3 ms·

Not being a power engineer, I'm still wondering if/when somebody is going to try to build a grid capacitor bank. Traditional advantages: long lifespan, high cha

Not being a power engineer, I'm still wondering if/when somebody is going to try to build a grid capacitor bank. Traditional advantages: long lifespan, high charge/discharge rate, low complexity. Traditional disadvantages: large, heavy. Maybe the cost doesn't work out or something, but seriously, a fixed land power station seems like the ideal use for caps. They're fundamentally DC, but so are batteries. I wonder if you could make yourself a massive LC tank, and bypass conversion losses...?

ChuckMcM 9y ago
Wouldn't it be! There are some interesting issues though with capacitors (my current favorite are graphene ones). At the end of the day, we're talking about charge waiting to be used sitting around in a capacitor. So looking at how much charge and what are the ramifications of that charge 'at rest' as it were, are really interest. For example, did you know that if you charged yourself negatively with a few dozen Columbs of charge (in a vacuum) you would float off the ground? You would find yourself repulsed from the earth (which has a huge store of electrons holding negative charge). If you tried it in air however you would find yourself the victim of an extremely vicious lightning attack as the positive charge in the air helped itself to your electrons to balance itself out! Part of the issue of trying to hold so much charge is that as raw charge, it actually exerts quite a bit of force. It can lift you off the ground, it can pull capacitors toward each other, it can deform the very structure you have created to contain it. The other issue is keeping it from its mate, if nature abhors a vacuum she really really hates two opposite charges separated by a small gap. You need a really good dielectric separating your charge. You can't use air unless you have a lot of separation and the more separation the less charge your capacitor can hold. So the ideal dielectric would be the evil twin of graphene, one atom thick and impervious to any charge transfer at all[1]. The compromise is you can't hold as much charge (which keeps the voltage potential down below the dielectrics failure point) and you are back to using a lot of space. Batteries in this case work because they store the charge in small chemical compounds that act as 'buckets.' The process of moving charge restructures those compounds taking an electron or two and leaving behind a different chemical structure or an empty bucket. This is a "win" because you never have to hold so much charge in one place that it is either acting mechanically on the mechanism (beyond the chemical bonds) or threatening to convert your insulating dielectric into an unwilling conductor. The down side is that assembling and disassembling the chemical compounds is an imperfect process at best. Ideally, a grid scale capacitor would have some sort of mechanical/physical way of automatically organizing trillions of small individual capacitors which could be processed through and drained or charged as the needs required. That would make them not quite 'ideal' capacitors, capable of delivering all of their charge in one go, but it would allow you to side step the mechanical and dielectric challenges. You would also avoid the challenge of having a chemical transition that was not be 100.0% reversible. I am always interested in reading papers on interesting research in this area. There are a couple of labs doing work on 'liquid' batteries where the electrolyte can be charged externally through one chemical process and discharged in the battery through another. Graphene capacitors are also pretty cool but they still need a better dielectric and a way of creating them cost effectively. Patterning a few billion connected in parallel seems to be a research target to try and avoid some of the high K issues. But there are many graphene labs and not many are focused on any one thing so it is harder to find. [1] Even with that you'll get tunneling as the electrons find their probability function landing on the other side of the dielectric, nothing is ever perfect!
cesarb 9y ago
Doesn't the grid already use capacitor banks and reactors for power factor correction?

ChuckMcM 9y ago

Wouldn't it be! There are some interesting issues though with capacitors (my current favorite are graphene ones). At the end of the day, we're talking about charge waiting to be used sitting around in a capacitor. So looking at how much charge and what are the ramifications of that charge 'at rest' as it were, are really interest. For example, did you know that if you charged yourself negatively with a few dozen Columbs of charge (in a vacuum) you would float off the ground? You would find yourself repulsed from the earth (which has a huge store of electrons holding negative charge). If you tried it in air however you would find yourself the victim of an extremely vicious lightning attack as the positive charge in the air helped itself to your electrons to balance itself out! Part of the issue of trying to hold so much charge is that as raw charge, it actually exerts quite a bit of force. It can lift you off the ground, it can pull capacitors toward each other, it can deform the very structure you have created to contain it. The other issue is keeping it from its mate, if nature abhors a vacuum she really really hates two opposite charges separated by a small gap. You need a really good dielectric separating your charge. You can't use air unless you have a lot of separation and the more separation the less charge your capacitor can hold. So the ideal dielectric would be the evil twin of graphene, one atom thick and impervious to any charge transfer at all[1]. The compromise is you can't hold as much charge (which keeps the voltage potential down below the dielectrics failure point) and you are back to using a lot of space. Batteries in this case work because they store the charge in small chemical compounds that act as 'buckets.' The process of moving charge restructures those compounds taking an electron or two and leaving behind a different chemical structure or an empty bucket. This is a "win" because you never have to hold so much charge in one place that it is either acting mechanically on the mechanism (beyond the chemical bonds) or threatening to convert your insulating dielectric into an unwilling conductor. The down side is that assembling and disassembling the chemical compounds is an imperfect process at best. Ideally, a grid scale capacitor would have some sort of mechanical/physical way of automatically organizing trillions of small individual capacitors which could be processed through and drained or charged as the needs required. That would make them not quite 'ideal' capacitors, capable of delivering all of their charge in one go, but it would allow you to side step the mechanical and dielectric challenges. You would also avoid the challenge of having a chemical transition that was not be 100.0% reversible. I am always interested in reading papers on interesting research in this area. There are a couple of labs doing work on 'liquid' batteries where the electrolyte can be charged externally through one chemical process and discharged in the battery through another. Graphene capacitors are also pretty cool but they still need a better dielectric and a way of creating them cost effectively. Patterning a few billion connected in parallel seems to be a research target to try and avoid some of the high K issues. But there are many graphene labs and not many are focused on any one thing so it is harder to find. [1] Even with that you'll get tunneling as the electrons find their probability function landing on the other side of the dielectric, nothing is ever perfect!

cesarb 9y ago

Doesn't the grid already use capacitor banks and reactors for power factor correction?