Author Topic: EM Drive Developments - related to space flight applications - Thread 5  (Read 940848 times)

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2364
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 3089
  • Likes Given: 2659
Another thought for you Shell. Haven't you talked about how cold your shop is now? Why not allow the test equipment get as cold as possible in the shop before testing so that the thermal gradient between the room and the heating frustum under load is as large as possible?
Not a great idea for these old bones. Isn't cold air denser?

Offline jmossman

  • Member
  • Posts: 73
  • San Jose, CA
  • Liked: 58
  • Likes Given: 180
It is appropriate to be concerned about the force effects of heating .
...
3) Balance the heat effect with an equal and opposite source, removing residual effects post processing.
...
Of the above, 3) has been proposed but gained no traction. Why is that? A Google search on "common mode error rejection in physical systems" will return more information than can be absorbed, but a lot of good guidance, too.

The common mode error compensation technique proposed previously is:

Make two, not just one, frustum, near identical.
Mount two frustums on opposite sides of the balance beam pivot.
Run the system normally and remove residual effects by post processing.
...
a) How expensive is it to make a second, near identical frustum? Several hundred $$ plus time.
b) How hard is it to make and mount the second frustum safely on the test stand? Not hard at all
c) How difficult will it be to take meaningful data with two 2 EM Drive effect sources? Are you suggesting 2 separate, powered magnetrons? If so, add some more $$ for a power supply, plus a controller to make them fire at the exact same time. Even with that, an imbalance will occur as air masses several feet apart are not identical, meaning there will still be residual thermals with one end versus the other. While its a good idea in theory, practical experience tell me that one could not guaranteed each side would heat and lift identically.

My take is ambient air perturbations are best left to only one source, the primary frustum. Delta displacement versus mag on/off times is extractable as has been demonstrated with earlier flight tests. As long as we're not in a vacuum, rotary or fulcrum is what we have to deal with in home lab testing.

The simplest option to help reduce thermal effects is to run a frustum/magnetron in two different configurations that are rotated by 180 degrees.  (i.e.  run#1-10: small end-up,  run#11-20: big end-up)

If a higher Q (i.e. solid frustum vs mesh frustum) significantly increases the Signal-to-Noise-Ratio (SNR), the added complexity of a 2nd frustum/magnetron may not be needed.

However, out of the various options that seem within the grasp of a DIY effort, I would agree with Aero that at least partial common-mode rejection seems like the simplest option to further reduce thermal effects (after the dual 180 degree rotations runs have been completed).  Even if the thermal forces are not 100% equal, I suspect that an identical magnetron/heatsink will have very similar thermal effects once it has reach a stabilized temperature.  Common-mode rejection in a system design is rarely expected to be perfect, but reducing the thermal effect on the balance beam movement would seem like a worthwhile effort.

My engineer "gut" feeling is that the key would be matching temperatures of the 2 independent magnetron/heatsink, rather than the ~kv power supply and simultaneous on/off.  It's likely a heat radiating phenomenon causing the lift, and could likely be well matched with a point heat source inside the magnetron controlled by a PID control loop tying the temperature to the "working" magnetron. (i.e. built-in magnetron heating element is probably what needs to be accurately controlled in the 2nd dummy frustum/magnetron setup).

EDIT:  Using a PID control loop to match temperature would also work well to provide a "control" run without adding the complexity of a 2nd dummy frustum/magnetron. (aka Seeshell's Glennfish's suggestion)

i.e.  1) measure temperature of magnetron/heatsink during a "powered run", 2) perform a "dummy run" using the same setup but using a PID control loop to control the magnetron heating element to attempt mimicing the thermal signature from a "powered run"

EDIT2:  Whoops, need to give Glennfish proper credit for the recent "control run" repost.  ;)
« Last Edit: 10/26/2015 07:11 PM by jmossman »

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2364
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 3089
  • Likes Given: 2659
A couple of points.

Thermal is the 800 pound gorilla sitting in my test lab (hope he gets cold). It's not going to go away even in vacuum where it becomes a 1800 pound gorilla.

I've come to the conclusion selecting and controlling the thermal profile where I can negate several aspects of it for short periods by profiling just one. Getting a clean linear thermal inclined line is going to be easier to map out any thrust effects by using a what is essentially a hot air thermal insulated balloon. I'll eliminate the chaotic bubbling off a Maggie thermal chimney like rfmwguy had to work with. Simply by taking all the heat and putting it into one effect not several.

