Flight laser transmitter - for Psyche mission has Mass 29 kg (100 W).

It also requires pointing up to 3 degrees of the Sun (towards the receiver telescope).

My question is about really small probes (less than 1kg) that can carry only a small flashlight or a 1 W laser.

  • 2
    $\begingroup$ Since you mentioned communication it might be required to specify the symbol rate (on/off rate or some equivalent parameter) since telescopes can detect fainter amd fainter objects given enough time to accumulate the photons. $\endgroup$
    – AJN
    Commented Apr 6, 2023 at 13:19
  • $\begingroup$ @AJN - this is a continuation of my previous question about micro probes that can take a pictures and transmit them with light emitter (without establishing two-way communications). Space-based telescope will know location and exact time of transmission from the probe. For example, if we are talking about Asteroid Belt probes, can LEO telescope detect flashlight in Asteroid Belt, or you will need to move this telescope(s) to the Mars orbit. $\endgroup$ Commented Apr 6, 2023 at 14:09
  • $\begingroup$ Voyager uses ~23W radio transmitters. $\endgroup$
    – Jon Custer
    Commented Apr 6, 2023 at 15:28
  • $\begingroup$ @JonCuster - how would you use radio transmitter on a probe less than 1kg? $\endgroup$ Commented Apr 6, 2023 at 16:28
  • $\begingroup$ That was for you to compare transmit powers and distances... $\endgroup$
    – Jon Custer
    Commented Apr 6, 2023 at 16:33

1 Answer 1


Deep Space Optical Communications (DSOC) - from what distance a flashlight1 can be detected2 by a space-based telescope3

  • 1or a 1 W laser, on a really small probes (less than 1kg)
  • 2data received from at a useful rate
  • 3of modest/reasonable size typically used for DSOC

tl;dr: At a million kilometers, with a thin, 2 cm diameter, f=10 cm Fresnel zone plate collimator and laser diode, into a 1 meter diameter receiving telescope, you should be able to pull of 1000 baud downlink with some very clever actuator engineering to keep the laser diode near the focus of the zone plate and doing micro-steering to compensate for your 1 kg spacecraft attitude errors.

We'll use a 1W laser diode because they are

  1. compact
  2. light-weight
  3. efficient (converting electrical power to optical power)
  4. don't require heavy drive/support electronics (open a diode laser pointer and see what's inside!)
  5. and the actual light source is only a few microns in diameter. No tungsten filament or white light LED flashlight will have such a nearly point-source property.

Why is #5 important?

We want a narrow beam so that very far away a telescope will be cable to capture a useful amount of it.

The width of a beam in the far field (in radians) is basically the source diameter divided by the collimator focal length. A laser pointer will have a laser diode with a divergence of something like 3 to 10 degrees and a small collimating lens (molded plastic and therefore usually aspheric) as a single tiny module.

These can give final divergences close to the theoretical 1.22 $\lambda/d$ where the wavelength is of order 1 micron and the diameter is of order 1 mm so about 1 milliradian or about 0.06 degrees.

Composite image of two common laser diodes packaged with small collimating lenses, cropped and resized from Wikimedia Commons' https://commons.wikimedia.org/wiki/File:Diode_laser.jpg and https://commons.wikimedia.org/wiki/File:Metal_covered_Laser_diode_switched_on.jpg

above: Composite image of two common laser diodes packaged with small collimating lenses, cropped and resized from Wikimedia Commons' Diode_laser.jpg and Metal_covered_Laser_diode_switched_on.jpg

That's good enough to be seen by eye from the ISS from Earth

but not good enough for us.

So your real challenge is to make the biggest aperture collimator for your 1 kg spacecraft.

That's not going to be a big old Newtonian or Cassegrain reflector telescope with a glass or silicon carbide mirror

or a thick bulk lens

it's going to be a planar diffractive device like a zone plate or Fresnel lens or optical counterpart of a Fresnel zone antenna. Basically, since you have essentially an almost single wavelength of light, you can collapse your thick lens or deep-dish mirror to a nearly flat device where only the fractional part of the phase surface is used.

Some excellent examples of this for radio waves can be found in

The simplest, easiest, lightest weight collimator lens I can think of is a thin glass disk (say a 300 um thick fused silica nearly-optically-flat wafer) with a Fresnel zone pattern either etched into it for a phase plate, or patterned as thin absorbing aluminum or gold (say 10 to 100nm thick) as an amplitude plate.

You need to be a clever engineer to find a way to make this thing pop up on deployment such that the laser diode is near the correct focal distance and somewhere near the optical axis (line drawn from the center of the zone plate perpendicular to it). Maybe a shape memory alloy or something even more clever.

This could be a few up to 10 cm. Let's be conservative and call it 2 cm diameter with a focal length of 10 cm. Such a zone plate will have about 1,000 zones with a 5 micron period at the edge - really quite easy to make in a modest university lab these days.

Your theoretical divergence (half-angle) from 1.22 $\lambda/d$ is now about 60 microradians or about 0.0035 degrees.

Sounds impossible to point now, doesn't it? With your 1 kg spacecraft? Well it's doable with a lot more clever engineering. You do have to get pretty good attitude control, but a tiny actuator using light-weight voice coil or other MEMS actuators moves the little laser diode around in 3D near the focal point of your zone plate, so you can do sub 1 degree pointing of the beam electronically to correct for spacecraft attitude imperfections.

Okay but from what distance can we get useful data?

At 1 million kilometers (109 meters) a 60 $\times 10^{-6}$ radian beam will be 60,000 meters in diameter.

Your reasonably sized 1 meter diameter receiving telescope will pick up (1/60,000)2 of that, or 3 $\times 10^{-7}$ milliwatts. In engineering terms that's -75 dBm which is not bad at all!

If we assume 1 eV per photon, that's 6 $\times 10^{18} \ \times 3 \times 10^{-10}$ or over a billion photons per second received. Expressed as a photocurrent that's a few tenths of a nanoampere.

If we assume that shot noise limits your bandwidth (rather than kBT thermal noise) then your bandwidth will be something like the square root of the photon rate, or of order 10 kHz.

So with your 1 Watt laser diode, a 2 cm diameter zone plate at 10 cm focal length and a distance of 1 million kilometers, you should be able to pull off something very roughly like 1000 baud.

Caveats galore!

I've used a nominal wavelength of 1 micron. You'll probably use something between 800 nm and 1.5 microns.

Several factors of 2, $\pi$, and 1/2 and $1/\pi$ have been swept under the rug here, I'm just trying to get an order of magnitude data point

Please note that you can't always scale data rate vs distance the same way that you can for radio because of the way photons are first converted to carriers and then the photocurrent is detected in conventional receivers. If you use a fancier direct conversion receiver it gets better.

  • 1
    $\begingroup$ Since OPs interest is for use in the asteroid belt, 1000 baud will get a 1 Mbyte picture down in a not terrible 1/4 of an hour, but 1 million km is only several times the earth to moon distance, somewhat short of the hundreds of millions needed to reach asteroids. $\endgroup$ Commented Apr 7, 2023 at 0:19
  • 2
    $\begingroup$ @TheMatrixEquation-balance there's nothing in the question post as written about that, I just addressed what was asked. I'll go along with the necessity for some "nearby" 1 to 10 million km receivers, getting signals all the way to Earth direct is too hard. Also, it really isn't clear if for a 1 watt signal from a 1 kg spacecraft that optical at 1 um wavelength would be better than radio at say 1 cm or 1 mm wavelength from an engineering perspective. A lightweight (metal on thin kapton or mylar) expandable reflector might be easier to implement than an optical one. $\endgroup$
    – uhoh
    Commented Apr 7, 2023 at 0:44
  • 1
    $\begingroup$ @GremlinWranger as mentioned above I just addressed the question post as written. I won't do a full engineering solution, I've posted some links that address how to scale data rate with distance for optical vs radio. Folks are welcome to start there and dig in. I like the idea (mentioned in the OP's comment right after yours) of using much heftier "local" optical receivers "nearby" in the asteroid belt that connect these 1 kg spacecraft to Earth via relay, sort-of TDRS for asteroids. $\endgroup$
    – uhoh
    Commented Apr 7, 2023 at 0:48
  • 1
    $\begingroup$ @uhoh - This is not part of the question, but, do you think tunable lasers on micro-probes and corresponding wavelength filters on optical communications telescope would help to route the communications traffic? $\endgroup$ Commented Apr 7, 2023 at 13:46
  • 2
    $\begingroup$ @TheMatrixEquation-balance For example ALL of the GPS satellites transmit on the same frequencies! They don't have different channels. Instead, they have different codes and the receiver uses correlators to pull out the separate GPS satellites' signals from a single amplified signal on-chip. See for example 66 GPS channels for 22 satellites - why the factor of 3? $\endgroup$
    – uhoh
    Commented Apr 7, 2023 at 14:15

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.