-
#2180 by Rodal on 17 Oct, 2014 17:31
-
@aero McCulloch's table
Demonstrator by ratio
w_big = 0.28 meters 0.28 meters AGREE
w_small = 0.18375 meters 0.04 meters DISAGREE
height = 0.2975 meters
Demonstrator by photo
w_big = 0.28 meters 0.28 meters AGREE
w_small = 0.0778 meters 0.04 metersDISAGREEWhatever works
height = 0.381 meters
Experimental by photo
w_big 0.16 meters 0.16 metersDISAGREAgree
w_small 0.0778 meters 0.08 meters AGREE
height 0.177 meters
You don't think Shawyer just extended the big end of the Experimental model to make the Demonstrator model, do you?
For the Experimental Thruster, using Fig. 6 of Shawyer's report (http://www.emdrive.com/IAC-08-C4-4-7.pdf) I get:
Large Base Diameter = 16 cm = 0.16 m
Small Base Diameter = 11 cm = 0.11 m
Length = 15.8 cm = 0.158 m
Where is the photograph you used for Shawyer's demonstrator drive? -
#2181 by aero on 17 Oct, 2014 17:36
-
Quote
Where is the photograph you used for Shawyer's demonstrator drive?
At the bottom of the page here: http://emdrive.com/ -
#2182 by Rodal on 17 Oct, 2014 17:43
-
@aero McCulloch's table
Demonstrator by ratio
w_big = 0.28 meters 0.28 meters AGREE
w_small = 0.18375 meters 0.04 meters DISAGREE
height = 0.2975 meters
Demonstrator by photo
w_big = 0.28 meters 0.28 meters AGREE
w_small = 0.0778 meters 0.04 metersDISAGREEWhatever works
height = 0.381 meters
Experimental by photo
w_big 0.16 meters 0.16 metersDISAGREAgree
w_small 0.0778 meters 0.08 meters AGREE
height 0.177 meters
You don't think Shawyer just extended the big end of the Experimental model to make the Demonstrator model, do you?
For the Shawyer Demonstrator Thruster, using the picture at the bottom of the page here: http://emdrive.com/ I get:
Large Base Diameter = 28 cm = 0.28 m
Small Base Diameter = 15.9 cm = 0.159 m (2 times @aero's estimate)
Length = 22.6 cm = 0.226 m
RelativeRatio = Large Base Diameter/Small Base Diameter = 1.76
Symmetrized Measure of Base Diameter Difference = ((RR-1)+(1/RR-1))= 0.329
(1/Small Base Diameter - 1/Large Base Diameter)=2.72 (1/m)
I think that the only length that is relevant in a truncated cone analysis is the length of the truncated cone itself and not the length of the cylinder attached to it, so "Length" means Length of truncated cone
For the Experimental Thruster, using Fig. 6 of Shawyer's report (http://www.emdrive.com/IAC-08-C4-4-7.pdf) I get:
Large Base Diameter = 16 cm = 0.16 m
Small Base Diameter = 11 cm = 0.11 m (+41% greater than @aero's estimate)
Length = 15.8 cm = 0.158 m (-11% less than @aero's estimate)
RelativeRatio = Large Base Diameter/Small Base Diameter = 1.45
Symmetrized Measure of Base Diameter Difference = ((RR-1)+(1/RR-1))= 0.142
(1/Small Base Diameter - 1/Large Base Diameter)=2.84 (1/m)
-
#2183 by frobnicat on 17 Oct, 2014 17:54
-
Without Brady b, Here we go :
6 entries
Thresholds : mean=0.50 stddev=0.35
a b L Q P F c mean stddev
---------------------------------------------------------------------
a b L Q P F c exterm mean stddev
---------------------------------------------------------------------
a^0 b^-2 L^1 Q^1 P^1 F^-1 c^-1 (a^-1 + b^-1)^-1 0.49 0.29
a^2 b^-1 L^1 Q^1 P^1 F^-1 c^-1 (a^ 1 + b^ 1)^-2 0.19 0.27
a^1 b^-1 L^2 Q^1 P^1 F^-1 c^-1 (a^ 1 + b^ 1)^-2 -0.43 0.24
a^-1 b^-2 L^-1 Q^1 P^1 F^-1 c^-1 (a^-2 + L^-2)^-2 0.24 0.31
a^-2 b^-2 L^0 Q^1 P^1 F^-1 c^-1 (a^-2 + L^-2)^-2 -0.37 0.16
a^-2 b^-2 L^2 Q^1 P^1 F^-1 c^-1 (a^-2 + L^-2)^-1 -0.10 0.16
a^-1 b^-2 L^1 Q^1 P^1 F^-1 c^-1 (a^-1 + L^-1)^-2 -0.08 0.26
a^-1 b^-2 L^2 Q^1 P^1 F^-1 c^-1 (a^-1 + L^-1)^-1 0.35 0.20
a^0 b^-2 L^1 Q^1 P^1 F^-1 c^-1 (b^-1 + L^-1)^-1 0.33 0.26
a^-1 b^-2 L^2 Q^1 P^1 F^-1 c^-1 (b^-1 + L^-1)^-1 -0.28 0.35
a^-1 b^1 L^2 Q^1 P^1 F^-1 c^-1 |a^-2 - b^-2|^1 -0.07 0.29
a^0 b^0 L^1 Q^1 P^1 F^-1 c^-1 |a^-1 - b^-1|^1 0.24 0.25 <-
a^-1 b^0 L^2 Q^1 P^1 F^-1 c^-1 |a^-1 - b^-1|^1 -0.38 0.17
a^-1 b^-2 L^2 Q^1 P^1 F^-1 c^-1 sqrt(b^-1 L^-1)^-1 0.49 0.29
Checked : 94140625
Validated : 21769
Checked is number of all the combinations, Validated is number of formulas that respect dimensional consistency for kg m s units. Mean and standard deviation thresholds are lowered to yield approximately same number of selected formulas as previous run with all seven data points.
Relative values for MiHsC
(a^0 b^0 L^1 Q^1 P^1 F^-1 c^-1 |a^-1 - b^-1|^1 ) / mean value (without log)
0.63 1.45 1.01 1.10 1.12 0.88
As per order Shawyer a & b, Juan 11 & 12, Brady a & c
This is quite good except Shawyer's results : should we skip those two outliers ?
Rodal you are right that Q is more significant without Brady b : the Qless F = 13100 P/c (a-b)^2/(ab) no longer appears easily when relaxing mean constraint up to 12 (fudge factor in 163000 and 1/163000).
a^1 b^1 L^0 Q^0 P^1 F^-1 c^-1 |a^-1 - b^-1|^2 -9.63 0.46
is to be found amongst 200 formulas of similar or better levels of deviation.
It is not robust. But I still like it !
Waiting for adjusted inputs...
-
#2184 by aero on 17 Oct, 2014 17:58
-
Quote
I think that the only length that is relevant is the length of the truncated cone itself and not the length of the cylinder attached to it, so "Length" means Length of truncated cone
I don't know. I speculate that knowing the big diameter, we might be able to back the small diameter out of Shawyer's performance model using his published data. He does say that the thrust agrees with his performance model and it does use diameters of the ends of the cavity.
-
#2185 by Rodal on 17 Oct, 2014 17:58
-
...
This is quite good except Shawyer's results : should we skip those two outliers ?
.....
What statistical argument can you use to state that those two are outliers? -
#2186 by Notsosureofit on 17 Oct, 2014 18:08
-
Quote
Where is the photograph you used for Shawyer's demonstrator drive?
At the bottom of the page here: http://emdrive.com/
Don't know if it's been mentioned, but the demo unit has a tuning plunger in the straight section. You can see the motor and drive gears etc hanging out on the rear. -
#2187 by frobnicat on 17 Oct, 2014 18:10
-
...
This is quite good except Shawyer's results : should we skip those two outliers ?
.....
What statistical argument can you use to state that those two are outliers?
It was a sarcastical argument, not a statistical argument. I'm not seriously considering to lower the data count to 4 samples.
-
#2188 by Rodal on 17 Oct, 2014 18:13
-
...
We still would like to hear whether there is an argument that can be made, explaining the formulas that model the experimental results either as:
Waiting for adjusted inputs...
A) an experimental artifact
B) a photon rocket
You can use the whole data (including the outlier) if you like to make the argument. For the photon rocket argument I don't understand how the photons get out of the EM Drive and how does it get to do better than a perfect photon rocket.
Can you make an experimental artifact argument? -
#2189 by Rodal on 17 Oct, 2014 18:15
-
OK, but I wear my humorous hat only with Kernosabe
It was a sarcastical argument, not a statistical argument. I'm not seriously considering to lower the data count to 4 samples....
This is quite good except Shawyer's results : should we skip those two outliers ?
.....
What statistical argument can you use to state that those two are outliers?
But seriously, that Brady (b) is a statistical outlier on the basis of its deviation from the statistical cluster formed by the data without it.
-
#2190 by Rodal on 17 Oct, 2014 18:30
-
Quote
I think that the only length that is relevant is the length of the truncated cone itself and not the length of the cylinder attached to it, so "Length" means Length of truncated cone
I don't know. I speculate that knowing the big diameter, we might be able to back the small diameter out of Shawyer's performance model using his published data. He does say that the thrust agrees with his performance model and it does use diameters of the ends of the cavity.
Do you agree that there is a cylindrical section to Shawyer's Demonstrator EM Drive that has the same diameter throughout? and that the diameter of this cylindrical section is the same as the diameter of the small base of the truncated cone? And therefore the smaller diameter at the end of the cavity is the same as the diameter of the small base of the truncated cone?
Did I miss something ?
If my logic above is not faulty, then our large discrepancy (a factor 2 between us ) is a matter of measurement and scaling.
One of us must be closest to the real answer ...
-
#2191 by Rodal on 17 Oct, 2014 19:27
-
Another explanation for the EM Drive experimental findings based on a theory that allows electromagnetic fields to modify the space-time metric of General Relativity:
http://inspirehep.net/record/1220790/files/arXiv%3A1302.5690.pdf
http://arxiv.org/pdf/1310.5029.pdf
"Some axisymmetric electromagnetic modes of a truncated conical cavity are then presented
and used as source in the weak-field approximation of the equations, previously obtained,
to determine the force on the cavity. It is found that a coupling of the same magnitude
as used in [5] between the scalar and the electromagnetic field results in a correct
magnitude and sign for the forces reported in asymmetric resonant cavities. ...The theory, however, does not seem to be completely satisfactory because in its linearized version it also predicts strong gravitational effects
by the Earth’s magnetic field, which are clearly not observed. A possible resolution of
this problem is considered in the last section."
"It was shown that the weak field approximation of a rather general scalar-tensor theory of
gravity, which includes an additional scalar with strong coupling to the electromagnetic
field, as proposed in[5], could account for the forces reported on asymmetric resonant
cavities. Although highly speculative, it is interesting that this was done using the
same coupling coefficient adjusted by[5] to explain discordant measurements of Newton
gravitational constant. It is also of interest that the inclusion of the external scalar
can help to reconcile the Solar System tests with values of the Brans-Dicke parameter
close to unity (see relation (15)). The weakest part of the theory seems to be that
there is no clear way of preventing large gravitational effects due to the magnetic field
of the Earth, as predicted by Eq. (17). A possible solution can be sought in non-linear
effects,...Note that if this is possible for a static magnetic field, it could possibly not be the
case for a static electric field outside its sources, because the difference of sign would not allow real solutions.
It is so expected that static magnetic and electric fields show no unusual
gravitational effects, while non-stationary electromagnetic fields do. "
-
#2192 by aero on 17 Oct, 2014 19:43
-
Quote
I think that the only length that is relevant is the length of the truncated cone itself and not the length of the cylinder attached to it, so "Length" means Length of truncated cone
I don't know. I speculate that knowing the big diameter, we might be able to back the small diameter out of Shawyer's performance model using his published data. He does say that the thrust agrees with his performance model and it does use diameters of the ends of the cavity.
Do you agree that there is a cylindrical section to Shawyer's Demonstrator EM Drive that has the same diameter throughout? and that the diameter of this cylindrical section is the same as the diameter of the small base of the truncated cone? And therefore the smaller diameter at the end of the cavity is the same as the diameter of the small base of the truncated cone?
Did I miss something ?
If my logic above is not faulty, then our large discrepancy (a factor 2 between us and a factor of 4 between my estimate and McCulloch's estimate) is a matter of measurement and scaling. Of the photograph you and I are using is not the Demonstrator Drive, and Prof. McCulloch has better information as to what the dimensions of the Demonstrator Drive are?
One of us must be closest to the real answer ...
I don't know. I did post asking Prof M if he had better information. I also emailed Shawyer and Juan's university asking the same question. We will see if anyone responds.
I had assumed that the tapered cavity extended within the cylindrical section but you may be right. However, as was pointed out by Notsosureofit, it looks like there is some sort of mechanism in the cylindrical section. If there is a moveable small end, then I can't guess what the interior of the cylinder looks like. -
#2193 by Notsosureofit on 17 Oct, 2014 19:53
-
The upper gizmo is the motor and at the bottom it's driving a variable resistor as position feedback. It probably was fed w/ a fixed frequency generator.
Hanging out the back is prob a limit switch. -
#2194 by Rodal on 17 Oct, 2014 19:57
-
....
Prof. McCulloch has answered that "I think I estimated 4cm using the design factor, probably wrongly. Your photographic method is better"
I don't know. I did post asking Prof M if he had better information. I also emailed Shawyer and Juan's university asking the same question. We will see if anyone responds.
I had assumed that the tapered cavity extended within the cylindrical section but you may be right. However, as was pointed out by Notsosureofit, it looks like there is some sort of mechanism in the cylindrical section. If there is a moveable small end, then I can't guess what the interior of the cylinder looks like.
So it is up to us now to decide. Can you please double check your diameter estimate? I double checked mine and it is 2 times your estimate for the small base of the Demonstrator -
#2195 by Notsosureofit on 17 Oct, 2014 20:11
-
For the O.D. of the copper straight section I get 18.8 cm if that helps (using the pot as ref)
15.3 cm using the RF connector, but not sure of the right connector and worse camera angle, sooo ? -
#2196 by Rodal on 17 Oct, 2014 20:14
-
For the O.D. of the copper straight section I get 18.8 cm if that helps (using the pot as ref)
Can you also please estimate the OD of the big diameter base end?
(Using your same method) -
#2197 by Notsosureofit on 17 Oct, 2014 20:28
-
For the O.D. of the copper straight section I get 18.8 cm if that helps (using the pot as ref)
Can you also please estimate the OD of the big diameter base end?
(Using your same method)
29.6 cm for the pot
24.1 cm for the connector
still might have some camera distortion in that picture
(of course I'm comparing to parts I have here) -
#2198 by RotoSequence on 17 Oct, 2014 20:46
-
Another explanation for the EM Drive experimental findings based on a theory that allows electromagnetic fields to modify the space-time metric of General Relativity:
http://inspirehep.net/record/1220790/files/arXiv%3A1302.5690.pdf
http://arxiv.org/pdf/1310.5029.pdf
The proposed experiment looks exceptionally simple. Has this experiment been conducted? If there were positive results, the fringe forums would be ablaze with excitement over the possibilities.
I suspect that an experiment may have already been performed in the ten months since the paper was published, but due to the unpopularity of null results in scientific publishing, the results haven't made it very far. -
#2199 by Rodal on 17 Oct, 2014 20:57
-
Another explanation for the EM Drive experimental findings based on a theory that allows electromagnetic fields to modify the space-time metric of General Relativity:
http://inspirehep.net/record/1220790/files/arXiv%3A1302.5690.pdf
http://arxiv.org/pdf/1310.5029.pdf
The proposed experiment looks exceptionally simple. Has this experiment been conducted? If there were positive results, the fringe forums would be ablaze with excitement over the possibilities.
Don't know, but I agree that the experiment is simple and if it would be positive we should have heard about it.
Also, if one is going to posit an interaction of the EM Drive magnetic field, the first interaction one should explore is the interaction between the microwave cavity's magnetic field with the Earth's magnetic field. This should be proportional to the Power Input. One could also posit PowerInput*Q. What I haven't determined is magnetic interaction with the Earth's field producing the required magnitude of forces, the direction, and the dependence on the diameters of the bases. Waiting for frobnicat to help me with that...
