Author Topic: EM Drive Developments Thread 1  (Read 765507 times)

Online Mulletron

  • Full Member
  • ****
  • Posts: 1113
  • Liked: 776
  • Likes Given: 1013
Re: EM Drive Developments
« Reply #3240 on: 11/14/2014 09:30 PM »
Please bear with me here, but here's an idea I had to test this at home, coupled with the simple torsion balance I linked to recently.

So I need a copper cavity. Those aren't easy to come by and I'm too lazy to buy copper sheet to build my own and I'd probably build it like crap anyway. But I remembered that I can get my hands on a little brass bell just about anywhere. Look in your Christmas decorations. Once a suitable bell is found, it is a straight forward exercise to drill a hole in it to mount an rf connector, fabricate a suitable loop probe and solder it into the rf connector cup, glue some dielectric material in it, and then cut out a copper sheet and solder the thing shut. A quick and dirty resonant cavity on the cheap.

Now you're going to have to find a signal generator. I guess you can rent one. I have piles of them where I work so no issues there. HP 83712B and 83752B sig gens are very very stable. They go up to 20ghz. If I remember right, you can get +20dbm out of them. I never tried pushing them that high. Never more than 0dbm for me.

Next by some homebrew miracle, find a resonant frequency close to what you calculated and try not to reflect all your power back into the sig gen.

The only right way is to use a directional coupler with power detectors or a network analyzer to measure fwd/reflected power.

If you can't get a sig gen. I know you can hack old Linksys WAP11 access points into CW mode @ around 2.4ghz.

Necessity is the mother of invention.

As far as a dielectric, look in your toolbox. You probably have PTFE tape in there.

You'd have to be pretty creative to sort out cable strain on your setup and you have to put the mess in the basement and isolate it from wind currents/people/pets, whatever. Not easy. I might do it. I'm still thinking about whether is is worth it or not personally to go through the trouble.
Challenge your preconceptions, or they will challenge you. - Velik

Offline Rodal

  • Senior Member
  • *****
  • Posts: 5838
  • USA
  • Liked: 5919
  • Likes Given: 5261
Re: EM Drive Developments
« Reply #3241 on: 11/14/2014 09:33 PM »
Please bear with me here, but here's an idea I had to test this at home, coupled with the simple torsion balance I linked to recently.

So I need a copper cavity. Those aren't easy to come by and I'm too lazy to buy copper sheet to build my own and I'd probably build it like crap anyway. But I remembered that I can get my hands on a little brass bell just about anywhere. Look in your Christmas decorations. Once a suitable bell is found, it is a straight forward exercise to drill a hole in it to mount an rf connector, fabricate a suitable loop probe and solder it into the rf connector cup, glue some dielectric material in it, and then cut out a copper sheet and solder the thing shut. A quick and dirty resonant cavity on the cheap.

Now you're going to have to find a signal generator. I guess you can rent one. I have piles of them where I work so no issues there. HP 83712B and 83752B sig gens are very very stable. They go up to 20ghz. If I remember right, you can get +20dbm out of them. I never tried pushing them that high. Never more than 0dbm for me.

Next by some homebrew miracle, find a resonant frequency close to what you calculated and try not to reflect all your power back into the sig gen.

The only right way is to use a directional coupler with power detectors or a network analyzer to measure fwd/reflected power.

If you can't get a sig gen. I know you can hack old Linksys WAP11 access points into CW mode @ around 2.4ghz.

Necessity is the mother of invention.

As far as a dielectric, look in your toolbox. You probably have PTFE tape in there.

You'd have to be pretty creative to sort out cable strain on your setup and you have to put the mess in the basement and isolate it from wind currents/people/pets, whatever. Not easy. I might do it. I'm still thinking about whether is is worth it or not personally to go through the trouble.

Please be very careful and safe if you are going to perform this experiment at home as we always appreciate your contributions and would be very upset if something happens to you !
« Last Edit: 11/14/2014 09:34 PM by Rodal »

Offline IslandPlaya

  • Full Member
  • ****
  • Posts: 582
  • Outer Hebrides
  • Liked: 163
  • Likes Given: 166
Re: EM Drive Developments
« Reply #3242 on: 11/14/2014 09:34 PM »
Please be careful, but hey, go for it!
You could change the Universe!

Online Mulletron

  • Full Member
  • ****
  • Posts: 1113
  • Liked: 776
  • Likes Given: 1013
Re: EM Drive Developments
« Reply #3243 on: 11/14/2014 09:46 PM »
Less than 1 watt won't hurt anybody.
Challenge your preconceptions, or they will challenge you. - Velik

Offline ThinkerX

  • Full Member
  • ***
  • Posts: 303
  • Alaska
  • Liked: 113
  • Likes Given: 59
Re: EM Drive Developments
« Reply #3244 on: 11/15/2014 12:18 AM »
Ok...

Ron...

In my most recent summary of the approaches taken on this thread to investigate EM Drive, I briefly described four approaches as to why we thought at one time or another it might work, and one explanation as to why it does not.  I was not putting forth the 'thermal artifact' explanation as a valid drive, but as a possibly valid explanation as to why the device does not work, yet might give the appearance of doing so.

Mulletron-

For your cavity, head to the plumbing section of the local hardware store, if such a place is to be found in your area.  There are angled copper couplings and caps which should come fairly close to your requirements.   But be very careful!

John, add proper cautionary video/image for Mulletron...

Offline JohnFornaro

  • Not an expert
  • Senior Member
  • *****
  • Posts: 9162
  • Delta-t is the salient metric.
  • Planet Eaarth
    • Design / Program Associates
  • Liked: 610
  • Likes Given: 314
Re: EM Drive Developments
« Reply #3245 on: 11/15/2014 12:37 AM »
John, add proper cautionary video/image for Mulletron...

[Raises finger.]  [Not that finger.  The index finger.]

Izzit because I'm that predictable, or izzit because the material is that predictable?  It's almost tooooo eeezy...

Sometimes I just flat out don't get it.

Offline Ron Stahl

  • Full Member
  • **
  • Posts: 210
  • Liked: 32
  • Likes Given: 2
Re: EM Drive Developments
« Reply #3246 on: 11/15/2014 03:14 PM »
Please bear with me here, but here's an idea I had to test this at home, coupled with the simple torsion balance I linked to recently.

So I need a copper cavity. Those aren't easy to come by and I'm too lazy to buy copper sheet to build my own and I'd probably build it like crap anyway. But I remembered that I can get my hands on a little brass bell just about anywhere. Look in your Christmas decorations. Once a suitable bell is found, it is a straight forward exercise to drill a hole in it to mount an rf connector, fabricate a suitable loop probe and solder it into the rf connector cup, glue some dielectric material in it, and then cut out a copper sheet and solder the thing shut. A quick and dirty resonant cavity on the cheap.

Now you're going to have to find a signal generator. I guess you can rent one. I have piles of them where I work so no issues there. HP 83712B and 83752B sig gens are very very stable. They go up to 20ghz. If I remember right, you can get +20dbm out of them. I never tried pushing them that high. Never more than 0dbm for me.

Next by some homebrew miracle, find a resonant frequency close to what you calculated and try not to reflect all your power back into the sig gen.

The only right way is to use a directional coupler with power detectors or a network analyzer to measure fwd/reflected power.

If you can't get a sig gen. I know you can hack old Linksys WAP11 access points into CW mode @ around 2.4ghz.

Necessity is the mother of invention.

As far as a dielectric, look in your toolbox. You probably have PTFE tape in there.

You'd have to be pretty creative to sort out cable strain on your setup and you have to put the mess in the basement and isolate it from wind currents/people/pets, whatever. Not easy. I might do it. I'm still thinking about whether is is worth it or not personally to go through the trouble.

