-
#2540 by Monomorphic on 03 Nov, 2017 00:36
-
More progress on the insulation. All components have been placed back onto the torsional pendulum after some testing with the antenna indicates that it is not self-resonating, and there is a reasonable field in the cavity.
I would still like to improve the heat sink on the main amplifier.
-
#2541 by tleach on 03 Nov, 2017 00:57
-
Yes, Thanks, Paul.
It is likely not enough for detailed calculations. (I forgot to ask about the quality factor, was it on the order of 7000 for these runs?) That's OK because I'm not sure how to run the calculations anyway. But it is enough to illustrate the thought and question that I want to ask about.
The thing is, the frustum is warmer at the big end than the small end so RF energy is being dissipated preferentially at the big end. Inside the frustum, doesn't that mean that the reflection coefficient is lower at the big end than the small end? If I'm reading the images right, the big end temperature rise is 6.6 K while the small end temperature is effectively unchanged. I will assume from this that the reflection coefficient at the small end is near enough to one as to make no difference while at the large end something on the order of 100 watts is being dissipated.
Now, I know that there are many ways to cause the reflection coefficient to differ end to end, one way is to use different end materials for example. Another way is to highly buff one end only. But that is not the case with this data. It seems to me from the temperature data, that another way to make the reflection coefficient differ is to operate one end of the frustum near the cut-off frequency.
The reflection coefficient is a wave phenomenon. Proceeding with the thought and using Q = energy stored/energy lost per second gives Q = power stored/power lost, so power stored = Q * power lost. With Q = 7000 and drive power (power lost) = 100W, power stored = a big number. 0.7 MW. If the reflection coefficient at the small end = 1, then the reflection coefficient at the big end seems to equal (power stored - power lost)/power stored which is (1 - 1/Q), or in this case 0.99986.
Taking a step further out on the limb, using radiation pressure, P = Ef/c (Ef is radiation intensity) and for a perfectly reflected wave, P = 2*Ef/c and introducing the reflection coefficient, Γ, calculate the internal pressures on the ends of the frustum as:
Pressure Force, P = 2 *Γ *Ef/c . I have arbitrarily introduced the reflection coefficient as it seems right to me.
But I must take another step. Ef is defined as radiation intensity, in units of N/m^2 and I have E on the ends of the frustum. I'm guessing that the frustum end diameters are: inside big dia. = 0.25552668 ; inside small dia. = 0.15875 from my meep control file for the Brady cavity. Those numbers give A_be = 0.0513 and A_se = 0.0198 m^2. This gives
big end intensity, Ef_be = 1.37E+7 and
small end intensity, Ef_se = 3.54 E+7
Factoring these parameters into the pressure force equation and using differential pressure, delta P = P_se - P_be gives
delta P = 2 * ( Ef_se - Γ * Ef_be)/c . This is now in units of N/m^2 though and I need total. Maybe the best shot is simply to use total stored energy and not factor areas into then back out of the equation and call it Thrust.
Thrust = 2 * 0.7E+6* (1 - Γ ) / c
This number calculates to 0.667 micro-N. That is different than 83.6 micro-Newtons from Paul's temperature image data.
Lots of problems here. It gives the wrong answer by 2 orders of magnitude. That could be due to miss treatment of the areas of the two ends, or perhaps the Q value is wrong, or maybe the experimental force number is not the right data point to use, or maybe the temperatures were measured too long after power-off allowing too much heat to be conducted away/lost from the ends which would give the wrong ratio of reflection coefficients. Or maybe there is nothing to it at all. Lots of models can easily fit one data point. Still, I would like some input from the forum.
I found some related stuff here - https://arxiv.org/pdf/0807.1310.pdf
I came across this a couple of weeks ago. I don't remember if I ever saw it brought up on the forums and the math is way beyond me, but the author's conclusion is that the radiation pressure always ends up canceling out once all of the factors are accounted for.
Might be worth it for you to take a look at if you're headed down this particular rabbit hole
http://www.gregegan.net/SCIENCE/Cavity/Cavity.htmlQuoteProof of zero force for any shape of cavity
Despite the asymmetry of our truncated spherical cone along the z-axis, the net force from radiation pressure on its walls is zero. How can we be sure, though, that there isn’t some other shape that will yield a non-zero net force?
To see what the net force will be in a resonant cavity of a completely arbitrary shape, we need to construct the stress tensor [10] for the electromagnetic field. This can be defined as a three-by-three matrix T with components:
Tij = (ε0 / 2) [ (E2 + c2 B2) δij – 2 EiEj – 2 c2 BiBj ]
The subscripts i and j range from 1 to 3, and correspond to the Cartesian x, y and z coordinates. The symbol δij is the Kronecker delta, equal to 1 if i=j, and 0 otherwise.
If we have a small area whose unit normal vector is n, then the force per unit area due to the electromagnetic field in a region that n points away from is equal to Tn, where we multiply the matrix T and the vector n in the usual way. A careful analysis of any scenario involving an electromagnetic field will yield a force in agreement with this formula [11] (but note that there are other sign conventions in use, where T is defined to be the opposite of the matrix given here, and the force is measured across a surface element facing in the opposite direction).
