while browsing for new propulsion methods I found the following interesting:


"Radiation pressure (also known as light pressure) is the mechanical pressure exerted upon any surface due to the exchange of momentum between the object and the electromagnetic field.

This includes the momentum of light or electromagnetic radiation of any wavelength that is absorbed, reflected, or otherwise emitted (e.g. black-body radiation) by matter on any scale (from macroscopic objects to dust particles to gas molecules)."

enter image description here

So my question is: If a solenoid and a metal plate are attached together and they replace the light with an electromagnetic field from a solenoid, is it possible to have net thrust as shown in picture below? (Eddy current Repulsion Magnetic field from Solenoid and Metal Plate): enter image description here

Perhaps it is better to Consider 2 cases :

Case 1 : Source of Light and Reflector plate attached together . Will there be Net Unidirectional thrust ?

Case 2 : Source of Magnetic field or Antenna and Metal reflector plate attached together . Will there be Net Unidirectional thrust ?

  • 1
    $\begingroup$ I think your drawing violates conservation of momentum, as a closed system (solenoid plus plate) it's total momentum won't change. If you let go of the plate, then maybe the plate will go one way and the solenoid the other. Otherwise it will just sit there. Also, An AC magnetic field might have a very small "electromagnetic" component in the near field, but mostly it's just a magnet. $\endgroup$
    – uhoh
    Commented Apr 15, 2023 at 12:40
  • $\begingroup$ You could certainly use an actual radio transmitter with it's tuned antenna and beam radio waves at a plate, and definitely the transmitting antenna would go one way and the plate would go the other, because the radiation pressure for radio-frequency electromagnetic waves is the same as that of light of the same power per unit area (to first order). But in all cases you have to conserve total momentum, so we can't avoid for every action there is an equal and opposite reaction no matter how cleverly we try. $\endgroup$
    – uhoh
    Commented Apr 15, 2023 at 12:43
  • 1
    $\begingroup$ @uhoh The total system is solenoid plus plate plus electromagnetic field. The EM field also has momentum, so it's possible to create a situation where the antenna and the plate (connected together) accelerate in one direction, and EM waves move in the opposite direction (and carry away some momentum with them). Of course, the thrust would be tiny: to get 1 millinewton of thrust this way, you need the power of at least 1 mN times $c$ = 300000 W. $\endgroup$
    – Litho
    Commented Apr 16, 2023 at 9:30
  • $\begingroup$ @Litho I've added a answer to elaborate. What you describe is just a "flashlight" or any source of directed photons, there's no plate to "push against" like the OP is illustrating and asking about. But you do need a radiation reflector because most simple antennas in free space will have some symmetry in their radiation patterns that cancels a unidirectional thrust. $\endgroup$
    – uhoh
    Commented Apr 16, 2023 at 21:04
  • 2
    $\begingroup$ This meme except in space? No. This doesn't work. $\endgroup$
    – SF.
    Commented Apr 17, 2023 at 9:46

1 Answer 1


tl;dr: Think "antenna" and "beam of photons" rather than "pushing a plate with an electromagnet".

If your spacecraft accelerates (in your frame of reference) because of a force you are generating, you are constantly changing its momentum in that frame. Since momentum is conserved you must also find a way to be constantly imparting the opposite momentum on to something else.

for every action there is an equal an opposite re-action.

The above invokes the 2nd and 3rd of Newton's laws of motion

In your case if you want your electromagnetic propulsion system to accelerate your spacecraft in one direction, you must constantly generate photons going in the other direction. Instead of a stream of fast and massive rocket exhaust carrying away momentum, you need a stream of photons.

The drawing below shows a modestly effective photon generating system.

You make a resonantly-tuned loop antenna for the frequency you're transmitting, which efficiently converts the time-changing magnetic field into a propagating electromagnetic field instead of an evanescent time-dependent magnetic field.

Standing in free space alone, it would radiate bidirectionally so there would be no net thrust. You need a reflector element which could be a 2nd, parasitic loop of a slightly different size to reflect the radiation going one way back the other, using constructive interference, or in this case a flat conducting plane (sheet of metal or mesh of wire).

Now you've made a "radio flashlight" that beams momentum-carrying photons in one direction so that your spacecraft is propelled in the other direction.

Just remember that an electromagnet that's switched on and off doesn't necessarily make much useful electromagnetic radiation in the far field. Unless $\mathbf{B}(\text{t})$ and $\mathbf{E}(\text{t}) \propto d \mathbf{B}(\text{t}) / \text{dt}$ are generated in the right ratio, the field is evanescent and won't produce your beam of photons in the far field.

So think "antenna" rather than "electromagnet".

loop antenna with a reflector element beaming photons in one direction

Side view of a resonantly-tuned loop antenna in front of reflector, spaced at a quarter wavelength. The reflector can be a sheet of metal or a wire grid. The quarter wavelength spacing produces constructive interference (radiation to the left) because the process of reflecting from a conducting surface produces a phase shift of about $\pi$, equivalent to a half-wavelength shift.

  • $\begingroup$ So the mentioned concept is possible to achieve unidirectional thrust , though the thrust will be insignifficant? $\endgroup$
    – Stoyan
    Commented Apr 18, 2023 at 18:22
  • $\begingroup$ @Stoyan Short answer: possibly. Any time we turn on a light switch or in your case turn on an electromagnet, there is a small amount of radiated radio photons. Your design focuses on the magnetic field which won't make any propulsion (unless you're somehow propelling the plate away from the magnet). If the tiny amount of photons as radio noise you make when turn the magnet on/off is stronger in one direction than the opposite, yes you'll get a very very teeny tiny unintended propulsive impulse each time. But depending on the detailed direction distribution, they might also cancel to zero. $\endgroup$
    – uhoh
    Commented Apr 18, 2023 at 20:12
  • $\begingroup$ Lets consider the case with photons and the reflective plate. The source of the photons (instead Solenoid) is Super powerful Laser. Again the Laser and the plate are attached together. Is it possible to achieve Unidirectional thrust. $\endgroup$
    – Stoyan
    Commented Apr 18, 2023 at 20:15
  • $\begingroup$ @Stoyan btw following your comment "...Consider the plate to be perfect reflector" consider BowlOfRed's reply "What exactly is the plate reflecting?" Unless you convert your solenoid to a resonant antenna and start radiating a lot of radio waves, it won't be reflecting much of anything. IF you make your solenoid resonant and excite it with radio waves, then yes you will get some kind of propulsion. But for a long skinny solenoid the radiation field and it's propagation direction is totally different.en.wikipedia.org/wiki/Ferrite_core#Ferrite_rod_aerial $\endgroup$
    – uhoh
    Commented Apr 18, 2023 at 20:19
  • $\begingroup$ Lets consider the case with photons and the reflective plate (The first Picture). The source of the photons (instead Solenoid) is Super powerful Laser. Again the Laser and the plate are attached together. Is it possible to achieve Unidirectional thrust. $\endgroup$
    – Stoyan
    Commented Apr 18, 2023 at 20:26

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.