I am not a fan of DIY propulsion experiments since it's very difficult to provide the instrumentation necessary to learn anything, and most replications for value cost between $1-2 million, though most of this is labor costs.  However, I'll be as supportive as possible.  Just please do as Dr. Rodal suggests and avoid frying yourself.  Have you worked with microwave before?

I think before you decide to build anything, you want to have a long talk with Paul March so you can avoid some of the pitfalls.  For example, I think copper is not the best choice because it will vaporize and arc.  Stainless steel is supposed to be better.  You might modify a steal cone as found here:

http://www.globalindustrial.com/p/material-handling/drum-barrel/drum-covers/drum-cone-t304-stainless-steel-60-degree-size-23-fits-23-3-8-od-rim

Can you do precision brazing?

If you have a 20Ghz signal generator and power generator, then you're looking at a very small resonator--about 1/10 the size of that at Eagle.  It is possible you could fit something like this on a Mettler H20.  I believe the H20 has a max load of 260g and resolution of 0.05mg, but I am terrible with numbers so do look the specs up if you are interested.  If the Mettler will work, I have an H20 I can have sent to you.  It's in a lab in PA and I'm not so we are at the mercy of the lab here, but should be no trouble.

Note that the Mettler is mostly metal, and you will have serious coupling issues.  So you will need a Mu Metal shield to wave around at the least, and a full magnetic shield would be better.  Note that Mu Metal is highly anisotropic and needs special annealing done at the factory in order to form a real shield.  You should not plan to anneal it yourself after forming.  So waving around a simple sheet of foil to test for coupling is a better, prior choice and wait to spend cash on a real shield if you can.

Not sure you can use a signal generator and power generator.  The cavity has a Q of something like 2,500?  so one expects it has a great deal of reflected power.  In any case, the reason Eagle used a continuous wave magnetron over a simple microwave oven magnetron, was normal vacuum tubes cannot stand the strain of a standing wave.  That might be true of your power amp as well.

Generally in these kinds of explorations, labor is the most expensive cost, so DIY means you stand to save a lot (but you get what you pay for).  Next is the cost of the instrumentation, primarily balance and vacuum, but also some very expensive scopes and such to look at what the power system is providing. Paul mentioned one particular instrument he found as ideal and extremely cost effective so you should scour the Eagle paper for mention of this, and write Paul if you don't find it.  Third in cost is the power system.  You will need a PLL circuit.  Eagle remarked several times what a huge step forward it was to provide resonance matching.  I've been trying to provide the same power circuitry to Woodard for years but even though PLL circuits are relatively easy for an EE to design and build, microwave circuits are not.  They are considered "black magic" and very difficult to engineer in the microwave region because of all the stray capacitances, etc.  So do indeed consider the power system a serious challenge no matter what choices you make.  Finally comes the thruster which is generally the simplest, cheapest part of these systems.  Don't be deceived about the complexity of the task by the simplicity of the thruster.

I designed an M-E replication lab a year ago so I have an idea of the challenge here.  It is between $1-2 million to do this as cheaply as possible with professional resources.  I would not want to try it on a hobbyist budget.
« Last Edit: 11/15/2014 03:19 PM by Ron Stahl »

Offline DIYFAN

  • Member
  • Posts: 48
  • Liked: 28
  • Likes Given: 149
Re: EM Drive Developments
« Reply #3247 on: 11/15/2014 04:56 PM »
So I need a copper cavity. Those aren't easy to come by and I'm too lazy to buy copper sheet to build my own and I'd probably build it like crap anyway. But I remembered that I can get my hands on a little brass bell just about anywhere. Look in your Christmas decorations. Once a suitable bell is found, it is a straight forward exercise to drill a hole in it to mount an rf connector, fabricate a suitable loop probe and solder it into the rf connector cup, glue some dielectric material in it, and then cut out a copper sheet and solder the thing shut. A quick and dirty resonant cavity on the cheap.

