations how to dohttp://www-elsa.physik.uni-bonn.de/Lehrveranstaltungen/FP-E106/E106-Erlaeuterungen.pdfThe defacto way to measure unloaded Q in the EMDrive world is via 1 port S11 -3db off the peak return loss dBs. Like it or not, it is the way the measurement is done.Its an invented technique that any RF engineer with say is nonconformal. An EM drive is an RF device and should be tested with integrity as opposed to unconventional methodology. IOW this is specsmanship. In legitimate RF systems, Qs of 100K+ will be viewed as nonsensical to the engineering community.Therefore, it is settled, Shawyer and Yang are likely using unconventional methodology to define Q and should consult with the British Standards Institute or other reputable body to resolve their specification claims.
More thoughts on getting an experiment that actually provides some useful information and order of magnitude above background.. do we really need to fire the magnetron for 120 seconds if we go to higher power? (i.e. if we get a pulsed magnetron at 1MW, fire it for mili-seconds at resonance (TM010)) we should see something; correct?). This ironically is easier to cool and implement than 100kW for 2 minutes.Also is there any thought to modelling resonance with ANSYS HFFS vs. COMSOL? Anyone on here any good at ANSYS HFFS?
Hmmm, If Shawyer and Yang are using non standard methodology to figure out Q. What does this non standard methodology tell us. For example is there a difference in result between the Shawyer methodology and the standard methodology? If so, with what parameters does the Q result begin to diverge? What is the implication of this divergence?Also, and most interestingly. Why has Shawyer and Yang opted for this non standard methodology?