EEVblog Electronics Community Forum
General => General Technical Chat => Topic started by: Psi on September 27, 2016, 12:02:11 pm
-
Is anyone else looking forward to the peer reviewed NASA paper on the EmDrive which has been accepted and is due to be published in December?
The sad thing is, once proved and verified, both sides will claim victory over the other.
Sceptic -> Believer: Told you so!.. See, it doesn't violate the laws of physics. You were wrong!
Believer -> Sceptic: Told you so!.. See, it works Look! You were wrong!
-
Have you read it? it's a joke!
-
Thunderf00t is doing a debunking of the EmDrive, that will be interesting. Can't say I've looked much into it. As far as I can tell the thrust achieved is so small it almost definitely is some form of (EM) leakage?
-
What I don't understand is why they base the thrust on N/W. They IMO must also consider the Q factor of the microwave cavity. So more reasonable unit for thrust would be something like N/(W*Q)
If only the input power is considered, a EMdrive should get the same thrust no matter whether it has a superconducting microwave cavity with great Q or one that's filled with e.g. water as a big dummy load with horrible Q. Yet the superconducting one should achieve much higher field strengths with much less input power than the water filled one.
Intentionally decreasing the Q factor is also something their experiment is lacking. (as far as I can remember from reading the paper) They easily could worsen Q by placing some absorber inside the cavity or by changing frequency (cavity out of tune) while measuring the trust.
-
<rant>
So....
- The whole theory is based in just 18 data points. Figure 19 in the paper.
- kW and W are continuously being swapped.
- 60W (kW?) and 80W(kW?) produce the same torque.
- Again, only 18 points in 2 years?
- They use a FieldFox as VNA. Its a good VNA but....
- Seriously, 18 points.
- The error bars don't make sense.
- YES, 18 points.
- No mention of any of the parts use in the RF path. No mention on what PLL, VCO, Amplifier, Couplers and cabity filters.
- Have you seen the images of the device? made in a Kitchen with hammer.
- The Experiment drawings... hahahhahaha
- 18 points... and look at the trend line, "hey, it goes up...." but again, 60W (or kW?) is the same as 80W(kW?)
- Thermal analysis? ahahahhahaha In my lab we have better equipment that what they show....
There are more things, like the errors discussion with the DC cables being "twisted" ....
</rant>
-
<rant>
So....
- The whole theory is based in just 18 data points. Figure 19 in the paper.
- kW and W are continuously being swapped.
- 60W (kW?) and 80W(kW?) produce the same torque.
- Again, only 18 points in 2 years?
- They use a FieldFox as VNA. Its a good VNA but....
- Seriously, 18 points.
- The error bars don't make sense.
- YES, 18 points.
- No mention of any of the parts use in the RF path. No mention on what PLL, VCO, Amplifier, Couplers and cabity filters.
- Have you seen the images of the device? made in a Kitchen with hammer.
- The Experiment drawings... hahahhahaha
- 18 points... and look at the trend line, "hey, it goes up...." but again, 60W (or kW?) is the same as 80W(kW?)
- Thermal analysis? ahahahhahaha In my lab we have better equipment that what they show....
There are more things, like the errors discussion with the DC cables being "twisted" ....
</rant>
60W had appropriately more thrust than 40W. 80W had a bit less than 60W. Why doesn't anyone mention that...only that the trend didn't continue at 80W. Given that the entire thing is poorly understood, it's more reasonable to assume that there's a roll off in power output for reasons having to do with their setup.
You guys should get over the "reactionless drive" hang up over this thing and start considering that NASA is a pretty serious organization and knows what they're doing. It's not reactionless. It's just that no one knows where the reaction is coming from yet. It's not so unusual. It took a very long time to discover, and then accept, the notion of Hawking radiation too. In retrospect, it's completely obvious and a straight forward application of physics that OF COURSE black holes must radiate, but at the time it wasn't so simple or obvious after all. I suspect the same will happen here. Someone just needs to apply the right physics in the right way, and the situation will resolve itself eventually.