Now, suppose we take one of the rows of T, and, treating it as a vector field, compute its divergence. For example, if we take the first row:
div T1 = ∂T11 / ∂x + ∂T12 / ∂y + ∂T13 / ∂z
It’s a tedious but straightforward calculation to evaluate this as:
div T1 = ε0 [ – E1 (div E) + E2 (curl E)3 – E3 (curl E)2 – c2 B1 (div B) + c2 B2 (curl B)3 – c2 B3 (curl B)2 ]
We get similar results for the divergence of the other rows, and we can package all three results quite compactly as:
div T = ε0 [ E × (curl E) + c2 B × (curl B) – (div E) E – c2 (div B) B ]
where “×” here indicates the vector cross product.
Maxwell’s equations tell us that div B is zero everywhere, and that in the absence of charges (as in the interior of our cavity) div E is also zero. The curls of B and E, in the absence of currents (which again holds true in the interior of our cavity), are linked to the rates of change of E and B:
c2 curl B = ∂E / ∂t
curl E = –∂B / ∂t
So we have:
div T = ε0 [ –E × (∂B / ∂t) + B × (∂E / ∂t) ]
= –ε0 ∂[E × B] / ∂t
= –(1/c2) ∂S / ∂t
where S = c2 ε0 E × B
The vector field S is known as the Poynting vector, and it describes the rate of flow of energy per unit area in an electromagnetic field. The quantity S/c2 gives the momentum per unit volume contained in the field.
If we apply Gauss’s Theorem to the integral over the walls of an arbitrarily-shaped cavity of any one of the rows of T, say Ti, we obtain:
(Net force)i = ∫wall Ti · dA = ∫interior div Ti dV = –(1/c2) ∫interior (∂Si / ∂t) dV
If the cavity contains a standing wave, then the fields will have a harmonic time dependence of the form sin(ωt) or cos(ωt), and over one complete cycle of the mode, a period of 2π/ω, all the fields will return to their origin values. So at each point in the interior of the cavity, we will have:
∫cycle (∂Si / ∂t) dt = Si(t0+2π/ω) – Si(t0) = 0
So, averaged over a complete cycle in the same way, each component of the net force on the wall will sum to zero. -
#2542 by aero on 03 Nov, 2017 02:46
-
Yes, Thanks, Paul.
It is likely not enough for detailed calculations. (I forgot to ask about the quality factor, was it on the order of 7000 for these runs?) That's OK because I'm not sure how to run the calculations anyway. But it is enough to illustrate the thought and question that I want to ask about.
The thing is, the frustum is warmer at the big end than the small end so RF energy is being dissipated preferentially at the big end. Inside the frustum, doesn't that mean that the reflection coefficient is lower at the big end than the small end? If I'm reading the images right, the big end temperature rise is 6.6 K while the small end temperature is effectively unchanged. I will assume from this that the reflection coefficient at the small end is near enough to one as to make no difference while at the large end something on the order of 100 watts is being dissipated.
Now, I know that there are many ways to cause the reflection coefficient to differ end to end, one way is to use different end materials for example. Another way is to highly buff one end only. But that is not the case with this data. It seems to me from the temperature data, that another way to make the reflection coefficient differ is to operate one end of the frustum near the cut-off frequency.
The reflection coefficient is a wave phenomenon. Proceeding with the thought and using Q = energy stored/energy lost per second gives Q = power stored/power lost, so power stored = Q * power lost. With Q = 7000 and drive power (power lost) = 100W, power stored = a big number. 0.7 MW. If the reflection coefficient at the small end = 1, then the reflection coefficient at the big end seems to equal (power stored - power lost)/power stored which is (1 - 1/Q), or in this case 0.99986.
Taking a step further out on the limb, using radiation pressure, P = Ef/c (Ef is radiation intensity) and for a perfectly reflected wave, P = 2*Ef/c and introducing the reflection coefficient, Γ, calculate the internal pressures on the ends of the frustum as:
Pressure Force, P = 2 *Γ *Ef/c . I have arbitrarily introduced the reflection coefficient as it seems right to me.
But I must take another step. Ef is defined as radiation intensity, in units of N/m^2 and I have E on the ends of the frustum. I'm guessing that the frustum end diameters are: inside big dia. = 0.25552668 ; inside small dia. = 0.15875 from my meep control file for the Brady cavity. Those numbers give A_be = 0.0513 and A_se = 0.0198 m^2. This gives
big end intensity, Ef_be = 1.37E+7 and
small end intensity, Ef_se = 3.54 E+7
Factoring these parameters into the pressure force equation and using differential pressure, delta P = P_se - P_be gives
delta P = 2 * ( Ef_se - Γ * Ef_be)/c . This is now in units of N/m^2 though and I need total. Maybe the best shot is simply to use total stored energy and not factor areas into then back out of the equation and call it Thrust.
Thrust = 2 * 0.7E+6* (1 - Γ ) / c
This number calculates to 0.667 micro-N. That is different than 83.6 micro-Newtons from Paul's temperature image data.
