Can Bipolar Plate Technology Create High Voltage Capacitors?

[CassieFordham]

I was reading about bipolar plate batteries, where the plates function as both the anode for one cell and cathode for the next cell, in series. It seems to me that the same idea should work to create a capacitor of higher voltage.

If I had a dielectric material that had dielectric strength of of 0.8 volts, and created a 9 plate, 8 dielectric capacitor like the picture below, would it work as a 6 volt capacitor?

 

[davenn]

welcome to PF

If I had a dielectric material that had dielectric strength of of 0.8 volts, and created a 9 plate, 8 dielectric capacitor like the picture below, would it work as a 6 volt capacitor?

a good question, what do you think and why ?
have you done any reading on capacitors in series ?Dave

 

[CassieFordham]

welcome to PF
a good question, what do you think and why ?
have you done any reading on capacitors in series ?Dave

Yes I've read about capacitors in series, that's why I think that this would be the same thing as 8x1.5 volt capacitors in series. when you connect the (+) of one capacitor to the (-) of another the leads become lie a single unit, so couldn't the middle plates be like (+) connected to (-) ?

 

[berkeman]

Yes I've read about capacitors in series, that's why I think that this would be the same thing as 8x1.5 volt capacitors in series. when you connect the (+) of one capacitor to the (-) of another the leads become lie a single unit, so couldn't the middle plates be like (+) connected to (-) ?

Yes you can stack capacitors in series to get higher voltage tolerance. What happens to the total capacitance when you stack 2 capacitors of the same value in series?

There is also a practical issue that may require high-value "bleed" resistors to be added in parallel with each capacitor. Can you think what that issue might be?

 

[CassieFordham]

Yes you can stack capacitors in series to get higher voltage tolerance. What happens to the total capacitance when you stack 2 capacitors of the same value in series?

There is also a practical issue that may require high-value "bleed" resistors to be added in parallel with each capacitor. Can you think what that issue might be?

2 capacitors in series have half the capacitance. In this case 1/(1/1.5+1/1.5) =0.75

With capacitors in series there can be a voltage imbalance, which potentially could exceed the tolerance of the dielectric. Would a voltage imbalance occur in this configuration, where both the plates and the dielectrics all have the same and dimensions to a high degree of precision?

 

[Baluncore]

If I had a dielectric material that had dielectric strength of of 0.8 volts, and created a 9 plate, 8 dielectric capacitor like the picture below, would it work as a 6 volt capacitor?

In theory you would have a 6.4 volt capacitor with a capacitance of C / 8.
A bipolar capacitor element with a dielectric strength of 0.8V is NOT a 1.5V capacitor element.

No matter how precise, there will be a voltage balance problem when the capacitor is hit by a pulse. Leakage current would unbalance the charge until one element failed, then all would fail due to increased voltage.
To protect the capacitor you would need to place a protection diodes across each element to prevent any element being over charged.

 

[CassieFordham]

In theory you would have a 6.4 volt capacitor with a capacitance of C / 8.
A bipolar capacitor element with a dielectric strength of 0.8V is NOT a 1.5V capacitor element.

No matter how precise, there will be a voltage balance problem when the capacitor is hit by a pulse. Leakage current would unbalance the charge until one element failed, then all would fail due to increased voltage.
To protect the capacitor you would need to place a protection diodes across each element to prevent any element being over charged.

Oops, I started the thread with a 0.8x 8 for a 6 volt capacitor , but then I mixed that up with 1.5x 8 for a 12 volt capacitor. Sorry for the confusion, If I would fix that in the previous posts if I still was able to edit them.

So it sounds like you are saying then, that this arrangement wouldn't work because there is no way to put protection diodes between cells with shared plates.

 

[Baluncore]

So it sounds like you are saying then, that this arrangement wouldn't work because there is no way to put protection diodes between cells with shared plates.

Correct.
Low loss capacitors with unpredictable leakage are never used in series without some form of voltage balancing protection. Why do you include the intermediate conductive plates when using a thicker dielectric would be the capacitive equivalent?

Tantalum pentoxide has a high dielectric constant and a high breakdown voltage. Deeper surface oxidation during manufacture gives a thicker oxide layer with a greater breakdown voltage, but a proportionally lower capacitance. When the oxide layer breaks down, the energy stored in the capacitor appears as heat and the capacitor explodes. Eye protection is necessary.
https://en.wikipedia.org/wiki/Tantalum_capacitor#Basic_principle

 

[CassieFordham]

Correct.
Low loss capacitors with unpredictable leakage are never used in series without some form of voltage balancing protection. Why do you include the intermediate conductive plates when using a thicker dielectric would be the capacitive equivalent?

Tantalum pentoxide has a high dielectric constant and a high breakdown voltage. Deeper surface oxidation during manufacture gives a thicker oxide layer with a greater breakdown voltage, but a proportionally lower capacitance. When the oxide layer breaks down, the energy stored in the capacitor appears as heat and the capacitor explodes. Eye protection is necessary.
https://en.wikipedia.org/wiki/Tantalum_capacitor#Basic_principle

The reason for the intermediate plates is that the dielectric is a paste containing an aqueous solution. So the chemical breakdown of water is an issue.

Tantalum pentoxide has a relative permittivity of 27. The paste has a relative permittivity many orders of magnitude greater than any metal oxide by it's self.

 

[Baluncore]

The reason for the intermediate plates is that the dielectric is a paste containing an aqueous solution. So the chemical breakdown of water is an issue.

If you can control the electrical conductivity of the water based paste you will form a resistive divider that will balance the voltage across all the cells. Electrolysis of the water and the paste will be important in setting the maximum cell voltage. Can electrolysis of the water be used to balance the voltage?

 

[Baluncore]

The paste has a relative permittivity many orders of magnitude greater than any metal oxide by it's self.

How do you insulate your high dielectric material in a wet paste if it does not have an oxide layer ?
Calcium copper titanate for example, is a metal oxide.

Water does not retain a dielectric constant of 80-ish when it is trapped in a fixed position such as a gel, biological structure or ice. Any filler between particles of dielectric will form an effective low series capacitance that can kill the high dielectric constant advantage.
Advocatus Diaboli.

 

[CassieFordham]

How do you insulate your high dielectric material in a wet paste if it does not have an oxide layer ?
Calcium copper titanate for example, is a metal oxide.

Water does not retain a dielectric constant of 80-ish when it is trapped in a fixed position such as a gel, biological structure or ice. Any filler between particles of dielectric will form an effective low series capacitance that can kill the high dielectric constant advantage.
Advocatus Diaboli.

I didn't say it doesn't contain a metal oxide. Google "super dielectric material"

 

[rbelli1]

Some metallized film capacitors are constructed just like this. For example http://www.wima.com/EN/WIMA_MKP_4.pdf. These are not polarized.

This is one of many. I chose this brand as an example because it has an easy to follow diagram.

BoB

 

[CassieFordham]

Some metallized film capacitors are constructed just like this. For example http://www.wima.com/EN/WIMA_MKP_4.pdf. These are not polarized.

This is one of many. I chose this brand as an example because it has an easy to follow diagram.

BoB

The example you showed has plates connected in parallel, my question is about about in series. Parallel= increased capacitance, series= increased voltage.

 

[rbelli1]

The example you showed has plates connected in parallel,

The diagram for MKP 4: 63 VAC, 160 VAC, 200 VAC, 280 VAC has parallel plates.

If you look at the MKP4:400VAC, 630 VAC, 700 VAC diagram there is a plate that is not electrically connected. This forms a pair of series capacitors internal to the single unit.

If you look at page 20 of http://www.kemet.com/Lists/ProductCatalog/Attachments/334/KEM_F3033_PHE450.pdf you will see this structure as well as two unconnected sections forming three capacitors in series.

I don't know how they keep the voltages from becoming unbalanced.

BoB

 

[Baluncore]

Maybe the gaskets or spacers that hold the plates apart, and keep the paste in place, can be fabricated from a semiconductor diode wafer. That way the bandgap voltage of diodes can prevent over voltage of each cell.
The conductive separator plates could have a pattern of chemically developed 'metal rectifier' such as copper oxide, grown to provide a very close spacing at the same time as the protective voltage clamping.

 

[CassieFordham]

Maybe the gaskets or spacers that hold the plates apart, and keep the paste in place, can be fabricated from a semiconductor diode wafer. That way the bandgap voltage of diodes can prevent over voltage of each cell.
The conductive separator plates could have a pattern of chemically developed 'metal rectifier' such as copper oxide, grown to provide a very close spacing at the same time as the protective voltage clamping.

Sounds very complicated and expensive. The whole reason for considering this design, was to make things simpler and cheaper.

 

[Baluncore]

Sounds very complicated and expensive.

Metal rectifiers are very, very, very low technology. MRs are a chemically coated plate like a battery, not a clean-room fabrication process. MRs have been used for about a century, while transistors with current gain have only been used for 60 years. MRs were bolted together in a stack. The problem was their low breakdown voltage, which might now be turned to an advantage. https://en.wikipedia.org/wiki/Metal_rectifier

Metal Oxide Varistors are used widely to make voltage protection devices today. You might consider something as simple as mixing a MOV powder into your capacitor paste as voltage protection.
https://en.wikipedia.org/wiki/Varistor#Composition_and_operation

Capacitor modules will need a voltage rating of about 48 volts to be economic for voltage inverters and energy storage. I very much doubt you will be able to make paste cells that will match better than ±20%, so there must be some form of voltage clamp. If you cannot connect all your capacitor cells in parallel and so use one string of voltage protection diodes, then you will have to build the voltage clamping into the cells as part of the paste or part of the enclosure surface.

I believe this paper is relevant to the technology you are considering. It includes several useful references.
https://www.researchgate.net/profile/Thomas_Cuff/publication/294259463_The_Copper_Oxide_Rectifier/links/56bf447a08ae44da37fa57dc/The-Copper-Oxide-Rectifier.pdf?origin=publication_detail

At its simplest, a wet type rectifier consisted of, for example an electrolytic
cell composed of an aluminum plate and a lead plate dipped into a solution of
water and borax.

Boric acid is used in super dielectric materials. Why not find a way to mix the SDM and MOV technologies in a bucket?