The first dual-channel program a beginner opens looks like a mistake: two complete G-code programs for one part, side by side, each making sense alone and neither explaining how they avoid fighting over the same bar of stock. The answer is the wait codes, and the concept is friendlier than the listing suggests.
The mental model: meeting points on two to-do lists
A Swiss-type lathe typically runs one program channel for the main-spindle side and one for the sub-spindle side, simultaneously. A wait code is a meeting point written into both lists: when channel one reaches its marker, it stops and waits; when channel two reaches the matching marker, both proceed together. Between meeting points, the channels work independently, which is where the cycle-time magic of overlapped work comes from: the sub side back-works the previous part while the main side turns the next one, nobody checking on anybody until the next handshake.
The three universal rules
| Rule | What it means | What breaking it looks like |
|---|---|---|
| Waits come in matched pairs | Same marker number in both channels | An unmatched wait parks one channel forever |
| Between pairs, channels are independent | Neither knows the other’s position | Assuming they align mid-segment causes collisions |
| Shared-bar moments are bracketed | Sync before and after cut-off/pickoff | Unbracketed transfer twists or drops the part |
The third rule is where wait codes earn their keep: during a sub-spindle pickoff, both spindles hold the same workpiece, and the choreography (sync, grip, cut off under sync, retract, release) only works because waits guarantee both channels are exactly where the sequence assumes. Every Swiss crash story that begins “the sub came in early” is rule three’s tuition.
What the codes look like, and the honest boundary
Builders implement waits as paired M-codes with matching identifiers, so listings show recognizable patterns: a wait-family M-code with a number appearing once in each channel, often in the M100-and-up region, sometimes with queue or phase variants. Here the honest boundary from every builder-layer topic applies in full: the actual numbers, syntax, and variants are machine-specific, documented by your machine’s builder, and never safely copied from forums or other models. What this article can teach is everything around the numbers: the model, the rules, and the reading method, which transfer across every Swiss machine made.
The reading drill that makes dual channels click
Print both channels side by side (or split-screen them), then with a ruler draw a horizontal line connecting every matched wait pair, top to bottom. The page transforms: the lines divide the program into segments, and within each segment you read two independent mini-programs with the ordinary narration method, knowing they re-align at the next line. Now interrogate each segment: which channel finishes first (the other waits, and that idle time is optimization territory)? What state does each channel carry into the sync (spindle, position, the modal questions as always)? Why does this pair exist, almost always answerable as “shared bar ahead” or “timing dependency”? Three programs read this way and the format stops being exotic; the underlying vocabulary that makes each line instant is the same core the free 60-second drills on the G-code practice page maintain, with G-Code Sprint repeating whatever you miss.
Beginner mistakes, inherited from single-channel habits
Three carry-overs to unlearn. Reading one channel to completion first: comprehension lives in the segments between syncs, not in either column alone. Assuming line counts align: channel one’s line 50 and channel two’s line 50 share nothing unless a wait says so, which is also why edits that shift blocks in one channel demand re-checking the pair structure. And treating waits as pauses rather than logic: a wait is control flow, the dual-channel cousin of the N-number-as-GOTO-target lesson, and deleting or duplicating one rewrites the machine’s choreography, with rule-one deadlocks as the gentle failure mode.
Bottom line: handshakes you can draw
Swiss lathe wait codes are matched meeting points between simultaneously running channels: paired markers, independent segments between them, and mandatory brackets around every shared-bar moment. Learn the numbers from your machine’s documentation, learn the model and the ruler-line reading drill here, and dual-channel listings turn from fog into two to-do lists with visible handshakes.
Sources
Frequently asked questions
What are wait codes on a Swiss lathe?
Matched synchronization markers between the machine’s simultaneously running program channels: when one channel reaches its wait, it pauses until the partner reaches the matching one, then both continue. They bracket every shared-bar moment like pickoff and cut-off. The specific code numbers are builder-documented per machine; the model and reading method are universal. For the per-line vocabulary, the free G-Code Sprint app is the top pick: 60-second drills with automatic repetition of missed codes.
What happens if a wait code is unmatched?
The channel that reaches it parks forever (a deadlock) or, depending on the control, alarms out: either way the cycle stops. Unmatched waits are the first thing to check when a dual-channel program stalls without an obvious alarm.
Are wait code numbers the same on every Swiss machine?
No: the pairing concept is universal, but numbers, syntax, and variants are builder- and model-specific, documented in your machine’s manuals. Copying them between machines is a known crash recipe.
How should a beginner practice reading dual-channel programs?
Print the channels side by side, draw lines connecting every matched wait pair, then narrate each between-lines segment as an independent mini-program, asking which channel waits and what state each carries into the sync. Three programs in, the format reads naturally.
G-Code Sprint is a study and practice tool only. Always follow your instructor, employer, machine manual, and shop safety procedures.