Could a Gyroscopic inertial thruster ever work?

[M Drive]

"A kid can stop the movement of the swing by shifting its center of mass"

Fair enough. It's pretty prevalent in my machine though. It can both accelerate and decelerate swing speed by doing cycles. However, when you turn off the gyroscopes (video 21) that phenomenon more or less disappears. This means that the pendulum test should be pretty easy to pass, and I probably do pass it in one or more of these videos.

"Can you do a quick overview which video contains what?"

Sure. Video 9 (00009.MTS) is a short introduction. Showing the setup and so on.

In 11 I try to get it to swing and then get it to stop swinging by doing cycles either when it's swinging to the left or right. At the end of the video I have to abort it as something went wrong with the gyros.

In 12 I do the same thing. Because I do cycles during the entire forward swing (left), it's hard to analyze. I also seem to stop it manually. Still shows the "braking" though.

In 13 I attempt to pass the pendulum test. I only do cycles when the dot is in the far most left position, meaning the dot should be easy to track once it passes over the middle line. At 1:25 I stop doing cycles and just let the camera run for 13 minutes, waiting for the dot to stop by itself. The camera is only able to record 15 minutes at a time, and the dot doesn't come to a complete standstill before that time, unfortunately.

In 14 I did the same thing, with the difference that I tried "braking" the backwards swing, meaning you wouldn't have to wait so long for the dot to come to a standstill. I started breaking at 1:56.

In 15 I tried seeing if the dot would move more to the right than left for some reason. Eventually I just try to see how much further the dot moves when you add cycles. Turns out it's 2-3 cm each cycle. Pretty significant.

In 16 I tried passing the pendulum test again, but I had to cancel both runs that showed the dot (zoomed in). You could probably still analyze the footage I got even though even though I never waited for the dot to come to a standstill.

In 17 I got a new laser pointer! This one is interesting and well worth analyzing. Dot is clearly visible and I try to make it easy to track. I basically avoid doing (shaky) cycles when the dot is about to cross the middle line. A lot of the video is just waiting for the dot to come to a standstill.

In 18 I just try to make the swing as big as possible, then stop the swing doing cycles when it moves backwards.

In 21 the gyros are off. I try to get it to swing and then get it to brake, but it's not really cooperating.

What program do you use for tracking? I don't really have one yet.

Edited April 10, 2014 by M Drive

[N_las]

Ok. Today I can analyse video 17. I do it as good as I can, and I will send you all results. If video 17 shows a pass of the pendulum test, I can do the same for other videos next week. If not, than i won't put more work into it.

Edit:

I only have the free software 'tracker'. But thats only to get the Data from pixel position of the dot in the video. To make the analysis of the data I use Matlab.

Edited April 10, 2014 by N_las

[Alchemist]

I see what gyroscopes do here: when you turn the cycling on, precession forces the gyroscopes to swing forward, moving the center of mass forward relatively to the platform - so the platform is shifted backwards and so the strings produce a bit more forward force (for the same center of mass position).

When you keep it on during forward swing and off during backward swing, you some more forward force during forward swing that accelerates the device, but no extra force during backward swings. When you inverse this, you get forward force during backward swing and so it slows the device down.

If you switch the gyroscopes off there is no center of mass shift due to precession and the effect doesn't work.

Unfortunately, that effect utilizes gravity and string reaction and is just what a child can do on a swing

[M Drive]

I've analyzed a couple of cycles in video 17 already.

I started at frame 1272 because at that point in the video the swinging (I only do one cycle at the end of each forward swing) had become stable. I figure if I count the amount of frames the dot stays on the left and right sides and compare them, you'll get a pattern.

Frames on the left. "1272 to left" means the dot passed the middle line going left at frame 1272. "61 frames until" means it spent 61 frames on the left side of the middle line before crossing over to the right side.

1272 to left 61 frames until

1333 to right 57 frames until

1390 to left 60 frames until

1450 to right 44 frames until

1494 to left 69 frames until

1563 to right 51 frames until

1614 to left 60 frames until

1674 to right 58 frames until

1732 to left 58 frames until

1790 to right 58 frames until

1848 to left 58 frames until

1906 to right 63 frames until

1969 to left 53 frames until

2022 to right 58 frames until

2080 to left 58 frames until

2138 to right 54 frames until

2192 to left 61 frames until

2253 to right 58 frames until

2311 to left

I used the free program forevid to see frames and skip forward frame by frame. http://www.forevid.org/

Now you can see that the average amount of frames that the dot stayed on the left side is higher than the average amount of frames it stayed on the right. All you have to do is calculate the average, so for the left side it's 61+60+69 ..... divided by 9. And for the right it's 57+44+51 .... divided by 9.

This gives you 59,7777 for the left side and 55,6666 for the right side. The dot spends 9.3% more time on the left side than the right.... assuming I did the math right.

Alchemist: Except I don't only do cycles when it's at the rightmost position. And the center of mass does shift when the gyros are off if you check out video 21. However, the fact that it can swing isn't the proof I'm after. Also, the fact that it can't get swing momentum nearly as well as when the gyros are on (see video 21) is pretty interesting.

Edited April 10, 2014 by M Drive

[N_las]

Well M-drive it isn't sufficent to only count the frames. For the average position, the distance from the middle is important to. Imagine the dot is 1 frame 10 cm left of the middle, and 2 frames 5 cm right of the middle. If you only see the number of frames, you think the dot wasn't in the middle on average, but of course it was.

The real analysis is loads more work than counting frames.

[M Drive]

The pendulum experiment I've had explained to me simply states that a light (laser or other) attached to the machine and pointed down needs to stay on one side more than the other. That can be determined by simply recording at what frames the light passes the middle line and counting the amount of frames it stays on either side.

Anyone else know?

[K^2]

This gives you 59,7777 for the left side and 55,6666 for the right side. The dot spends 9.3% more time on the left side than the right.... assuming I did the math right.

And the standard deviation is 4.24 on the left and 5.45 on the right. Both of which are greater than your difference.

As a matter of fact, if I take each of these pairs, subtract left - right, and take standard deviation of that, which is statistically more meaningful, since amplitude is going to change through the experiment, I get standard deviation of 8.08. That's twice the average of this difference. This is what I basically expect from random chance.

If you think this is a real, systematic difference, and if this statistics persists, you'd need to repeat this whole experiment about 40 more times. If this is a real effect, that will bring the error down, putting your result outside of 3 sigmas. But it's looking random so far, and I'm pretty sure the source of randomness are oscillations of the base and not any sort of bias.

Edited April 10, 2014 by K^2

[N_las]

The pendulum experiment I've had explained to me simply states that a light (laser or other) attached to the machine and pointed down needs to stay on one side more than the other. That can be determined by simply recording at what frames the light passes the middle line and counting the amount of frames it stays on either side.

Anyone else know?

'Stay on one side more than the other' is a simplification. The average position is the important point.

And as K^2 pointed out, you have to take statistics into account. The correct analysis of an experiment is usually much more work than the performing the experiment. I will post my results in a few hours.

[M Drive]

K^2: Yes, after today I'm even more convinced it's real. The fact that it easily accelerates in only one direction, just like it does on the ground, starting and stopping a swinging motion with ease only motivates me further. Not to mention the fact that this ability almost completely disappears when you turn the gyros off. I wasn't once able to "brake" it with the gyros off, and the way it just wouldn't "do" anything with the gyros off, even though it's supposed to mimic a person swinging on an actual swing... yeah, I don't really see why I should have much of any doubt.

I completed counting the frames on video 17 if anyone wants.

[K^2]

The fact that it easily accelerates in only one direction

That's actually entirely expected of your setup. It's pretty typical of the way most people swing on the swing set.

And again, the critical test is that of average displacement. Your setup does not pass that test.

[M Drive]

About average displacement. In 3 of the videos from today, I only do cycles for a short time, then literally leave it on for 6-14 minutes, just swinging back and forth (no cycles), waiting for it to stop due to air resistance (or whatever resistance the nylon strings would put up).

Doesn't this mean the average will be greatly reduced if I account for all that dead time? I mean, given infinite time, the displacement would be literally zero, even if it had a jet engine strapped to it.

How do I avoid this?

[K^2]

You can count the average on just the time the machine is running. The idle time can (and, in fact, should) be used to check the neutral position.

[Technical Ben]

A swinging arm rotating on a grooved pivot will swing one way easier than the other, due to friction acting stronger in one direction. The energy being turned into heat etc when against the "grain". So careful what you conclude from "strange" results. 100% of every experiment, observation and calculation so far has shown there is no such thing as free energy, and no such thing as a propellent less propulsion (if that's the correct technical term).

[TheGatesofLogic]

A swinging arm rotating on a grooved pivot will swing one way easier than the other, due to friction acting stronger in one direction. The energy being turned into heat etc when against the "grain". So careful what you conclude from "strange" results. 100% of every experiment, observation and calculation so far has shown there is no such thing as free energy, and no such thing as a propellent less propulsion (if that's the correct technical term).

Not entirely correct, see this here: http://www.emdrive.com/yang-juan-paper-2012.pdf but unfortunately no similar principle can be applied in the instance of gyroscopic inertial thrust, since it is by definition a closed system, and can't exploit relativistic principles the way this can. Any forces you may think you are creating in this experiment will result in nothing but mechanical strains without a net gain in momentum.

[Z-Man]

'Stay on one side more than the other' is a simplification. The average position is the important point.

Just posting to agree with this. It is quite possible to stay on one side of the center line 90% of the time without producing thrust.

And if you want to be really, really precise, it also is possible to create a slight deviation of the average position if you shift a mass up and down a lot in sync with the pendulum motion. But the effect is very small and I don't think it can contribute here.

Sadly, MEGA does not like me very much and I can't watch the videos. Here's how I would deal with the averaging of the dead time at the end: Take K^2's suggestion to use the dead time at the end for center calibration. For the actual measurement, average over individual periods, that is from one time where the dot passes the center line from left to right to the next. That will give you a nice series of numbers you can analyze statistically. Do they sum up to something statistically significantly different from zero? There are bound to be deviations from zero for each individual period due to your device not being in resonance with the pendulum frequency (presumably), it is important that they enter the analysis, and they do that way.

[N_las]

Ok, I am done with my analysis. The brigth laser dot in video 17 was easy to track. It was very hard to put the perspective into account and calculate accurate length values. Keep in mind that the distance between two drawn lines is 20 mm.

I took the phase after the machine was turned off to calculate the rest position of the dot: 3.0±0.9 mm to the right of the middle line. (so 3.3 ÃÆ’ from the line) It could very well be that the rest position IS at the middle line, but the video doesn't allow for a better estimation.

The median position of the laser dot (during the time the machine was on) was 9.3±1.8 mm TO THE RIGHT of the middle line.

Taking both values into account, the average deflection from the rest position (during the time the machine was on) was 6.3±2.0 mm TO THE RIGHT. This is just 3.2 ÃÆ’ (borderline) from the rest position, so one could argue if the measurment is significant or if its just an statistical artifact.

In summary:

Maybe there is a very small effect that favors a position to the right of the line (maybe has something to do with the cable of the controls). But there is no clear proof, that there even is a deflection.

There is clearly no evidence whatsoever (in video 17) that the machine has an average deflection to the left (the machine is intended to deflect to the left). So there is no evidence for a reactionless propulsion.

There is a small margin of error, because i didn't tracked the dot until the very end, because even with autotracker it would take to much time. ( the autotracker isn't reliable and needs babysitting). But I tracked EVERY frame of the machine-on-phase and every frame of a good portion of the machine-off-phase.

Sometimes the dot leaves the picture. This also creates a small error.

If anybody is interested, i can upload a table containing all tracked x-pixel and y-pixel values, the corresponding time values and my calculated real-x values (in mm, taking the perspective into account).

EDIT:

Here a link of my tracking Data, if you compare with the video in real time, you see it is accurate.

http://imgur.com/a/yLBvD/embed

Edited April 10, 2014 by N_las

[M Drive]

Wow, thanks so much N_las!

The fact that the median position of the dot is to the right of the line, even if the machine is designed to "thrust" to the left is easily explained and supported by the graph.

You can clearly see the wave being 'cut off' as a cycle (push) is being performed in the graph. This is because the dot doesn't track the center of mass, but rather the position of the wagon. The center of mass is static relative to the wagon, but only when no cycles are being performed.

As a cycle is being performed the gyros will shift forward, causing the wagon to move back as a reaction. And as the laser pointer is attached to the wagon, of course the laser pointer will move to the right as the gyros move left (generally speaking).

The machine obviously gains momentum, as you can see the waves getting bigger and bigger. The question is whether this is due to normal swinging caused by shifting the center of gravity or something else. My argument for it being something else is video 21, where I'm able to mimic the motion the gyros do when they're on, basically moving the center of gravity in the same way. If you look at all the videos, you'll see a pretty noticeable difference from when the gyros are on.

What I really need to do is do more runs with the gyros off to have something to compare with.

"If anybody is interested, i can upload a table containing all tracked x-pixel and y-pixel values"

Sure! The more data I have at hand the better.

Edited April 11, 2014 by M Drive

[M Drive]

Uploaded them all to Youtube as one big video.

Not going to implement it to the forum as the timestamps are only available on Youtube (press "Show more").

[Technical Ben]

Not entirely correct, see this here: http://www.emdrive.com/yang-juan-paper-2012.pdf but unfortunately no similar principle can be applied in the instance of gyroscopic inertial thrust, since it is by definition a closed system, and can't exploit relativistic principles the way this can. Any forces you may think you are creating in this experiment will result in nothing but mechanical strains without a net gain in momentum.

An electric drive still has "propellent" if I count magnetic fields and other types of "energy", as it can be considered in quantum packets and fields. It still follows the same restrictions, we just put the size of the propellent below what we can see with the naked eye (not literally, but it's close enough an explanation).

That is, a real Electro Magnetic drive still expels energy. We could even build a photon drive I guess, but it still expels photons. A closed system drive is impossible.

[TheGatesofLogic]

not quite, the device in this study does not interact with an external electromagnetic field to impart momentum, but rather gains momentum from a system which ought to be closed from a newtonian standpoint. The thrust force appears when newtonian mechanics are overtaken by special relativity due to light being the resonant working-fluid. Other than this no known system can operate and produce momentum gain without imparting momentum on some external energy medium or mass. The key difference here is that this does not have ANY reactant, and no unequal radiation pressure produced by the system would have the ability to impart this kind of thrust on the testing device.

[M Drive]

I'm still amazed by video 18 and 21 in the video I posted above. The difference is significant.

In video 18 (starts at 25:52) you see the machine has no problem starting a swing, and then stopping a swing, all within a few cycles. Yet, in video 21 (starts at 30:01) when I turn the gyros off this ability is almost completely lost.

I'm thinking if I just gain a little height on the wires, I'll be able to keep the center of gravity on one side of the line at all times. The longer the wires, the slower the swing. The dot moves back about 2-4 centimeters each time a cycle is being performed, and gains upwards of 2-3cm in the forward swing (as can be seen in N_las graph). I might be able to squeeze in another cycle if I do cycles continuously.

I've also figured out that in video 17 (the one N_las tracked the dot on), while it may have passed the pendulum test, the experiment is skewed to favor the dot being on the right side. First off, I didn't run the gyros at their optimal speed (16 volts instead of 18.4 volts) for some reason, but also, because the laser pointer isn't mounted near the center of gravity (when the gyros are in their starting position), it'll move to the right side of the line prematurely if the machine is wobbling (twisting).

I figure I should do at least one "control experiment" with the gyros off for every experiment with them on. I can also have the gyros in the forward position (like in video 21) and have them precess towards the screwdriver, just like they are when the gyros are off. All I really have to do is reverse the rotation direction of the screwdriver, and the gyros will want to precess towards it. This means the center of gravity of the gyros will move in more or less exactly the same way both when the gyros are on and off, meaning it'll be easier to see the differences!

[N_las]

I've also figured out that in video 17 (the one N_las tracked the dot on), while it may have passed the pendulum test, the experiment is skewed to favor the dot being on the right side. First off, I didn't run the gyros at their optimal speed (16 volts instead of 18.4 volts) for some reason, but also, because the laser pointer isn't mounted near the center of gravity (when the gyros are in their starting position), it'll move to the right side of the line prematurely if the machine is wobbling (twisting).

I don't now how you come to the conclusion, that it may have passed the pendulum test. Don't you think, that if your machine was capable of producing thrust, that there was a very significant displacement to the left? For example, if I would place a hair dryer on the pendulum, that would produce thrust from its air blowing capacity, it would easily have a displacement of over 10 ÃÆ’.

From my analysis, it isn't even clear if there IS any displacement to the right. There could be, but it may be just as well a statistical error.

EDIT:

Your whole taking about gyro off and on: It really doesn't matter. The effect in which the machine builds up the swing momentum might be a very complicated combination of things, that depend on the gyro. So it may be lost if the gyros are off.

But thats like placing a disabled kid on a swing, and measure the swing movement. And then place an abled kid on the swing, and measure a greater movement. It dosen't have anything to do with reactionless propulsion. The one kid just is a better swinger.

Edited April 12, 2014 by N_las

[M Drive]

Didn't you read my reply to you yesterday? Post 342, second post on this page?

Counting frames, that is, the amount of time the dot spends on the left side, is a viable method for calculating displacement. Calculating median position of the laser dot isn't, at least not in this version of the M Drive. (Z-man did bring up that it's possible to keep the dot on one side 90% of the time, but I assume it means shifting the center of gravity in a way so as to fool the dot. The dot is static in relation to the center of gravity when there are no cycles being performed on my machine though.)

This version of the M Drive requires the wagon part (that the laser pointer is attached to) to move. A hair dryer or fan attached to the wagon wouldn't. If I were to construct another version of the M Drive with 4 gyros, where the other 2 rotate the opposite direction and move back as the other 2 moves forward, and does this in perfect sync, then it'd act like if you attached a hair dryer to the wagon, and the dot from the laser pointer would actually measure the center of gravity.

Ironically what this means is that if you can't be sure there's displacement to the right, then it's almost guaranteed to have displacement (of the center of gravity) to the left.

Edited April 12, 2014 by M Drive

[rtxoff]

one just simply can not produce thrust without rockeeets :-)

seriously... there is no impuls without sacrifying mass....

it's simple as that... that machine would't do anything beside wobbling in a 0g vacuum enviroment....

[N_las]

I don't now what to tell you, but thats not how evidence works.

You can't start with:

' There might be a small displacement to the right ' And ' I may have an explanation for the displacment to the right '

And conclude:

' There is almost guaranteed a displacement to the left '

You have to make it more like this. Start with:

' There might be a small displacement to the right ' And ' I may have an explanation for the displacment to the right '

Now:

' I try to proof that the explanation for the displacement to the rigth is correct', ' I try to calculate the real average displacement if the explanation is taken into account.'

If then a significant average displacement to the left is observed, you have good evidence that you passed the pendulum test.

You said:

'Counting frames, that is, the amount of time the dot spends on the left side, is a viable method for calculating displacement.'

But you lie to yourself. Counting frames CAN'T be a viable methode for calculating a displacement, since the time the dot spends on any side hasn't directely to do with the displacement.

And even if counting the frames would be a viable methode. From the Data you provided a few posts earlier, the dot is 59.8±1.4 frames on the left side, and 55.7±1.8 frames on the right side. Yet in your post and in the describtion of the youtube video, you claim the dot spends 9.3% more time on the left side than the right. You do realize that the data doesn't show that, do you? The difference between left and right is 4.1±2.3 frames (ÃÆ’ = 1.8) That means there is NO evidence for a difference between the frame number. (For comparison, effects measured at CERN do need a ÃÆ’>5 to be consider real. That would mean the frame difference between left and right should be something like 4.1±0.8 to support your claim, that there is a difference in left and right side. But ÃÆ’>3 can be used as very weak evdence)

If you counted the whole video, you can post the data and I can make the calculations for you, to set your mind at ease. But as it looks now, frame counting dosen't support your machine.

And the standard deviation is 4.24 on the left and 5.45 on the right. Both of which are greater than your difference.

K^2 made an error here. He calculated the corrected sample standard deviation and got the right result.

http://en.wikipedia.org/wiki/Standard_deviation#Corrected_sample_standard_deviation

But one have to divide by the square of the sample size again, to get the standart error of the mean.

http://en.wikipedia.org/wiki/Standard_error_of_the_mean#Standard_error_of_the_mean

But as shown above, that doesn't change that your frame counting doesn't show a difference between left and right

EDIT:

And don't you think, that if your machine works, there would be a huge displacement to the left, that would easily overshadow the incorrect movement of the dot in relation to the center of mass? The average displacement to the right was 6.3±2.0 mm. Don't you think that a real working machine would show very very clear results? As it stands, we are arguing over displacements in the range of mm. Do you really think the effect your machine has (if it works) would be so insignificant? Or are you saying to yourself the displacement to the left is easily like a few cm, but it is hidden behind the incorrect movement of the dot?

If I understand you correctelly, the correction I made in this graph by hand would show the actual movement of the center of mass:

http://imgur.com/ljhx6Oj

An average position to the left with this correction isn't clearly obvious, and I can't easily do the analysation with this hand drawn picture. And even if I could make the analysation, it would be limited evidence, because this isn't real data, but just hand drawings.

But look at the graph, are you really convinced, that there is an average displacement? It looks to me just like an ordinary swing movement, without any anomalies.

In the describtion of your Youtube Video, you write:

' These experiments show that the machine is able to accelerate in one direction only. If you start a swing and accelerate as the machine is swinging backwards, the swinging motion will decelerate, that is, brake. This is interesting because it suggests this is a functioning reactionless drive. '

And in an earlier post, you wrote:

' It can both accelerate and decelerate swing speed by doing cycles. However, when you turn off the gyroscopes (video 21) that phenomenon more or less disappears. This means that the pendulum test should be pretty easy to pass, and I probably do pass it in one or more of these videos. '

Can you explain why? How does the observation suggest a functioning reactionless drive? Stopping and starting a swing motion doesn't lead to the conclusion of reactionless drives. So being better at stopping and starting the motion if the gyros are on hasn't anything to do with reactionless drive or the pendulum test.

An comparison: Imagine for an object traveling to the north and south would be considered impossible. You test it, and find out that you can go to the west. And you measure, that the object can go to the east. From that you think it suggests it could go the the north. Then you cripple your machine. Now it can't really go to the west and east anymore. From that you conclude the north-south test would e pretty easy to pass.

Edited April 12, 2014 by N_las