Comments in chat have noted the recent post in the Centauri Dreams blog titled The Dipole Drive: A New Concept for Space Propulsion.

In contrast to the single positively charged screen employed by the electric sail, the dipole drive is constructed from two parallel screens, one charged positive, the other negative, creating an electric field between them with no significant field outside.

Also near the end:

Because the dipole drive does not interact with plasma outside of the zone between its screens, the issue of Debye shielding of its screen system to outside charges is not a concern.

My question is about what the phrase "no significant field" (as well as "does not interract... outside of...") means. Each screen is an equipotential, and the value of the integral of $\mathbf{E} \centerdot d\mathbf{l}$ is going to be the same whether you integrate along a short line between the plates, or the long way around, where the field is much weaker but the distances are just as much longer.

Is this just an oversight, or does this have some real impact on how this device would actually perform for a realistic size, plasma density, and potential difference in the scenarios discussed.

Note that adding a second pair of grounded grids outside the "dipole" pair to try to enforce E=0 outside just adds additional forces in the additional gaps that might tend to cancel the effectiveness. I don't see a quick fix here.

Question: Is this description of the "dipole drive" and how it would work physically correct?

The new "dipole drive" as reported in Centauri Dreams

The new "dipole drive" as reported in Centauri Dreams

  • $\begingroup$ Thanks to @KimHolder for noting this new blogpost in the Pod Bay! $\endgroup$
    – uhoh
    Commented Jul 1, 2018 at 13:25

2 Answers 2


I think you're right. A finite pair of charged screens, with zero net charge, can't create an infinite potential well; the potential far away from the device must return to zero, which means that the propelled particles must get decelerated to their original velocities, taking back the momentum they initially provided.

It seems that Zubrin has taken a familiar approximation — that the field near a charged plane surface is constant (and therefore the fields of the two screens cancel out, outside the gap between them) — and forgotten its key assumption: it applies only at small distances (relative to the size of the screens). But the particles don't stay that close for very long at all. As they move away, they start to feel the dipole field around the spacecraft. Although this field is weak and falls off fast, as $\frac{1}{r^3}$, the distance the particle travels in this field, as you point out, is much greater.

(I'd love to be wrong about this. It seems like such a clever idea. But I can't find a flaw in this reasoning. Maybe someone else can.)

Edited to add: I think it's instructive to consider what would happen if the screens were actually infinite. Suppose that the density of plasma entering the gap is uniform, and that power sources are also placed with uniform spacing between the screens, so that the input power per unit area is constant and the density of ejected particles on the two sides is also constant. This device would work! — because the field really would be identically zero except in the gap between the screens; it would no longer matter how far away the particles got. Interestingly, the charge density on the screens would increase without bound; as protons were ejected out the negative side and electrons out the positive side, the power sources would have to continuously add charge to the screens so that the charge difference between the half-universes on each side of the gap would remain constant (as it must for the voltage across the gap to be held constant).

But clearly, on a finite device, the charge on the screens can't increase without bound; the ejected particles quickly get far enough away that their charge no longer counteracts that on the screen they came through. Zubrin's mistake was to confuse the analysis for a finite section of an infinite device with that for a finite device.

  • $\begingroup$ Don’t you mean they must get decelerated to their original speeds (i.e. magnitude of velocity), not velocity? The direction of the particle’s velocity vector can change in the process, no? And if so, doesn’t this impart momentum? $\endgroup$ Commented Jan 10, 2019 at 10:19
  • $\begingroup$ An analogy to a gravitational flyby may be helpful here: the spacecraft’s energy relative to the planet remains constant, but a momentum transfer still occurs because the direction of the spacecraft’s travel changes. $\endgroup$ Commented Jan 10, 2019 at 10:21
  • 1
    $\begingroup$ Your criticism appears to be valid in the case that the proton travels straight through, as depicted in some of the diagrams. But in the reflection case I don’t believe it is. $\endgroup$ Commented Jan 10, 2019 at 15:14

I want to admit, that Zubrin’s Dipole Drive consept will not work even in condition if you have infinite electrodes surface. Situation even more complicated than it is discribed by ScottBurson . Scott forgot that during infinite capacitor charging process, external vacuum electrical field will fully compensate internal electrical field, two parts of the Universe will be isolated by this infinit grid :-), only quantuum (tunnel) currents / noise will exchange between half-universes.

In addition Supposed electrodes are not composed by solid conducting surface, it is some type of grid, that has huge intergrid gaps, electric field and electrodynamic interaction between moving charged particles and finit sized non-solid grid electrodes is much more complicated. Summary energy balance will be always zero, no movement as sequence. You should pay attention also to expected DC Voltage,DC Current and power calculations, that are very speculative and have no any connection to real physic. Conductivity of interstellar plasma is very small, so described in Dipole Drive DC currents are not real, orders higher that real life. Youshould mention also supposed interstelar plasma conposition electrones and positive charged ions , where are negative charged ions?

  • 1
    $\begingroup$ Welcome to space stack exchange. This would be a mucn better answer if you could expand it and explain your various points at more length. Long answers, equations, etc. are encouraged here. $\endgroup$ Commented Jan 23, 2019 at 10:30
  • $\begingroup$ The problem with Zubrins concept, that he wrongly use well known electristatic equations to his concept. $\endgroup$
    – AlexT
    Commented Jan 25, 2019 at 17:56
  • $\begingroup$ 1. For example Zubring expects kinetic momentum from positive charged ions due to higher mass of positive particle , he wrongly apply kinematic laws related to rigid bodies impact to the case when charged particle are free flying through electric grids in electric field. Lorenz (electrostatic) forces depend only on particle charge and do not depend on charged particle mass. Summary, there should be no any momentum applied on dipole when it is located in neutrally charged media (plasma). $\endgroup$
    – AlexT
    Commented Jan 25, 2019 at 18:10
  • $\begingroup$ 2. Second problem DC current calculation is totally wrong, even if we will accept Zubrin’s numbers rekated to plasma density of interstellar vacuum, DC current that is caused by DC voltage applied between two grids can be caused only by charged particles that impacts conducting surface of the grid, most particles are free passing by both grids, so cannot be accouted in DC current calculations in any way, despite some electromagnetic disturbance as sequence AC currents will be caused by passing by particles, but it has not connection to the equations Zubrin use. $\endgroup$
    – AlexT
    Commented Jan 25, 2019 at 18:12

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.