Lots of problems here. It gives the wrong answer by 2 orders of magnitude. That could be due to miss treatment of the areas of the two ends, or perhaps the Q value is wrong, or maybe the experimental force number is not the right data point to use, or maybe the temperatures were measured too long after power-off allowing too much heat to be conducted away/lost from the ends which would give the wrong ratio of reflection coefficients. Or maybe there is nothing to it at all. Lots of models can easily fit one data point. Still, I would like some input from the forum.
I found some related stuff here - https://arxiv.org/pdf/0807.1310.pdf
I came across this a couple of weeks ago. I don't remember if I ever saw it brought up on the forums and the math is way beyond me, but the author's conclusion is that the radiation pressure always ends up canceling out once all of the factors are accounted for.
Might be worth it for you to take a look at if you're headed down this particular rabbit hole
http://www.gregegan.net/SCIENCE/Cavity/Cavity.htmlQuoteProof of zero force for any shape of cavity
Despite the asymmetry of our truncated spherical cone along the z-axis, the net force from radiation pressure on its walls is zero. How can we be sure, though, that there isn’t some other shape that will yield a non-zero net force?
To see what the net force will be in a resonant cavity of a completely arbitrary shape, we need to construct the stress tensor [10] for the electromagnetic field. This can be defined as a three-by-three matrix T with components:
Tij = (ε0 / 2) [ (E2 + c2 B2) δij – 2 EiEj – 2 c2 BiBj ]
The subscripts i and j range from 1 to 3, and correspond to the Cartesian x, y and z coordinates. The symbol δij is the Kronecker delta, equal to 1 if i=j, and 0 otherwise.
If we have a small area whose unit normal vector is n, then the force per unit area due to the electromagnetic field in a region that n points away from is equal to Tn, where we multiply the matrix T and the vector n in the usual way. A careful analysis of any scenario involving an electromagnetic field will yield a force in agreement with this formula [11] (but note that there are other sign conventions in use, where T is defined to be the opposite of the matrix given here, and the force is measured across a surface element facing in the opposite direction).
Now, suppose we take one of the rows of T, and, treating it as a vector field, compute its divergence. For example, if we take the first row:
div T1 = ∂T11 / ∂x + ∂T12 / ∂y + ∂T13 / ∂z
It’s a tedious but straightforward calculation to evaluate this as:
div T1 = ε0 [ – E1 (div E) + E2 (curl E)3 – E3 (curl E)2 – c2 B1 (div B) + c2 B2 (curl B)3 – c2 B3 (curl B)2 ]
We get similar results for the divergence of the other rows, and we can package all three results quite compactly as:
div T = ε0 [ E × (curl E) + c2 B × (curl B) – (div E) E – c2 (div B) B ]
where “×” here indicates the vector cross product.
Maxwell’s equations tell us that div B is zero everywhere, and that in the absence of charges (as in the interior of our cavity) div E is also zero. The curls of B and E, in the absence of currents (which again holds true in the interior of our cavity), are linked to the rates of change of E and B:
c2 curl B = ∂E / ∂t
curl E = –∂B / ∂t
So we have:
div T = ε0 [ –E × (∂B / ∂t) + B × (∂E / ∂t) ]
= –ε0 ∂[E × B] / ∂t
= –(1/c2) ∂S / ∂t
where S = c2 ε0 E × B
The vector field S is known as the Poynting vector, and it describes the rate of flow of energy per unit area in an electromagnetic field. The quantity S/c2 gives the momentum per unit volume contained in the field.
If we apply Gauss’s Theorem to the integral over the walls of an arbitrarily-shaped cavity of any one of the rows of T, say Ti, we obtain:
(Net force)i = ∫wall Ti · dA = ∫interior div Ti dV = –(1/c2) ∫interior (∂Si / ∂t) dV
If the cavity contains a standing wave, then the fields will have a harmonic time dependence of the form sin(ωt) or cos(ωt), and over one complete cycle of the mode, a period of 2π/ω, all the fields will return to their origin values. So at each point in the interior of the cavity, we will have:
∫cycle (∂Si / ∂t) dt = Si(t0+2π/ω) – Si(t0) = 0
So, averaged over a complete cycle in the same way, each component of the net force on the wall will sum to zero.
I respect Greg Egan, and I am familiar with the quoted paper. In this case the statement - quoting
"If the cavity contains a standing wave, then the fields will have a harmonic time dependence of the form sin(ωt) or cos(ωt), and over one complete cycle of the mode, a period of 2π/ω, all the fields will return to their origin values. So at each point in the interior of the cavity, we will have:
∫cycle (∂Si / ∂t) dt = Si(t0+2π/ω) – Si(t0) = 0
So, averaged over a complete cycle in the same way, each component of the net force on the wall will sum to zero."
does not agree in the general case with the paper
arXiv:0807.1310v5 [physics.class-ph] 21 Nov 2008
The Lorentz Force and the Radiation Pressure of Light
Tony Rothman∗ and Stephen Boughn†
Which I linked. Again https://arxiv.org/pdf/0807.1310.pdf
Check equation 2.3 and figures 2 and 3 of this link. The authors show that the pressure parallel to the z-axis averages to zero only in the special case where the phase angle equals pi/2. Even in that case, there is a non-zero average force in the off-axis direction. But if you read the linked paper from the beginning you will quickly understand why knowledgeable people with training in EM physics make this mistake.
-
#2543 by tleach on 03 Nov, 2017 03:07
-
I respect Greg Egan, and I am familiar with the quoted paper. In this case the statement - quoting
"If the cavity contains a standing wave, then the fields will have a harmonic time dependence of the form sin(ωt) or cos(ωt), and over one complete cycle of the mode, a period of 2π/ω, all the fields will return to their origin values. So at each point in the interior of the cavity, we will have:
∫cycle (∂Si / ∂t) dt = Si(t0+2π/ω) – Si(t0) = 0
So, averaged over a complete cycle in the same way, each component of the net force on the wall will sum to zero."
does not agree in the general case with the paper
arXiv:0807.1310v5 [physics.class-ph] 21 Nov 2008
The Lorentz Force and the Radiation Pressure of Light
Tony Rothman∗ and Stephen Boughn†
Which I linked. Again https://arxiv.org/pdf/0807.1310.pdf
Check equation 2.3 and figures 2 and 3 of this link. The authors show that the pressure parallel to the z-axis averages to zero only in the special case where the phase angle equals pi/2. Even in that case, there is a non-zero average force in the off-axis direction. But if you read the linked paper from the beginning you will quickly understand why knowledgeable people with training in EM physics make this mistake.
Thanks for pointing out that link again. I was reading on my phone at Dairy Queen while "helping" my daughter eat her strawberry sundae... Didn't even see the link, let alone read it, let alone understand it. So, I'll at least try to read it before I put my foot in my mouth again. -
#2544 by aero on 03 Nov, 2017 03:22
-
@ tleach
No foot in mouth - Egan's conclusion is completely understandable in light of the information from the linked paper. It is rare that anyone questions the basics that they learned their freshman year. Fortunately, it is also rare that they were taught an erroneous, or more charitably, an oversimplified solution to a basic phenomenon. -
#2545 by graybeardsyseng on 03 Nov, 2017 03:29
-
Indeed, that 'they quietly fade away' is not a good sign. I would say it is because making the microwave system work is difficult enough, but setting up a good experiment to measure the forces reliably and doing clever measurements and report about them in a clear way, is even harder.
I work on it until i'm frustrated and sick of it, then I usually take a break for a couple of weeks. Other pauses are that I don't have the time because of other projects or vacation.
I'm dealing with two issues right now before I can continue: more natural convection and an antenna self-resonating problem.
Adding the insulation to the draft enclosure greatly reduced the natural convection. But as the on-board electrical components begin to heat up, they cause natural convection of ~3uN after about 15 minutes. So most recently I've added a hefty heatsink to the on-board computer and moved it to the top of the torsional pendulum beam rather than attached to the side. I also want to add a better heatsink to the RF amplifier - perhaps even use the same phase change wax NASA is using. I also need to wrap all the aluminum below and to the sides of the pendulum arm with insulation, which I will probably do today.
In the image below I have the cavity removed so I can work on the antenna. Roger Shawyer thinks my antenna design, which the Polish group is also using, is self-resonating at a certain frequency rather than exciting the cavity. This makes a lot of sense as Jakub and I are both showing the same ~2.409Ghz -40dB return loss. This is highly unlikely since we have very different geometric dimensions. This was unlucky for me as the spherical end-plate frustum is designed to resonate at 2.405Ghz with mode TE013. As that may be too close to the resonate frequency of the antenna, it looks like I will need a different antenna. The antenna should still work in the flat end-plate frustum as that cavity was designed to resonate at 2.45Ghz. My simulations of Jakub's cavity show TE012 at 2.369Ghz, so he is going to try that next instead of 2.404Ghz - though it looks like the small end is a little below cut-off (second image below).
Jamie - are you using a loop antenna? Sorry but I have forgotten.
If so it can likely be analyzed as a single coil inductor - probably treated as a helical transmission line since typically above 1Ghz lumped components must usually be addressed as transmission lines. In any case it should be relatively simple to move the self resonance frequency (SRF) with a bit of capacitance (treat this as an open circuit transmission line).
If you are using a 1/4 lambda spike or wheel antenna that is somewhat different.
Herman
graybeardsyseng -
#2546 by dustinthewind on 03 Nov, 2017 04:29
-
this site seems to have been updated
any views on what Richard Banduric proposes?
http://electricspacecraft.org/
Hi Amit. I was thinking about the possibility. At the moment I think the propulsion (spinning charged disk with non-spinning charged disk) may be possible to describe with a classical understanding of E&M though I remain uncertain of the propulsion possibility. Any motion through an existing electric field gives a magnetic field. First I have my reasons to believe this magnetic field does not rotate via my thesis experiments, but 1st I will later introduce a possible alternative via a time helix leading to non-charge conservation (I don't think this likely - there are some tight restrictions on charge conservation - thermal charge of the atom for instance) and 2nd - pancaking of charge electric fields with out pancaking of the disk.
This magnetic field via moving through the static charge field describes the relativistic aspects of the charges that give off their electric field. Motion through this induced magnetic field via motion through the electric field (-v x E = B) then (v x B = E) describes the Lorentz contraction of the object holding the charges.
