Why the focuser knob cannot always help

An eyepiece shows a sharp star, but attaching a camera leaves a large bright spot. The focuser knob has already reached its stop. Does that mean the camera or telescope is broken?

A sharp image requires the relevant optical planes to coincide physically. A camera without its own lens must place its sensor where the telescope forms an image. An eyepiece has another task: adapting the beams for the eye. If mechanical travel cannot position the parts correctly, turning the knob alone cannot help. Celestron: reaching the focal plane.

Let us follow light through a simple lens, then an eyepiece and camera. The numbers and diagrams are teaching examples: they explain how to check distances, rather than describe a physically tested set of parts.

Rays from one tip must come together

Imagine an arrow in front of a converging lens. Its tip sends light to different parts of the lens. For a sharp image, these different rays must meet near one corresponding point beyond it. Another object point produces another image point. This forms a whole picture, rather than one common focus for every detail.

A real image can be intercepted by a screen or sensor: light actually passes through its location. In the diagram, the object is farther from the lens than its focal length. Its image is smaller and inverted: rays from the tip arrive below the axis. OpenStax: constructing a lens image.

Rays from one tip form one image
ObjectLensImage−F+Fu = 300 mmv = 150 mm

Teaching thin lens: f = 100 mm, u = 300 mm, v = 150 mm, m = −0.5. Longitudinal distances share one scale; transverse sizes are enlarged. The lens curve is a symbol, not a surface prescription. Diffraction and aberrations are omitted.

OpenStax · 2.4

The three lines are convenient representatives of the whole beam. One arrives parallel to the axis and passes through the rear focus after the lens; a second crosses the centre of the thin lens without changing slope; a third passes through the front focus and leaves parallel to the axis. Light arrows always point from object to image. This is a paraxial approximation, not a trace through two real curved glass surfaces.

Inversion alone is not a focusing defect. A camera image can be rotated; in visual observation, orientation depends on the entire optical train. A mirror-reversed image cannot be corrected by rotation alone.

A distant object and a nearby target need different positions

A very distant point sends an almost parallel beam across a small aperture; a nearby point sends a noticeably diverging one. A camera focused on a star therefore need not sharply image a nearby object without a change in sensor position.

For a distant source, the image lies near the rear focus. Neighbouring sky directions correspond to neighbouring points in the focal plane. In a real wide-field system, the best image surface may be curved; a “plane” is an approximation near the axis or the result of deliberate correction.

As an object approaches a converging lens from a large distance, its real image moves farther behind the lens. For a hypothetical 1000 mm focal length and an object 20 m away, the result is about 1052.63 mm: 52.63 mm beyond the infinity-focus position. That change may exceed the whole focuser travel, even though the instrument works normally with stars.

Branch · The thin lensOptional · you can skip this

Signed distances: 1/f = 1/u + 1/v

Choose one convention: light travels left to right; a real object before the lens has u > 0; a converging lens has f > 0; a real image beyond it has v > 0, and a virtual image before it has v < 0. Then

(1)/(f) = (1)/(u) + (1)/(v), m = − (v)/(u).

u,v,f are distances in the same units, and m is dimensionless linear magnification. Its sign describes arrow orientation. For the diagram, f = 100 mm and u = 300 mm:

v = (fu)/(u − f) = 150 mm, m = − 0.5.

Unit check: mm²/mm gives mm. As u→∞, v→f. But if u = 50 mm with the same f = 100 mm, we obtain v = − 100 mm and m = + 2: rays beyond the lens diverge, and their backward extensions define an enlarged upright virtual image before it. A bare screen cannot catch it; an eye or additional camera optics can turn these beams into a real image. Evident: real and virtual images, and principal planes.

For a target close to the front focus, the denominator u − f is small, so the image requires a large distance. At u = f, output is parallel and no finite v exists. The model assumes a thin lens in air, small angles and proximity to the axis; it does not calculate aberrations, diffraction or a compound objective across its whole field.

Where should focal length be measured from?

In the teaching thin lens, all refraction is notionally placed at its centre. A real objective has thickness and several elements; a reflecting system has a folded path. A “1000 mm focal length” label therefore does not require a tube exactly one metre long.