In fact, if you think about it for a bit, a working EM drive could actually end up being our first direct experimental evidence for a many-worlds interpretation, and could well lead to a much deeper understanding of things like conservation of momentum.
Looking forward to seeing some reports from other labs to confirm/deny.
-
Care to post a link to what is being discussed?
-
Care to post a link to what is being discussed?
Already some discussion and links in this thread:
https://www.eevblog.com/forum/chat/measurement-of-impulsive-thrust-from-a-closed-radio-frequency-cavity-in-vacuum/ (https://www.eevblog.com/forum/chat/measurement-of-impulsive-thrust-from-a-closed-radio-frequency-cavity-in-vacuum/)
-
60W had appropriately more thrust than 40W. 80W had a bit less than 60W. Why doesn't anyone mention that...only that the trend didn't continue at 80W. Given that the entire thing is poorly understood.
Where? Why do they plot a trend then? Did you see the error bars?... Explain those reasons, because they can't....
it's more reasonable to assume that there's a roll off in power output for reasons having to do with their setup.
Really? Tell me more about it. Oh, wait, it's not explained, ANYWHERE.
At least when CERN published that Neutrinos were faster than light, they asked the community for their error and not assert everything. :horse:
-
Were is this paper people refer to?
-
Were is this paper people refer to?
The following link, seems to discuss it and has a link to the paper, at the bottom.
But the paper seems to be dated 2015, which is why I deleted my original post/link, in case it is wrong/misleading.
http://www.sciencealert.com/leaked-nasa-paper-shows-the-impossible-em-drive-really-does-work (http://www.sciencealert.com/leaked-nasa-paper-shows-the-impossible-em-drive-really-does-work)
tl;dr
NOT sure if this is the right link or not, sorry.
Possibly WRONG link (as it says 2015) ?
https://drive.google.com/file/d/0B7kgKijo-p0ibm94VUY0TVktQlU/view (https://drive.google.com/file/d/0B7kgKijo-p0ibm94VUY0TVktQlU/view)
-
60W had appropriately more thrust than 40W. 80W had a bit less than 60W. Why doesn't anyone mention that...only that the trend didn't continue at 80W. Given that the entire thing is poorly understood.
Where? Why do they plot a trend then? Did you see the error bars?... Explain those reasons, because they can't....
You obviously didn't actually read the paper.
-
https://www.youtube.com/watch?v=jCAqDA8IfR4 (https://www.youtube.com/watch?v=jCAqDA8IfR4)
-
This video supposedly has the inventor of the drive explaining how it is supposed to work:
https://youtu.be/wBtk6xWDrwY (https://youtu.be/wBtk6xWDrwY)
I think this probably has multiple basic errors.
The fundamental one seems to me is he is using a formula for radiation pressure relating power and group velocity to radiation pressure. He argues that since the group velocity is different at the two ends of the cavity, there should be a net force. But this assumes the power is the same at both ends, which may not be true.
An example I thought of is just take a coaxial resonator that is half air line and half filled with dielectric. The group velocity on the dielectric end is slower. So according to his formulas, this should produce thrust. But actually it is just two lines in series, and the velocity at each end of each line is the same, so the force cancels out totally.
He also makes some kind of argument that since things are moving at the speed of light and because of special relativity, this is not really a closed system.
I'm sure his explanation was quickly debunked. Then came more possible explanations of virtual particles, quantum vacuum, pilot waves, etc. :phew:
-
You obviously didn't actually read the paper.
(http://arc.aiaa.org/na101/home/literatum/publisher/aiaa/journals/content/jpp/0/jpp.ahead-of-print/1.b36120/20161117/images/medium/figure19.gif)
Oh really?, Enlighten me.
DOI:10.2514/1.B36120
-
Some Physicist on a physics subreddit explained very good were the errors are:
https://www.reddit.com/r/Physics/comments/5ewj86/so_nasas_em_drive_paper_is_officially_published/ (https://www.reddit.com/r/Physics/comments/5ewj86/so_nasas_em_drive_paper_is_officially_published/)
Short version: Most likely they just measured the thermal expansion.
You can see that it produces "thrust" even after it is turned off http://i.imgur.com/n6KMlFK.png (http://i.imgur.com/n6KMlFK.png) (picture from the reddit top comment)
and it pretty much looks like a heating/cooling curve...
-
Video snipped...
Unfortunately that video made me older, but not wiser.
I didn't watch all of it but I don't believe it debunks the claim of thrust. Surely that is the interesting part.
He just raves on for _ages_ about the amount of thrust is too small to be useful on earth - because the inventor made some dumb claim about flying cars.
-
Video snipped...
Unfortunately that video made me older, but not wiser.
I didn't watch all of it but I don't believe it debunks the claim of thrust. Surely that is the interesting part.
He just raves on for _ages_ about the amount of thrust is too small to be useful on earth - because the inventor made some dumb claim about flying cars.
In the beginning, yes. He starts off by debunking the concept of the EmDrive /given/ that it works and provides the thrust published in the paper. He later describes the likely cause of the measured thrust, namely heat. Same principle as the radiometer.
-
Until real physicists are able to conduct actual experiments that leave no question, or somebody actually just puts one in orbit and demonstrates that it really works will any of this be news worthy.
-
Video snipped...
Unfortunately that video made me older, but not wiser.
I didn't watch all of it but I don't believe it debunks the claim of thrust. Surely that is the interesting part.
He just raves on for _ages_ about the amount of thrust is too small to be useful on earth - because the inventor made some dumb claim about flying cars.
In the beginning, yes. He starts off by debunking the concept of the EmDrive /given/ that it works and provides the thrust published in the paper. He later describes the likely cause of the measured thrust, namely heat. Same principle as the radiometer.
Yeah I did see that bit. Still so many minutes of waffle for a possibly relevant clip about a radiometer is a big red debunking fail in my book. We already knew that it would be inefficient and that the thrust/W would be very low.
The radiometer only works in a partial vacuum - yet I thought the emdrive has already demonstrated thrust in a vacuum?
OTOH, for a thinking mans debunking, the reddit stuff Akra mentioned raises some good questions.
-
You obviously didn't actually read the paper.
(http://arc.aiaa.org/na101/home/literatum/publisher/aiaa/journals/content/jpp/0/jpp.ahead-of-print/1.b36120/20161117/images/medium/figure19.gif)
Oh really?, Enlighten me.
DOI:10.2514/1.B36120
What can I help with? I mean that sincerely. A lot of the criticisms are directly addressed in the paper, probably far better than I can. I'm not sure what there is to say about the graph you posted other than the error bars are large, but that's not unusual with stuff like this.
-
I'm not disagreeing with Thunderf00t's conclusion, but he was driving me nuts with his arguments in that video.
"Efficiency" in a rocket is not a measure of power to thrust conversion efficiency, it's mass efficiency. A rocket that's good at power conversion (high thrust) is usually a good booster but a bad space drive.
There are two measures of note with rockets: Specific Power (Thrust over weight) and Specific Impulse (Propellant Mass Efficiency). The latter being a mere computational convenience to relate propellant mass to exhaust velocity. Ideally you want both high specific power and high specific impulse, but even conceptually, those are rare (e.g., Zubrin's bad-for-flying-cars nuclear salt water rocket https://en.wikipedia.org/wiki/Nuclear_salt-water_rocket (https://en.wikipedia.org/wiki/Nuclear_salt-water_rocket)). Usually you get one or the other and choose your application based on that. Higher specific impulse drives obviously use more energy to generate a unit of thrust, but do so with less actual reaction mass. This allows vehicles like the Dawn space probe (electric ion rocket yielding well in excess of 10km/s for only 425kg of propellant in a 1200kg wet-mass vehicle!) to have a remarkably high deltaV with very little onboard propellant*. Albeit, all this at the cost of very low thrust (around 2000 day total burn time).
So a photon rocket, far from being "inefficient" in his words, has the theoretical limit of exhaust velocity (the speed of light) and hence, would be the most mass efficient design possible. Of course, apart from matter/antimatter reactions, there is little that can generate the power needed to take advantage of it in a useful way.
Likewise he kept saying the EM Drive "generates thrust" over and over, as a matter of comparison. No, it fricking doesn't. It's purest BS. No one would like that thing to actually work more than me, but it's a clear violation of the conservation of momentum. Likewise, the quantum vacuum it's supposedly pushing against (like all quantum fields at their vacuum expectation value, even the Higgs) is a frameless phenomena. You can't push against it!
I will happily, happily eat my shorts if the "effect" is anything other than experimental error.
*Note: if you run the numbers on Dawn, you find a max deltaV of 13km/sec, but that would be in running the propellant tanks all the way dry, and they didn't do that.
-
Likewise he kept saying the EM Drive "generates thrust" over and over, as a matter of comparison. No, it fricking doesn't. It's purest BS. No one would like that thing to actually work more than me, but it's a clear violation of the conservation of momentum. Likewise, the quantum vacuum it's supposedly pushing against (like all quantum fields at their vacuum expectation value, even the Higgs) is a frameless phenomena. You can't push against it!
I don't believe anyone is claiming that is "pushes" against a quantum vacuum, and I also don't think anyone is claiming that it violates conservation of momentum or is "reactionless", even though it keeps getting called a reactionless drive. When people set up straw men (I'm not referring to you, but others who write about this), I tend to instantly tune out.
But even if it did, so what? That's the point of doing experiments, publishing and then other people do similar experiments and improved experiments. The entire history of physics has been experimental data showing completely unexpected, and "impossible" results. Parity conservation comes to mind as a major whoops. I don't suspect tossing out conservation of momentum will be necessary here, though, even if it works. Either way, you have to allow nature to reveal herself, whether we like it or not.
I'll wait and see what an improved experimental setup comes up with. It looks to me like they did a reasonable job of tracking down sources of error and got something other than a null result, albeit with a lot of uncertainty. But any result other than zero would be spectacular. Fact is that it doesn't seem so simple to demonstrate one way or another whether it works or not, but it's also consistent that you run experiments just so you can demonstrate NO effect at certain energies, and you can expand the bounds where you're absolutely sure that current theories operate as expected.
-
Just for fun,
Here's the inventor Shawyer's theory of how it works:
http://www.emdrive.com/theorypaper9-4.pdf (http://www.emdrive.com/theorypaper9-4.pdf)
Here's a very simple debunking of Shawyer's theory, based on photons bouncing around inside a box:
http://johncostella.webs.com/shawyerfraud.pdf (http://johncostella.webs.com/shawyerfraud.pdf)
Here's a specific solution to the EM fields showing there is no net force, and a general proof that for any shape of cavity there is no net force generated:
http://gregegan.customer.netspace.net.au/SCIENCE/Cavity/Cavity.html (http://gregegan.customer.netspace.net.au/SCIENCE/Cavity/Cavity.html)
So no matter how you look at it with conventional physics, the force on the sidewalls cancel the forces on the ends. It doesn't generate thrust.
That just leaves some other unknown explanation. So far the most likely is experimental error.
-
I don't believe anyone is claiming that is "pushes" against a quantum vacuum,
Does the principal investigator of the project, Harold White count? http://www.astronomy.com/news/2016/11/emdrive-thruster-cleared-first-hurdle (http://www.astronomy.com/news/2016/11/emdrive-thruster-cleared-first-hurdle)
and I also don't think anyone is claiming that it violates conservation of momentum or is "reactionless", even though it keeps getting called a reactionless drive. When people set up straw men (I'm not referring to you, but others who write about this), I tend to instantly tune out.
If there is no exchange of momentum with something that leaves the engine, either a reaction mass or momentum-carrying massless particles like photons, then it is by definition a reactionless drive.
They may not be claiming that it is (they're claiming it's not) but their explanations of how it's not hold no water.
-
You obviously didn't actually read the paper.
(http://arc.aiaa.org/na101/home/literatum/publisher/aiaa/journals/content/jpp/0/jpp.ahead-of-print/1.b36120/20161117/images/medium/figure19.gif)
Oh really?, Enlighten me.
DOI:10.2514/1.B36120
If they made the measurements at two points instead of three, they'd have a perfect linear regression fit. Just saying. :)
-
Were is this paper people refer to?
From the other thread:
PDF:
http://arc.aiaa.org/doi/pdf/10.2514/1.B36120 (http://arc.aiaa.org/doi/pdf/10.2514/1.B36120)
HTML:
http://arc.aiaa.org/doi/10.2514/1.B36120 (http://arc.aiaa.org/doi/10.2514/1.B36120)
Dave, could you merge the 2 threads?
-
I don't believe anyone is claiming that is "pushes" against a quantum vacuum,
Does the principal investigator of the project, Harold White count? http://www.astronomy.com/news/2016/11/emdrive-thruster-cleared-first-hurdle (http://www.astronomy.com/news/2016/11/emdrive-thruster-cleared-first-hurdle)
and I also don't think anyone is claiming that it violates conservation of momentum or is "reactionless", even though it keeps getting called a reactionless drive. When people set up straw men (I'm not referring to you, but others who write about this), I tend to instantly tune out.
If there is no exchange of momentum with something that leaves the engine, either a reaction mass or momentum-carrying massless particles like photons, then it is by definition a reactionless drive.
They may not be claiming that it is (they're claiming it's not) but their explanations of how it's not hold no water.
re: "pushing off"
The author is describing a different system with some relevant results (a Q thruster), and then proposes that this might use the same mechanism. That's much different than drawing a conclusion.
re: reactionless drive
That's not by definition what a reactionless drive is. If it was, a simple propeller couldn't work...walking wouldn't work either. All that conventional physics requires is if something goes ---> that way with some momentum, something(s) else goes <---- that way with the opposite momentum.
Anyhow, as I said it's really irrelevant how it works, assuming it does, and what it potentially violates. At this stage, the question is one of existence, not anything more mundane than that, and it'll be what it is whether or not it offends your sensibilities in the process.
-
re: "pushing off"
The author is describing a different system with some relevant results (a Q thruster), and then proposes that this might use the same mechanism. That's much different than drawing a conclusion.
There is no evidence that a Q-Thruster works either. The only "relevant" results are is that there is no real evidence it works, and the "principal" they have suggested it works on is not a thing.
With regard to the quantum vacuum, you understand what "frameless" means, right?
re: reactionless drive
That's not by definition what a reactionless drive is. If it was, a simple propeller couldn't work...walking wouldn't work either. All that conventional physics requires is if something goes ---> that way with some momentum, something(s) else goes <---- that way with the opposite momentum.
Yes, propulsion without momentum exchange is indeed the definition of a reactionless drive.
Walking does require an exchange of momentum. The mechanism is gravity pulling you down coupled by friction to the body you are standing on. Try walking in zero g in a vacuum. Likewise, a propeller doesn't work in a vacuum. It uses air as reaction mass. There are many different ways of trading momentum with another object, but the EmDrive isn't one of them. It's reactionless.
Anyhow, as I said it's really irrelevant how it works, assuming it does, and what it potentially violates. At this stage, the question is one of existence, not anything more mundane than that, and it'll be what it is whether or not it offends your sensibilities in the process.
Apart from the swell line they drew through their noise (did you look at the graph?), I'm not seeing any evidence it works in any fashion. You're aware the paper invokes de Broglie/Bohm physics (toward the end) which invokes hidden variables and explicit non-locality, right? From a purely Bayesian reasoning standpoint, they are making a colossal unproven claim explained by other colossal unproven claims. While I concede it can't be ruled out that physics has been horribly horribly wrong about everything, and yet bizarrely still works as advertised, right now the EmDrive is a homeopathic rocket. There's little reason to think it won't stay that way.