I've been thinking about the use of Dyson Spheres as a long range communication device. The sphere would effectively have an and open and closed state for a collection of panels which could be swapped in rapid succession.

What I'm thinking is that with a setup like this it could be used to transmit binary information at the speed of light in an omnidirectional fashion. I'm thinking from a fixed point of view this would look a lot like a pulsar.

Is this realistic? What kind of details should I add here?

  • $\begingroup$ So, basically, using a blinking light as a form of communication? $\endgroup$ – Frostfyre Mar 11 '16 at 15:50
  • $\begingroup$ Sounds like you are trying to build a giant one of these $\endgroup$ – cobaltduck Mar 11 '16 at 15:57
  • 1
    $\begingroup$ @jamesqf If the source of the data can be placed at the star's core or near the surface of the star then the signal can be made to arrive across the sphere within a certain amount of time to improve bitrate to an arbitrary amount (assuming the panels on the sphere are "dumb") but once each panel on the sphere knows how much delay to use you can once again get bitrate limited only by the switching rate of these panels $\endgroup$ – Steven Lu Mar 11 '16 at 18:16
  • 2
    $\begingroup$ @jamesqf You'd have very high latency, but I don't see why bitrate would have to be any lower than the maximum switching speed of the panels. Just rig it so that once the coordinating signal reaches the farthest point on the sphere from the source, then all the switches activate simultaneously. (And there's no reason you'd have to wait for this before transmitting the next bit. You'd just need a buffer of some kind to keep track of previously received bits on the panels nearest to the source.) $\endgroup$ – Ajedi32 Mar 11 '16 at 18:43
  • 2
    $\begingroup$ Ah, yes, stars, the backlight of choice for Dyson-scale LCDs... $\endgroup$ – Daniel Wagner Mar 11 '16 at 20:58

From your description, it sounds like you're looking for a stellar heliograph or signal lamp.

I think they problem here is that you're relying on a mechanical device to open a "window" for light to pass through. As Lacklub suggested, you could use several techniques to increase the speed of transfer. However, you're still going to need to ensure that the giant signal lamp is pointed exactly at the receiver. This is more complex with mechanical devices.

Instead, I would suggest looking at laser communication. Given that you're already gathering energy from your Dyson sphere, a laser seems like a reasonable transmission device. It has the advantage of being much more reliable and regular than opening a "window". You can also encode more information per second of burst as you can turn the laser off and on faster than you can open a panel.

It all makes sense now...

Edit: Schwern brings up a good point that this is not omnidirectional as asked for in the OP. If the OP doesn't need truly omnidirectional communication, but rather communication to multiple sources, than you could of course use multiple lasers. Outside of a SETI context, I think the creators would have an idea of where they want to "broadcast" to, and could thus harness all of the light going to empty space. The amount of power wasted by turning the Dyson sphere "off" is literally astronomical.

  • 1
    $\begingroup$ The OP asked for omni-directional transmission (for some reason) which rules out lasers. $\endgroup$ – Schwern Mar 11 '16 at 19:00
  • $\begingroup$ A good point, edited my post accordingly $\endgroup$ – Kys Mar 11 '16 at 19:29
  • $\begingroup$ Blacking out an LCD screen would be about as fast as turning off a laser. $\endgroup$ – coblr Mar 11 '16 at 21:08
  • $\begingroup$ But could a structure made of LCD panels also act as a Dyson sphere? $\endgroup$ – Kys Mar 11 '16 at 21:31

You would need to think about how big the panels are, and how fast you're moving them. If the panels are the size of the star, for example, then it will take more than two seconds to move them onto the hole - and that is travelling at near the speed of light. If it's smaller than the star, then you're dimming it for the destination.

There are a few things that you could do to increase information density in this transmission, and still be (nearly) as bright as a star:

  1. Frequencies. You can only allow certain frequencies of light through, ie. by putting chemicals in the way, to increase the amount of information each second. This forces the destination to be using spectroscopy to retrieve the information.

  2. Smaller panels. You can have a grid of panels that each know all of the information that is going out: they can be thin hexagons, for example. Have them be flat when they block the light, then turn to their thin side to let the light through. With a very large array of panels, you could have each panel be small and flip rapidly.

  3. Polarization. The sun emits light of a continuous spectrum of polarization. There are a few polarizations that you can use to increase the information density, but this one is trickier for quantum mechanical reasons.

If you want to keep this as bright enough to see, and want a decent information transfer speed, I would recommend going with 2, and maybe 1. You may even be able to cover the whole sphere with panels like 2. Of course, you might want to find a reason why using a laser isn't a better alternative.