For the cavity itself, metal 3d printing is now ubiquitous and affordable.

http://gpiprototype.com/services/metal-3d-printing.html

https://www.solidconcepts.com/technologies/direct-metal-laser-sintering-dmls/?gclid=CMCy2rWzmMECFQqCfgod3A4AXw

(and many others)

A simple CAD design and a submission to a 3d printing company can yield a well-formed prototype cavity within 1-2 weeks. The prototype can be formed of a variety of metals. The prototype could be tested in a non-superconducting configuration first to get a baseline. Then, the inner portion of the cavity could be lined with YBCO film, cooled to liquid nitrogen temperatures, and tested in a superconducting configuration.

It even appears that some universities have access to 3d printers that are capable of printing using superconducting materials.  Perhaps this would make a good senior project or thesis for an ambitious student.  I always wished I'd taken more advantage of my university's resources when I had the chance.
http://www.tamuk.edu/engineering/departments/mien/3D%20Printers/index.html
« Last Edit: 11/15/2014 06:42 PM by DIYFAN »

Online Mulletron

  • Full Member
  • ****
  • Posts: 1113
  • Liked: 776
  • Likes Given: 1013
Re: EM Drive Developments
« Reply #3248 on: 11/15/2014 06:25 PM »
Quote
Have you worked with microwave before?
Lol yes. I know what I'm doing. Radars and satcom are what I do. The biggest issue here is trying to couple rf from a 40 pound sweep generator to the test article because of the cable strain. Rf cables are heavy and rigid and using them would screw everything up. If I can't figure this out, I'm not doing it. I'd rather have a small rf generator balanced right there on the thing, running on batteries, but I'm not spending money buying one. I want to use an actual sweep generator so I can tune it and provide an rf sweep. But the cable issue is daunting. I'm thinking of using an xbee pro or putting an old wifi access point in CW mode. I have those. The best I can get in any case is +20dbm, so the test article needs to be small and light. The low powers involved would mean keeping the thing running for hours or days to see if any rotation happens.
« Last Edit: 11/15/2014 06:32 PM by Mulletron »
Challenge your preconceptions, or they will challenge you. - Velik

Offline Ron Stahl

  • Full Member
  • **
  • Posts: 210
  • Liked: 32
  • Likes Given: 2
Re: EM Drive Developments
« Reply #3249 on: 11/15/2014 06:40 PM »
The solution for the trouble you note used on the ARC Lite is to use liquid metal (Galinstan) contact pots, where the pot is stationary and a small cylindrical cross section probe rotates inside the liquid.  The pots are arranged directly above the C-flex bearings so there is no displacement at all of the probe inside the Galinstan.  This is what gives the ARC Lite such fantastical resolution as compared to all the previous torsion balances, such as the one at the Austrian Research Center.

As to this notion of a superconducting resonator, I would just note that in order to line the chamber with YBCO, at the least the outside of the resonator would need to be immersed in liquid nitrogen and linked to a dewar.  This is a total nightmare for the balance and contrary to the opine over at Reddit, there is no way to do something like this for $2k.  I doubt it can be done for $2M.

Online Mulletron

  • Full Member
  • ****
  • Posts: 1113
  • Liked: 776
  • Likes Given: 1013
Re: EM Drive Developments
« Reply #3250 on: 11/15/2014 06:53 PM »
The solution for the trouble you note used on the ARC Lite is to use liquid metal (Galinstan) contact pots, where the pot is stationary and a small cylindrical cross section probe rotates inside the liquid.  The pots are arranged directly above the C-flex bearings so there is no displacement at all of the probe inside the Galinstan.  This is what gives the ARC Lite such fantastical resolution as compared to all the previous torsion balances, such as the one at the Austrian Research Center.

As to this notion of a superconducting resonator, I would just note that in order to line the chamber with YBCO, at the least the outside of the resonator would need to be immersed in liquid nitrogen and linked to a dewar.  This is a total nightmare for the balance and contrary to the opine over at Reddit, there is no way to do something like this for $2k.  I doubt it can be done for $2M.

