New Mars Forums

Official discussion forum of The Mars Society and MarsNews.com

You are not logged in.

Announcement

Announcement: We've recently made changes to our user database and have removed inactive and spam users. If you can not login, please re-register.

#1 2019-07-28 19:15:41

kbd512
Moderator
Registered: 2015-01-02
Posts: 3,128

CNT Flywheels vs Batteries for Energy Storage

We've had a lot of argumentation about using solar panels and batteries vs nuclear reactors to provide 24/7 power.  I've looked at a lot of different technologies that could reliably provide 24/7 power the way nuclear reactors do, but there aren't many feasible concepts out there when mass and cost are issues.  Mass may be a far greater issue for space applications than cost, but cost will determine suitability for use here on Earth.  Reliability / durability obviously affects both applications.

The batteries we actually know how to make fall woefully short of the energy density required to replace liquid hydrocarbon fuels and even the stuff in the labs doesn't approach 50% of the equivalent energy density required to achieve parity with liquid hydrocarbons.  The battery advocates will wave their magic wand and proclaim that better technology will appear in the future.  That may be true, but using the rate of energy density increase over the past decades from incremental refinements would seem to indicate that it will be many decades before batteries achieve energy densities comparable with liquid hydrocarbon fuels.  Absent a massive increase in battery energy density from some as-yet-unknown technological windfall, few people alive today will be alive to witness these miraculous new battery technologies.  Betting the future on batteries that don't exist, even in a lab, seems like a poor strategy if increased use of solar and wind is desired.

My takeaway from this is that practical energy storage is an exquisitely difficult problem to solve, but we may still have something that could service the requirement while providing reliable energy storage with drastically reduced mass and cost and service life that's multiple orders of magnitude over what batteries have ever demonstrated.  That's a win-win for both humanity and our various space applications technology development programs.

In Japan and China, as well as the US, there has been renewed experimentation with flywheel technology.  The Japanese have created a practical superconducting Carbon Fiber flywheel system.  The Chinese are working on a CNT variant with multi-kW/kg storage potential.  Commercially available Lithium-ion batteries can manage ~250Wh/kg, ignoring the mass required for thermal regulation / power conditioning / power distribution.  A flywheel constructed of CNT, instead of Carbon Fiber, would have a low-end energy density of 1kWh/kg and a high-end energy density of 10kWh/kg for the complete system, rather than just the energy storage component of the system.  The flywheel rotor itself could have an energy storage density of more than 50kWh/kg, so most of the mass is in the containment and drive system.  For comparison purposes, kerosene stores ~4.5kWh/kg of usable energy for a high efficiency simple cycle gas turbine.

The efficiency of storing and extracting power from flywheels, using electric motors / generators, can exceed 98%.  In point of fact, there are already non-superconducting electric motors for aircraft that surpass that level of efficiency in electrical-to-mechanical power conversion.  Unlike batteries, there are no-rated limited electrochemical reactions to contend with.  Unlike super capacitors, there are no charge-separation limited electrostatic reactions to contend with.  If the motor / generator can sustain the load placed upon it without significantly increasing resistance in its windings, then the flywheel combined with a motor / generator can store / supply the power.  The Carbon Fiber flywheels have already demonstrated more than 1 million cycles with no appreciable degradation in energy storage capability.  A Lithium-ion battery could theoretically deliver just as many cycles, but only if you were willing to discharge the battery in single-digit percentages before recharging it.  Flywheels and, to a somewhat lesser extent, super capacitors don't work that way.  It doesn't matter if you discharge them to 100% before recharging them.  There's very little capacity loss.  Since there are no moving parts that physically touch each other in the Japanese prototype, service life becomes a function of how many starts and stops the flywheel can achieve until the resin that bonds the CF plies to the flywheel fails.

I will point out that achieving those 1kWh/kg to 10kWh/kg energy density figures quoted above is not without technological challenge, as both use advanced fabrication techniques and superconducting electromagnetic bearings, but there's nothing technologically infeasible about implementing such systems.  The Japanese Carbon Fiber prototype already uses superconducting bearings and its flywheel is encased in a vacuum chamber to inhibit losses from air resistance.

The only significant downside to flywheel energy storage technology is that flywheels lose energy, just like batteries and super capacitors.  For a system that will see daily charge / discharge cycles, this is not a show-stopper, nor even much of a speed bump.  All practical solar / wind plus energy storage systems will work this way, on Earth or Mars and pretty much anywhere else.  The materials in the flywheels are under incredible strain, so even the new designs with no parts in physical contact with each other have some number of cycles they can realistically achieve before the flywheel is structurally damaged.  However, a flywheel constructed from CNT would be as near to immortal as any other non-nuclear power storage technology that we know of.

I've included a research paper from University of Texas on this topic for reference purposes:

Low-Cost Flywheel Energy Storage for Mitigating the Variability of Renewable Power Generation

Info on the Japanese Carbon Fiber Composite Superconducting Bearing Flywheel System:

World's Largest Superconducting Flywheel Power Storage System Test Machine Completed and Test Operation Started

The astute will note that the Japanese flywheel rotor alone only stores ~25Wh/kg, which is quite similar to a Lead-acid battery.  That's true, but you won't charge / discharge the Lead-acid battery a million times before it fails and no Lead-acid battery weighing 4 tons could deliver a 300kW burst of energy because that would destroy the battery.  A Lithium-ion battery could do that a few times, but that would also swiftly destroy the battery from both thermal and electrochemical effects of discharging at that rate.  A Carbon Fiber flywheel could pull that off 3 times a day for nearly the next millennia.

Why might we want CNT flywheels instead of batteries for Mars operations, even at a mere 1kWh/kg worth of energy storage?

A series of 7 1ton flywheel energy storage systems would have a combined mass of around 7,000kg and could both store enough energy and discharge fast enough to power an electromagnetic launch to return to orbit.  A solar farm with power output levels sufficient to produce enough rocket fuel to return something with the mass of Starship to orbit, after 2 years of continuous propellant production, could also send someone in a small capsule to an awaiting ship in orbit around once per day.  Needless to say, the rest of the time when that solar farm isn't storing power to return people to orbit, that power becomes available for useful tasks such as construction / drilling / mining / refining / farming / science experiments.

To accelerate to Mars orbital velocity using some sort of electromagnetic acceleration loop would require around 14kWh of electricity per kg of payload.  A survivable 3g acceleration to 4km/s, during which the vehicle occupant suddenly feels like he or she weighs about what he or she normally weighs on earth, takes approximately 136 seconds to complete.  Given the length of linear track required, we'd be looking at a circular launch track.  Obviously many more tons of materials for the launch loop would be required, but Carbon for CNT track and wiring can be locally sourced from the Martian atmosphere.  Assuming a minimalist capsule with a launch mass of 500kg, including a single occupant, that works out to about 7MWh worth of electricity per shot.

Launching to LEO would require about 28MWh/ton, but if the Sun was supplying the energy I doubt we'd have to worry about running out.  As always, the durability and longevity of the power transfer equipment is of extreme importance.  That's where the flywheel and motor / generator really shines.  If solar power can be supplied at an all-inclusive cost of around 10 cents per kWh, then each ton delivered to orbit requires about $2,800 worth of electricity.  That pegs the price of a ticket to orbit, or half-way to anywhere else, in a 500kg personal space capsule, at about the same price as a first class airline ticket to the other side of the world.  That seems infinitely more reasonable than paying for the cost of a very nice house to move to Mars.  Businesses could afford to send workers to other planets at those prices.

Offline

#2 2019-07-28 20:29:39

SpaceNut
Administrator
From: New Hampshire
Registered: 2004-07-22
Posts: 17,490

Re: CNT Flywheels vs Batteries for Energy Storage

https://en.wikipedia.org/wiki/Flywheel_energy_storage
A spinning mass supported by magnetic bearings to reduce spinning friction with components in a vaccumn.

The energy storage goes up with the square of the speed, so it is better to have a faster flywheel that weighs less.

If you have a 400kg flywheel doing 500rpm(51 rads/sec) you get about 1067kJ of stored energy if you take a 4kg flywheel and spin it at 5000rpm (517 rads/sec)you get the same amount of energy. and at 10,000 4271kJ4 4x more, at 20,000 rpm 16x more.

Steel is good for about 3,000 rpm while the fiberglass and other carbon will go higher for sure.

https://www.greenoptimistic.com/cheap-f … -20130305/
https://www.21stcentech.com/gizmos-gadg … real-hoax/
https://www.scientificamerican.com/arti … el-design/

Looks like it can be built with lower efficencies for fun
https://www.bounceenergy.com/blog/2017/ … -flywheel/

The motor and generator are permanent magnet design with the magnets moving to make the motor generator brushless removing drag losses.
https://www.diyelectriccar.com/forums/s … 88622.html

A Review of Flywheel Energy Storage System Technologies and Their Applications

Offline

Board footer

Powered by FluxBB