  • $\begingroup$ I was already thinking #2, actually having the whole sphere built this way possibly with tens of thousands (or more) panels. I was originally thinking lazer light, but it seems the loss of output versus full spectrum wouldnt be worth it. $\endgroup$ – Tim Brigham Mar 11 '16 at 16:06
  • $\begingroup$ @TimBrigham Ah, ok. After reading your post again, I realized I misread. Laser light can be switched on and off very quickly compared to conventional mechanical switches, and you can just have several lasers of different frequencies to get the required information density. And full spectrum is only a benefit if you use it. $\endgroup$ – Lacklub Mar 11 '16 at 16:10
  • $\begingroup$ By using a grid of panels, you're making a giant Aldis Lamp. $\endgroup$ – Schwern Mar 11 '16 at 19:02
  • $\begingroup$ @Schwern Exactly. But they can't be secured just at the ends: if they were then the mechanical stress to turn them, which can only go at the speed of light (even less: the speed of sound in the material) would cause waves in the material, and result in slow actuation. So they need to be actuated all along the length of the material. $\endgroup$ – Lacklub Mar 11 '16 at 19:16
  • $\begingroup$ @Lacklub You're assuming these panels have to be enormous and have to be long. Instead use a grid of smaller, square(ish) panels. Like how E Ink uses a grid of tiny particles to make pictures. $\endgroup$ – Schwern Mar 11 '16 at 19:36

I'm not sure how useful it would be. Most people close enough to read the signal would be better communicated with via radio.

You're of course limited to the speed of light for your communications which also happens to be the same speed as radio waves and other electromagnetic communications.

The one thing it would have over a radio signal is it would have the power of the star generating the signal, instead of equipment. But at this time they have a Dyson Sphere, collecting all of the energy it can from the sun. So while it is a neat idea to have a Morse code signaler using a sun blinking in and out. It doesn't seem like the most economical use of time or energy.


This is no different from optical pulses in optical fiber cable - it's just that now, space is the transmission medium. Just follow the protocol.

From Agrawal, Manish (2010). Business Data Communications. John Wiley & Sons, Inc. p. 54.

  1. The data is coded as binary numbers at the sender end
  2. A carrier signal is modulated as specified by the binary representation of the data
  3. At the receiving end, the incoming signal is demodulated into the respective binary numbers
  4. Decoding of the binary numbers is performed

...Only to find out that it doesn't work.

See, fiber optic cables have what's called low attenuation, given certain guidelines such as the length of the cable and the bit rate.

In space, you have data loss from things like

  • Wave interference
  • Gravity - passing celestial body, will bend the light
  • Space dust
  • The points of failure involving the mechanical energy spent moving your panels
  • ...

If you want a reliable mode of communication at the speed of light, there's another one: Radio.

It's likely that because radio would be developed long before you have a Dyson Morse Code Sphere, you likely have better error checking technology developed on the receiving end. Longer wavelengths are also better than visible light (shorter than radio) because they don't scatter as easily (See Rayleigh scattering), which would overcome more obstacles you'd find space.

  • 4
    $\begingroup$ radio IS light, light IS radio. The only difference is the frequency of the electromagnetic radiation. Gravity affects radio waves just as much as it affects light beams. $\endgroup$ – Ben Voigt Mar 11 '16 at 19:14
  • $\begingroup$ @BenVoigt By saying that radio travels at the speed of light, I was not saying that radio is not light. It travels at the speed of light because it is light. I just thought it was so obvious that I didn't have to explicitly say it. And your point is understood, but I have a feeling that receiving and interpreting radio is more well formulated than a many light year long version of a fiber optic cable. $\endgroup$ – The Anathema Mar 11 '16 at 23:42
  • $\begingroup$ @BenVoigt Yes I did. There's another one: Radio. I listed radio as opposed to flipping a switch on and off a bunch of times. Did you read the question? $\endgroup$ – The Anathema Mar 11 '16 at 23:47
  • $\begingroup$ @BenVoigt Amplitude modulation and frequency modulation, neither of which involve mechanical energy moving several mile long panels with millions of points of failure, which could affect the level of attenuation and the amount (and quality) of information that can be sent. $\endgroup$ – The Anathema Mar 11 '16 at 23:55
  • $\begingroup$ @BenVoigt I have addressed your issues in my answer. $\endgroup$ – The Anathema Mar 12 '16 at 0:02

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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