This is simply going to take some testing, note a candle outputs about 40 watts of heat.


Offline zen-in

  • Full Member
  • ****
  • Posts: 536
  • California
  • Liked: 472
  • Likes Given: 371

...

Dumb question that may save you some time.

Keeping in mind that there are at least three sources of thermal "lift"
1. convective air flow, hot air rising on the walls
2. buoyancy, hot air inside weighs less than colder air outside
3. evacuating air, expanding air evacuates through any openings

Would it not be easier to set up a control run and a test run?

In the control run you totally whack your antenna (dielectric?) position internally so that your resonance is as close to zero as practicable, and measure in detail.

In the test run, you position your antenna (dielectric?) internally so that your resonance is as high as you can make it, and measure in detail.

Then subtract control data from test data and hope that you can carefully replicate everything else between the two runs.  Your only variable being (if you're really good), antenna position and resonance.

This assumes that the resonance discussions I've been seeing are related to whatever it is to be measured.

I don't believe there is any fool-proof way of nulling out thermal effects.   You can drill holes so the hot air escapes horizontally but you still have hot air accumulating in the fustrum and providing lift.   In over a year of being a spectator to this em-drive pursuit I have only seen small forces that are almost indistuinguishable from thermal effects.   So my position has always been that experimenters should characterize thermal effects at some point in their data collection so that a comparison can be made.   So far I haven't seen anyone do this.   Maybe it seems too pessimistic to do a null experiment where only heat is applied to the fustrum, but that is what we call Science.   If you don't do any counter experiments you will never know if the effect you are observing is from something more mundane.

You can follow breadcrumbs but look at where that got Hansel and Gretel.
« Last Edit: 10/26/2015 08:28 PM by zen-in »

Offline aero

  • Senior Member
  • *****
  • Posts: 2929
  • 92129
  • Liked: 783
  • Likes Given: 272
A couple of points.

Thermal is the 800 pound gorilla sitting in my test lab (hope he gets cold). It's not going to go away even in vacuum where it becomes a 1800 pound gorilla.

I've come to the conclusion selecting and controlling the thermal profile where I can negate several aspects of it for short periods by profiling just one. Getting a clean linear thermal inclined line is going to be easier to map out any thrust effects by using a what is essentially a hot air thermal insulated balloon. I'll eliminate the chaotic bubbling off a Maggie thermal chimney like rfmwguy had to work with. Simply by taking all the heat and putting it into one effect not several.

This is simply going to take some testing, note a candle outputs about 40 watts of heat.

I thought you had planned to use an antenna in your frustum with coax to the magnetron at the pivot point? That alone will remove most of the thermal effects that rfmwguy saw.

So do I understand that you intend to bag the frustum and the end of the balance beam with insulation so that heat won't conduct through the sides of the bag? I guess it would then be logical to run a pressure release tube back to the pivot and let the expanding air jet out to the side. That should reduce ballooning leaving only temperature change inside the constant volume balloon to deal with. And of course air currents in your lab which now have the bag to blow around.

I guess your bag will be about what, 0.3 m3. Thermal capacity of air, and the ideal gas law are well understood. Does someone want to calculate the bag air temperature and density profile as a function of Watts of drive power delivered to the frustum? That will tell us how long it will be before the bag melts.

Then you can watch your lift due to buoyancy to know when your bag will melt.  :-\
Retired, working interesting problems

Offline Bob Woods

  • Full Member
  • **
  • Posts: 284
  • Salem, Oregon USA
  • Liked: 390
  • Likes Given: 1248
Another thought for you Shell. Haven't you talked about how cold your shop is now? Why not allow the test equipment get as cold as possible in the shop before testing so that the thermal gradient between the room and the heating frustum under load is as large as possible?
Not a great idea for these old bones. Isn't cold air denser?

Yes, it is in general. But gas escaping past your seal on the small end will be producing thrust counter to the     thrust vector of the drive itself. If a cold room dissipates heat from all that copper surface area faster...