To describe a compound system near the axis, we introduce principal planes, equivalent optical reference planes. Focal lengths and the corresponding object and image distances are measured from them. They need not coincide with the front edge of glass, a cell rim or an accessible mark on the housing. The mechanical distance from a rear mounting surface to the sensor is another quantity, specified by the particular instrument’s drawing.

First establish which surface the manufacturer uses as a reference. Otherwise, the correct number measured from the wrong place produces an incorrect camera position. Our thin-lens drawing merges the principal planes; it cannot serve unchanged as an assembly drawing for a multi-element objective.

The sensor occupies the plane; the eyepiece works with it

In prime-focus imaging, a camera without its own lens places the light-sensitive sensor at the telescope’s real image. No eyepiece is needed. If the sensor lies before or behind the required plane, rays from one star cover a larger region and form a defocused spot.

Visual observation differs. In the illustrated positive-eyepiece arrangement, the eyepiece sits beyond the intermediate image and acts as a magnifier. When the objective’s rear focal plane coincides with the eyepiece’s front focal plane, each image point produces a separate, nearly parallel outgoing beam. The eye can view it while focused at a distant scene; the eye’s own optics form a real image on the retina. OpenStax: eyepieces and parallel output.

One focus, two ways to obtain a result

Camera without its own lens: sensor at F

FSensor

Eyepiece and eye: parallel output beams

FEyepieceExit pupil

The objective is off-frame to the left. Equivalent thin elements: fₒ = 1000 mm, fₑ = 25 mm; the eyepiece is at x = 1025 mm. One source direction, paraxial model. The pupil line is an image of the entrance, not another star-image plane; the eye is symbolic. Transverse scale is enlarged.

OpenStax · 2.8 · Nikon

That is why the position “eye behind the eyepiece” cannot simply be given to a bare sensor. A phone taking pictures through an eyepiece with its own lens uses a different, afocal arrangement: the phone lens supplies the subsequent focusing optics. Prime-focus imaging, imaging through an eyepiece and eyepiece projection require different distances; one mounting adapter does not make them equivalent.

Where to place the eye, and how much sky is visible

Move your eye sideways or along the axis behind an eyepiece, and part of the field may darken or vanish. The star’s focus is not necessarily responsible. The eyepiece also forms an image of the entrance aperture: the exit pupil. The eye’s pupil should be near the appropriate location to receive beams across the field.

Eye relief is the distance from the outer surface of the eyepiece’s last lens to that location. It is neither the eyepiece focal length nor the distance to the intermediate image. A real multi-element eyepiece determines it through its optical design. Nikon: eye relief.

Output-beam diameter matters too: if it exceeds the eye’s actual pupil, some light does not enter. There is no universal size for every reader and lighting condition. True field of view is the angular patch of sky; apparent eyepiece field describes how wide the viewing window appears. They are different quantities. Nikon: pupils and two field definitions.

For a camera, field also depends on sensor dimensions and image scale, and at the edges on useful optics and beam restrictions by mounts. A sharp central star does not guarantee sharpness and illumination across a large sensor.

Branch · Magnification and pupilOptional · you can skip this

25 mm on the eyepiece and 1.25 inches on its barrel mean different things

For two equivalent thin elements in the illustrated infinity-focus setting, angular magnification is

M = − (f_(o))/(f_(e)).

f_(o) is objective focal length and f_(e) eyepiece focal length in the same units. The minus sign indicates inversion of directions in this straight-through arrangement; it does not include an added diagonal or prisms. With f_(o) = 1000 mm and f_(e) = 25 mm, M = − 40. This is angular magnification, rather than the linear m of the nearby arrow.

With aperture D = 100 mm, the focal ratio is N_(f) = f_(o)/D = 10, or f/10. This is a geometrical ratio for a distant object, not a promise of a particular SNR. Edmund Optics: defining f-number. The simple afocal model gives exit-pupil diameter

p = (D)/(∣M∣) = (f_(e))/(N_(f)) = 2.5 mm.

Halving f_(e) doubles the magnitude of magnification and halves pupil diameter. It does not double objective resolution. The physical eyepiece mounting diameter does not appear in these equations.

Diagonals, mounting sizes and sensor distance

The focuser supports the eyepiece or camera and allows axial motion. Travel length matters, alongside tilt, play, load support and positional repeatability. Other instruments focus by moving the primary mirror; the next calculation describes fixed optics and a moving rear assembly.

A diagonal folds the light path for a convenient viewing position and adds path between objective and eyepiece. A mirror or ordinary prism diagonal and an image-erecting prism are different optical systems. Some configurations retain mirror reversal; orientation also depends on tube rotation and eye position. Alignment should not be judged solely by whether the image’s “top” matches the object’s top. Celestron: orientation with a diagonal.

1.25″ and 2″ usually describe nominal insertion-barrel diameters—31.75 and 50.8 mm—rather than focal length. A 25 mm eyepiece may have a 1.25″ barrel. An adapter must match dimensions and connection type; equal outside diameter does not establish equal threads, length or optical-plane position. Celestron: mounting diameters.

The camera’s sensor is inside its housing. Distances must reach the sensor surface, rather than the front window or outer cover. A reducer or field flattener also has a separate specified distance from a manufacturer-defined reference surface to the sensor. Focuser adjustment may sharpen the centre while leaving the wrong corrector spacing, spoiling the edges. “55 mm” is not universal across systems. Celestron: reducer working distance and compatibility.

Reachability can be checked before trying to focus

Take a hypothetical system without diagonal or corrector. The reference is a fixed rear mounting surface with the focuser fully retracted. The distant-object image lies 60 mm behind that reference. The focuser can extend 30 mm. Adapters and the camera’s internal sensor depth lie between its moving mounting surface and the sensor.

Does sensor travel cross the required plane?
FABCD0306090

Sensor position from the fixed reference, mm

Case A
L to sensor, mm: 37.5; Required t, mm: 22.5. Focus reachable
Case B
L to sensor, mm: 77.5; Required t, mm: -17.5. Outside travel
Case C
L to sensor, mm: 20; Required t, mm: 40. Outside travel
Case D
L to sensor, mm: 35; Required t, mm: 25. Focus reachable

Hypothetical fixed optics without diagonal or corrector: zF = 60 mm, t = 0…30 mm, zsensor = t + L. L includes adapters and the camera’s mounting-to-sensor distance. This checks an interval, not a kit specification. Dashed B/C are unreachable; solid A/D are reachable.

Celestron · back focus

In A, the sensor can pass through the required position, so focus is reachable. In B, even fully retracted, it is already too far behind: an extension makes matters worse. In C it cannot reach far enough outward; a moderate extension shifts the available interval as in D. Every case may have the correct barrel diameter and still retain this longitudinal problem.

Branch · Distance budgetOptional · you can skip this

One equation instead of guessing about an extension

Let z_(F) be the required plane position, t the focuser extension and L the distance from its moving mount to the sensor. All coordinates share one fixed reference; positive points rearward, away from the telescope:

z_(sensor) = t + L, t_(req) = z_(F) − L.

For z_(F) = 60 mm, 0 ≤ t ≤ 30 mm and L = 37.5 mm, required travel is t = 22.5 mm, within range. Add 40 mm: the new L = 77.5 mm requires unavailable t = − 17.5 mm. Conversely, a short train with L = 20 mm needs 40 mm extension; adding 15 mm gives L = 35 mm and reachable t = 25 mm.

Check: the focal plane must belong to the interval [L,L + 30]. Every term is in mm. This is simple geometry for a straight mechanical train; glass, a folded diagonal path, moving mirrors and correctors require their optical data. A component’s physical length cannot automatically be equated with its optical effect.

First the planes, then fine adjustment

When focus is missing, separate the questions. Is the target distant enough for the chosen setup? Is the sensor plane correctly identified? Which direction is still required, and is travel left? Is the corrector spacing preserved? If the eye sees sharply but the field darkens as it moves, also check exit pupil and eye relief. If the centre is sharp while edges are not, one knob may not correct the geometry of the whole field.

The eyepiece helps the eye view a formed image, while the camera records it at a defined location. They are not interchangeable plugs for the same hole. Revisit the shared principle in “How does a telescope work?”. The course next examines optical materials and precision, then Galilean and Keplerian arrangements separately: these show why not every eyepiece sits beyond a real intermediate image.