There is a short Wikipedia article Focal Plane Arrays that enumerates some projects, but my question is more along the lines of what is (at least) nearly complete or in "first light" phase, even if not commissioned yet.

I'd like to differentiate between focal plane arrays (an array of multiple feeds and amplifiers) used in single dish telescopes, and those integrated into multiple dish arrays, because I'm particularly curious about single dishes being used for spatial information, or even true imaging. Ideally, the answer will give some information for each case.

In the case of single dish instruments, are the elements - roughly speaking - used as pixels? Despite the longer wavelength, it's still optics and it is a telescope. If there are N individual, uncoupled feeds, does one build up an image roughly N times faster? Is the relative phase between the feeds ever used (for single dish instruments)?

For those like me who aren't already familiar with focal plane arrays, here is a random picture from one of the links I found in a quick internet search. It's from the Parkes 21cm Multibeam Receiver, has (had?) 13 receivers and sat at the focus of the 64m dish. The photo is dated 1997 - I have a hunch there's been some development in this technique in the intervening 20 years.

Is this actually the only one?

Parkes Radio Telescope 21cm Focal Plane Array in 1997

Parkes Radio Telescope 21cm Focal Plane Array in 1997

Edit: The Parkes array is still in use as shown below:

enter image description here

Above: Superposition of the half-power beam widths of the 13-element array of the Parkes 21-cm Mutlibeam Receiver, as used in [a study of a Fast Radio Bursters]http://arxiv.org/abs/1602.07477.

It's likely the image is a screen shot from from The host galaxy of a fast radio burst Nature volume 530, pages 453–456 (25 February 2016), Keane et al. I can't find my archived copy now, but instead see Phys.org's New fast radio burst discovery finds 'missing matter' in the universe

where the caption there reads:

This image shows the field of view of the Parkes radio telescope on the left. On the right are successive zoom-ins in on the area where the signal came from (cyan circular region). The image at the bottom right shows the Subaru image of the FRB galaxy, with the superimposed elliptical regions showing the location of the fading 6-day afterglow seen with ATCA. Image Credit: D. Kaplan (UWM), E. F. Keane (SKAO).

  • $\begingroup$ I've asked a somewhat related question. I got started on this after reading this good question which might benefit from an aditional answer. $\endgroup$
    – uhoh
    Commented Jun 15, 2016 at 2:31
  • $\begingroup$ This is essentially a 13-pixel camera, but since the pixels are small a double exposure can provide a tiny "image" with 26 individual pixels. Thus the astrophotography tag. $\endgroup$
    – uhoh
    Commented Mar 18, 2017 at 6:20
  • $\begingroup$ Slightly off-topic, why is "Subaru" listed a few times in that picture? $\endgroup$ Commented Oct 22, 2018 at 13:03
  • $\begingroup$ @MagicOctopusUrn those are optical images, and there are also optical telescopes with the names WISE and Subaru so I'm going to guess that those are the sources of the images. $\endgroup$
    – uhoh
    Commented Oct 22, 2018 at 13:19
  • 1
    $\begingroup$ I wasn't familiar with either of those telescopes, and they weren't linked or mentioned, thanks for the links; reading them now. $\endgroup$ Commented Oct 22, 2018 at 13:20

1 Answer 1


I am not a professional astronomer, so take this answer with a grain of salt, but just from visiting a few facilities, I know some.

For single dish applications, the 100m telescope in Effelsberg, Germany, uses a 7-beam receiver -- fewer than the Parkes array you mention, and I dont't know of any single dish setup with a larger number of beams.

Regarding arrays, the Westerbork Synthesis Array in the Netherlands uses the APERTIF arrays in most of its individual 25m dishes. With 121 elements per array, this seems to be on the higher-end side of granularity.


From astron.nl/dailyimage for 31-01-2017 First image with Apertif: a new life for the Westerbork radio telescope

The first images made with the upgraded telescope that demonstrate this new 'wide-angle' capability is shown here. The first image shows the dwarf galaxy Leo T. The image is colour-coded and shows the gas (in blue) in this galaxy together with many distant radio galaxies in the background shown in orange. For comparison, the field of view of the previous Westerbork system and the size of the full moon are also indicated.

To make this new capability possible, ASTRON developed and built the hardware in-house. 121 small receivers are used in each telescope whose signals are combined electronically to produce the large field of view.

The upgraded telescope will also be used to search for and study new variable sources in the radio sky. With the new Apertif receivers, observations of large parts of the sky can be done much faster, and projects that used to be impossible as they would take tens of years can now be done in a much shorter time. Westerbork is therefore poised to make many new discoveries in the radio sky.

Apertif first image

  • $\begingroup$ This is great, thank you! This is quite a large focal plane array for a radio telescope. If you are interested, have a look at Why are the ALMA Correlator's ADCs only 3-bits? as well. $\endgroup$
    – uhoh
    Commented Dec 18, 2017 at 15:15
  • $\begingroup$ I hope you don't mind; I just happened to run across this image today and I remembered your answer, so I added APERTIF's first image. Thanks again for your help! $\endgroup$
    – uhoh
    Commented Oct 19, 2018 at 16:03
  • $\begingroup$ The comments about Apertif are not correct. Apertif is a phased array feed, not a focal plane array. While it does have 121 receiving elements, there are electronically combined to form "beams". The number of beams will be significantly less. Note the ASKAP interferometer also has PAFs. These have 188 elements forming 36 (dual polarisation) beams on the sky. In the ASKAP case, the 36 beams are equivalent to the Parkes 13 beams. Parkes is currently building a 72 dual polarisation beam receiver. $\endgroup$
    – Chris
    Commented May 27, 2021 at 10:30
  • $\begingroup$ @Chris Thanks for the heads-up! I'll take a bit of time to review this and will likely delete this answer if I can confirm your remark. $\endgroup$
    – jstarek
    Commented May 28, 2021 at 11:57
  • $\begingroup$ @jatarek: I would suggest you edit/update rather than delete your answer - while the question asks about focal plane arrays, PAFs are actually an equivalent from the perspective of what the question is trying to ask. $\endgroup$
    – Chris
    Commented May 30, 2021 at 23:50

You must log in to answer this question.

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