### Author Topic: EM Drive Developments Thread 1  (Read 1022631 times)

#### aero

• Senior Member
• Posts: 3004
• 92129
• Liked: 801
• Likes Given: 285
##### Re: EM Drive Developments
« Reply #2540 on: 10/24/2014 06:38 PM »
Quote
What is the "DesignFactor" you compute for smallDiameter=0.17 m, bigDiameter=0.28 m and length=0.345 m

For DesignFactor = 0.844  and bigDiameter=0.28 m and length=0.345 m

I get smallDiameter = 0.1289 m which is much larger than the values you showed, and not so far off from 0.17 m
All my dimension estimates were from measuring the Photos. And my design factor equation is not working.
If I plug Lg1=.28 and Lg2=.17 into my current (broken?) spread sheet formula, I get Df =-0.337700803. That is negative and length doesn't enter into my version of the design factor equation.
Retired, working interesting problems

#### Rodal

• Senior Member
• Posts: 5911
• USA
• Liked: 6121
• Likes Given: 5457
##### Re: EM Drive Developments
« Reply #2541 on: 10/24/2014 06:50 PM »
This is the equation (obtained from inverting the Shawyer designFactor formula) for the smallDiameter in terms of the other variables:

If you plug in:

c = 299792458 m/s
rfFrequency = 2.45 * 10^9 1/s
bigDiameter=0.28 m
demoCavityLength=0.345 m
DesignFactorNumber = 0.844

you should get:

smallDiameter = 0.128887 m

Equation  © Rodal 2014
« Last Edit: 10/24/2014 07:09 PM by Rodal »

#### Rodal

• Senior Member
• Posts: 5911
• USA
• Liked: 6121
• Likes Given: 5457
##### Re: EM Drive Developments
« Reply #2542 on: 10/24/2014 07:03 PM »
And this is what a plot of SmallDiameter (m) in the vertical axis vs. DesignFactor in the horizontal axis looks like
Observe that for DesignFactor = 0, SmallDiameter = BigDiameter, which makes sense
« Last Edit: 10/24/2014 07:35 PM by Rodal »

#### aero

• Senior Member
• Posts: 3004
• 92129
• Liked: 801
• Likes Given: 285
##### Re: EM Drive Developments
« Reply #2543 on: 10/24/2014 07:12 PM »
I like the value you calculate for the small diameter. But can that length be right? I suppose so. Using that length, big diameter and the taper from the photo I calculate small diameter about 0.1 meter. Close enough considering the uncertainty in just exactly where Shawyer is measuring when he tells us the diameter of the thruster and we don't know the wall thickness either. The equations are all related to the internal dimensions while photos show external dimensions and it's all based on assuming we know what Shawyer measured to get the published number.

Oh. Would you post your design factor equation in it's normal form? I'd like to put it into my spreadsheet to see what the curve looks like and to confirm your number.
Retired, working interesting problems

#### Rodal

• Senior Member
• Posts: 5911
• USA
• Liked: 6121
• Likes Given: 5457
##### Re: EM Drive Developments
« Reply #2544 on: 10/24/2014 07:14 PM »
I like the value you calculate for the small diameter. But can that length be right? I suppose so. Using that length, big diameter and the taper from the photo I calculate small diameter about 0.1 meter. Close enough considering the uncertainty in just exactly where Shawyer is measuring when he tells us the diameter of the thruster and we don't know the wall thickness either. The equations are all related to the internal dimensions while photos show external dimensions and it's all based on assuming we know what Shawyer measured to get the published number.

Oh. Would you post your design factor equation in it's normal form? I'd like to put it into my spreadsheet to see what the curve looks like and to confirm your number.
What do you mean by normal form of equation?  It looks very normal to me.

If you like to put the equation in some other form, I'll be glad to check your interpretation, if you define your variables so that I understand your interpretation.
« Last Edit: 10/24/2014 07:34 PM by Rodal »

#### aero

• Senior Member
• Posts: 3004
• 92129
• Liked: 801
• Likes Given: 285
##### Re: EM Drive Developments
« Reply #2545 on: 10/24/2014 07:36 PM »
I mean in the form Design factor = expression.
Retired, working interesting problems

#### Rodal

• Senior Member
• Posts: 5911
• USA
• Liked: 6121
• Likes Given: 5457
##### Re: EM Drive Developments
« Reply #2546 on: 10/24/2014 07:47 PM »
I mean in the form Design factor = expression.
Here is Shayer's DesignFactor in terms of the other variables, notice that it goes to zero when BigDiameter=SmallDiameter
« Last Edit: 10/24/2014 07:48 PM by Rodal »

#### Rodal

• Senior Member
• Posts: 5911
• USA
• Liked: 6121
• Likes Given: 5457
##### Re: EM Drive Developments
« Reply #2547 on: 10/24/2014 08:39 PM »
New blog page from Dr. McCulloch explaining Unruh-radiation inertia-modification in simple terms:

http://physicsfromtheedge.blogspot.it/2014/10/mihsc-motion-from-logic.html

#### Rodal

• Senior Member
• Posts: 5911
• USA
• Liked: 6121
• Likes Given: 5457
##### Re: EM Drive Developments
« Reply #2548 on: 10/24/2014 08:48 PM »
Table comparing Shawyer and McCulloch's predictions with measurements

c=299792458 m/s

lengths in meter
frequency in 1/second
power in watts
force in milliNewtons

predicted force (either Shawyer or McCulloch) followed (in parenthesis) by ratio of prediction divided by measurement

ShawyerForce = (2 * PowerInput * Q / c ) * designFactor

McCullochForce = ( PowerInput * Q / c ) * ((cavityLength/smallDiameter) - (cavityLength/bigDiameter))

Force/PowerInput of a Photon Rocket = 1 / c

Therefore, the Force/PowerInput ratios divided by the Force/PowerInput of a Photon Rocket are:

Shawyer/PhtnRckt = 2 * Q * designFactor
McCulloch/PhtnRckt = Q * ((cavityLength/smallDiameter) - (cavityLength/bigDiameter))

(* Shawyer Experimental *)
rfFrequency=2.45*10^9;
cavityLength=0.156;
bigDiameter=0.16;
smallDiameter=0.11;
designFactor= 1.23205

power =  850
Q = 5900

measured force = 16
ShawyerForce = 41 (2.56)
McCullochForce = 7  (0.438)

(* Shawyer Demo *)  BEST PREDICTED VALUES
rfFrequency=2.45*10^9;
cavityLength=0.345;
bigDiameter=0.28;
smallDiameter=0.17;
designFactor =0.400601

power =  1000
Q = 45000

measured force = 147
ShawyerForce =  120 (0.816)
McCullochForce = 123 (0.837)

(* Shawyer Demo *)
rfFrequency=2.45*10^9;
cavityLength=0.345;
bigDiameter=0.28;
smallDiameter=0.129;
designFactor =0.842138

power =  1000
Q = 45000

measured force = 147
ShawyerForce =  253 (1.72)
McCullochForce = 216 (1.47)

All Brady cases have the following dimensions:

cavityLength=0.332;
bigDiameter=0.397;
smallDiameter=0.244;

(* Brady a *)
rfFrequency=1.933*10^9;
designFactor =0.31201

power =   16.9
Q = 7320

measured force =  0.09
ShawyerForce =  0.26 (2.89)
McCullochForce = 0.22 (2.44)

(* Brady b *)  STATISTICAL OUTLIER
rfFrequency=1.937*10^9;
designFactor =0.311024

power = 16.7
Q =  18100

measured force = 0.05
ShawyerForce =  0.63 (12.6)
McCullochForce = 0.53 (10.6)

(* Brady c *)
rfFrequency=1.88*10^9;
designFactor =0.325756

power = 2.6
Q = 22000

measured force = 0.06
ShawyerForce =  0.12 (2.00)
McCullochForce = 0.10 (1.67)

Formula for designFactor in term of other variables
« Last Edit: 10/24/2014 11:50 PM by Rodal »

#### aero

• Senior Member
• Posts: 3004
• 92129
• Liked: 801
• Likes Given: 285
##### Re: EM Drive Developments
« Reply #2549 on: 10/24/2014 08:56 PM »
Using your equation I also get 0.129 m for the demo small end. Now we only need confidence in the 0.345 m length measurement. Where did that value originate?
Retired, working interesting problems

#### frobnicat

• Full Member
• Posts: 518
• Liked: 500
• Likes Given: 151
##### Re: EM Drive Developments
« Reply #2550 on: 10/24/2014 09:41 PM »
Dimensions are bigger than those used in my program. Are things settled ? Sorry I haven't followed the designFactor parameter : "Formula for designFactor in term of other variables" is it dependent or independent ??

I may have time this week-end to do new runs on updated values and added parameters, but I don't want composite (dependent) parameters as this would add duplicates. Independent parameters (like length) can be safely added. If special functions need to be added to generate relevant dependant parameters it's also possible. Can anyone take the responsibility and time to summarize a consensus in the following format ?
(changing the number of columns and labels as see fit). Will make available the results as spreadsheet.

`#define Nrec 6t_data data_in[Nrec] ={    //                                 w_big  w_small  lambda      Q     power   force    {"Shawyer (2008) a",        1.0 ,  16    ,  8    , C/2.45  ,  5900 ,  850   , 16       },    {"Shawyer (2008) b",        1.0 ,  28    ,  4    , C/2.45  , 45000 , 1000   , 214      },    {"Juan (2012) TE011",       1.0 ,  28    ,  4    , C/2.5   , 32000 , 1000   , 214      },    {"Juan (2012) TE012",       1.0 ,  28    ,  4    , C/2.45  , 50000 , 1000   , 315      },    {"Brady et al. (2014) a",   1.0 ,  24.75 , 16.5  , C/1.933 ,  7320 ,   16.9 ,   0.0912 },    {"Brady et al. (2014) c",   1.0 ,  24.75 , 16.5  , C/1.88  , 22000 ,    2.6 ,   0.0554 },};`

I'm trying to build a case for Brady a with thermal air jet : do we agree on inner volume approx. that of a truncated cone length 0.33  big diameter .4 small diameter .24  (I'm not into a few %) that is approximately 27 litres (27000 cm^3), equivalent to a cube of 30cm side ?

#### Rodal

• Senior Member
• Posts: 5911
• USA
• Liked: 6121
• Likes Given: 5457
##### Re: EM Drive Developments
« Reply #2551 on: 10/24/2014 09:43 PM »
.....
I'm trying to build a case for Brady a with thermal air jet : do we agree on inner volume approx. that of a truncated cone length 0.33  big diameter .4 small diameter .24  (I'm not into a few %) that is approximately 27 litres (27000 cm^3), equivalent to a cube of 30cm side ?

I will do what you ask (unless somebody else wants to do it), but I think that it would be better to wait until we settle on dimensions, there are still issues to be discussed with John (both beams) and aero.
« Last Edit: 10/24/2014 09:50 PM by Rodal »

#### frobnicat

• Full Member
• Posts: 518
• Liked: 500
• Likes Given: 151
##### Re: EM Drive Developments
« Reply #2552 on: 10/24/2014 09:46 PM »
I will do what you ask (unless somebody else wants to do it), but I think that it would be better to wait until we settle on dimensions, there are still issues to be discussed with John (both beams) and aero.

All right, better have best data.
« Last Edit: 10/24/2014 09:47 PM by frobnicat »

#### aero

• Senior Member
• Posts: 3004
• 92129
• Liked: 801
• Likes Given: 285
##### Re: EM Drive Developments
« Reply #2553 on: 10/24/2014 10:11 PM »
Ok - What dimensions are we currently questioning?
Retired, working interesting problems

#### Rodal

• Senior Member
• Posts: 5911
• USA
• Liked: 6121
• Likes Given: 5457
##### Re: EM Drive Developments
« Reply #2554 on: 10/24/2014 10:23 PM »
Ok - What dimensions are we currently questioning?
All of them.  Take a look at my table above.  The predictions are closest for the Shayer demo when using the AutoCad drawing from JohnFornaro.

They are off by more than a factor of 2 if we estimate the Shawyer demo small diameter based on the other Fornaro dimensions and the 0.844 designFactor.

John Fornaro based his dimensioning for the NASA Eagleworks on the back beam cross section, and not the front one, so the NASA Eagleworks dimensions should be reviewed.

#### aero

• Senior Member
• Posts: 3004
• 92129
• Liked: 801
• Likes Given: 285
##### Re: EM Drive Developments
« Reply #2555 on: 10/24/2014 10:38 PM »
Ok - What dimensions are we currently questioning?
All of them.  Take a look at my table above.  The predictions are closest for the Shayer demo when using the AutoCad drawing from JohnFornaro.

They are off by more than a factor of 2 if we estimate the Shawyer demo small diameter based on the other Fornaro dimensions and the 0.844 designFactor.

John Fornaro based his dimensioning for the NASA Eagleworks on the back beam cross section, and not the front one, so the NASA Eagleworks dimensions should be reviewed.

Those dimensions do not include length, which Prof. M now needs. I hope your records are better than mine because I don't have lengths recorded, (or the other dimensions, for that matter.)
Retired, working interesting problems

#### Rodal

• Senior Member
• Posts: 5911
• USA
• Liked: 6121
• Likes Given: 5457
##### Re: EM Drive Developments
« Reply #2556 on: 10/24/2014 10:48 PM »
...I hope your records are better than mine because I don't have lengths recorded, (or the other dimensions, for that matter.)
The "search" function of this thread is really awful.  Too bad that this thread does not use Google as a search engine
« Last Edit: 10/24/2014 10:49 PM by Rodal »

#### frobnicat

• Full Member
• Posts: 518
• Liked: 500
• Likes Given: 151
##### Re: EM Drive Developments
« Reply #2557 on: 10/24/2014 10:51 PM »
I use the microwave at home and I have never noticed the air getting warm inside it, unlike the air inside my oven.   That's why to get a crust on a pie, I use the oven.

Yes but you don't use microwave oven empty (you shouldn't) and putting an absorbing water containing thing in it degrades its Q factor.
Quote
The quality factor of an empty microwave oven (30 × 30 × 20 cm^3), with penetration depth of the walls δ≈1μm, is of the order of 10^4, while when we put a glass of water inside the chamber the quality factor is of the order of 10^2. If we put more water in the chamber, the quality factor would be lower and the absorption would be greater.
From this publication page 8.

Can't believe I'm posting about details of household microwave ovens on advanced concepts...

.../...
So, the air inside the EM Drive would have to get heated by convection heat transfer, and the same considerations as in the message http://forum.nasaspaceflight.com/index.php?topic=29276.msg1275630#msg1275630  hold: the copper temperature increases only 1 deg F, and the air needs to get heated by convection from this tiny temperature differential.

If this is the only way air inside cavity can be heated then yes.

Quote
The different consideration, is that rather than relying on natural convection circulation, considering the gas law P V=n R T, and since the volume inside the cavity stays the same, as the temperature of the air increases, the pressure increases, and this may produce an air jet at the gap between the bases and the cone.  This would have the advantage of explaining the force always being directed axially regarding of orientation of the EM Drive.

That's the idea.

Quote
However:

1) it still would not explain the impulsive response in 2 seconds (at NASA Eagleworks) since heating of air inside the cavity due to convection heat transfer is much slower than that

Unless...

Quote
and
2) the temperature rise of only 1 deg F is so tiny, that, without doing any calculation my intuition would be that this would produce a very small change in pressure and probably not enough to have the EM Drive act as a jet.  On the other hand, the forces measured at NASA Eagleworks are also extremely small (50 microNewtons)

That's my intuition also, too small, but working on quantitative estimations to be sure.

