Description
Astigmatism is an optical aberration that manifests when the optical system has two different focal planes along two perpendicular axes: rays passing through one meridian converge at one distance, while rays through the orthogonal meridian converge at a different distance.
The focal plane no longer exists as a single point but as two sagittal and tangential focal lines separated in depth, with a "circle of least confusion" between them.
At best focus, stars show a blurry cross or fine ellipse PSF, and their shape changes radically depending on the direction of defocus: elongated along one axis on one side of focus, elongated perpendicularly on the other side.
This is the absolute diagnostic signature of astigmatism and what distinguishes it from every other defect. It can be optical (polishing defect in a lens or mirror, pinched optic) or mechanical (stress on a poorly supported mirror, Schmidt plate tension, pressed secondary mirror), and affects poorly collimated or pinched Newtons as well as SCTs with a misaligned secondary, or refractors with a lens pinched in its cell or a poorly polished element.
Visual signature
At best focus, stars are not round but show a slight ellipse, sometimes invisible at the center of the field and more pronounced toward the edges.
The critical diagnostic test is done by deliberately defocusing: on one side of focus (for example intra-focal), stars become oval oriented in one direction (for example horizontal); on the other side (extra-focal), they become oval oriented perpendicularly (vertical).
This 90-degree rotation of the ellipse on either side of focus is the pathognomonic signature of astigmatism.
In a frame taken at exact focus, bright stars may show a faint cross pattern or fine ellipse.
At the eyepiece, the star test shows asymmetric diffraction with oval rings on one side of focus and rings rotated 90 degrees on the other.
In a stacked image, the signature is uniform across the whole field if the astigmatism is central optical, or more pronounced on one side if the origin is mechanical (local pinch).
More details in our article on astigmatism.
Differential diagnosis
To be distinguished from poor seeing (stars round and wide, no directional elongation, variable FWHM).
Not to be confused with coma (comet tails pointing toward the center of field, more pronounced at the periphery, no orthogonal rotation on defocus).
Different from incorrect backfocus (deformation at all 4 corners symmetrically, central stars sharp).
Not to be mixed with sensor tilt (asymmetry between opposing corners, no orthogonal elongation on defocus).
Different from tracking error (stars elongated in one direction, all oriented identically, no variation on defocus).
The defocused star test is the key examination: take two images equidistant intra- and extra-focal, astigmatism is present if and only if the ellipse has rotated 90 degrees between the two.
Also check whether this is a pinched lens from an over-tightened retaining ring (a frequent mechanical cause, to be distinguished from an intrinsic optical defect).
Probable causes
- Polishing defect in a lens or mirror (native astigmatism, persistent at any temperature)
- Lens pinched by an over-tightened retaining ring (mechanical stress)
- Newton primary mirror poorly supported: retention clips too tight (radial stresses)
- Poorly designed or failing mirror cell (3-point support insufficient on a large mirror)
- Schmidt plate tension on a poorly mounted SCT
- Misaligned or badly bonded secondary mirror
- Decentered lens from a drop, shock, or careless disassembly
- Adapter or ring applying mechanical stress to the focuser or cell
- Strong thermal differential between front and back face of a thick mirror (transient thermal stress)
- Thick filter inserted without clearance in its holder, pressing on the optical surface
Course of action
- Perform a star test on a bright star: take intra- and extra-focal frames equidistant from focus, compare ellipse rotation
- Check the tightness of the primary mirror retention screws (Newton): they should be in light contact, never tightened
- On the mirror cell, check for any hard point or uneven support
- For lenses, remove the retaining ring, clean, reassemble without over-tightening (leave ~0.1 mm of play)
- On SCT, NEVER touch the Schmidt plate retention screws without specific expertise
- If astigmatism appears only in cold weather: suspect thermal stress from a poorly designed cell
- Test by rotating the focuser 90 degrees: if the astigmatism rotates with it, the source is the focuser; if not, it is in the main optics
- For suspected intrinsic optical astigmatism, contact the manufacturer or an optician (Howie Glatter, Mike Lockwood) for assessment or refacing
- On an affected image, deconvolution with a measured PSF can marginally compensate, but does not restore a failing optical system
- Document the defect (intra/extra-focal images) before any intervention so you can compare after adjustment
The Doc's advice
Astigmatism is the most didactic aberration: the defocused star test gives you the diagnosis in 30 seconds, no software, no calculation. Put a star in the center, defocus by 5-10 mm in one direction then the other, and if the oval rotates 90 degrees between the two, that is astigmatism. For causes, always start with mechanical before optical: 80% of amateur astigmatism comes from a screw too tight somewhere, not a polishing defect. Back off your primary mirror clips a quarter turn and retest: there is a one-in-two chance you have fixed everything.
Think you can see this defect in your image?
Run a diagnosisFrequently asked questions
How do you distinguish optical astigmatism from mechanical astigmatism?
The test is temporal stability and the effect of rotating the tube. Optical astigmatism (polishing flaw, intrinsically defective lens) is constant, reproducible session after session, and rotates with the optics if you spin the tube. Mechanical astigmatism (pinch, thermal stress) can vary with temperature, tube orientation (gravity deforms the cell), or disappear after loosening a screw. Practically: remove the focuser and retest; if the astigmatism disappears, the constraint was in the focuser; if it persists but rotates when you rotate the camera, it is central optics; if it persists without rotation, it is somewhere downstream in the optical train (filter, corrector).
Can astigmatism be corrected in post-processing?
Only partially, and only for mild astigmatism. Deconvolution with a precisely measured PSF (DynamicPSF on unsaturated stars) can compensate weak astigmatism by reducing asymmetry in the restored PSF. BlurXTerminator also handles mild astigmatism better than earlier tools. However, for significant astigmatism (FWHM elongated >50% along one axis relative to the other), no algorithm can reconstruct the missing information: stars will remain visibly asymmetric. Hardware correction remains the only real solution.
My astigmatism only appears in cold weather. Is that normal?
This is a classic case of thermal stress. When temperature drops rapidly, mechanical components (metal cell, aluminum tube) contract faster than the glass mirror (different thermal expansion). A poorly designed cell then applies non-uniform pressure on the mirror, inducing transient astigmatism. Solutions: thermalize the instrument for 1-2 hours before the session to reach thermal equilibrium, actively ventilate the mirror (Newton) for homogenization, and verify that mirror supports conform to best practice (9- or 18-point cell for a large mirror, elastomer pads). A failing cell can be replaced with an amateur-built one (open Plop designs available online).
What level of astigmatism is acceptable before intervention?
It depends on the imaging goal. On a wide-field setup intended for astro-landscape work, slight corner astigmatism (eccentricity <0.5 as measured by SubframeSelector) may be acceptable because it is barely visible. On high-resolution galaxy or globular cluster work, where tightly packed stars must be point-like, the tolerance drops to eccentricity <0.3 across the whole field. Practical measurement in PixInsight: DynamicPSF measures the star axis ratio; a ratio >1.2 between major and minor axes indicates visible astigmatism. For demanding imagers, aim for <1.1.