I guess the question of the seal sticks most in my mind. What internal temperature rise will be enough to cause the seal to leak and will you be generating that much heat?/   

Offline glennfish

  • Full Member
  • ***
  • Posts: 388
  • Liked: 281
  • Likes Given: 169

Sure, profiling is the key to picking the best way to gain data but now I get to pick which way will give me cleaner data with some chicken wire and a piece of insulation. I think it's a good investment,. What do you think?

Shell


Chicken wire is a nice faraday cage on a good day.  When my brother was doing his thesis, a faraday cage saved his life.  I like chicken wire since that day.

The insulation I think only deals with convection.  I ran a hot-air balloon simulation a while back (I'll try to find and post the spreadsheet), and I would roughly characterize the effects as

1.  Hot air balloon - 60%
2.  Convection - 30%
3.  Air leaks - 10%  (could go either direction)

Lots depends on the geometry, that strange epoxy you've fallen for, and how uniform the heating is throughout the frustum, and things I'm horribly unqualified to imagine.

I guess my concern is, when you turn it on at normal atmospheric pressure, it will absolutely positively rise like a banshee on Halloween for many reasons having nothing to do with what you're looking for whether pointed up or down.  The banshee index (new technical term Bi) will vary depending on how you point the gizmo, and someone who knows how to calculate surface areas of truncated cones could probably calculate that.

I am a very lazy person.  IMHO, being lazy keeps you from working harder than you should.

I can't imagine all the ways heat can mess things up, but I know how to subtract two data sets, especially in the case where you can repeat either a 100 times if you need to.  :) 

I could imagine a good engineer spending months compensating for all the thermal thingees and still miss one or two or thirty.

If, on the other hand, you say "Let There Be HEAT", and you get that clearly characterized, then when you say, "Let There Be HEAT AND Lift", well, IMHO, you could generate a mind blowing data set if there was something after the "AND".

Surely there's a few turns of the knob that could totally destroy your ideal resonance?  :)  That would be a simple control/test experiment and would answer the question, "Does Resonance Impact the Observed Thrust in an non-vacuum environment?"  If there's no difference, you toss me over the bridge for wasting your time.  If there is, then you have to replicate both conditions many many times, and then be beaten repeatedly at Reddit for sins against Jackson, Chapter 8.

Anyway, my free advice.  :)

Offline rfmwguy

  • EmDrive Builder (retired)
  • Senior Member
  • *****
  • Posts: 2166
  • Liked: 2684
  • Likes Given: 1124
These are great discussion on thermal problems, which I lived with. I did get a thermal camera attachment and plan to static-fire NSF-1701 and observe the heating patterns soon. I fully expect the majority to be directly off the magnetron itself (based on my IR spot temp measurements) The double core, copper shaving filled "heat trap" I was musing about might just work long enough to reduce the large lift transitions I saw and Glenn had to disect (to do the mag on/off comparisons).

Thermal migration (conduction) through an inner wall, thru shavings and to the outer wall then the air itself will be much slower than the magnetron frame being exposed to the air directly. Will be fun to experiment with.

p.s. For those not following earlier threads, Shell's copper shavings are insulation for an inner and outer can that surrounds the magnetron. Think of it as a muffler, only its a heat sink that delays the 170 deg C heat from hitting the surrounding ambient air. - Crazy? Yep, but don't tell anyone  :o

Offline Prunesquallor

  • Full Member
  • *
  • Posts: 174
  • Currently, TeV Brane Resident
  • Liked: 157
  • Likes Given: 73
Perhaps it makes more sense to create a bad vacuum (e.g. one tenth of atmospheric pressure) with a simple and cheap pump in an enclosure with the test article in it? We just want to increase SNR considerably, right? I don't think that a hard vacuum is needed. We just want to get rid of most of the buoyancy for now. My 2 cents ;) .

Edit: Actually, it should easily be possible to create a pretty hard vacuum the cheap way:

1) 3D-print a metallic enclosure with cooling channels for liquid nitrogen in the walls, perhaps even just the bottom/floor of the enclosure.
2) Put an automated, complete test article in it (sorta like a space probe)
3) Shut the enclosure and fill it with pure CO2, so that all other gases are pushed out of the enclosure
4) Seal the enclosure and start pumping liquid nitrogen through the wall channels. The CO2 freezes out, until there's only solid dry ice left (maybe best only on bottom/floor of enclosure)
5) You got vacuum  8)

What do you guys think about this method?

I suspect that even at LN2 temperatures, CO2 will sublime before a very good vacuum is attained.
Retired, yet... not

Offline aero

  • Senior Member
  • *****
  • Posts: 2929
  • 92129
  • Liked: 783
  • Likes Given: 272
Shell,

After my most recent post, I had another thought for you to consider.

If you run a pressure relief line from the bag to the pivot point, what is to prevent you from making that pressure relief line, say 4" diameter sewage line, then make another in parallel as a cool air feed line. Use a computer cooling fan to blow cool air into the bag through the one pipe which forces the warm air to escape via the other pipe.

Done carefully and with a small fan, the fan turbulence should settle out as the cooling air transits the inlet pipe reaching laminar flow conditions at the bag. The cool air picks up the heat and exits out the other pipe at the pivot. That way there should be little concern about pressure in the bag, leaks from the bag, and buoyancy is much reduced.

And to reduce turbulence even more, configure the fan to draw air out of the tube/bag/tube and blow it sideways at the pivot, that is, in any direction except up and down. You could even turn your fan around to see which direction of air flow you prefer. Or maybe you could draw the air through the bag horizontally and perpendicular to the beam and forgo pipes to the pivot point.

Point is, now that you have encapsulated the heat source there are several things you can do to cool your frustum without introducing a lot of extra thermal or turbulence noise.
Retired, working interesting problems

Offline Prunesquallor

  • Full Member
  • *
  • Posts: 174
  • Currently, TeV Brane Resident
  • Liked: 157
  • Likes Given: 73

...

Dumb question that may save you some time.

Keeping in mind that there are at least three sources of thermal "lift"
1. convective air flow, hot air rising on the walls
2. buoyancy, hot air inside weighs less than colder air outside
3. evacuating air, expanding air evacuates through any openings

Would it not be easier to set up a control run and a test run?

In the control run you totally whack your antenna (dielectric?) position internally so that your resonance is as close to zero as practicable, and measure in detail.

In the test run, you position your antenna (dielectric?) internally so that your resonance is as high as you can make it, and measure in detail.

Then subtract control data from test data and hope that you can carefully replicate everything else between the two runs.  Your only variable being (if you're really good), antenna position and resonance.

This assumes that the resonance discussions I've been seeing are related to whatever it is to be measured.

I don't believe there is any fool-proof way of nulling out thermal effects.   You can drill holes so the hot air escapes horizontally but you still have hot air accumulating in the fustrum and providing lift.   In over a year of being a spectator to this em-drive pursuit I have only seen small forces that are almost indistuinguishable from thermal effects.   So my position has always been that experimenters should characterize thermal effects at some point in their data collection so that a comparison can be made.   So far I haven't seen anyone do this.   Maybe it seems too pessimistic to do a null experiment where only heat is applied to the fustrum, but that is what we call Science.   If you don't do any counter experiments you will never know if the effect you are observing is from something more mundane.

You can follow breadcrumbs but look at where that got Hansel and Gretel.

I agree.  If there is a way to fire up all the heat-generating components, but guarantee that thrust is not generated (short antenna??, detune frustum??), measure the pure thermal effects and then subtract this signal from a (presumably) thrust-producing configuration, it should prove informative.
Retired, yet... not

Offline tchernik

  • Full Member
  • **
  • Posts: 267
  • Liked: 311
  • Likes Given: 631
Esteemed DIYers:

I was thinking and some probably dumb idea came to me: maybe trying to get rid of buoyancy is not worth the effort?

You see, this frustum is and wants to behave like a hot air copper balloon.

It will never rise on its own because it's too heavy, but it still stubbornly tries to push any balance upwards, regardless of what you do, because it will be filled with hot air from the maggie's very function.

And any dynamic cooling you do would most likely make the noise level on the forces worse.

While at the same time, buoyancy is a tamed beast, because you know where it wants to go every time: upwards.

So you simply measure the strength of the upwards force in function of the temperature, and you subtract that from  your other experiments.

Because you will try to prove it pushes downwards, rightwards, leftwards, backwards, etc, in every angle. That is, you want to prove the thrust of the device is vectorial.

Proving vectorial thrust would indeed make things much more interesting, because less things can explain that, and the methods to disprove these additional effects can take you beyond the chore of simply proving it isn't thermal buoyancy in action.

