Description
Field rotation is a distortion where stars trace circular arcs centered on a single point in the image: the field slowly rotates around that point during a single exposure or across the session.
Two main origins. On an azimuthal (alt-az) mount without a field rotator, the field rotates by design during tracking, faster as the target gets higher. On an equatorial mount, an imperfect polar alignment introduces a slow residual rotation.
The signature is unambiguous: stars are point-like at the rotation center and increasingly arced as one moves away from it, with the arc length growing with radius and duration.
This is a mechanical and geometric defect. Distinct from tracking drift (identical trails everywhere, with no rotation center).
Visual signature
Stars form concentric arcs around a single point in the image (the rotation center), like miniature star trail traces.
The arc amplitude is zero at the rotation center and grows with distance: stars are point-like near the center, with progressively longer arcs toward the far edges.
The effect increases with time: visible on a long single exposure, or accumulated across a session stacked without rotation-based registration.
On an alt-az mount, the rotation rate depends on the target's position in the sky; on an equatorial mount, it betrays a polar alignment error.
Differential diagnosis
Do not confuse with tracking drift: drift produces identical trails in length and direction everywhere, whereas field rotation produces arcs whose size and orientation depend on position (zero at the rotation center).
Distinct from periodic error: periodic error modulates elongation cyclically on the RA axis, with no rotation center.
Separate from guiding oscillations: those produce uniform zigzag stars, not arcs organized around a point.
Confirmation: if you identify a point in the field where stars remain point-like while they arc around it, it is field rotation.
Do not confuse with a stacking misalignment: if individual raws are sharp but stars double only after stacking, the problem is registration (a rotation between frames not compensated for can indeed be the cause). Also distinct from a simple framing issue, where the object is poorly centered without star distortion.
Probable causes
- Alt-az mount without a field rotator
- Imperfect polar alignment on an equatorial mount
- Target high in the sky (fast rotation on alt-az)
- Long exposures accumulating rotation
- Field rotator absent, poorly calibrated, or at the limit of its travel
- Stacking a long session without rotation-based registration
Course of action
- On alt-az, install a motorized field rotator
- On equatorial, improve polar alignment (drift align, SharpCap)
- Shorten exposures to limit accumulated rotation
- Stack with rotation-based registration (adapted alignment)
- Prefer an equatorial mount for long-exposure deep sky
- Calibrate and verify the travel of the field rotator if present
- Avoid targets too high on alt-az (maximum rotation at the zenith)
The Doc's advice
Field rotation: first understand where it comes from. On an alt-az mount (motorized Dobsonian, AZ-GTi in azimuthal mode), it is mechanical and unavoidable without a field rotator: either install one, or expose short and stack with rotation-based registration. On an equatorial mount, field rotation signals a polar alignment to redo: redo your alignment properly (drift align or SharpCap) and it disappears. The trap is mistaking it for a guiding problem: guiding holds the center but does not correct rotation around that center. For deep sky, a well-aligned equatorial mount remains the solution.
Think you can see this defect in your image?
Run a diagnosisFrequently asked questions
Can deep-sky imaging be done with an alt-az mount?
Yes, but with limits. Without a field rotator, field rotation appears on long exposures: the solution is to expose short (a few tens of seconds) and stack many subs with software that performs rotation-based registration. Good results can be obtained on bright targets. For long exposures on faint subjects, a motorized field rotator is better, or switching to an equatorial mount. Alt-az remains excellent for visual and planetary use, where rotation is not an issue.
Why do I have arced stars even though I am using an equatorial mount?
Because your polar alignment is imperfect. When the mount axis is not exactly aligned on the celestial pole, a slow residual field rotation remains in addition to drift. The larger the alignment error and the longer the exposure, the more pronounced the arcs. The solution is to redo a careful polar alignment (SharpCap Polar Align or drift align). Guiding does not correct this rotation: it holds the center, but the field rotates around it.
What is a field rotator?
It is a motorized accessory placed between the telescope and the camera that rotates the sensor in exactly the opposite direction of the field rotation induced by an azimuthal mount. It thus keeps the orientation of the sky fixed on the sensor throughout the exposure. Essential for long-exposure deep-sky imaging on an alt-az mount, it must be calibrated for the target's position. On a correctly polar-aligned equatorial mount, it is unnecessary, as the tracking axis already compensates for the rotation.
Field rotation or tracking drift: how to tell them apart?
Look at the geometry of the trails. Field rotation draws concentric arcs around one point in the field: stars are point-like at that center and increasingly arced moving away from it. Tracking drift, by contrast, gives straight trails of identical length and direction across the whole field, with no center. If you find a sharp point surrounded by stars that arc around it, it is field rotation; if everything trails uniformly in one direction, it is drift.