Is this Voltage Variable Capacitor available commercially?

[Leo Freeman]

Is the "Voltage variable capacitor" described in patent US 20070242410 A1 a commercially available component? Has anyone heard of this method of changing the dielectric constant using a biasing voltage? It has been nearly 13 years since the patent was filed, but I've never heard of it. It would be very useful if it did work as described.

Thanks in advance for any clues or leads.
Leo

 

[Paul Colby]

I'm not aware of any material that changes dielectric constant in any reasonably sized applied field. Polarizability of a material will saturate in very (very) high fields but these fields are on the order of MeV not volts.

 

[Leo Freeman]

Thanks for your reply. It is a bit of a mystery; no-one seems to have heard of this idea, but the patent definitely claims it works. I can't imagine why someone would spend the money patent a fake device?

 

[Paul Colby]

Yes, it would work I just don't know of a material which would oblige. Are you familiar with varactor diodes? They use changing junction capacitance.

 

[Leo Freeman]

Yes, I just read about varicaps. They have two leads only, though, and apparently limited to the pFarad range. This patent, however, seems to describe a triode-like device, and not using semiconductors.

 

[Paul Colby]

It's interesting that the FET was originally patented in the 1920's. The author knew it would work but there were no materials or processes to support an actual device. Basically he was envisioning a solid state triode. Along comes the 50's and we have FETs

 

[Leo Freeman]

Wow, That was a long-term gamble! Do you think Leeder is trying the same trick, just with capacitors? I noticed he said in the patent: "As a practical matter, there are no limitations as to the size or geometry of the capacitor of the present invention or the type of dielectric material used." [emphasis mine] But do you think it would require a semiconductor material, after all?

The way I originally came across this patent was: A few days ago, I had the idea of varying the dielectric constant of an air-dielectric capacitor by inserting another set of capacitor plates, made of grid mesh, between the main capacitor plates. The outer plates' electric field could be canceled by controlling the voltage on the inner plates, thus effectively increasing the capacitance of the outer plates. I looked on Google images for anything that suggested this idea, and found the above image from the patent.

 

[berkeman]

described in patent US 20070242410 A1

That looks to be a patent application, not a granted patent. There have certainly been some things that were granted a patent that really should not have, but this one looks like it didn't make it to that stage anyway. I also don't see it being assigned to any company...

 

[jim hardy]

the patent number is 7411774it's available here but you may need a software TIFF reader .

.http://patft.uspto.gov/netacgi/nph-...50&s1=7411774.PN.&OS=PN/7411774&RS=PN/7411774

Write him and ask if he knows of a source ?

 

[Leo Freeman]

That looks to be a patent application, not a granted patent. There have certainly been some things that were granted a patent that really should not have, but this one looks like it didn't make it to that stage anyway. I also don't see it being assigned to any company...

Thanks for that. Yes, at first I was amused by Paul's account of the patenting of the FET, but now I'm angry; that sort of "patent squatting" on vague half-baked ideas should not be allowed. It's corruption, I say.

 

[Leo Freeman]

the patent number is 7411774
Write him and ask if he knows of a source ?

Thank you very much, Jim, for that info! Can we assume, then, that this device actually works?
If so, why do people waste time with varicaps and cumbersome mechanically-variable capacitors? How come this important device is not more widely known to the electronics community? (I hope there's a great conspiracy story in here! hehehe!)

 

[jim hardy]

Thank you very much, Jim, for that info! Can we assume, then, that this device actually works?

The patent only asserts that dielectric constant is affected by charge on the grid

i'm not a materials expert so can't refute that there exists some material that acts that way
but my Missouri roots say "Show Me"
and i am doubtfull that it works as he describes.

Basics of capacitor dielectrics is here
http://physics.info/dielectrics/

Myself i think he's built a capacitive voltage divider,
and deluded himself and a patent examiner that he's modulating permittivity of some material (that he didn't name).

But one can find papers on dielectric constant versus voltage in some exotic materials, eg
http://www.sciencedirect.com/science/article/pii/S027288421501932X

Abstract
Bi2/3Cu3Ti4O12 (BCTO) ceramics were successfully prepared by traditional solid-state reaction method (BCTO-SS) and sol–gel method (BCTO-SG). Pure perovskite phase and dense structure were obtained in BCTO ceramics prepared by both methods. BCTO-SG ceramics showed a large dielectric constant of ~1.1×104 while BCTO-SS ceramics exhibited a low dielectric constant of ~3200. At 100 kHz, the dielectric constant of BCTO-SS ceramics decreased with applied voltage increasing, while the dielectric constant of BCTO-SG ceramics increased with applied voltage increasing. Further study of the relationship between dielectric constant and voltage suggested BCTO-SG ceramics had larger defect concentration than BCTO-SS ceramics. The investigation of complex impedance indicated that the electrical properties of grain boundaries for all BCTO ceramics were evidently affected by applied voltages and the electrical properties of grains were independent of applied voltages. In addition, the non-Ohmic properties of BCTO ceramics were studied in detail. The non-linear coefficients of BCTO-SS and BCTO-SG ceramics were 1.65 and 1.01, respectively. The breakdown electric fields of BCTO-SS and BCTO-SG ceramics were found to be 1.21 and 0.48 kV/cm, respectively. The potential barrier heights of BCTO-SS and BCTO-SG ceramics were calculated to be 0.549 and 0.485 eV, indicating that the potential barriers at the grain boundaries for BCTO-SS and BCTO-SG ceramics are the Schottky-type barrier.

