I was reading about telescopes and the Hubble Telescope for example has a 2.4m mirror which reflects lights to a sensor. Other type of telescopes use lenses to focus light to the imaging sensor.

I was wondering, is it possible to have a telescope without a mirror or lens? So in Hubble's case, instead of having a 2.4m mirror reflecting light to a small sensor, why not have a big imaging sensor (same size as the mirror - 2.4m). Would this type of telescope have similar capabilities as the mirror one?

I know that we use mirrors because it's way cheaper and easier than building large sensors but I'm curious if a mirror-less telescope would be better / worse or just the same.

  • $\begingroup$ Even if we use a sensor how could you confirm there is no lens or mirror or lens within the device $\endgroup$ – Utkarsh futous Apr 1 '17 at 3:37
  • 2
    $\begingroup$ if you do not have a mirror or a lens, how do you project the image into the sensor? Grab a sensor and put it in front of you, do you think it will capture an image of you? $\endgroup$ – user126422 Apr 1 '17 at 3:37
  • $\begingroup$ Why do you absolutely have to project the image into the sensor? Even without projecting anything, photons will still hit the sensor, and it will register those. $\endgroup$ – Pema Apr 1 '17 at 4:36
  • $\begingroup$ There will be no image, photons from everywhere will hit a given pixel, not just from a specific point in your face. $\endgroup$ – user126422 Apr 1 '17 at 5:45
  • $\begingroup$ What you are describing is the world's most expensive photocell: a device which will tell you the general light level in front of it. Such things are useful, for instance, in camera light meters, but not as cameras -- indeed in many cameras the aperture in the lens, even wide open, is much smaller than the sensor area: I have one where the ratio is about 20. $\endgroup$ – tfb Apr 1 '17 at 6:23

Systems other than mirrors can provide telescope function.

There's four other ways to make a telescope.

  1. Like Galileo did, we can use lenses instead of mirrors (this does not work well at large scale).

  2. One can employ diffraction structures (holograms, or gratings, or zone plates), but there are significant problems in doing this for a broad range of wavelengths.

  3. More primitive, is the pinhole camera, and variations using multiple holes, so-called "coded apertures"; that is how some X-ray imaging is done, but it takes a lot of work.

  4. Last, is to find a black hole or neutron star, and take advantage of the gravitational bending of light in its vicinity. That's more of a discovery than a construction project, but several useful systems have come into view.Hubble gravity lens


A phase(d) array antenna is an imaging device without either a mirror or a lens. Of course, it needs a reference oscillator that is phase coherent with the incoming wave so when the two are mixed in the array, sensor element by sensor element, only the I and Q ("inphase" and "quadrature") components of the incoming wave are processed. I admit, this is conceptually easier for a radar antenna that generates its own illumination but I can imagine that, for example, by some magic if you could delay the incoming wave illuminating the array but ahead capture its carrier to phase lock your own oscillator then you could play the same game locally...


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.