Home / Learn / Science / Sky watching / Lesson 6.1
All lessons

Module 6: Astrophotography · Lesson 6.1 · about 25 min

Cameras, pixels and image scale

Image scale, in arcseconds per pixel, links your camera to your telescope. Match it to your seeing: about 2 to 3 pixels across a star's blur gives sharp stars without wasting light.

Key points, meanings and formulas on one page: choose "Save as PDF" in the print box.

Module 6 · about 25 minutes · read first: 4.3 Focal length, f-ratio and magnification, 3.3 Seeing and transparency · maths: small angles. You’ll be able to explain how a camera sensor records light, work out image scale and field, and judge whether a camera suits a telescope and your seeing.

Image scale, in arcseconds per pixel, links your camera to your telescope; matching it to your seeing, with about 2 to 3 pixels across a star's blur, gives sharp stars without wasting light.

1. Why this matters

Astrophotography is where amateurs push hardest and where professional methods start. Getting the camera and telescope matched is the first decision, and image scale also sets the precision of measurements: astrometry, photometry and double star measures all depend on it.

2. How a sensor records light

A camera sensor is a grid of pixels, each a tiny light bucket. Photons knock electrons loose in the silicon (the photoelectric effect); the camera counts them after the exposure. The fraction of photons that become electrons is the quantum efficiency (often 50% to 90% today, far above film’s few percent). Colour cameras put a mosaic of red, green and blue filters over the pixels; monochrome astronomy cameras take separate exposures through filters.

3. Image scale and sampling

The telescope’s focal length spreads the sky over the sensor. One pixel then covers a small patch of sky: the image scale.

Try it: match the camera to the telescope
image scale
pixels across the star
sampling
Each square is one pixel; its brightness is the starlight falling on it. Under about 1.5 pixels across the FWHM the star turns into a blocky square (undersampled: detail lost); over about 4 the light is spread thin and each pixel gets little signal (oversampled: fainter, noisier for the same exposure). The 2 to 3 pixel rule of thumb comes from the Nyquist sampling idea.

Things to notice:

  • A short focal length or big pixels make each pixel cover more sky: wide fields, but stars can collapse to single blocky pixels.
  • A long focal length with small pixels spreads each star over many pixels: fine detail if the air allows, but each pixel gets less light.
  • Aim for about 2 to 3 pixels across the star’s FWHM for your typical seeing (often 1″ to 2″ per pixel for deep-sky imaging).

Match a camera to three telescopes, for 3″ seeing:

The camera. 3.76 µm pixels. Seeing blurs stars to about 3″ wide.
On a 250 mm lens. s = 206.265 × 3.76 / 250 = 3.1″ per pixel: stars are one pixel wide and look blocky (under-sampled), but the field is huge.
On a 530 mm refractor. 1.46″ per pixel: about 2 pixels across each star. A good match.
On a 2,000 mm SCT. 0.39″ per pixel: about 8 pixels across each star. The light is spread thin for no extra detail (over-sampled).
Fix it. Bin 2 × 2 on the SCT (0.78″ per pixel) or add a focal reducer.
Maths and principle

Image scale for pixel size p (µm) and focal length F (mm):

s=206.265×pFarcseconds per pixel

(206,265 arcseconds in a radian; the 1,000 between µm and mm turns it into 206.265.)

Field of view along a sensor side of L mm: θ≈57.3∘×L/F (or exactly 2arctan⁡(L/2F)).

Sampling: with star FWHM ε (from seeing and optics), aim for

ε3≲s≲ε2

Fewer than about 1.5 pixels across the FWHM is undersampled; more than about 4 oversampled.

Try it: 4.63 µm pixels, 1,000 mm: s=206.265×4.63/1000=0.95′′/px. With 2.5″ seeing: 2.6 pixels across. Good.

s = 206.265 p / F
Image scale in arcseconds per pixel (p in µm, F in mm).
θ ≈ 57.3° L / F
Field of view along a sensor side of L mm.
ε / s ≈ 2 to 3
Good sampling: 2 to 3 pixels across the star blur ε.
Physical meaning
Pixels are the tiles of a mosaic. Tiles much bigger than the details lose them; tiles much smaller than the blur waste light, because each tile gets so little that noise wins. The sweet spot puts a few tiles across each blurred star.

“More megapixels always mean sharper astrophotos.” Sharpness is limited by seeing and optics. If your pixels already put 2 to 3 across each star, smaller pixels only spread the light thinner and add noise.

4. Worked example

A full-frame camera (36 × 24 mm, 6 µm pixels) on a 600 mm lens, typical seeing 2.5″.

  • Image scale: 206.265×6/600=2.06′′/px; stars span 2.5/2.06=1.2 pixels: slightly undersampled, so stars look a little blocky at 100% zoom, but fine on a screen.
  • Field: 57.3×36/600=3.4∘ by 57.3×24/600=2.3∘: the whole Andromeda Galaxy fits (Lesson 6.6).

Planetary imagers do the opposite: they use long focal lengths (often with a Barlow lens) to reach about 0.1″ to 0.2″ per pixel, because they catch moments of steady seeing in very short frames.

5. Practice

More practice (answers below):

  1. Image scale of 2.9 µm pixels at 250 mm?
  2. Which focal length gives 1.0″/px with 3.76 µm pixels?
  3. Why might a camera good for wide fields be poor for planets?
Answers
  1. 206.265×2.9/250=2.39′′/px.
  2. F=206.265×3.76/1.0=776 mm.
  3. At a wide-field image scale, a planet only a few tens of arcseconds across covers just a few pixels.
  • Image scale s = 206.265 × pixel size (µm) / focal length (mm).
  • Aim for 2 to 3 pixels across the seeing blur.
  • Field of view ≈ 57.3° × sensor size / focal length.
  • Binning or reducers fix over-sampling.

Try it with real numbers: work out image scale, field and sampling for your camera and telescope with the telescope calculator, and measure star FWHM in your own images with the star FWHM checker.

New words: pixel, quantum efficiency, image scale, sampling, undersampled, oversampled, Barlow lens.

Check yourself

1A camera with 3.76 µm pixels on a 400 mm telescope has what image scale?
2Your seeing is 3″ and your image scale 4″/px. What is the problem?
3What does a camera pixel measure?

Sources

  • Howell, S. B. (2006), Handbook of CCD Astronomy, 2nd ed., Cambridge University Press (the CCD equation, calibration frames)
  • Berry, R. and Burnell, J. (2005), The Handbook of Astronomical Image Processing, 2nd ed., Willmann-Bell

Your ticks stay in this browser. Save them as a file on the course home.

Updated 2026-11-22. Numbers in the examples are made up to show the idea unless a source is given. Simulations run in your browser on made-up data; nothing you type is sent anywhere. Astronomy only: nothing here is astrology.