I'd track him down and ask.

 

[Leo Freeman]

Thank you again for that information.

Leeper says in the patent: "It is known that altering the electromagnetic field within or surrounding a dielectric material will alter the relative permittivity of the dielectric. It is upon this principle that the present invention is based." It sounds fairly emphatic.

Also, "Charging the conductive grid with a negative potential causes the dielectric constant of the capacitor to decrease in value thereby reducing the capacitor's value. Placing a positive potential on the grid causes the dielectric constant to increase, thereby raising the capacitor's value."
How could this polarity principle hold true for all types of dielectrics?

Also, this dielectric does not necessarily have to be a semiconductor, hence the presence of the embedded conductive grid in some embodiments of the device. If a semiconducting dielectric is used, it would not need the grid to charge it, just a contact electrode.

So I wonder if the dielectric material is just incidental, it is the inner electric field that counts.
Is it possible that it is just the electric field from the metal grid which interacts with and varies the plate field? If there were two grids, like an internal capacitor, wouldn't that affect the plate field in the same way that a normal dielectric does?

 

[jim hardy]

You're on the same thought path i am.
A capacitor stores energy in the volume between its plates, actually in the electric field permeating that volume. One can write equations and integrate to calculate that energy.
His grid does something to that field.

So I wonder if the dielectric material is just incidental, it is the inner electric field that counts.
Is it possible that it is just the electric field from the metal grid which interacts with and varies the plate field?

That's what i think.

 

[jim hardy]

while there is reported in highfalutin' journals some effect in exotic materials, I've yet to see any basis for his claim it's polarity sensitive.

https://www.researchgate.net/publication/254236527_Relation_of_Dielectric_Permittivity_and_Electric_Field_Dependence_of_Polarization_in_Some_Relaxors_with_Perovskite_Structure

doesn't look to me like it's a practical idea as yet.

old jim

 

[Paul Colby]

http://www.uni-leipzig.de/~ids2012/tutorials/richert2010.pdf

Here is a talk on non-linear dielectric properties. I would think if a material were commonly known to change by a factor of 2 (like in the patent) by applying a manageable field it would appear as an example. Fields like $10^{10}$ volts per meter appear in some of the plots accompanied by very small dielectric changes. This is not surprising given basic physics. Materials are made up of tightly bound charges. Saturating their polarization would be a feat. I certainly can't prove the non-existence of such a material but I certainly have reason to doubt such exists.

 

[Leo Freeman]

while there is reported in highfalutin' journals some effect in exotic materials, I've yet to see any basis for his claim it's polarity sensitive.
doesn't look to me like it's a practical idea as yet.

It certainly doesn't look workable by relying on some property of the dielectric. I wonder what was the point of patenting the device before a suitable material could be found?

 

[Leo Freeman]

I certainly can't prove the non-existence of such a material but I certainly have reason to doubt such exists.

Thanks for all this. It has illuminated the murky world of patent law. My take-home message in all of this is that an inventor doesn't have to prove his invention actually works in order to get a Patent!

 

[Baluncore]

The inventor is not really “patent squatting” as the patent expires shortly and no one else can patent it again for at least 50 years. He may not have paid all the annual renewal fees, so it may have lapsed. The patent should expire in a couple of years anyhow.

But that does not tell us what the material is. I think it is actually a concentrated delusion of grandure mixed with the whisp of dream; an idea worth a million dollars, but totally secret.

There are some neat materials out there. For example, ferrite materials that have frequency dependent dielectric constants. You could make a tuned circuit that was resonant over a very wide band with very little loss.

 

[Leo Freeman]

Is it possible that the dielectric may not the the thing which really changes the capacitance? If an air capacitor had another inner set of plates made of mesh, like this:

wouldn't the inner electric field modify the outer plate field (and hence, vary the capacitance) just like a dielectric would? I can't see the difference between a dielectric displacement field reducing the plate field, and an "artificial" field from the mesh plates doing the same. I choose mesh for the inner plates so that they can "let the outer field through", if that makes any sense!

 

[Baluncore]

Each parallel mesh lies on an equipotential plane in the dielectric. All you are doing is building a stack of higher value capacitors in series that must be switched.

You might consider a variable capacitance multiplier, or maybe using something like a gyrator to simulate a capacitor.
Why do you really need a variable capacitor?

 

[Paul Colby]

A capacitor with a central conductor (a mesh like you describe) is equivalent to two capacitors in series. Add a wire to the connection point and you have a three terminal device. It's not very interesting device, however, because it's linear. What makes the patent interesting (for some value of interest) is the non-linearity of the mythical dielectric material.

 

[Baluncore]

Yes, I just read about varicaps. They have two leads only, though, and apparently limited to the pFarad range. This patent, however, seems to describe a triode-like device, and not using semiconductors.

Varicap diodes are used for tuning VHF and UHF RF circuits so they have low values of capacitance.

If instead you use high current power diodes you will find that they have many nF of capacitance. By connecting two power diodes in common anode or common cathode you get a symmetrical two terminal varicap that can be tuned by variation of the DC bias on the common terminal.

Try cheap power diodes such as 1N4001. I have used the stud diodes out of vehicle alternators as variable capacitance diodes as they have a huge area and so a high capacitance. I suspect that the Schottky diode pairs used in switching power supplies are designed for high speed and so have a lower capacitance.

You might also consider using a pair of MOSFETs. The older high current low voltage MOSFETs often have a huge capacitance.

 

[Baluncore]

This is the symmetrical circuit used with varicap diodes.

 

[Paul Colby]

This is the symmetrical circuit used with varicap diodes.

In effect this does what the patent does but differently. The bias changes the distance between the "plates" by varying the depletion layer.

 

[Leo Freeman]

Each parallel mesh lies on an equipotential plane in the dielectric. All you are doing is building a stack of higher value capacitors in series that must be switched.
You might consider a variable capacitance multiplier, or maybe using something like a gyrator to simulate a capacitor.

Thanks for the tip about the gyrator! This is turning into a learning experience for a non EE like me.
I'm not looking so much for a variable capacitor, but to simulate a "super-dielectric", using applied potentials to force a capacitor soak up a huge amount of charge, by canceling any field within the plates.

Why do I really need a variable capacitor?
I have an intuitive hunch about a possible circuit, that is based on Kelvin's Thunderstorm apparatus.
It might be able to soak up the positive charges in the atmosphere. I have to sit down and try to sketch out this vague idea.

 

[Leo Freeman]

A capacitor with a central conductor (a mesh like you describe) is equivalent to two capacitors in series. Add a wire to the connection point and you have a three terminal device. It's not very interesting device, however, because it's linear. What makes the patent interesting (for some value of interest) is the non-linearity of the mythical dielectric material.

Thank you for this info. I can see how a single mesh would make two capacitors in series, but what about two meshes?
Two meshes forming an inner capacitor, with a polarity that's the reverse of that of the outer plates?

Discrete capacitors in circuits have their own self-contained electric fields that are not visible to the outer circuit. But I'm still uneasy about the scenario where the field of one capacitor directly influences the field of another.
If I can settle this, you would have taught me more than all my Uni electronics courses combined.

 

[Baluncore]

When you insert a mesh into a capacitor you do not increase the energy storage because you must also reduce the voltage. The total energy that can be stored is proportional to the volume of the dielectric.

Capacitance is proportional to the area of the dielectric and inversely proportional to the thickness. Unfortunately, the thickness of the dielectric also decides the voltage rating. There lies the problem.
The energy stored in a capacitor is E = ½ C V².

Split a capacitor along the V/2 equipotential and you get two capacitors.
They have twice the capacitance each, but only half the voltage rating.
The total possible energy stored after the split is then; E = 2 * ½ * 2C * (V/2)²
Which is unfortunately the same as the original E = ½ C V².

Likewise, if you cut the capacitor in half, parallel with the field lines you make no change to the voltage, the total capacitance or the energy stored.

I can see how a single mesh would make two capacitors in series, but what about two meshes?

Two meshes forming an inner capacitor, with a polarity that's the reverse of that of the outer plates?

The voltage rating becomes V/3, the capacitance is 3C, and there are then three capacitors so;
E = 3 * ½ * 3C * (V/3)² = ½ C V² again.

 

[Leo Freeman]

When you insert a mesh into a capacitor you do not increase the energy storage because you must also reduce the voltage. The total energy that can be stored is proportional to the volume of the dielectric.
...
The voltage rating becomes V/3, the capacitance is 3C, and there are then three capacitors so;
E = 3 * ½ * 3C * (V/3)² = ½ C V² again.

I get it now. Thank you for explaining this so patiently!
So the fact that the inner mesh plates are actively charged by an external voltage, it still doesn't have any effect on the capacitance of the outer plates? How about using an electret as a dielectric?

Well, if Nature is not going to give me something for nothing, then I'll just have to take it by force!