The rotating disk is rotating through the electric field from the disk below so it sees a relativistic length contraction of the non-rotating disk but each charge in this rotating disk sees a different length contracted non-rotating disk because each charge has a different direction of velocity. So the "oval" of the non-rotating Lorentz contracted disk changes in orientation depending on the observing charge.
As a result the entire rotating disk experiences a greater electric field than the disk below. The reason the disk below should not see the increased E field is because it is not rotating through the above disks non-rotating magnetic field which means no relativistic effects.
As a result there is a non-symmetric extra force on the rotating disk where there is a lack of an equal and opposite force on the non-rotating disk.
If such a configuration did experience a thrust it would indicate this may be what is happening.
However, if there is no extra thrust force then there are some other possibilities. The first I consider shaky which I will go over first. The 1st possibility is a time helix, which may introduce extra charge. This may possibly describe a rotating magnetic field which I would be fascinated to observe. I think only possible with rotating space time but that is for another time.
The rotating disk with charge would appear to the non-rotating disk below to acquire extra charge via extra charge from the past (time helix). This would indicate some form of non-charge conservation and it may cancel/partly-cancel out thrust effects but that would need to be determined by experiment.
I may have observed such a time helix possibly in my thesis experiment. It didn't effect my measurement of if the magnetic field was axially rotating, so I didn't bother with it. At the time I didn't know exactly what it was but now I realize it may have been such a helix or the 2nd possibility. (I did a prediction on the 2nd possibility and it was close didn't do the time helix.)
I had a solenoid surrounded by a very large low capacitance capacitor (with high resistance volt-meter) and when I would turn on the current in the non-rotating solenoid it would get what appeared to be a very observable change in charge. This change in charge remained as long as I had the current running and would drop when I stopped the current so it wasn't -dB/dt because it was a constant current. If I ran the current in the opposite direction I again see a similar jump in charge but at a different magnitude. Only the negative charges were moving around the solenoid so 1st possibility time helix. The 2nd possibility I estimated and it came close but I had difficulty resolving the change in magnitude depending on current direction so I left the prediction out if I remember correct.
This 2nd possibility may also cancel thrust which I think is more likely than a time helix. It is the Lorentz contraction of the electric fields of the charges in the spinning disk. As a result this intensifies the vertical electric field of the disk with out changing the charge (so charge conservation remains.) Basically it deflects the disks moving charges electric field lines via what Edward Purcell called electric field pancaking. This changes the orientation of the electric field lines of a charge to intensify the electric field above and below the disk while lowering it at the sides. While the charges in the non-rotating disk will "not" see a Lorentz contracted spinning disk they will see Lorentz contracted electric fields from the "individual" charges in the spinning disk.
This would correspond to a magnetic field that follows each individual charge in the rotating disk that describes the pancaking of their electric fields possibly canceling thrust or only partially. One might describe this as a rotating magnetic field but I like to think of it as radiating away from the charge as its orientation/position changes via light speed limits. This may have been the cause of the strange change in charge I observed during my experiment with a non axially rotating solenoid (only starting up the current).
Hopefully this isn't too convoluted. This was only concerning their charged spinning/non-spinning disk propulsion. I will need to look into them further to see if what they are actually proposing is this or something more complicated. Not really sure it would be related to EM drives or Mach effects.
-Dustin -
#2547 by meberbs on 03 Nov, 2017 04:32
-
I respect Greg Egan, and I am familiar with the quoted paper. In this case the statement - quoting
You either linked the wrong paper or completely misunderstood the paper. The paper is dealing with a single electron, not a closed cavity. Eagan's result and that paper are consistent. Trying to say otherwise is making a very similar mistake to what the paper is pointing out about the "Freshman argument," applying basic results without going through the full details.
"If the cavity contains a standing wave, then the fields will have a harmonic time dependence of the form sin(ωt) or cos(ωt), and over one complete cycle of the mode, a period of 2π/ω, all the fields will return to their origin values. So at each point in the interior of the cavity, we will have:
∫cycle (∂Si / ∂t) dt = Si(t0+2π/ω) – Si(t0) = 0
So, averaged over a complete cycle in the same way, each component of the net force on the wall will sum to zero."
does not agree in the general case with the paper
arXiv:0807.1310v5 [physics.class-ph] 21 Nov 2008
The Lorentz Force and the Radiation Pressure of Light
Tony Rothman∗ and Stephen Boughn†
If you look at your own result you will find not coincidentally that you got 2 times a photon rocket. This is from not including the reaction of the antenna, and probably a factor of 2 that I think is likely related to how you inserted gamma which you admitted was arbitrary, and I think isn't quite correct.
I don't really feel like delving into any more details of your math right now since Maxwell's equations quite clearly will not produce a force consistent with a working emDrive. -
#2548 by ThatOtherGuy on 03 Nov, 2017 09:16
-
More progress on the insulation. All components have been placed back onto the torsional pendulum after some testing with the antenna indicates that it is not self-resonating, and there is a reasonable field in the cavity.
that's very good news, indeed
!I would still like to improve the heat sink on the main amplifier.
This would probably mean going for a phase-change heat sink, but this in turn would mean modifying the amplifier since if I'm not wrong, the heatsink is part of the RF amp structure... and making change to it may cause issues
<sigh> I was looking at this but I'm not sure it (or something like it) may be appropriate for your device and won't introduce noise (edit: more stuff here)
On the other hand, as I already wrote, you may consider adding something to cool it down quickly, for example, you may have a couple fans mounted on the top of the box with pipes pointing toward the heatsink, a fan could then blow cold air toward the heatsink while the other could pump out heat; such a thing, used after each test may help cooling down the amplifier faster so allowing to run more test in a given interval of time ... or am I missing something
?
-
#2549 by ThatOtherGuy on 03 Nov, 2017 09:25
-
Jamie - are you using a loop antenna? Sorry but I have forgotten.
If so it can likely be analyzed as a single coil inductor - probably treated as a helical transmission line since typically above 1Ghz lumped components must usually be addressed as transmission lines. In any case it should be relatively simple to move the self resonance frequency (SRF) with a bit of capacitance (treat this as an open circuit transmission line).
If you are using a 1/4 lambda spike or wheel antenna that is somewhat different.
Herman
graybeardsyseng
Herman, check out this
https://forum.nasaspaceflight.com/index.php?topic=42978.msg1698864#msg1698864
if I'm not wrong the above is the antenna Jamie is currently using
-
#2550 by SeeShells on 03 Nov, 2017 10:30
-
Jamie - are you using a loop antenna? Sorry but I have forgotten.
If so it can likely be analyzed as a single coil inductor - probably treated as a helical transmission line since typically above 1Ghz lumped components must usually be addressed as transmission lines. In any case it should be relatively simple to move the self resonance frequency (SRF) with a bit of capacitance (treat this as an open circuit transmission line).
If you are using a 1/4 lambda spike or wheel antenna that is somewhat different.
Herman
graybeardsyseng
Herman, check out this
https://forum.nasaspaceflight.com/index.php?topic=42978.msg1698864#msg1698864
if I'm not wrong the above is the antenna Jamie is currently using
I'm not sure if that is the loop Jamie is using. He did a lot of different testing with different styles. Although he should be using a
Loop antenna that couples to the magnetic field. like the loop in EagleWorks test frustum. (see attached)
It should have both ends of the loop coupled to the coax feed.
https://www.nonstopsystems.com/radio/frank_radio_antenna_magloop.htm
This is a very nice site for referencing loop antennas.
My Very Best,
Shell
-
#2551 by kenny008 on 03 Nov, 2017 10:48
-
More progress on the insulation. All components have been placed back onto the torsional pendulum after some testing with the antenna indicates that it is not self-resonating, and there is a reasonable field in the cavity.
Wonderful progress!
I would still like to improve the heat sink on the main amplifier.
I believe a phase change heat sink has the potential of introducing additional errors, depending on the design. The ones I see in my searches show the heat sink material flowing and moving within the device as it heats up. Wouldn't this movement change the balance on your beam a tiny amount? I wouldn't think you'd want any mass flowing on any part of the apparatus attached to your beam. -
#2552 by Monomorphic on 03 Nov, 2017 11:24
-
Jamie - are you using a loop antenna? Sorry but I have forgotten.
If so it can likely be analyzed as a single coil inductor - probably treated as a helical transmission line since typically above 1Ghz lumped components must usually be addressed as transmission lines. In any case it should be relatively simple to move the self resonance frequency (SRF) with a bit of capacitance (treat this as an open circuit transmission line).
If you are using a 1/4 lambda spike or wheel antenna that is somewhat different.
It's a simple antenna design. The polish group is also using the same design. It had the same coupling as a standard loop in FEKO.
-
#2553 by graybeardsyseng on 03 Nov, 2017 12:59
-
Jamie - are you using a loop antenna? Sorry but I have forgotten.
If so it can likely be analyzed as a single coil inductor - probably treated as a helical transmission line since typically above 1Ghz lumped components must usually be addressed as transmission lines. In any case it should be relatively simple to move the self resonance frequency (SRF) with a bit of capacitance (treat this as an open circuit transmission line).
If you are using a 1/4 lambda spike or wheel antenna that is somewhat different.
Herman
graybeardsyseng
Herman, check out this
https://forum.nasaspaceflight.com/index.php?topic=42978.msg1698864#msg1698864
if I'm not wrong the above is the antenna Jamie is currently using
I'm not sure if that is the loop Jamie is using. He did a lot of different testing with different styles. Although he should be using a
Loop antenna that couples to the magnetic field. like the loop in EagleWorks test frustum. (see attached)
It should have both ends of the loop coupled to the coax feed.
https://www.nonstopsystems.com/radio/frank_radio_antenna_magloop.htm
This is a very nice site for referencing loop antennas.
My Very Best,
Shell
Shell -
EXCELLENT reference for mag loops - I had not seen that site before but it is now in my favorites for both EMdrive and Ham Radio.
Concur completely with your comments!
I remember that open "loop" antenna (i.e. only one end connected) but as you mention I thought Jamie had decided to go with a full loop. As you said - a true loop antenna has both ends connected to the coax - one to the center conductor and one to the shield. If it is open as in
it is NOT a loop and it must be analyzed differently - probably as a vertical (or so-called spike) with a capacity hat or a primitive form of wheel antenna (with only one lobe). see below for a picture of a more common 3 lobe wheel antenna.
In any case that antenna is NOT a loop and should not be analyzed as a loop for SRF determination. Nor is the pattern of radiation anything like a loop.
Normally the self resonance frequency is where the capacitance (usually parasitic) and the inductance of an inductor are in resonance such that the antenna (or any inductor really) will have a very high impedance and appear like an open circuit. I don't personally recall seeing a SRF calculated for such an antenna - I will research that as time permits.
Herman -
#2554 by graybeardsyseng on 03 Nov, 2017 14:49
-
KJamie - are you using a loop antenna? Sorry but I have forgotten.
If so it can likely be analyzed as a single coil inductor - probably treated as a helical transmission line since typically above 1Ghz lumped components must usually be addressed as transmission lines. In any case it should be relatively simple to move the self resonance frequency (SRF) with a bit of capacitance (treat this as an open circuit transmission line).
If you are using a 1/4 lambda spike or wheel antenna that is somewhat different.
It's a simple antenna design. The polish group is also using the same design. It had the same coupling as a standard loop in FEKO.
Jamie -
I hadn't seen your post when I answered Shell. Couple of comments - in no way was I trying to say that this isn't a good antenna (the one you and the Polish group are using). it is - I was just addressing calculation of SRF and whether or not it constituted a loop antenna. And I'm not sure why FEKO treats it as a loop. I am going to try digging deeper into this design a bit.
Herman
-
#2555 by SeeShells on 03 Nov, 2017 14:58
-
Herman,Jamie - are you using a loop antenna? Sorry but I have forgotten.
If so it can likely be analyzed as a single coil inductor - probably treated as a helical transmission line since typically above 1Ghz lumped components must usually be addressed as transmission lines. In any case it should be relatively simple to move the self resonance frequency (SRF) with a bit of capacitance (treat this as an open circuit transmission line).
If you are using a 1/4 lambda spike or wheel antenna that is somewhat different.
Herman
graybeardsyseng
Herman, check out this
https://forum.nasaspaceflight.com/index.php?topic=42978.msg1698864#msg1698864
if I'm not wrong the above is the antenna Jamie is currently using
I'm not sure if that is the loop Jamie is using. He did a lot of different testing with different styles. Although he should be using a
Loop antenna that couples to the magnetic field. like the loop in EagleWorks test frustum. (see attached)
It should have both ends of the loop coupled to the coax feed.
https://www.nonstopsystems.com/radio/frank_radio_antenna_magloop.htm
This is a very nice site for referencing loop antennas.
My Very Best,
Shell
Shell -
EXCELLENT reference for mag loops - I had not seen that site before but it is now in my favorites for both EMdrive and Ham Radio.
Concur completely with your comments!
I remember that open "loop" antenna (i.e. only one end connected) but as you mention I thought Jamie had decided to go with a full loop. As you said - a true loop antenna has both ends connected to the coax - one to the center conductor and one to the shield. If it is open as in
it is NOT a loop and it must be analyzed differently - probably as a vertical (or so-called spike) with a capacity hat or a primitive form of wheel antenna (with only one lobe). see below for a picture of a more common 3 lobe wheel antenna. In any case that antenna is NOT a loop and should not be analyzed as a loop for SRF determination. Nor is the pattern of radiation anything like a loop.
Normally the self resonance frequency is where the capacitance (usually parasitic) and the inductance of an inductor are in resonance such that the antenna (or any inductor really) will have a very high impedance and appear like an open circuit. I don't personally recall seeing a SRF calculated for such an antenna - I will research that as time permits.
Herman
You're spot on, better than I could have said. Although that open loop antenna design like Jamie's design will generate a Circular Wave Pattern similar to a helical but I'm not sure how the near field generation on the open loop will look or couple into the magnetic field of the frustum which is important for a TExxx mode.
That's the reason for a ~1/10 WL closed loop antenna is to couple into the TE013 H-Field in the frustum.
It needs to be a current driven coil to produce the highest magnetic fields. You externally "tune" these closed loops with a small variable capacitor. A Magnetic Loop Antenna is basically just a resonant circuit using an copper wire coil and an adjustable capacitor in this case ~ a 1-10pf (pico-farad) adjustable. If the coil has a circumference of much less than, say, 1/10th of a wavelength, then the efficiency of the antenna will suffer. If the circumference approaches Ľ of a wavelength or more the antenna is accurately characterized as an electrical loop antenna, with characteristics similar to those of a dipole.
Jamie's isn't wrong but I don't think FEKO allows you to create a "tuned circuit" current loop antenna to give you a good VWSR.
My Very Bet,
Shell
-
#2556 by SeeShells on 03 Nov, 2017 15:01
-
Same here Herman. FEKO just didn't allow Jamie to see all the options to couple into the frustum.
KJamie - are you using a loop antenna? Sorry but I have forgotten.
If so it can likely be analyzed as a single coil inductor - probably treated as a helical transmission line since typically above 1Ghz lumped components must usually be addressed as transmission lines. In any case it should be relatively simple to move the self resonance frequency (SRF) with a bit of capacitance (treat this as an open circuit transmission line).
If you are using a 1/4 lambda spike or wheel antenna that is somewhat different.
It's a simple antenna design. The polish group is also using the same design. It had the same coupling as a standard loop in FEKO.
Jamie -
I hadn't seen your post when I answered Shell. Couple of comments - in no way was I trying to say that this isn't a good antenna (the one you and the Polish group are using). it is - I was just addressing calculation of SRF and whether or not it constituted a loop antenna. And I'm not sure why FEKO treats it as a loop. I am going to try digging deeper into this design a bit.
Herman
Shell -
#2557 by Monomorphic on 03 Nov, 2017 15:17
-
The antenna excites the TE01x modes and is very simple to construct. It was designed and impedance matched in FEKO before it was constructed. There are a number of different antennas that will work for TE01x modes, but few that are as easy to build.
-
#2558 by graybeardsyseng on 03 Nov, 2017 15:20
-
Shell - Concur absolutely - good discussion! Thanks!
Herman,Jamie - are you using a loop antenna? Sorry but I have forgotten.
If so it can likely be analyzed as a single coil inductor - probably treated as a helical transmission line since typically above 1Ghz lumped components must usually be addressed as transmission lines. In any case it should be relatively simple to move the self resonance frequency (SRF) with a bit of capacitance (treat this as an open circuit transmission line).
If you are using a 1/4 lambda spike or wheel antenna that is somewhat different.
Herman
graybeardsyseng
Herman, check out this
https://forum.nasaspaceflight.com/index.php?topic=42978.msg1698864#msg1698864
if I'm not wrong the above is the antenna Jamie is currently using
I'm not sure if that is the loop Jamie is using. He did a lot of different testing with different styles. Although he should be using a
Loop antenna that couples to the magnetic field. like the loop in EagleWorks test frustum. (see attached)
It should have both ends of the loop coupled to the coax feed.
https://www.nonstopsystems.com/radio/frank_radio_antenna_magloop.htm
This is a very nice site for referencing loop antennas.
My Very Best,
Shell
Shell -
EXCELLENT reference for mag loops - I had not seen that site before but it is now in my favorites for both EMdrive and Ham Radio.
Concur completely with your comments!
I remember that open "loop" antenna (i.e. only one end connected) but as you mention I thought Jamie had decided to go with a full loop. As you said - a true loop antenna has both ends connected to the coax - one to the center conductor and one to the shield. If it is open as in
it is NOT a loop and it must be analyzed differently - probably as a vertical (or so-called spike) with a capacity hat or a primitive form of wheel antenna (with only one lobe). see below for a picture of a more common 3 lobe wheel antenna. In any case that antenna is NOT a loop and should not be analyzed as a loop for SRF determination. Nor is the pattern of radiation anything like a loop.
Normally the self resonance frequency is where the capacitance (usually parasitic) and the inductance of an inductor are in resonance such that the antenna (or any inductor really) will have a very high impedance and appear like an open circuit. I don't personally recall seeing a SRF calculated for such an antenna - I will research that as time permits.
Herman
You're spot on, better than I could have said. Although that open loop antenna design like Jamie's design will generate a Circular Wave Pattern similar to a helical but I'm not sure how the near field generation on the open loop will look or couple into the magnetic field of the frustum which is important for a TExxx mode.
That's the reason for a ~1/10 WL closed loop antenna is to couple into the TE013 H-Field in the frustum.
It needs to be a current driven coil to produce the highest magnetic fields. You externally "tune" these closed loops with a small variable capacitor. A Magnetic Loop Antenna is basically just a resonant circuit using an copper wire coil and an adjustable capacitor in this case ~ a 1-10pf (pico-farad) adjustable. If the coil has a circumference of much less than, say, 1/10th of a wavelength, then the efficiency of the antenna will suffer. If the circumference approaches Ľ of a wavelength or more the antenna is accurately characterized as an electrical loop antenna, with characteristics similar to those of a dipole.
Jamie's isn't wrong but I don't think FEKO allows you to create a "tuned circuit" current loop antenna to give you a good VWSR.
My Very Bet,
Shell
Here is a link to an IEEE article I have found which discusses open loop antennas for use at high UHF frequencies. I have skimmed but have to head out of office in just a minute so I am posting them if anyone wants to dive in more fully.
http://repo.lib.hosei.ac.jp/bitstream/10114/3846/1/37_TAP%28Circularly%29.pdf
I still haven't found a reference to finding the SRF for such a loop. I will try to model such an antenna tonight with NEC and see what I can find.
I don't know nor have access to FEKO so I really can't comment more about that.
Herman
-
#2559 by meberbs on 03 Nov, 2017 15:27
-
Can people please trim quotes, especially removing embedded pictures? The massive quote trees make this thread really hard to read.
From the first post in the thread:In order to minimize bandwidth and maximize information content, when quoting, one can use an ellipsis (...) to indicate the clipped material.
!
<sigh> I was looking at 
?