Description
Sensor tilt occurs when the sensor plane is not exactly perpendicular to the optical axis: one side of the sensor is closer to the focal plane than the other, so focus is not uniform across the field.
The signature is asymmetric along a diagonal: one or two corners remain sharp and pinpoint while the opposite corners show stretched or soft stars. The defect worsens toward the edges in a preferred direction.
The origin is mechanical: adapters or rings that are loose or misaligned, a focuser flexing under the camera's weight, a sensor mounted off-axis inside the camera, or a top-heavy imaging train that tilts.
This is a correctable defect using a tilt plate. It must be distinguished from coma or an incorrect backfocus (symmetric degradation in all four corners) and from a collimation error (directional but uniform deformation across the field).
Visual signature
Sharpness is unequal from corner to corner: typically one corner (or two diagonally opposite corners) is sharp and pinpoint, while the diagonally opposite corner shows elongated or bloated stars.
The defect follows a diagonal focus gradient across the field: a line can be found where stars are sharp, with degradation increasing on either side.
The center remains relatively sharp; it is the asymmetry between opposite corners that diagnoses tilt, as opposed to a symmetric defect (coma, backfocus) that degrades all four corners equally.
When refocusing, the best focus position differs depending on which corner is examined, and no single global focus setting renders all four corners sharp simultaneously.
Differential diagnosis
Do not confuse with coma or incorrect backfocus: those defects degrade all four corners symmetrically, whereas tilt leaves one corner sharp and the opposite corner degraded (diagonal asymmetry).
Distinguish from a collimation error: miscollimation produces a directional but uniform deformation across the whole field, while tilt creates a sharpness gradient between opposite corners.
Separate from astigmatism: astigmatism stretches stars in two axes depending on focus position, regardless of corner, without the sharp/soft diagonal gradient that characterizes tilt.
Confirmation: if no global focus setting can render all four corners sharp at the same time, the defect is tilt.
Probable causes
- Adapters or extension rings loose or mounted off-square
- Focuser flexing under the weight of the camera
- Sensor mounted non-perpendicularly inside the camera body
- Top-heavy imaging train drooping under its own weight
- Mechanical play anywhere in the imaging train
- Incorrect backfocus mimicking or compounding the tilt
Course of action
- Check and tighten all rings, ensuring they sit squarely and are fully engaged
- Eliminate focuser play (add a brace, upgrade to a more rigid focuser)
- Adjust tilt using a dedicated tilt plate (push-pull screws) in small increments
- Use a measurement tool (ASTAP, CCDInspector, NINA) to pinpoint which corner needs correction
- Lighten or support an overly heavy imaging train
- Verify and set correct backfocus before attributing the defect to tilt
- Take a test frame after each adjustment to confirm progress
The Doc's advice
With tilt, hunt down the obvious mechanical culprits before touching any fine-adjustment screws. Start by ruling out the basics: rings tightened squarely and fully engaged, no play in the focuser, imaging train not so heavy it causes the focuser to droop. A large proportion of apparent tilt cases turn out to be nothing more than a poorly mounted ring. Only then should you tackle fine correction with a push-pull tilt plate, guided by a measurement tool (ASTAP, CCDInspector, NINA's curvature sensors) that tells you which corner to raise. Work in small increments, one test frame between each step. And check your backfocus first -- a wrong spacing can mimic or amplify tilt.
Think you can see this defect in your image?
Run a diagnosisFrequently asked questions
How do I know if I have sensor tilt?
The typical symptom is unequal sharpness between opposite corners: one corner (or two diagonally) shows pinpoint stars while the opposite corner shows elongated or soft stars, with the center remaining acceptable. The decisive test: no global focus adjustment can render all four corners sharp at the same time, because the sensor plane is not perpendicular to the optical axis. Tools such as ASTAP, CCDInspector, or NINA's curvature measurement features measure and quantify tilt corner by corner.
Tilt or backfocus error: how do I tell them apart?
By symmetry. An incorrect backfocus (wrong corrector-to-sensor spacing) degrades all four corners in the same way, symmetrically around the center. Tilt is asymmetric: one corner is sharp and the diagonally opposite corner is degraded. If all four corners are bad but equally so, look at backfocus first; if degradation follows a diagonal (one side good, the other bad), it is tilt. Always set backfocus correctly first, because a wrong spacing can mimic or amplify tilt.
How do I correct sensor tilt?
Work through it in order. First, eliminate the obvious mechanical issues: tighten all rings squarely, remove any focuser play, and make sure the imaging train is not heavy enough to cause flexure. Many tilt problems are nothing more than a poorly mounted ring. Only after that should you use a push-pull tilt plate to tilt the sensor slightly in the right direction, guided by a measurement tool and with a test frame between each small adjustment. Patience is key: correct in fine increments.
What tool should I use to measure tilt?
Several options give an objective corner-by-corner measurement. ASTAP provides a tilt and curvature analysis from a single frame. CCDInspector (commercial) has historically been the reference tool for mapping FWHM across the field. In NINA, the aberration measurement tool and curvature plugins (Hocus Focus, other plugins) display sharpness by zone. All work on the same principle: measuring star size in each corner to identify which corner needs to be raised. This avoids blind adjustments and dramatically speeds up convergence.