Quote
However, although molecules with mirror symmetry like oxygen, and nitrogen have no permanent dipole moments, it is possible to induce a dipole moment by the application of a strong external electric field. This is called polarization and the magnitude of the dipole moment induced is a measure of the polarizability of the molecular species.  One would have to calculate whether the Electric Fields could be strong enough to produce polarization of the air molecules inside the cavity to the extent that the microwave can heat the air molecules so that a pressure would be generated enough to produce a jet with the measured microNewton forces.  Also whether the air inside the cavity could be humid enough to contain enough water molecules for microwave heating to produce this effect.

Can't tell for a polarisation of oxygen and nitrogen, but usual air sure has water content, a good fraction of a % by weight. As water molecules are dispersed they could heat the rest of the air mass, if only they can couple enough to get a fraction of microwave energy. We are inside a cavity that's designed to have a high Q, the losses on the walls are minimised (by being copper of or gold and presumably by being polished) but at the end all the power injected in has to go somewhere. The power pumped into makes a significant stored energy level before losses equal input, photons go back and forth a lot of time (high Q) and have all that much chances to interact with water molecules.

I have a hard time finding accessible information of water vapor coupling with GHz microwaves. There is a nice graph on this wikipedia entry stating that "The pure rotation spectrum of water vapor extends into the microwave region" but the graph is apparently for liquid water (and shows a decrease in coupling when hotter).

So surely water vapor in usual lab air gets heated by microwaves, and this heating (and subsequent heating of air) would be fast, no conduction or convection required + small thermal inertia. By how much would it be heated, in a high Q empty "microwave oven" at about 20 W  (Brady a) is the question I can't answer.

That said, apparently Brady a reported Q of 7320 has nothing exceptional compared to above citation of 10^4 for empty microwave ovens.
« Last Edit: 10/25/2014 10:07 AM by frobnicat »

#### aero

• Senior Member
• Posts: 3004
• 92129
• Liked: 801
• Likes Given: 285
##### Re: EM Drive Developments
« Reply #2558 on: 10/24/2014 11:49 PM »
Just for giggles I calculated the sensitivity (numerically) of Force to the dimensional parameters. I used Prof. M's new equation on Shawyer a" to calculate force. I used delta X = 2%.

dF/dX          dF/dX
for X     mN/meter    mN/cm
=s    47.523     0.475
=w-big    101.979     1.020
=w-small    215.756     2.158

No real surprise but in general centimeter sized errors in dimensions are noticeable.
Retired, working interesting problems

#### Rodal

• Senior Member
• Posts: 5911
• USA
• Liked: 6121
• Likes Given: 5457
##### Re: EM Drive Developments
« Reply #2559 on: 10/25/2014 12:32 AM »
Just for giggles I calculated the sensitivity (numerically) of Force to the dimensional parameters. I used Prof. M's new equation on Shawyer a" to calculate force. I used delta X = 2%.

dF/dX          dF/dX
for X     mN/meter    mN/cm
=s    47.523     0.475
=w-big    101.979     1.020
=w-small    215.756     2.158

No real surprise but in general centimeter sized errors in dimensions are noticeable.

I calculated the derivatives dF/dX for Shawyer's design factor with respect to all the same three variables.

dF/dX (where X is either the smallDiameter or the bigDiameter) goes to infinity as the small diameter approaches the big diameter.

The value of the derivatives depend on all three variables.

Let me know if you are interested, I can post the formulas.

I could plot them, but since the derivatives are a function of all three variables, in a 3-D plot I need to arbitrarily fix one of the variables to an arbitrary value to be able to plot the derivative as a function of the other two.

For example,  I can plot dF/dsmallDiameter as a function of small diameter and big diameter for a given value of cavity length.

The derivatives of Shawyer's design factor have a similar dependence as the ones you found for McCulloch's.
« Last Edit: 10/25/2014 12:53 AM by Rodal »

Tags: