Reflector telescopes (normally?) have a parabolic primary mirror that reflects the light to a secondary mirror which is placed at the center of the telescope. The construction holding that mirror and the mirror itself leave artifacts in the acquired image especially in images with high contrast (as can prominently be seen in the "first" JWST image) JWST Telescope alignment image

Now, since the mirror is parabolic anyway, would it be possible to use a part of the parabola that is shifted to one side so that the secondary mirror (and the construction holding it) could be placed outside of the field of view of the primary mirror? I guess, the mirror would have to be elongated to give a round FOV and thus require more material.

I'm 100% sure that many people much smarter than me have had this idea (and it seems to be used on satellite dishes around the world) but have not seen an optical telescope employing this system.

Can you tell me what the downsides of such a system are and why it is not used? Is it just not worth the more complicated shape and bigger, oblong mirror required?


2 Answers 2


If your secondary is off-axis, then your primary has to be off-axis as well. That is, the primary will have to be a parabolic section rather than a pure spherical section. Not only is that a ton more difficult to manufacture, but you'll get some coma in the image as a result.

There are some off-axis telescopes in existence built this way, but in general there are better ways to deal with spider diffraction. The patterns can be removed with software since the 'scope designer knows the exact pattern that the physical structure generates. To some extent specific "spiral-like" spiders will generate much less diffraction spiking.

  • $\begingroup$ Cool. Thank you for this answer. Is the coma due to the asymetric nature of such a mirror (I amagine a full parabolic mirror as kind of a "cup" and this mirror would in my mind just be round or oblong section of that cup that is not in the center). $\endgroup$
    – kruemi
    Commented May 2, 2022 at 14:17
  • $\begingroup$ Also, did you mean by "spherical section" just a circular section? $\endgroup$
    – kruemi
    Commented May 2, 2022 at 14:19
  • 1
    $\begingroup$ Oh, and thank you for using the term "off-axis"! Really! I'm now deep diving on the topic since thanks to you I can now find lieterature... Awesome! $\endgroup$
    – kruemi
    Commented May 2, 2022 at 14:35
  • $\begingroup$ Are you suggesting that high-end telescopes like the JWST are spherical? $\endgroup$ Commented May 3, 2022 at 15:25
  • 1
    $\begingroup$ @Peter-ReinstateMonica I should have written "rotationally symmetric." However, even for designs incorporating a spherical primary, if you go off-axis instead, I'm pretty sure you have to go for a conic section (parabola usually) $\endgroup$ Commented May 3, 2022 at 16:59

To supplement @CarlWitthoft's insightful answer I'll mention some links related to off-axis operation:

The last of which says:

The Nancy Grace Roman Space Telescope, formerly named the Wide Field Infrared Survey Telescope (WFIRST), employs a folded three-mirror anastigmat featuring an ellipsoidal primary, hyperboloidal secondary, and ellipsoidal tertiary.9 An earlier design used an off-axis three-mirror anastigmat.10

9Pasquale, Bert A.; et al. (17 September 2018). "Optical Design and Predicted Performance of the WFIRST Phase-B Imaging Optics Assembly and Wide Field Instrument". Conference Proceedings of the SPIE. 107450K. doi:10.1117/12.2325859

10Content, D.A.; Goullioud, R.; Lehan, J.P.; Mentzell, J.E. (14 September 2011). "Optical design trade study for the Wide Field Infrared Survey Telescope [WFIRST]" (PDF). Conference Proceedings of the SPIE. 8146. doi:10.1117/12.898528.

So in an alternate timeline the Nancy Grace Roman Space Telescope might have been exactly as you propose, without any secondary obstruction!

We can certainly point to the GAIA space telescope as a completely unobstructed view off-axis telescope, and this is likely of absolute necessity! In order to make extremely precise astrometric measurements of millions of relative star positions some of which are in dense neighborhoods, it's important to have very very clean point spread functions without zillions of overlapping diffraction spikes.

From this answer to Why does the Gaia space telescope have two main mirrors:

GAIA optical bench and dual primary mirrors

above: GAIA's Silicon Carbide Optical Bench, with the two objective mirrors of it's twin telescopes pointing 106.5° apart. From Spaceflight 101, image credit: ESA/Astrium.

Some examples of three mirror anastigmat-type layouts:

"Design and development of a freeform active mirror for an astronomy application" http://dx.doi.org/10.1117/1.OE.53.3.031311

(a) Layout of a F4 Three-Mirror Anastigmat (TMA) design that resembles the Cook design with the freeform primary, spherical secondary, and tertiary. It has a field-of-view (FoV) of 8 deg ×12 deg (the initial design with three conical sections provides 8 deg ×8 deg FoV). (b) Layout of the freeform based two-mirror system (10 deg ×10 deg).

I don't know about instrumentation attached to space telescopes (some drawings of systems aboard Hubble in this answer to Where exactly is the modification that first corrected the spherical aberration in Hubble's primary mirror? include on-axis internal "telescopes" which change divergence and magnification) but spectrometers will often use off-axis transfer optics. Here's an extreme example from Why does X-shooter use double passes through prisms for Echelle cross-dispersion instead of gratings? which references X-shooter, the new wide band intermediate resolution spectrograph at the ESO Very Large Telescope:

Figure 6. shows the NIR spectrograph. "Double pass" means the light passes through the prisms going towards the Echelle and again after being back-diffracted. That's twelve additional optical surfaces!

Figure 6. The NIR spectrograph optical layout.

Figure 6. The NIR spectrograph optical layout.

See also Sentinel-2: Multispectral Instrument (MSI) design and system performance (also here):

from Sentinel-2: Multispectral Instrument (MSI) design and system performance "Layout of the Three-Mirror Anastigmat (TMA) telecentric mirrors and dichroic beam splitters for VNIR and SWIR spectral bands."

Layout of the Three-Mirror Anastigmat (TMA) telecentric mirrors and dichroic beam splitters for VNIR and SWIR spectral bands.

  • 1
    $\begingroup$ Wow, thank you for that great answer with all the references. So it seems that it's mainly practical reasons why Primary and Secondary are on axis in most setups. It does not pos a problem in most cases and is just not worth the effort. While as far as I can see fom instrument schematics, inside instruments off-axis mirrors seem the be the norm? $\endgroup$
    – kruemi
    Commented May 3, 2022 at 7:04
  • 1
    $\begingroup$ Yes, I think Carl's answer is complete and this is just some additional background. The off-axis optics mirrors are much more difficult to make but it's a lot easier when they are physically small. $\endgroup$
    – uhoh
    Commented May 3, 2022 at 9:30

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .