Can the reaction wheels also be used to store energy?

My understanding is that the parts of a typical flywheel energy storage are pretty much the same as the parts of a typical reaction wheel subsystem -- both have a flywheel, electric motor/generator, etc.

Is it possible for a spacecraft to use a flywheel to store and later supply energy, and also (perhaps at an even later time) re-use the same subsystem as a reaction wheel for attitude control?

Is it possible for a spacecraft to use a sufficient(*) number of flywheels to simultaneously control attitude and store energy, and later simultaneously control attitude and supply energy?

(*) 3 flywheels are not sufficient, but my understanding is that many spacecraft already have 4 or more reaction wheels ( Optimal placement of 4 reaction wheels? ).

• possible duplicate of Would a Centrifugal Battery have a distinct advantage or disadvantage in zero or low gravity? Yes, they can. Whether that is a good idea is another matter. Jun 12, 2014 at 19:34
• @RossMillikan: This question is asking about the 4 (or more) reaction wheels that some spacecraft have. I'm not sure what a "centrifugal battery" is, but the responses to that question seem to imply it's something that (so far) no spacecraft has. How can one question, about something some spacecraft have, be a duplicate of a question about something that (so far) no spacecraft has? Jun 13, 2014 at 2:37
• As you suggest, you can couple two reaction wheels to store both energy and angular momentum around one axis. Then if you have enough pairs (4 for redundancy) you can store energy and momentum around all three axes. The simple answer to your question is "Yes-you can reuse reaction wheels to do energy and momentum storage." The next question is whether it is a good idea. That was the source of my link. Jun 13, 2014 at 2:46
• I think [user:James Jenkins] suggested a pair of two parallel counter-rotating reaction wheels, and [user:Peter Wone] suggested using several such parallel pairs for attitude control. I see now that 3 such parallel pairs (6 wheels) could, in theory, simultaneously control attitude and store energy. Which leaves both parts my original question unanswered: Can the reaction wheels of any real spacecraft be used to supply energy? And if so, can the 4 or more wheels of any real spacecraft -- no two of which are parallel -- simultaneously control attitude and supply energy? Jun 13, 2014 at 4:17
• It is mathematically possible to do both attitude and energy with four wheels, but I am not aware of the wheels on real spacecraft being sufficiently large to store significant energy. I have seen presentations suggesting it. Jun 13, 2014 at 13:36

It would be highly problematic for reaction wheels to serve dual purposes as reaction control devices and energy storage mechanisms. It might be possible, but the implementation would be extremely complicated.

Starting with a simpler case of a three-wheel design, a desired spacecraft attitude and/or slew rate uniquely determines the wheel speed. That is, given a set of initial conditions that include spacecraft attitude and current wheel speed, the wheel speeds at the desired attitude and rate is completely determined.

This means reaction wheels would make crummy energy storage devices, because if you want to control attitude you wouldn't be able control how much energy was stored in the wheels, and conversely if you needed to store or use energy, it would change the spacecraft orientation.

Perhaps if the spacecraft was not 3-axis controlled that would be acceptable. Likewise, with enough reaction wheel redundancy you might have enough degrees of freedom to control energy storage and attitude simultaneously, but the math involved sounds like Ph. D work to me.

Lastly, batteries aren't really that expensive, supply very clean power, have high energy density, are reliable, and can be recharged with solar arrays very efficiently. In contrast, wheels tend to be expensive because they have to be precisely balanced and have moving parts; this also causes wheels to have generally lower reliability than solid-state subsystems. I'm not inclined to replace a battery with a wheel.

• By using the wheels to charge batteries, one could de-couple the opportunity of spinning them down from the time when the energy is needed (like hybrid cars make the combustion engine run more economically by smoothing out its load). Spinning something very fast doesn't add mass to the payload to be launched. Mayby not only reaction wheels, but most of the payload could be given spinning kinetic energy before launch, to be converted to electricity or heat, or trajectory change, in space? Jun 19, 2014 at 9:39
• You really don't want to be spinning large mass flywheels up before launch. Manoeuvres are difficult enough! Jun 19, 2014 at 19:43
• Several reasons you wouldn't want to spin up huge flywheels on the launch pad. 1) launches are often delayed and wheel speed would slow due to friction, requiring frequent top-off 2) if any significant momentum was stored the rockets controls engineers would kill you because you'll make it harder to fly the rocket 3) keeping the wheels spinning probably requires the payload to be on during launch—often this is a no-no. Jun 20, 2014 at 1:28

When you have two reaction-wheels on the same axis, you can accelerate one in one direction and the other in the opposite direction. The net torque would be 0, but you would have energy stored in both. When you then want to still use them as reaction-wheels, you just have to transfer momentum from one of the wheels to the other to get a net torque.

However, keep in mind that it is mechanically impossible to create a flywheel which is completely free of friction, even in microgravity. That means that this contraption is only suitable for short-term energy storage.

• +1 for point about friction. In the scheme you're proposing you'd need to limit each wheel to less than it's maximum speed to give you head space to transfer momentum to. Or you could spin up both wheels to full, but then not only would you bleed energy through heat friction, but in order to control attitude you'd need to have a place to dump electrical power on-demand—like say a battery, in which case why bother. There is a reason these are separate subsystems: the cost of combining greatly exceeds the potential benefit. Jun 20, 2014 at 1:24