I was thinking about liquid metal all day but couldn't remember what it was called. I know that's what Eagleworks used. I don't want to mess with mercury so thanks for the tip on the Galinstan. I think, with some bench time, I could build a f-f TNC or SMA bridge.

This setup needs to be only a few ounces at most and completely balance from low test fishing line or thread to even have a chance of not being a waste of time. So I have to solve the rf cable problem first.
« Last Edit: 11/15/2014 06:55 PM by Mulletron »
Challenge your preconceptions, or they will challenge you. - Velik

Offline Star One

  • Senior Member
  • *****
  • Posts: 8017
  • UK
  • Liked: 1281
  • Likes Given: 168
Re: EM Drive Developments
« Reply #3251 on: 11/15/2014 07:01 PM »
Shouldn't we be hearing more from EagleWorks sooner rather than later, weren't they supposed to be doing more tests by now?

Offline Ron Stahl

  • Full Member
  • **
  • Posts: 210
  • Liked: 32
  • Likes Given: 2
Re: EM Drive Developments
« Reply #3252 on: 11/15/2014 07:07 PM »
I'm sure they do tests nearly every day, but they don't release results without reason--such as fundraising.

Mullet, I suggest make provision for more than the necessary 2 pots as Woodward later added separate power channels for other reasons.  If for instance you put an M-E Thruster on there and want to put DC Bias on parts of the stack and not on other parts, you need separate channels for this.  Just make the brace taller than you need.  And I do recommend acrylic with urethane sheeting strategically placed to remove any stray resonances.  Dr. Rodel's stray resonance modeling was impressive, but putting the sheeting in is wise for many reasons.

What do you intend to isolate the balance from seismic?  Putting the whole thing on a platform sitting atop an inner tube is a cheap trick that seems to work well.
« Last Edit: 11/15/2014 07:29 PM by Ron Stahl »

Offline DIYFAN

  • Member
  • Posts: 48
  • Liked: 28
  • Likes Given: 149
Re: EM Drive Developments
« Reply #3253 on: 11/15/2014 07:08 PM »
. . . there is no way to do something like this for $2k.  I doubt it can be done for $2M.

With all due respect, I suggest it is the DIY vs. the Institutional mindset.  I remember my high school teacher using liquid nitrogen in the classroom near students to freeze and smash banana peels into bits.
« Last Edit: 11/15/2014 07:23 PM by DIYFAN »

Offline Ron Stahl

  • Full Member
  • **
  • Posts: 210
  • Liked: 32
  • Likes Given: 2
Re: EM Drive Developments
« Reply #3254 on: 11/15/2014 07:19 PM »
I remember that too.  The issue is that this stuff needs to rotate freely on the balance arm.  Just as Mullet is looking at how to feed power through the balance fulcrum without causing resistance that would show up as a spurious source, you would need to feed liquid nitrogen through and that is far more problematic.  It's a serious problem.  You'd likely need to place the entire cryo system on the mobile portion of the balance and then you'd have a huge period to cope with, and nitrogen boil off to account for.  It opens a whole new can of nasty worms.  The coolant cycling through the system would generate far larger forces than the thruster, and even though they ought to cancel, that cancelation might be time dependent based on how the coolant cycles.  So you could entirely swamp the system with spurious sources you can't get rid of.  It's a terrible solution for a tiny budget.
« Last Edit: 11/15/2014 07:23 PM by Ron Stahl »

Offline DIYFAN

  • Member
  • Posts: 48
  • Liked: 28
  • Likes Given: 149
Re: EM Drive Developments
« Reply #3255 on: 11/15/2014 07:24 PM »
I remember that too.  The issue is that this stuff needs to rotate freely on the balance arm.  Just as Mullet is looking at how to feed power through the balance fulcrum without causing resistance that would show up as a spurious source, you would need to feed liquid nitrogen through and that is far more problematic.  It's a serious problem.  You'd likely need to place the entire cryo system on the mobile portion of the balance and then you'd have a huge period to cope with, and nitrogen boil off to account for.  It opens a whole new can of nasty worms.  The coolant cycling through the system would generate far larger forces than the thruster, and even though they ought to cancel, that cancelation might be time dependent based on how the coolant cycles.  So you could entirely swamp the system with spurious sources you can't get rid of.  It's a terrible solution for a tiny budget.


If a resonant frequency can be found with a small superconducting test article, it won't be a matter of detecting tiny force measurements on a delicate balance.  If Shawyer type predicted forces bear out, then it would be more a matter of making sure the test article doesn't take off.

Offline Ron Stahl

  • Full Member
  • **
  • Posts: 210
  • Liked: 32
  • Likes Given: 2
Re: EM Drive Developments
« Reply #3256 on: 11/15/2014 07:57 PM »
Okay.  Lets say you want to test for those magnitude of forces.  Are you suggesting an experimental apparatus that can only detect very large forces?  Say you put it on something like a bathroom scale.  If you get no reading, what do you do then?  Do you then build a proper balance or do you count this as a null result?

Surely it is cheaper to use a bathroom scale but there are reasons science seldom does this sort of thing.  If one is to learn even from null results, one needs more precision than is cheap and easy, and most times science learns more from its mistakes than its successes.

Offline Ron Stahl

  • Full Member
  • **
  • Posts: 210
  • Liked: 32
  • Likes Given: 2
Re: EM Drive Developments
« Reply #3257 on: 11/15/2014 08:08 PM »
Can anyone here tell me why I can't find all but one of my outgoing mail?  I've been told it is going out, but I can't see any of it, and I don't see a box to check to save outgoing mail or anything like this.  Can't figure out what sort of bone-head mistake I'm making here. . .

Offline Ron Stahl

  • Full Member
  • **
  • Posts: 210
  • Liked: 32
  • Likes Given: 2
Re: EM Drive Developments
« Reply #3258 on: 11/15/2014 08:12 PM »
If a resonant frequency can be found with a small superconducting test article, it won't be a matter of detecting tiny force measurements on a delicate balance.  If Shawyer type predicted forces bear out, then it would be more a matter of making sure the test article doesn't take off.
I should note too, that YBCO's superconductivity is highly frequency dependent.  I don't think it works past low VHF. Certainly not microwave.

Offline DIYFAN

  • Member
  • Posts: 48
  • Liked: 28
  • Likes Given: 149
Re: EM Drive Developments
« Reply #3259 on: 11/15/2014 08:21 PM »
If a resonant frequency can be found with a small superconducting test article, it won't be a matter of detecting tiny force measurements on a delicate balance.  If Shawyer type predicted forces bear out, then it would be more a matter of making sure the test article doesn't take off.
I should note too, that YBCO's superconductivity is highly frequency dependent.  I don't think it works past low VHF. Certainly not microwave.

Shawyer discloses using YBCO film both in his patent application and in reported prototype testing, with positive results. 

There is no doubt in my mind that high-precision testing in suitable scientific labs is of paramount importance in exploring this astonishing effect. 

There is also no doubt in my mind that, if proven to work, the DIY community will eventually take this up on a tiny budget and expand the field at a pace that will make our heads spin.  Just look to the DIY drone community and what they accomplished in so short a period of time.  I would submit that if it were not for such a community, you would not see the high caliber of commercial drones that you now see in the marketplace.  Don't underestimate the diligent tinkerer with a shop in the back yard, and time to spare. 

What is needed more than ever is additional confirmation of the effect.  Perhaps some solid data on a small superconducting test article.  Maybe it is the well-respected scientific labs that need to first produce such a test article and prove it is worth the time to investigate further.  Then, thanks to the Internet and tightly-knit communities that can quickly rally behind an idea, the sky becomes the limit . . . (or perhaps the moon, or mars, um, okay I'll stop).
« Last Edit: 11/15/2014 08:23 PM by DIYFAN »

Tags: