Cannae/EmDrive

[Northstar1989]

I know this may seem like old news to some of you, but back in August, NASA's Advanced Propulsion Physics Laboratory tested a device known as the "Cannae Drive" - which appeared to be the second propellantless thruster ever validated. This built on earlier tests of the "EM Drive" a similar, but bulkier and better-performing drive, which was tested back in January (but the results were not released until NASA had also tested the Cannae Drive in August) and appeared to work as well. I suggest reading the Wikipedia article for a more complete history of the devices.

http://en.wikipedia.org/wiki/EmDrive

When this hit the media, there were two waves of articles. The first wave simply expressed astonishment at the exciting discovery and its possible implications.

The second wave largely criticized the device- based either on the "it's too good to be true" line of logic (with no evidence to back up these criticisms other than that the device *appeared* to violate known laws of physics- but remember, a law of science is only a law of science *until we find an exception to it*. Theory, must explain results, rather than the other way around..) or on pulling apart the abstract of the NASA paper (which was released ahead of the full article) and making dangerous assumptions about the paper (and that the tests were performed improperly) which turned out not to be true.

As an actual scientist in real life, I was quite upset by all of this. First, very few scientists ASK for this kind of publicity- in fact the Chinese professor who validated the EmDrive ahead of NASA actually said "the publicity was very unwelcome". Second, the buffoonery of automatically assuming an experiment is wrong, simply because it defies established notions of how things works, simply baffled me.

I spent a period of time working in an equally controversial (perhaps more so) field to theoretical physics- stem cell research. And the degree to which people outside of the field would *consistently* mis-understand and mis-represent results in the field (especially when they challenged established notions, such as how aging and cancer work), never ceased to irritate or amaze me.

So, having some sympathy for the poor physicists whose results and credibility were being unjustly attacked (I can *guarantee* you many of them were hesitant to publish their data for precisely this reason), and the idea behind an EmDrive actually being one I had independently thought of myself one day when pondering what would happen when you shined light (or microwaves) in an asymmetrical resonant cavity (although clearly I never did anything with this idea- since I never majored in physics, and figured there was probably some reason I didn't know of it wouldn't work...), I largely kept my quiet (after a few ill-fated attempts to defend the scientists' credibility to people who were simply unwilling to listen), and waited for more facts to come out about the drive, and for myself to have time to look them up...

Well, anyways, long explanation aside, there *were* more facts to the story. Some of them didn't take very long to come out at all. For instance, just a week or so after the major wave of criticism, Wired released the following Q&A to clear up common misconceptions about the drive:

http://www.wired.co.uk/news/archive/2014-08/07/10-qs-about-nasa-impossible-drive

For instance, one of the most COMMON misconceptions about the NASA test, which was fairly obvious to me even back when the abstract was released, was that people were mis-understanding what was going on with the "Null" device.

Basically, the situation was this: the lead scientist behind the "Cannae Drive" (Fetta), had certain theoretical ideas about how such a device should/would work, and what were the reasons for that. These led to him engraving "slots" into one of the ends in his resonant cavity of his device, which he THOUGHT were necessary for the drive to work. However NASA tested versions of the device with *and* without the slots, and BOTH produce thrust. Many *INCORRECTLY* interpreted this to mean that the "control" had produced thrust as well- which would have obviously invalidated the results. THIS WAS NOT IN FACT THE CASE.

Anyone who had bothered even to *carefully* read the abstract, and maybe educate themselves a little about Fetta's underlying theory, would have immediately realized that the no-slots device only invalidated the need for SLOTS in the device. In fact, NASA tested a *THIRD* device- this one with a resistive, rather than resonant, cavity. THIS was the control- and as expected THIS produced absolutely no thrust. All NASA proved with the slots vs. no-slots device is that Fetta's underlying explanation for how his device worked was wrong, and there was absolutely need to carve slots into the device in order to make it work (in fact, this slightly *decreases* the performance of the device). The *actual* control produced no thrust- so the test was valid. To quote the scientists at NASA:

"Finally, a 50 ohm RF resistive load was used in place of the test article to verify no significant systemic effects that would cause apparent or real torsion pendulum displacements. The RF load was energised twice at an amplifier output power of approximately 28 watts and no significant pendulum arm displacements were observed."

As I pointed out earlier, THEORY FOLLOWS FACT. This is *absolutely essential* to understanding science/technology, and a key lesson for any good scientist. That means, we don't start with a theory, and then twist the facts to support it (in fact this is one major mistake many researchers in biology make- which better scientists such as those I trained under as a student at friggin' Cornell University, a school I am *VERY* proud of being able to call my alma mater, had to *constantly* point out is the WRONG way to do things...) Rather, we *must* start out with observed results, and adjust our hypotheses (which is any ideas we have before those results) to come up with a theory that can explain our results.

To take a famous, but probably fictitious story from the life of the very scientist whose "laws" the EmDrive results are purported to violate; Isaac Newton didn't *start* with the theory of gravity, and then use the fall of an apple to justify it- rather he started with the fall of an apple, and then came up with the theory to *explain* observed results.

The same needs to hold with the Canne/EmDrive. If the device appears to violate the Conservation of Momentum by using no propellent, we can't simply state that it's impossible because it violated preconceived notions- this is *exactly* the kind of thinking that used to trap us in the faulty scientific theories of the late Middle Ages (where many results that violated the prevailing theories of the day were rejected simply because they did not conform to faulty expectations- it was not until the introduction of the Scientific Method that this kind of thinking *truly* got thrown out the window...) Rather, we need to carefully analyze the results and experimental methods with a critical eye. If no fault can be found with them, we have to *consider* that our preconceived notions (that there are no exceptions to the Conservation of Momentum, for instance), may in fact be false...

Now note I'm NOT saying the Cannae Drive *did* really prove Newton was wrong, or that all our existing theory should be thrown out the window- only that the criticism leveled against it was, largely, illegitimate. The masses of people who jumped on the band wagon about the no-slots version of the Cannae Drive is just one example of what I'm talking about.

So, I would like to revive discussion of this apparently "old" news here. Largely, because the Cannae/EmDrive HASN'T gone away. The media (with its 3-second attention-span, and the scientific literacy of a 5 year old to match) simply lost interest in the subject, and many people implicitly assumed that the ill-founded criticism against it was valid, and the whole thing had been dropped.

This is, of course, not the case. NASA's "Eagleworks" division continues to work on improved versions of the testing apparatus (and higher-powered versions of the device itself), while additional labs at NASA, and across the country (including John Hopkins University, for starters) line up to test the devices after NASA is done with it. There will be *quite* a few more articles on the Em/Canne Drives in the not-too-distant future, it hasn't been discredited in any way *yet*, and may not ever be. It is quite possible that 30 or 40 years from now, we may indeed be flying in EmDrive airplanes (most likely powered by wireless power transmission, as the EmDrive requires no propellent, but DOES require large amounts of electricity to produce usable thrust levels), spaceplanes, and even rockets to Mars. If you can circumvent the need for propellent, you can circumvent the Rocket Equation, and then the sky (truly) no longer becomes the limit in a *very* real way...

Regards,

Northstar

P.S. If you're curious, there are *several* possible theoretical explanations for the EmDrive. The Chinese team justified it with a certain interpretation of Maxwell's Equation. Fetta defended it with a theory that required the use of slots, and turned out to be wrong. And perhaps my *favorite* explanation, because it means there actually *is* a propellent, even if it never enters or leaves the drive in a conventional manner, is that it pushes against particles that spontaneously pop in and out of existence anywhere in the universe as predicted by the Heisenberg Uncertainty Principle- particles that have sometimes (incorrectly) been referred to as the "quantum vacuum plasma" (they don't behave as a plasma, at least not in the classical sense of the word- but there *ARE* such particles, this much was already well-studied and extensively validated *long* before the invention of the EmDrive...)

Edited October 22, 2014 by Northstar1989

[Red Iron Crown]

I thought the general consensus among the scientific community was "more experimentation is necessary", which is appropriate when an anomalous result is obtained.

The media are going to do what media always do, sensationalize one way or the other to sell copies or bait clicks.

[Dodgey]

Thanks for the write up Northstar, with so much testing being done on this it makes me very excited for the future.

[KerikBalm]

You are aware you can have multiple controls in the same experiment.

*one* of their controls produced thrust, and one didn't

The next question is if the one that didn't was an appropriate control.

Consider a very simple PCR reaction... lets say you are screening colonies for a successful transformation.

You could have a negative control that is just water instead of a PCR mix - then when its negative, and all your other colonies are positive, do you conclude they were all successfully transformed?

Do you include a PCR mix without the fwd primer? without the rv? without the polymerase? without the DNTPs? without the template?

Did you do a ligation, and in one you have ligase+DNA insert+ backbone, and as negative control you also try insert+ backbone but no ligase, another that is backbone and ligase but no insert?

Did you include any colonies from those as negative controls in your PCR screen?

A positive result for any of those could mean a problem with your PCR screen.

Good experiments have good controls.

Just because one negative control works, doesn't mean anything if it wasn't a good negative control.

Did they put that Rf load in the same cavity? was the RF load device otherwise the same as the RF test device

Did they alter the shape so that its not resonant?

Could they replace the microwave emitter with a LED emitting visible light (presumably, given the different wavelength, it wouldn't be resonant)?

The paper only states:

"the test article was replaced by an RF load to verify that the force was not being generated by effects not associated with the test article."

There are all kinds of other explanations (magnetic interactions, interactions with the air, etc).

They had no controls to rule out other possible explanations.

They did have a prediction specific to the effect they are trying to prove, and the negative control for that prediction failed.

Their RF test load is not worth much as a control.

The device produced a force. The experiments show the force production does not behave like predicted by their explanation.

Thus the experiment provides no evidence to support their explanation of the force.

The force remains unexplained, and they didn't have proper controls to narrow it down.

Does the cavity have to be resonant for this force to appear?

Does it have to emit microwaves?

Etc...

They had 1 negative control for their explanation of a reactionless thruster that failed,

and another negative control that wasn't worth much of anything - only showing that they could measure thrusts.

The source of the thrust remains completely unexplained on the basis of this experiment.

Therefore it is BS (at least scientifically speaking) to claim this as evidence to overturn current theories/"laws" which are well supported by evidence.

0 evidence against these "laws"

Much evidence for them.

I see no problem with these criticisms of the paper (and more particularly, the claims being passed around based upon this paper).

The experiments were not very well performed - regardless of your assertions to the contrary.

The controls were nearly worthless, and the experimental conditions werent very good either (not even a vacuum test, come on!)

Science requires good controls, you don't just have 1 control for an entire experiment, you should have many controls.

Without good controls, the results can't be interpreted with any certainty.

And that is the case we have now: uninterpretable results.

[Dkmdlb]

Consider probability here, ie what are the chances?

I'd say not very good. I think the correct view to have is to be very skeptical of these results.

[Mazon Del]

Dkmdlb: Science doesn't quite work that way. The question of "What are the chances that this breaks known law X?" should ONLY be asked for the purposes of attempting to arrange the order of operations for experiments. IE: Should we approach this from angle A or B first? The question should NEVER be used to determine if we want to accept the results of a test.

That said, yes, a healthy bout of skepticism is always welcome when making extraordinary claims. Way back when CERN thought that maybe neutrinos moved FTL, when they posted their results their reaction was more "Can someone please disprove this? Please?" than "HAH! LOOK AT IT!"

KerikBalm: If the RF load test was not the correct method to test if the effect of the system as a whole, that does not mean that the "null" device instantly becomes the new control. The point of the null device was not "Does the cannae/EM drive work?". It was "Does this man's (Feta) interpretation of how it works provide a better model of how it works?" The answer was no.

Now I see you point all this out for the most part. The controls were quite worthwhile as they were not existant to look for answers to the questions you insist they were looking for, they answered the questions to which they were designed to the satisfcation of NASA, and frankly if they are satisfied then that carries a lot of weight with me.

The statement about having only one control being bad is extremely relative. It entirely depends on your experiment in what you are searching for and the questions you are asking.

Right now the results stand at: The drive works, but we still don't have a good idea WHY it works. More funding and testing is necessary.

More funding has been granted, more testing is occuring.

Within the next few weeks (if they stuck to their original timeline) NASA is supposed to be finished with their test article that they will begin to "mass produce" and sell/give to various labs and universities around the country.

Northstar1989: I think it a little less likely that the drive powered planes will utilize energy transmitted from the ground, however there is nothing to say at this moment that they couldn't use something like Lockheed's Fusion plant, if it turns out that it is real anyway.

[lajoswinkler]

Is it so hard to use the search engine? No need for one more thread.

[magnemoe]

Dkmdlb: Science doesn't quite work that way. The question of "What are the chances that this breaks known law X?" should ONLY be asked for the purposes of attempting to arrange the order of operations for experiments. IE: Should we approach this from angle A or B first? The question should NEVER be used to determine if we want to accept the results of a test.

That said, yes, a healthy bout of skepticism is always welcome when making extraordinary claims. Way back when CERN thought that maybe neutrinos moved FTL, when they posted their results their reaction was more "Can someone please disprove this? Please?" than "HAH! LOOK AT IT!"

KerikBalm: If the RF load test was not the correct method to test if the effect of the system as a whole, that does not mean that the "null" device instantly becomes the new control. The point of the null device was not "Does the cannae/EM drive work?". It was "Does this man's (Feta) interpretation of how it works provide a better model of how it works?" The answer was no.

Now I see you point all this out for the most part. The controls were quite worthwhile as they were not existant to look for answers to the questions you insist they were looking for, they answered the questions to which they were designed to the satisfcation of NASA, and frankly if they are satisfied then that carries a lot of weight with me.

The statement about having only one control being bad is extremely relative. It entirely depends on your experiment in what you are searching for and the questions you are asking.

Right now the results stand at: The drive works, but we still don't have a good idea WHY it works. More funding and testing is necessary.

More funding has been granted, more testing is occuring.

Within the next few weeks (if they stuck to their original timeline) NASA is supposed to be finished with their test article that they will begin to "mass produce" and sell/give to various labs and universities around the country.

Northstar1989: I think it a little less likely that the drive powered planes will utilize energy transmitted from the ground, however there is nothing to say at this moment that they couldn't use something like Lockheed's Fusion plant, if it turns out that it is real anyway.

This, EmDrive is an high risk, high payoff sort of deal, fusion is another. In short its a bit like the lottery, low chance for winning big but that would change everything. Some chance of getting something interesting in any direction but most of the time you lose.

The best way to test this is an cheap and fast ........ test, either its ........ or its interesting.

[Mazon Del]

Well, the best way is to do small tests to indicate if there is enough of something there to warrant serious expenditure of funds/resources. Sometimes your "small test" can be done via math in lieu of a proper test (the Micro Warp Field Interferometer is an example of this).

[Hodo]

I was reading about this on CNN, and on a science journal. I find this drive fascinating for so many levels. If it does not require a thrust exit port, and can be placed at the Center of Mass of the craft you could truely build a "flying saucer". If the claims are correct, that you could generate 3kN of thrust from 1kW of power, you could conceivably have a vehicle the same size as a Nissan Leaf, which weighs in at 1500kg, that could fly for 48hrs on its on board battery, which is rated for 24kW/hours.

Pretty amazing if you think about it just from a transport standpoint.

[Mazon Del]

If I remember right Hodo, those are the claims coming from the British guy about what he thinks his Mk 2 will be able to do. Unfortunately, given that the Mk 2 utilizes superchilled superconductors and the like, we probably won't get that sort of performance for cars. Maybe in mass transportation such as airliners and such though. I could relatively easily see things like flying supertankers and the like (finally) becoming real.

[Red Iron Crown]

you could generate 3kN of thrust from 1kW of power

Source? The numbers I've read are much, much lower, around 0.1N/kW as a theoretical maximum.

[Northstar1989]

Source? The numbers I've read are much, much lower, around 0.1N/kW as a theoretical maximum.

He's referring to the (expected) performance of the superconducting version. Check the Wiki page- there are links there where you can find those numbers.

Regards,

Northstar

[Hodo]

He's referring to the (expected) performance of the superconducting version. Check the Wiki page- there are links there where you can find those numbers.

Regards,

Northstar

Exactly, that is why I said, it "could".

If you were to take the current power output, it is still more than addiquate for something like a probe in space or even a small UAV test vehicle and even a small car. Seeing as you don't need a great deal of power to get a car to roll forward, it doesnt have to be fast, just able to move.

Even if you were to able to achieve 1/3rd of the speculated power of 30kn, it could be enough to fly a Cessna, or even a Piper Cub. At 1/6th of the power output rating, or 5kn, it would be able to move a 1000lb object at a TWR 1:1. Seeing as the battery packs in a Nissan Leaf actually weigh less than 500lbs it is more than enough to get a probe or a UAV to move quite well through the air or even underwater.

EDIT-

NASA's test was at 17W of input power and net an output of 116microN. Which was 2% of the power input of the original test by Shaw, and .7% of the power the Chinese said they did. So if the power scale were consistant, it is concievable that you could with current setup max out at 6800-7000micro Newton when 1kW is applied.

Edited October 22, 2014 by Hodo

[SuperFastJellyfish]

There's a very informative 164 page thread over at NSF with lots of good info/debate on this subject if you have a few days to kill.

[Agumander]

Is there anything special about the manufacture of the test device such that a mechanically-inclined person couldn't build it in their garage? It seems like a resonant cavity shouldn't be that hard to build... but I guess there's a reason the magnetron from a microwave oven won't work as an emitter for an EM drive?

[Mazon Del]

From my understanding as far as the physical and electronics side is concerned, no. There isn't anything stopping you from doing so in the physical sense. Exactly how the objects are shaped and whatnot (for dimensioning purposes) I do not know. Presumably if you get access to the full report (I think it is available on a pay site somewhere) you could do it.

[KerikBalm]

If the RF load test was not the correct method to test if the effect of the system as a whole, that does not mean that the "null" device instantly becomes the new control. The point of the null device was not "Does the cannae/EM drive work?". It was "Does this man's (Feta) interpretation of how it works provide a better model of how it works?" The answer was no.

They had an idea about how this thing may produce a force. The point of the null was to test that idea.

Comparing the slotted to the null tells them their idea was wrong. Therefore, they lack any test for an explanation of how the force is produced.

they answered the questions to which they were designed to the satisfcation of NASA, and frankly if they are satisfied then that carries a lot of weight with me.

And it carries zero weight with me... or do you still believe that NASA paper about life substituting Arsenic for Phosphorus in its DNA?

Within the next few weeks (if they stuck to their original timeline) NASA is supposed to be finished with their test article that they will begin to "mass produce" and sell/give to various labs and universities around the country.

Citation please

Right now the results stand at: The drive works, but we still don't have a good idea WHY it works.

No, the results are: a force was measured.

Force and thrust are not the same thing. There are a number of explanations for the force that, if true, would mean that it would not be able to produce thrust in space, and that it would not be a working drive.

There is a *force*, we don't know what causes the *force*

Depending on the cause of the *force*, it may or may not be suitable for space propulsion.

To claim that this proves the drive works is to make a false claim.

there is nothing to say at this moment that they couldn't use something like Lockheed's Fusion plant, if it turns out that it is real anyway.

Lockheed doesn't have a fusion plant. They have an idea for a research project that may lead to one. Their press release made no such claims. One media group made claims out of thin air, and others parroted them. If you read what Lockheed says now, compared to what it was saying in Feb 2013, you'd come to the conclusion that the project isn't working out like they expected - there has been no breakthrough.

[Red Iron Crown]

He's referring to the (expected) performance of the superconducting version. Check the Wiki page- there are links there where you can find those numbers.

Ah, missed that. So there's either sci-fi handwavium high temperature superconductors involved, or the figures also need to take into account keeping the superconductor cooled to its operational range. Too bad.

I really want the em-drive to be true, but I'm still pretty skeptical.

[Northstar1989]

I really want the em-drive to be true, but I'm still pretty skeptical.

Keep watching. NASA hasn't forgotten about it, or given up on it (in fact, there's no reason to- all the results have pointed towards it being real/functional so far...) so you probably shouldn't either.

Regards,

Northstar

[Dodgey]

One thing which I haven't seen discussed much is the implications of such a drive should it turn out to be real. Due to its being propellant less as we all know it's delta v is effectively infinit (in the laymans sense). So let's say for the sake of conversation that next year NASA sends up a satellite with one of these drives attached and once and for all determines that yes it does work, what then?

I remember reading a statement from NASA which stated that if this drive did work then we could achieve an 8 month round trip to Mars including a one month stop on the surface.

So what would happen to the space industry, how big of an expansion would we see and how quickly?

[Scotius]

Station keeping of satellites, and maybe even ISS would become a lot simpler and cheaper.

[Rakaydos]

A truely reactionless drive does not need to be mounted at the "back" of a spaceship- you can stack thrusters like cordwood, as long as the center of thrust lines up with the center of mass.

A ship's Thrust to weight ratio, therefore, becomes measured in drive efficency and the mass of the ship devoted to propulsion.

[dharak1]

Could you build walls out of them then or even the hull of the ship? Seems insanely useful but if it's reactionless how do you direct the thrust to get where you want to go.

[magnemoe]

Could you build walls out of them then or even the hull of the ship? Seems insanely useful but if it's reactionless how do you direct the thrust to get where you want to go.

The limited factor is power, solar or reactors and both are heavy. the drives will also produce heat you have to handle.