Here’s to successfully modified frustum which now happily resonates at 2,331 MHz. I removed 8 mm from the small end, resulting in a shorter central length and a bigger diameter of the small end, thus pushing the small end cut-off frequency further down and away from resonance. New cut-off freq is 2,256 MHz which is 75 MHz below test frequency.
These resonance cavities turn out to be remarkably predictable. Simulating new dimensions (in COMSOL) was showing a freq shift of +20 Mhz (2323 MHz-> 2343 Mhz). Actual as-measured freq moved by +19 MHz (2312 MHz->2331 Mhz). 80(?) years-old technologies rule.
New dimensions:
D_big: 264 mm (as before)
D_small: 162 mm (+4 mm)
L_center: 196 mm (-8 mm)
TE012 freq (MHz): Simulated: 2343, Actual: 2331.
Df: 0.69
Small end cut-off: 2,256 MHz.
Q factor (at -3 dB S11): 2300 (went down from the original 3100, not sure if this is due to coupling mismatch under new dimensions, or oxidation from minor torch work, or just aging since the last time it was measured).
If anyone knows why this frustum should not be producing thrust then speak now or forever hold your peace.
Ladies and Gentleman, you can now make your bets. (I have already simulated mode frequency shifts for difference thicknesses of HDPE disks at the small end, so this gives a good hint as to what my own prediction for this upcoming test is…)
How have you ensured the joint from end plate to side wall will not introduce issues with eddy currents not being able to occur across the joint?
In all 3 of Roger's EmDrive builds there are flanges and many bolts to ensure a high pressure physical joint like that which occurs when jointing waveguide flanges.
Shell and others achieved this by soldering the entire end plate to side wall joint. A bad joint here may be what is causing low Q.
Have you also ensured there are no scratches or finger prints on all the internal surfaces and that everything has a mirror like finish?
Here’s to successfully modified frustum which now happily resonates at 2,331 MHz. I removed 8 mm from the small end, resulting in a shorter central length and a bigger diameter of the small end, thus pushing the small end cut-off frequency further down and away from resonance. New cut-off freq is 2,256 MHz which is 75 MHz below test frequency.
...
If anyone knows why this frustum should not be producing thrust then speak now or forever hold your peace.
Ladies and Gentleman, you can now make your bets. (I have already simulated mode frequency shifts for difference thicknesses of HDPE disks at the small end, so this gives a good hint as to what my own prediction for this upcoming test is…)
How have you ensured the joint from end plate to side wall will not introduce issues with eddy currents not being able to occur across the joint?
In all 3 of Roger's EmDrive builds there are flanges and many bolts to ensure a high pressure physical joint like that which occurs when jointing waveguide flanges.
Shell and others achieved this by soldering the entire end plate to side wall joint. A bad joint here may be what is causing low Q.
Have you also ensured there are no scratches or finger prints on all the internal surfaces and that everything has a mirror like finish?
Ok... so not only one has to worry about the cut-off freq of the small end, but also about currents through the wall joints at TE012 mode? At which side? I thought Shell only had one side connected, and the other one floating... Has she chosen the right side? And we don't know the exact number of bolts Roger used in his designs... And my frustum most certainly does not have all the same scratches at all the right places... There is no hope.
You're right though. Bolts are very helpful to "ensure a high pressure physical joint". High air pressure and a small gap somewhere are absolutely key here. All I need to do to see "thrust" is to use some caulk on those wall plate gaps. Just not on all of them.
...This is a rather sad conclusion to arrive to. It is very logical though taking into account the entire story. Well, there still exists the EW claim that one needs "either dielectric or a magnetron" (further details or justification be damned), so maybe those using a magnetron will be more lucky.
Here’s to successfully modified frustum which now happily resonates at 2,331 MHz. I removed 8 mm from the small end, resulting in a shorter central length and a bigger diameter of the small end, thus pushing the small end cut-off frequency further down and away from resonance. New cut-off freq is 2,256 MHz which is 75 MHz below test frequency.
...
If anyone knows why this frustum should not be producing thrust then speak now or forever hold your peace.
Ladies and Gentleman, you can now make your bets. (I have already simulated mode frequency shifts for difference thicknesses of HDPE disks at the small end, so this gives a good hint as to what my own prediction for this upcoming test is…)
How have you ensured the joint from end plate to side wall will not introduce issues with eddy currents not being able to occur across the joint?
In all 3 of Roger's EmDrive builds there are flanges and many bolts to ensure a high pressure physical joint like that which occurs when jointing waveguide flanges.
Shell and others achieved this by soldering the entire end plate to side wall joint. A bad joint here may be what is causing low Q.
Have you also ensured there are no scratches or finger prints on all the internal surfaces and that everything has a mirror like finish?
Ok... so not only one has to worry about the cut-off freq of the small end, but also about currents through the wall joints at TE012 mode? At which side? I thought Shell only had one side connected, and the other one floating... Has she chosen the right side? And we don't know the exact number of bolts Roger used in his designs... And my frustum most certainly does not have all the same scratches at all the right places... There is no hope.
You're right though. Bolts are very helpful to "ensure a high pressure physical joint". High air pressure and a small gap somewhere are absolutely key here. All I need to do to see "thrust" is to use some caulk on those wall plate gaps. Just not on all of them.
...This is a rather sad conclusion to arrive to. It is very logical though taking into account the entire story. Well, there still exists the EW claim that one needs "either dielectric or a magnetron" (further details or justification be damned), so maybe those using a magnetron will be more lucky.
What I shared is just part of the check list requirements I have set for my build. You may be able to get some thrust BUT you need to figure out how to get a decent loaded Q.
Calc the skin depth and then imagine your whole interior surface as a very thin skin depth film that needs to be a contiguous non broken film to allow the high Q resonant wave to develop and generate thrust. Any breaks in that skin depth thick contiguous film (which may be scratches) may destroy Q and thrust.
I don't make the skin depth thick, non broken film rule. Maybe ask a microwave engineer what will happen if you losely join waveguide flanges or leave a bit of dust or dirt on the pressure faces? With our frustums there is a lot of end plate to side wall surface to ensure, at the interior surface, there is no join line to dispute eddy currents across the joint.
My goal is to achieve the specific thrust of 385mN/kW that Roger predicted for my frustum. Which means I need to do everything possible to achieve the predicted unloaded Q of approx 88k.
Dr. Rodal: thanks very much for correcting my misunderstanding re: frustrum wall currents, I appreciate the time you took to do so.
Would Ohmic heating still need to reconcile with the (induced) wall currents? More directly, can you give us all the scale of the expected wall currents in a 100W RF frustrum? I'm reluctant to re-visit the back of my envelope for a while, as you might imagine. I followed the link to the post you gave, but didn't spot anything which looked like it would answer the question directly.
R.
Just finished reading of Star-Drive post on page 41, thread 5 again.
http://forum.nasaspaceflight.com/index.php?topic=38577.800I was wondering if he might tell us more on the Lorentz force issue, but I guess we need to wait three more months for the peer paper.
Mr. Star-Drive can you give some pointers to the people here or is it forbidden to do that at the moment? It seems you have gone silent once again after some of your posts reached the news again...
Here’s to successfully modified frustum which now happily resonates at 2,331 MHz. I removed 8 mm from the small end, resulting in a shorter central length and a bigger diameter of the small end, thus pushing the small end cut-off frequency further down and away from resonance. New cut-off freq is 2,256 MHz which is 75 MHz below test frequency.
...
If anyone knows why this frustum should not be producing thrust then speak now or forever hold your peace.
Ladies and Gentleman, you can now make your bets. (I have already simulated mode frequency shifts for difference thicknesses of HDPE disks at the small end, so this gives a good hint as to what my own prediction for this upcoming test is…)
How have you ensured the joint from end plate to side wall will not introduce issues with eddy currents not being able to occur across the joint?
In all 3 of Roger's EmDrive builds there are flanges and many bolts to ensure a high pressure physical joint like that which occurs when jointing waveguide flanges.
Shell and others achieved this by soldering the entire end plate to side wall joint. A bad joint here may be what is causing low Q.
Have you also ensured there are no scratches or finger prints on all the internal surfaces and that everything has a mirror like finish?
Ok... so not only one has to worry about the cut-off freq of the small end, but also about currents through the wall joints at TE012 mode? At which side? I thought Shell only had one side connected, and the other one floating... Has she chosen the right side? And we don't know the exact number of bolts Roger used in his designs... And my frustum most certainly does not have all the same scratches at all the right places... There is no hope.
You're right though. Bolts are very helpful to "ensure a high pressure physical joint". High air pressure and a small gap somewhere are absolutely key here. All I need to do to see "thrust" is to use some caulk on those wall plate gaps. Just not on all of them.
...This is a rather sad conclusion to arrive to. It is very logical though taking into account the entire story. Well, there still exists the EW claim that one needs "either dielectric or a magnetron" (further details or justification be damned), so maybe those using a magnetron will be more lucky.
I thought Shell only had one side connected, and the other one floating... Has she chosen the right side?
First off good luck. To address your question both of my endplates are electrically connected. The large bottom plate is soldered and then a bead of silver epoxy to seal it airtight.
The small top endplate floats is right but it floats on a Beryllium flexible gasket.
Here are a few on ebay.
http://www.ebay.com/sch/i.html?_odkw=Beryllium+Copper+EMI+Gasket+Strip&_osacat=0&_from=R40&_trksid=m570.l1313&_nkw=Beryllium+Copper+EMI+Gasket+Strip&_sacat=0(mod corrected url)
This does several things. Allows me to keep the small endplate conductive to the frustum walls while sliding up and down inside of the tune chamber and also allows heated air out of the cavity through the gaps. The very top of the tune cavity is sealed off and I have a small airline connected to it running back down the beam to vent heated air.
I really want to be able to tune the chamber through the resonance points plotting any thrusts I get and put this cutoff or no cutoff issue to bed. This means an extended full power run. Currently I've been working on a little DC variable speed motor on the end of the micrometer at the large end to slowly slide the small plate up and down.
Good luck on your run and your data. Very nice work.
Shell
Here’s to successfully modified frustum which now happily resonates at 2,331 MHz. I removed 8 mm from the small end, resulting in a shorter central length and a bigger diameter of the small end, thus pushing the small end cut-off frequency further down and away from resonance. New cut-off freq is 2,256 MHz which is 75 MHz below test frequency.
...
If anyone knows why this frustum should not be producing thrust then speak now or forever hold your peace.
Ladies and Gentleman, you can now make your bets. (I have already simulated mode frequency shifts for difference thicknesses of HDPE disks at the small end, so this gives a good hint as to what my own prediction for this upcoming test is…)
How have you ensured the joint from end plate to side wall will not introduce issues with eddy currents not being able to occur across the joint?
In all 3 of Roger's EmDrive builds there are flanges and many bolts to ensure a high pressure physical joint like that which occurs when jointing waveguide flanges.
Shell and others achieved this by soldering the entire end plate to side wall joint. A bad joint here may be what is causing low Q.
Have you also ensured there are no scratches or finger prints on all the internal surfaces and that everything has a mirror like finish?
Ok... so not only one has to worry about the cut-off freq of the small end, but also about currents through the wall joints at TE012 mode? At which side? I thought Shell only had one side connected, and the other one floating... Has she chosen the right side? And we don't know the exact number of bolts Roger used in his designs... And my frustum most certainly does not have all the same scratches at all the right places... There is no hope.
You're right though. Bolts are very helpful to "ensure a high pressure physical joint". High air pressure and a small gap somewhere are absolutely key here. All I need to do to see "thrust" is to use some caulk on those wall plate gaps. Just not on all of them.
...This is a rather sad conclusion to arrive to. It is very logical though taking into account the entire story. Well, there still exists the EW claim that one needs "either dielectric or a magnetron" (further details or justification be damned), so maybe those using a magnetron will be more lucky.
What I shared is just part of the check list requirements I have set for my build. You may be able to get some thrust BUT you need to figure out how to get a decent loaded Q.
Calc the skin depth and then imagine your whole interior surface as a very thin skin depth film that needs to be a contiguous non broken film to allow the high Q resonant wave to develop and generate thrust. Any breaks in that skin depth thick contiguous film (which may be scratches) may destroy Q and thrust.
I don't make the skin depth thick, non broken film rule. Maybe ask a microwave engineer what will happen if you losely join waveguide flanges or leave a bit of dust or dirt on the pressure faces? With our frustums there is a lot of end plate to side wall surface to ensure, at the interior surface, there is no join line to dispute eddy currents across the joint.
My goal is to achieve the specific thrust of 385mN/kW that Roger predicted for my frustum. Which means I need to do everything possible to achieve the predicted unloaded Q of approx 88k.
Here is a thermal pic of Tajmar's build. It's self explanatory. He had a dismal Q of 50.
New Report of Anomalous Thrust from EmDrive Replicator (Artefact)
I saw this report yesterday that reported See-Shell's report of thrust. Artefact is the English verison of Artifact. 
http://www.e-catworld.com/2016/01/04/new-report-of-anomalous-thrust-from-emdrive-replicator-artefact/
Mr Li,
I have my suspicion (only) that the source of that post in an lenr forum is a poster from another emdrive forum. That emdrive forum just posted a link to the article asking the rhetorical question if it is good that lenr people are taking an interest In emdrive. There is no common technology between the two. My belief is its mischief.
NSFs policy is not to mix the two topics and I'm sure the lenr forum's management feels the same way...IF it is moderated. The downfall of many forums is the lack of rules and moderation. There is little I need to do here because users self-moderate. If they cannot, I step in. If that doesn't work, staff steps in. When they do, its usually a quick resolution.
p.s. thanks for the review of kurts setup. It still surprises me that 2mN of lorentz could occur, but you are our resident lorentz expert and we listen carefully
(edit - corrected to mN)
...
p.s. thanks for the review of kurts setup. It still surprises me that 2mg of lorentz could occur, but you are our resident lorentz expert and we listen carefully
typo: it is 2mN, instead of 2 mg in Zellerium's test (in http://emdrive.wiki/Experimental_Results)
101.97 times larger
Doc, you got me again...2 mN is correct...guess I picked the wrong week to stop sniffing glue*
*(a quote from the movie Airplane for those needing assistance)
Never give away the movie in the same post.
Mr Li,
I have my suspicion (only) that the source of that post in an lenr forum is a poster from another emdrive forum. That emdrive forum just posted a link to the article asking the rhetorical question if it is good that lenr people are taking an interest In emdrive. There is no common technology between the two. My belief is its mischief.
...
It seems nobody caught the joke so an explanation is needed. I posted the link because it was supposed to be humorous. This is because "Artefact" was the author of a comment to another post that got picked up by Frank Acland and posted there. It is clearly stated on that page that "The following comment was posted by Artefact", and a google search ( artefact see-shell site:e-catworld.com, second hit ) revealed that user "Artefact" posted the comment at this page:
http://www.e-catworld.com/2013/10/26/always-open-e-cat-world-thread/comment-page-64/It seems you may need to "load more comment" until page 64 to reach that comment though.
http://www.sciencedirect.com/science/article/pii/S0370269315008023
In conclusion, our analysis shows that the quantum nature of cosmological spacetime unavoidably affects the propagation of test particles, producing (apparent) Lorentz-violating effects.
Interesting concept.
Shell
Nice article shell. Lets face it, "old" physics won't get us to the stars, might as well investigate "new" physics
Here’s to successfully modified frustum which now happily resonates at 2,331 MHz. I removed 8 mm from the small end, resulting in a shorter central length and a bigger diameter of the small end, thus pushing the small end cut-off frequency further down and away from resonance. New cut-off freq is 2,256 MHz which is 75 MHz below test frequency.
These resonance cavities turn out to be remarkably predictable. Simulating new dimensions (in COMSOL) was showing a freq shift of +20 Mhz (2323 MHz-> 2343 Mhz). Actual as-measured freq moved by +19 MHz (2312 MHz->2331 Mhz). 80(?) years-old technologies rule.
New dimensions:
D_big: 264 mm (as before)
D_small: 162 mm (+4 mm)
L_center: 196 mm (-8 mm)
TE012 freq (MHz): Simulated: 2343, Actual: 2331.
Df: 0.69
Small end cut-off: 2,256 MHz.
Q factor (at -3 dB S11): 2300 (went down from the original 3100, not sure if this is due to coupling mismatch under new dimensions, or oxidation from minor torch work, or just aging since the last time it was measured).
If anyone knows why this frustum should not be producing thrust then speak now or forever hold your peace.
Ladies and Gentleman, you can now make your bets. (I have already simulated mode frequency shifts for difference thicknesses of HDPE disks at the small end, so this gives a good hint as to what my own prediction for this upcoming test is…)
This is great !!!This is truly scientific experimentation ! We finally have a test to verify or nullify a formal hypothesis ! (*)
It looks like we are going to have a test (when considering also your prior test) to verify or nullify Shawyer's strange prescription that the cut-off equations (we learnt at school only apply to open waveguides) somehow applies to a closed cavity like the EM Drive (*).
Looking forward to the experimental results !!!!
______
(*) To this date I have not seen any experimental verification (no tests) in any report by Roger Shawyer to support his cut-off prescription
Also notice:
TE012 freq (MHz): Simulated: 2343, Actual: 2331 Which makes sense as the Finite Element method converges from above, always stiffer than the converged solution, and therefore higher natural frequency than the converged solution
It is also interesting that
Q factor (at -3 dB S11): 2300 (Q went down from the original 3100, as the small diameter was increased above cut-off)
....
If anyone knows why this frustum should not be producing thrust then speak now or forever hold your peace.
Ladies and Gentleman, you can now make your bets. (I have already simulated mode frequency shifts for difference thicknesses of HDPE disks at the small end, so this gives a good hint as to what my own prediction for this upcoming test is…)
Here is my bet: No thrust. Why? Because I believe EmDrive effect does not exist.
....
If anyone knows why this frustum should not be producing thrust then speak now or forever hold your peace.
Ladies and Gentleman, you can now make your bets. (I have already simulated mode frequency shifts for difference thicknesses of HDPE disks at the small end, so this gives a good hint as to what my own prediction for this upcoming test is…)
Here is my bet: No thrust. Why? Because I believe EmDrive effect does not exist.
What are you betting ? How much $$$ ?
Is someone assigning odds? to properly value the bets and payoffs?
Are there puts and calls options as well ? I bet that there are many betting on straddles
I bet he sees something. No straddler here.
This retired lady here could use a little extra cash. My $2.00 PowerBall was a bust (who would have figured) and the way I see it, this is much much better odds at winning.
Shell
....
If anyone knows why this frustum should not be producing thrust then speak now or forever hold your peace.
Ladies and Gentleman, you can now make your bets. (I have already simulated mode frequency shifts for difference thicknesses of HDPE disks at the small end, so this gives a good hint as to what my own prediction for this upcoming test is…)
Here is my bet: No thrust. Why? Because I believe EmDrive effect does not exist.
And my bet, while the same, is slightly more optimistic: it seems that the only existing report to date of somebody being able to observe thrust from an RF cavity without dielectric and without using 700W of
heat DC power RF magnetron energy is Cannae Superconducting. They have (presumably) got like 10 mN from 10W at 10M Q. Well, if the effect does exist and it scales linearly with Q then the most optimistic estimate for one operating with 30W and 10K unloaded Q is ... ~30 uN
This is most likely below the resolution of my setup.
However, I am now pretty convinced (ok, not yet but I most likely will be after the next test) that any effects on the order of ~400 mN / KW cannot be replicated under pure single freq RF power using solely information available in the public domain. Note I am not saying they don't exist
I am saying there is no compelling reason to believe that they exist as of today...
EDIT: the above should read "... any effects on the order of ~400 mN / KW / ~50K of unloaded Q
This is most likely below the resolution of my setup.