Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Using magnetic fields in place of heat shielding

  1. Jul 21, 2011 #1
    Using Magnetic Fields in Place of Heat Shielding
    Joshua “FireStorm” King
    Note: All Graphs are in SI base units
    Note 2: I strongly recommend downloading the document version of this post as the equations are actually readable there.
    RTF Format: View attachment FireStorm-Magnetic Heat Shielding.rtf.zip
    Docx Format: View attachment FireStorm-Magnetic Heat Shielding.docx.zip

    Introduction: The Problem
    I'm a real space geek; when I was sitting in my AP physics class learning about magnetic fields, I thought "Why not use magnetic fields instead of bulky one-use heat shields." I mean, let’s face it: heat shields kind of suck; they are external to the spacecraft, and thus easily damaged by debris (See Columbia disaster). They also can really only be used once, which is fine in the case of a single use vehicle, but what about re-usable craft such as the shuttle, or any interplanetary craft we may build in the future.

    Introduction: My Solution
    In theory, magnetic fields should be just as effective as a hunk of metal or ceramic; maybe even more so. When a charged particle (such as the protons and electrons in the plasma disassociated when a spacecraft reenters the atmosphere) encounters a magnetic field it encounters a force perpendicular to both the magnetic field and velocity. This causes the particle in question to follow a circular path. In other words, the magnetic field causes the particle to turn around (half circle) in some radius R. Once many particles start doing this they should start colliding with other incoming particles, creating dense zone of plasma where the pressure from the magnetic field turning particles around counters exactly the pressure of the onrushing plasma.

    Aero Braking: spacecraft could use an effectively giant ‘magnetic sail’ so to speak to slow itself down in a planet’s exosphere; the atmosphere up there should already be ionized from solar radiation, thus one could easily form a sail with many thousands of times the cross section of the spacecraft.
    Aero Capture: Similar to Aero Braking, Aero Capture attempts to slow the spacecraft down to orbital velocity in one pass. This can be a very hit or miss maneuver as it can be very difficult to calculate accurately. By varying the strength of the magnetic field as the spacecraft plunges into the atmosphere, one could control precisely the delta v that the spacecraft receives, and end up in a better orbit.
    Reentry: Keeps the intensely hot plasma created during atmospheric reentry away from the relatively fragile spacecraft. No heat shield would be required (or at least much less shielding would be needed). Craft would be reusable, as there is nothing to ablate. The heat shield is the atmosphere. Also, aerodynamic considerations don’t need to be made for hypersonic flight, because the craft will be encased in a bubble of plasma. Once the craft slows down sufficiently, the air will stop ionizing, and the spacecraft can handle in atmosphere like a plane, or deploy parachutes, etc.

    Derivations: how to find external B field, turning radius of particles
    The turning radius of a given particle (assuming a roughly uniform magnetic field) is:
    a_ctp=v^2 / r
    Remember that equation for a second

    Now, knowing the magnetic field inside a superconducting magnet can easily reach 25T, we can use that number to calculate the B field outside the solenoid.
    dB/dl = μ_0/4π I/x^2
    B_(solenoid )= μ_0 nI
    B_(ext )= B_(close wire) – B_(far wire)
    B_ext =(2r)μ_0/4π In(( 1)/( x – r )^2 – 1/( x + r )^2 )
    B_(ext )= r^2/π B_(sol ) ( 2x + r )〖( x + r )〗^(-2) 〖( x – r )〗^(-2)
    Thus we now have an expression that will tell us the external magnetic field, for a solenoid of radius r, internal field strength Bsol at distance X. Graphing that with Bsol = 25T and r = 3m gives us:


    One can see that VERY significant magnetic fields would exist within several meters of the solenoid. A half Tesla magnetic field would exist 5 meters from the solenoid. Were the solenoid literally part of the exterior hull, that would mean that, at the hull(assume .25meters), we’d have a field of 278 Tesla. Let’s calculate the turning radius of a proton under those conditions:
    r_turning=mv/(q 〖r_sol〗^2/π B_sol ( 2x + r_sol )〖(x+r_sol )〗^(-2) 〖( x–r_sol )〗^(-2) )


    We can now further see that particles coming with 10m of the craft will have a turning radius on the order of .1 mm. 5 meters from the hull the turning radius of particles is .13mm.

    Derivation: Conclusions
    In my mind at least, this means that the magnetic field will end up pushing against the flow of charged particles, inducing drag much like a conventional heat shield. Based on the above calculations, even a modest strength solenoid would have an enormous effect on the plasma at anything above 1000m/s.

    How practical is this? If there aren’t any glaring problems, why hasn’t it been used?
    What shape would the “plasma bubble” take? Would one form, or would the plasma be more dispersed?
    How effective is this as a heat shield? Does it protect the spacecraft completely?
    What would the coefficient of drag be for varying angles of attack?
    Given that particles spiral along magnetic field lines, what happens at the open ends of the solenoid?
    What would be a good layout of electromagnets for stability, control, and redundancy on an actual craft?

    Approaching the Answers
    I definitely want to open this idea up to discussion. Pros and Cons, is my math right, etc. I would also like to find a good magneto hydrodynamic simulator that will let me see the effects of hypersonic plasma flow over the solenoid; I can vary the angle of attack of the solenoid(s) and see how protected the spacecraft would be. I think a simulator will be necessary to answer a majority of these questions to my satisfaction.
    Last edited: Jul 21, 2011
  2. jcsd
  3. Jul 22, 2011 #2


    User Avatar
    Staff Emeritus
    Science Advisor

    I don't believe this will work. The spacecraft depends on the friction of the air to slow down. In effect, you have a large mass of non-ionized air that forms a cushion against the primary shock front, causing most of the heated air to flow around the vehicle. Magnetic fields will not stop this mass of air from forming, and I think it is also heated to a certain amount, so it doesn't look like there is any benefit. There are many considerations when designing a heat shield. Not all of them are one use only. Many of the shuttle tiles themselves can withstand multiple uses since they are not ablative.

    Also, your method is very complex and requires lots of power, cooling, and support equipment to operate. Have you considered the size, weight, and reliability of all these?

    I recommend reading here: http://en.wikipedia.org/wiki/Atmospheric_entry
  4. Jul 22, 2011 #3
    My design here isn't so much to replace heat shields as to give a level of redundancy the current shuttle design lacks; in other words, if the shield is damaged, there's no way home.

    I actually considered this for a long time before posting (I had this idea when school was in session), and have consulted half of Wikipedia on the issue. What you describe is exactly what I envision. A pocket of cooler air protects the craft from the supersonic shock-wave and associated superheated air. The friction of air on air slows the spacecraft. It would the the same as an egg shaped ball reentering the atmosphere, only the surface is plasma.

    You are however correct there; research fail on me.:redface:

    I was actually thinking of that and you might be able to pull if off with less mass, or at least similar mass than a heat shield. I've done extensive research into superconducting magnets. They can be cooled easily by emerging then in LN2. Also, the require NO power once operating (resistive losses in a true superconductor are literally zerohttp://en.wikipedia.org/wiki/Superconductor#Elementary_properties_of_superconductors"). As far as weight goes, all a superconductor is is a ring of copper with superconductor wire wound within. The wire has high resistivity until cooled below the superconducting threshold by liquid nitrogen on-board, at which point a very small potential will let you build up a VERY strong magnetic field.
    Last edited by a moderator: Apr 26, 2017
  5. Jul 23, 2011 #4


    User Avatar
    Staff Emeritus
    Science Advisor

    Or we could simply make the current heat shield more durable and avoid the complex set of parts that this would require, along with the power source to power it.

    I really don't understand all the details of the re-entry shockwave and such, but I have to ask, if we already accomplish what you are suggesting with air, what's the point? Isn't the spacecraft still going to heat up just as much as it does now? That mass of air in front of the ship doesn't stay cool as far as I know.

    You need something to power the cooling system and most of the weight will be in the support systems and structure holding the coils in place.

    I honestly don't know if this would work. Maybe someone else can jump in and tell us if this magnetic field would be an effective shield?
    Last edited by a moderator: Apr 26, 2017
  6. Jul 23, 2011 #5
    Personally I envision this for a craft that will probably have a nuclear reactor in it anyways, so that's power out of the way. We're already carrying cryogenic fluid for the engines, and the cost of the superconductor itself is very low. Should be less than an MRI machine at $50000. I'd think that makes it a very cost effective device, assuming my figures are all on target. If I were to go up in the thing, I'd rather be sure I'm making it home with my expensive spaceship in one piece (Always have a plan B[and preferably C]).

    To make an analogy, if I'm going to point a blowtorch at your face, would you rather it be 10cm away, or 1m? It's the same thing. Peak temperature is the same, but the hull itself if farther away from it. That should reduce conductive heat transfer by a large amount.

    I picture this on a craft with a nuclear reactor, so power is no object, and as far as structural weight goes, it should experience the same amount of force as it would were it a conventional heat shield, thus support requirements should be equal between the two. I think.

    That would be swell. I also emailed this idea to my former physics teacher to hear his 2 cents, and another good friend of mine. I'll quote their thoughts when I get 'em. If no one points out any great flaws I'd love to see what NASA thinks.
  7. Jul 23, 2011 #6
    My friend pointed out that putting a solenoid sideways in a flow of plasma would be similar to putting earth's magnetic field in the path of oncoming solar wind. In effect, the magnetic field on the leading edge would squash, and the tail would elongate. He also found this graphic:
    As you can see, most of the oncoming plasma is slowed where labeled "bow shock." Earth is well protected from solar wind by it's magnetic field except at the poles. In theory, a spacecraft with magnetic field should experience the same effects: Production of a supersonic shock-wave (where most of the heat is produced) which is being directed around the ship by the magnetic field. The only significant contact with plasma would be the mini aurora along the axis of the solenoid. I definitely need to do more research in this area though.

    Some of his other concerns were the thermal distribution of ions hitting the field, black-body radiation, and the orientation of the magnetic field. I believe I addressed all of his concerns to his satisfaction. I explained that the axis of the solenoid would be perpendicular to the flow direction. Also, the craft should be surrounded by a layer of cool(er) gas which rides inside the shock-wave. The issue of thermalization is entirely dependent on the temperature at the shock-wave. Using values from the space-shuttle however, I believe it will be a non-issue.
  8. Aug 6, 2011 #7
    I dislike giving discouraging opinions to an inventor, but... Here are some difficulties I see.

    Air is slightly ionized in the shock wave. During the heat flux peak, much oxygen is dissociated into individual neutral atoms, some nitrogen is, but very little is ionized. Neutral atoms and molecules would pass through the magnetic shield.

    By repelling the incoming molecules, you compress hence heat them. I believe a magnetic shield does it just like an aerodynamic shock wave. Importantly, the gas+plasma heats the spacecraft by radiation, not through direct contact, and here the shock by a magnetic shield wouldn't improve over the aerodynamic shock.

    Also: a traditional ablative heat shield contains carbon which, as the shield ablates, makes the shock wave opaque to light, so the spacecraft gets heated only after light from the hottest part of the shock bow gets absorbed and re-emitted many times, reducing much the heat flux. Any non-ablative shield must give up this help - this is done presently.

    A superconducting coil can produce 25T but not easily, and I know none producing 278T. Progress in this area is very difficult. Anyway, magnetic forces destroy coils, and more than some 50T (more now?) is achieved only during pulses - a few pulses before the coil explodes. I wanted to recommend a fabulous introduction but don't find it any more on the Web, it was by F. Sonneman from CERN, dealing mainly with quench analysis.

    Improvements (or alternatives!) to heat shields would be highly welcome for sure, but the present is already better than when the Shuttles were designed 30 years ago. After landing, the X-37B's shield looked rather healthy, at least as seen from a good distance.

    If you like heat shields and reentry, you may find puzzling ideas at saposjoint.net, forum science, subforum technology, topic X-37B. The atmospheric agility for spacecraft will probably emerge in the coming years; the ablative spike before the delta wing, I don't know. You'll have to log in to see the sketches, as Sapo changed it on his site, alas.

    Marc Schaefer, aka Enthalpy
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook