Plain-language guide
How autofocus stepping works
The calculator gives you three numbers and sends you back to your imaging software. Here is what those numbers mean, in plain terms, and why getting them right is the difference between a sweep that nails focus and one that wanders off.
01 What your software is actually doing
Autofocus doesn't “see” a sharp star the way you do. Instead it takes a short exposure, measures how fat the stars are, nudges the focuser, and takes another. Do that a handful of times and the star sizes trace out a V: fat on the far side, shrinking as you approach focus, smallest at the bottom, then fattening again as you pass it.
The software finds the bottom of that V and parks the focuser there. Your whole job is to hand it a clean V to work with — and that is exactly what the three numbers set up.
02 The three numbers, in plain English
Think of measuring the depth of a valley by walking down one side and up the other, stopping to take a reading every few paces. You want your readings spread evenly across the valley — not all bunched at the bottom, and not so wide that you skip straight over it.
Step size
How far the focuser moves between each photo — the length of one pace down the hillside.
Starting offset
How far out of focus to begin, so the first photo is already well up one side of the V.
Number of photos
How many readings you take on the way through — more dots make a cleaner V.
Get the step size wrong and it shows immediately: too small and every photo looks nearly identical, so the software can't tell which way is “downhill.” Too large and you leap over the bottom of the V and never really land on it.
03 Why the stars have to blur — on purpose
This is the part that trips people up: for the sweep to work, your first and last photos need stars that are noticeably fat. If you never move far enough out of focus, the stars barely change from shot to shot and the V is too shallow to read.
It's like finding the bottom of a salad bowl by feeling only the very centre — it all feels flat. You have to reach out to the sloped sides to sense which way is down. That's why the calculator deliberately starts you well out of focus, with stars blurred into clear, fat discs, before working back in.
04 Why a big scope needs big steps
A long, slow telescope with large camera pixels is forgiving — you can move the focuser a fair way before the picture changes much, so it needs large steps. A short, fast scope is touchy: a tiny move swings the stars from sharp to soft, so it needs small steps. Same idea as a light switch versus a dimmer.
The calculator works this out from your scope and camera so you don't have to. Don't be alarmed if the number looks huge. One customer's big 14″ setup came out around 7,000 steps — that's completely normal for that rig, not a mistake. A little refractor might land in the low hundreds. Both are correct.
05 Rings instead of blobs (mirrors)
If your scope has a mirror in the middle — most SCTs, RCs and Newtonians do — a defocused star shows up as a ring (a “donut”), not a filled disc. That's the shadow of the secondary mirror, and it's completely expected.
It matters here because a donut needs a bit more defocus to read cleanly, so the calculator accounts for it automatically when you pick a mirror-based scope. A plain refractor gives solid discs and needs no such allowance.
06 SCTs & Maks: focusing by moving the mirror
Most Schmidt-Cassegrains and Maksutovs don't focus with a drawtube at all — the focus knob slides the big primary mirror back and forth inside the tube. That changes one important thing about the numbers, so it's worth knowing before you calibrate.
Because the secondary mirror magnifies everything, a tiny nudge of the primary moves the focus point a long way — on a typical SCT the image shifts roughly 25 times farther than the mirror itself does. (That isn't a guess: the standard reference, Rutten & van Venrooij, measured almost exactly 25× on a 200 mm f/10 SCT.) So the figure the calculator needs is how far the image moves per step — something you can't get by measuring the mirror.
Measuring it is simple once you know the trick: tick “This motor drives an SCT / Mak primary-mirror focuser” in the calculator, command a known number of steps, then slide the camera in or out until the stars are sharp again and note how far it moved. That's your focal-plane travel — and the calculator takes it from there.
07 7 photos or 9?
More photos means more dots on the V, which makes the “bottom” easier to pin down — at the cost of a little extra time each run. Seven is the sensible default. Reach for nine if your curve looks noisy, if you're chasing the last bit of precision, or if conditions are marginal.
Two reassurances: the step size doesn't change when you add photos — you're just extending the sweep a bit further out each side. And there's no prize for going higher; past nine you're mostly spending time, not buying accuracy.
08 Putting the numbers into your software
Whatever you run — N.I.N.A., SGP, ASIAIR or the like — the three numbers map onto the autofocus settings the same way. Names differ slightly between programs, but the meaning is identical:
- Step size goes in the field called “step size” (sometimes “steps”).
- Number of photos is the count of exposures or points per run.
- The starting offset is handled for you once the first two are set — the software moves out, then steps back through.
09 Worth knowing
- Focus drifts as the night cools. Metal shrinks. Re-run autofocus every so often, or let your software trigger it on a temperature change.
- Slack in the focuser can spoil a sweep. If your V looks lopsided or jumps around, backlash is the usual suspect — there's a separate guide for measuring and dialling that out.
- Filters shift focus. Narrowband and colour filters don't all focus at the same point; refocus after a filter change unless your software offsets it for you.
- The numbers are a strong starting point, not gospel. If your rig consistently likes a slightly different step, trust what you see on the V and adjust.