For example, for disproving a thermal rocket effect by heated gas ejection, you could use a smoke test, for showing that the convection currents around the device don't make a rocket-like effect. Or the contrary, that they are indeed making such an effect, therefore falsifying the anomalous vectorial thrust hypothesis.

Which would also be progress, regardless of our crushed hopes.

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2364
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 3089
  • Likes Given: 2659
A couple of points.

Thermal is the 800 pound gorilla sitting in my test lab (hope he gets cold). It's not going to go away even in vacuum where it becomes a 1800 pound gorilla.

I've come to the conclusion selecting and controlling the thermal profile where I can negate several aspects of it for short periods by profiling just one. Getting a clean linear thermal inclined line is going to be easier to map out any thrust effects by using a what is essentially a hot air thermal insulated balloon. I'll eliminate the chaotic bubbling off a Maggie thermal chimney like rfmwguy had to work with. Simply by taking all the heat and putting it into one effect not several.

This is simply going to take some testing, note a candle outputs about 40 watts of heat.

I thought you had planned to use an antenna in your frustum with coax to the magnetron at the pivot point? That alone will remove most of the thermal effects that rfmwguy saw.

So do I understand that you intend to bag the frustum and the end of the balance beam with insulation so that heat won't conduct through the sides of the bag? I guess it would then be logical to run a pressure release tube back to the pivot and let the expanding air jet out to the side. That should reduce ballooning leaving only temperature change inside the constant volume balloon to deal with. And of course air currents in your lab which now have the bag to blow around.

I guess your bag will be about what, 0.3 m3. Thermal capacity of air, and the ideal gas law are well understood. Does someone want to calculate the bag air temperature and density profile as a function of Watts of drive power delivered to the frustum? That will tell us how long it will be before the bag melts.

Then you can watch your lift due to buoyancy to know when your bag will melt.  :-\
What a day, TV took a big you know what and just got that fixed. Went out to pick up a few things and my truck died (some digital sensor) and is now in the mechanics and because of that the day was shot. Finally got a ride home, thank goodness I like cars (have a few) so I'm not with out wheels.

So sorry ppl no new pics from the Crazy Eddie Lab. Tomorrow is another day and I'll start early.

Still reading all the fine comments by everyone on the thermal problem and I've decided to profile the dickens out of it with active heat (light bulb ~100 watts) and active without thermal shielding and with shielding. And of course flip the frustum 180.

Aero, yes the magnetron is away from the frustum and I'm just running the microwaves down a Coax to the frustum, the only heat will be from the microwaves actions heating the frustum which will be about 100 watts. I'll need to do real world testing to make sure the math works out as to the 100 watts.

Interesting idea on the cooling tubes aero, I kind of like it.
 


Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2364
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 3089
  • Likes Given: 2659

Sure, profiling is the key to picking the best way to gain data but now I get to pick which way will give me cleaner data with some chicken wire and a piece of insulation. I think it's a good investment,. What do you think?

Shell


Chicken wire is a nice faraday cage on a good day.  When my brother was doing his thesis, a faraday cage saved his life.  I like chicken wire since that day.

The insulation I think only deals with convection.  I ran a hot-air balloon simulation a while back (I'll try to find and post the spreadsheet), and I would roughly characterize the effects as

1.  Hot air balloon - 60%
2.  Convection - 30%
3.  Air leaks - 10%  (could go either direction)

Lots depends on the geometry, that strange epoxy you've fallen for, and how uniform the heating is throughout the frustum, and things I'm horribly unqualified to imagine.

I guess my concern is, when you turn it on at normal atmospheric pressure, it will absolutely positively rise like a banshee on Halloween for many reasons having nothing to do with what you're looking for whether pointed up or down.  The banshee index (new technical term Bi) will vary depending on how you point the gizmo, and someone who knows how to calculate surface areas of truncated cones could probably calculate that.

I am a very lazy person.  IMHO, being lazy keeps you from working harder than you should.

I can't imagine all the ways heat can mess things up, but I know how to subtract two data sets, especially in the case where you can repeat either a 100 times if you need to.  :) 

I could imagine a good engineer spending months compensating for all the thermal thingees and still miss one or two or thirty.

If, on the other hand, you say "Let There Be HEAT", and you get that clearly characterized, then when you say, "Let There Be HEAT AND Lift", well, IMHO, you could generate a mind blowing data set if there was something after the "AND".

Surely there's a few turns of the knob that could totally destroy your ideal resonance?  :)  That would be a simple control/test experiment and would answer the question, "Does Resonance Impact the Observed Thrust in an non-vacuum environment?"  If there's no difference, you toss me over the bridge for wasting your time.  If there is, then you have to replicate both conditions many many times, and then be beaten repeatedly at Reddit for sins against Jackson, Chapter 8.

Anyway, my free advice.  :)
Your free advise is priceless, thank you.

On the tuning. I  picked up a couple of reversable DC motors that can drive the micrometer on the bottom after I attach it to it. I can allow me to sweep through peak resonate modes while monitoring the thrust levels, this can tell me a lot, like does max thrust occur at center frequency or just a little off, or does it vary when mixing two or more modes together?

There will be heat but the less of a issue I can make of it the better for detailing out the effects I just mentioned.

Shell

Offline RotoSequence

  • Full Member
  • ****
  • Posts: 906
  • Liked: 648
  • Likes Given: 829
Characterizing the thermal characteristics of the frustum with a heat source (light bulb), both upside down and right side up, looks like a really good idea. The scientists at CERN have been doing a lot of work to characterize ordinary Proton-Proton collision products at the Large Hadron Collider at the new, 13 TeV collision energies. When they've characterized the ordinary results, they can easily see the extraordinary ones.
« Last Edit: 10/27/2015 01:50 AM by RotoSequence »

Offline TheTraveller

I agree.  If there is a way to fire up all the heat-generating components, but guarantee that thrust is not generated (short antenna??, detune frustum??), measure the pure thermal effects and then subtract this signal from a (presumably) thrust-producing configuration, it should prove informative.

Problem is as the freq moves from that of frustum resonance, the VSWR climbs and Rf starts being reflected back to the Rf gen and not absorbed by the frustum. Can't see a way to get the Rf inside the frustum without the Rf freq being at frustum resonance.

Still believe the best way to deal with the thermal effects of buoyancy is to mount the frustum horizontal as EW did and I plan to do plus working to increase the Force generated for a fixed amount of heat.

From this data it looks like Roger's Experimental EMDrive generated about 200uN of buoyancy for 14mN of Force at 850Ws Rf. You can also see his sealed enclosure and the simple scale setup he used way back in 2002.

To me this suggest the approx amounts of buoyancy that may need to be dealt with and more importantly a guide to the Force generation that may be possible with a flat end plate design with no electronic freq tracking but with mechanical tuning at each end plus waveguide impedance matching to get a really good VSWR. Or close to what Shell is building.

It might also be helpful to look at the schematic of the setup he build and how he dealt with the heat buildup.

Roger passed this info on to us TO USE as I believe he feels it is the most appropriate info at our stage of DIY build. His only reason to do so was to again lay a trail of beard crumbs that if followed will eventually lead to most of us being able to make DIY EMDrives with 10mN level Force generation (lower Q flat end plate 750W maggie units or higher Q spherical end plate 100W solid state Rf amp units) or about where he was from 2002 to around 2009.
« Last Edit: 10/27/2015 02:37 AM by TheTraveller »
It Is Time For The EmDrive To Come Out Of The Shadows

Offline Bob Woods

  • Full Member
  • **
  • Posts: 284
  • Salem, Oregon USA
  • Liked: 390
  • Likes Given: 1248
I agree.  If there is a way to fire up all the heat-generating components, but guarantee that thrust is not generated (short antenna??, detune frustum??), measure the pure thermal effects and then subtract this signal from a (presumably) thrust-producing configuration, it should prove informative.

Problem is as the freq moves from that of frustum resonance, the VSWR climbs and Rf starts being reflected back to the Rf gen and not absorbed by the frustum. Can't see a way to get the Rf inside the frustum without the Rf freq being at frustum resonance.

Still believe the best way to deal with the thermal effects of buoyancy is to mount the frustum horizontal as EW did and I plan to do plus working to increase the Force generated for a fixed amount of heat.

From this data it looks like Roger's Experimental EMDrive generated about 200uN of buoyancy for 14mN of Force at 850Ws Rf. You can also see his sealed enclosure and the simple scale setup he used way back in 2002.

To me this suggest the approx amounts of buoyancy that may need to be dealt with and more importantly a guide to the Force generation that may be possible with a flat end plate design with no electronic freq tracking but with mechanical tuning at each end plus waveguide impedance matching to get a really good VSWR. Or close to what Shell is building.

It might also be helpful to look at the schematic of the setup he build and how he dealt with the heat buildup.

Roger passed this info on to us TO USE as I believe he feels it is the most appropriate info at our stage of DIY build. His only reason to do so was to again lay a trail of beard crumbs that if followed will eventually lead to most of us being able to make DIY EMDrives with 10mN level Force generation (lower Q flat end plate 750W maggie units or higher Q spherical end plate 100W solid state Rf amp units) or about where he was from 2002 to around 2009.

Thanks for the PDF explanations between the pictures.

Offline demofsky

  • Full Member
  • *
  • Posts: 170
  • Liked: 101
  • Likes Given: 1736
p
A couple of points.

Thermal is the 800 pound gorilla sitting in my test lab (hope he gets cold). It's not going to go away even in vacuum where it becomes a 1800 pound gorilla.

I've come to the conclusion selecting and controlling the thermal profile where I can negate several aspects of it for short periods by profiling just one. Getting a clean linear thermal inclined line is going to be easier to map out any thrust effects by using a what is essentially a hot air thermal insulated balloon. I'll eliminate the chaotic bubbling off a Maggie thermal chimney like rfmwguy had to work with. Simply by taking all the heat and putting it into one effect not several.

This is simply going to take some testing, note a candle outputs about 40 watts of heat.

I thought you had planned to use an antenna in your frustum with coax to the magnetron at the pivot point? That alone will remove most of the thermal effects that rfmwguy saw.

So do I understand that you intend to bag the frustum and the end of the balance beam with insulation so that heat won't conduct through the sides of the bag? I guess it would then be logical to run a pressure release tube back to the pivot and let the expanding air jet out to the side. That should reduce ballooning leaving only temperature change inside the constant volume balloon to deal with. And of course air currents in your lab which now have the bag to blow around.

I guess your bag will be about what, 0.3 m3. Thermal capacity of air, and the ideal gas law are well understood. Does someone want to calculate the bag air temperature and density profile as a function of Watts of drive power delivered to the frustum? That will tell us how long it will be before the bag melts.

Then you can watch your lift due to buoyancy to know when your bag will melt.  :-\
What a day, TV took a big you know what and just got that fixed. Went out to pick up a few things and my truck died (some digital sensor) and is now in the mechanics and because of that the day was shot. Finally got a ride home, thank goodness I like cars (have a few) so I'm not with out wheels.

So sorry ppl no new pics from the Crazy Eddie Lab. Tomorrow is another day and I'll start early.

Still reading all the fine comments by everyone on the thermal problem and I've decided to profile the dickens out of it with active heat (light bulb ~100 watts) and active without thermal shielding and with shielding. And of course flip the frustum 180.

Aero, yes the magnetron is away from the frustum and I'm just running the microwaves down a Coax to the frustum, the only heat will be from the microwaves actions heating the frustum which will be about 100 watts. I'll need to do real world testing to make sure the math works out as to the 100 watts.

Interesting idea on the cooling tubes aero, I kind of like it.

I really, really liked the original idea of having a "relief line" to the beam pivot.  Pumping in cooling air seemed like it might introduce vibrations. 

However,  a passive line may have a resonance introduced as cooler air forces its way back to the fustrum via the relief line.

So maybe having a cooling stream pumped to the fustrum and introducing a positive pressure and hopefully consistent signal might be the best approach after all.  Hm...

This whole approach requires some thought but one way or another I really think areo has a breakthrough idea here!  :D

Offline SeeShells

  • Senior Member
  • *****
  • Posts: 2364
  • Every action there's a reaction we try to grasp.
  • United States
  • Liked: 3089
  • Likes Given: 2659
" http://forum.nasaspaceflight.com/index.php?topic=38577.msg1439633#msg1439633
This whole approach requires some thought but one way or another I really think areo has a breakthrough idea here!  :D "

Yep he does, aero is one sharp man. It's called a controlled micro environment. Add a a flex cooling line and a fan... bingo!

The cooling hose only connects to the fulcrum beam and the fan isn't attached to the beam but sits on the base of the testing stand. Done correctly this would have little or no impact on the movement of the beam and remove most of the thermal balloon effects.
http://masterduct.com.tempdomain.com/SearchProducts/SearchResults/tabid/116/CategoryID/22/List/1/Level/a/ProductID/73/language/en-US/Default.aspx
I think it is something that just might work... what do you think?

Shell

Added: Sorry this is such a crude drawing and I'm sure there are mods that need to be considered, air insertion points and attachments to the beam and materials. I was simply excited.

One more thing... http://www.homedepot.com/p/Tripp-Lite-Portable-Cooling-Unit-or-Air-Conditioner-3-4-kW-120-Volt-60-Hz-12K-BTU-SRCOOL12K/203796126

Oops another: http://www.homedepot.com/p/Speedi-Products-3-in-x-20-ft-Standard-White-Vinyl-Flexible-Hose-EX-SVH-03/202907361
« Last Edit: 10/27/2015 12:22 PM by SeeShells »

Offline glennfish

  • Full Member
  • ***
  • Posts: 388
  • Liked: 281
  • Likes Given: 169
" http://forum.nasaspaceflight.com/index.php?topic=38577.msg1439633#msg1439633
This whole approach requires some thought but one way or another I really think areo has a breakthrough idea here!  :D "

Yep he does, aero is one sharp man. It's called a controlled micro environment. Add a a flex cooling line and a fan... bingo!

The cooling hose only connects to the fulcrum beam and the fan isn't attached to the beam but sits on the base of the testing stand. Done correctly this would have little or no impact on the movement of the beam and remove most of the thermal balloon effects.
http://masterduct.com.tempdomain.com/SearchProducts/SearchResults/tabid/116/CategoryID/22/List/1/Level/a/ProductID/73/language/en-US/Default.aspx
I think it is something that just might work... what do you think?

Shell

Added: Sorry this is such a crude drawing and I'm sure there are mods that need to be considered, air insertion points and attachments to the beam and materials. I was simply excited.

One more thing... http://www.homedepot.com/p/Tripp-Lite-Portable-Cooling-Unit-or-Air-Conditioner-3-4-kW-120-Volt-60-Hz-12K-BTU-SRCOOL12K/203796126

Well, there is that old HVAC law that relates input temperature to output temperature and calculates airflow required to achieve X BTU of cooling or heating.

How many BTU of heat are you producing?  What degree of temperature reduction is required?  Is the airflow directed at convection, lift, outgassing, all?   

Methinks active cooling adds complexity and additional error sources.  You will STILL  have to characterize the thermal effects because active cooling can't get all of them.

If you still have to characterize thermal, your life is simpler if you don't bother with active cooling.

I like simple.  Fewer parts to break or calibrate.

In my minds eye, you have two data sets derived from say, 100 - 10second runs per data set.

Data set one is with the frustum in bad resonance mode.   You get a set of lift curves and come up with a range of "normal" lift curves with error bars around sample points.

Data set two is with the frustum in good resonance mode, and again, you get a set of lift curves and come up with a range of "normal" lift curves with error bars around sample points.

Now you have a comparison that can be made.  If the two curves differ outside the range of their respective error bars, then something needs to be explained.  If the only change is the position of a dialectric and its impact on resonance, then either its position and the corresponding change on the center of mass has to be reviewed, or, resonance creates more heat than no resonance, or it's vacuum chamber time.   That would definitely call for an independent replication in any case.

If you add a fan and vent tubes and intentional air flow, all critiques will focus on the magnetic character of the fan, the stability of airflow (you need a second set of data to prove the airflow is constant at all times), you need a third set of data to monitor the temperature at multiple locations in the system, you need to verify that there were no vertical or horizontal breaches in your duct-work before and after each run, etc. etc. etc.  You could easily spend most of your of time defending your air conditioner whether you got positive or negative results.

Simplicity, young weedhopper, simplicity.

Still thinking about horizontal vs. vertical ... if you have a vertical design, you're close to first light.  If you have to go horizontal, all kinds of new issues enter the mix.
« Last Edit: 10/27/2015 12:49 PM by glennfish »

Tags: