Special combined machine tools are built from several independent machining units
This page explains how those units divide work, how transfer and indexing keep position between them, and where the concept stops paying off. It is written for engineers and buyers who need to decide whether a part family fits a combined setup or a standard machining center.

In this article
- 1
- 2
- 3
- 4
- 5
- 6
What special combined machine tools actually are
A special combined machine tool is a single production system assembled from several independent machining units. Each unit keeps its own spindle, its own feed axes, and often its own control loop. A base, a column, or a bed carries them, and a part moves from unit to unit instead of moving to a second machine.
That is the whole idea. You are not buying one very large machine. You are buying a group of machines that share a frame, a fixture strategy, and a sequence. A milling unit, a drilling unit, a tapping head, and a boring head can all sit on the same bed and work the same part.
The distinction matters at the drawing stage. On a general-purpose machining center, one spindle does everything and the part is repositioned between operations, or the table rotates. On a combined system, the part often stays clamped once while different units come to it, or it indexes along a line of stations. Clamp once, cut many faces.
Independent units also mean independent failure. If one head goes down, the others keep running, and a spare head can be swapped in. That redundancy is one reason the layout survives in high-volume plants where a stopped line costs more than the extra iron.
The trade is real, though. More units mean more spindles to align, more thermal sources, and more calibration work. The design only pays off when the part family is stable and the annual volume is high enough to amortize that overhead.
- 1Shared bed, separate spindlesEach unit keeps its own drive and control loop.
- 2One clamp, many facesThe part moves less; the units move more.
- 3Built for a part familyNot for one-off or rapidly changing geometry.
How independent units divide the work
The usual split follows the feature list. Rough milling goes to a heavy unit with a large tool and low spindle speed. Finish boring goes to a unit with a fine feed and a rigid quill. Tapping, drilling, and chamfering usually get their own heads because their cycle times differ by an order of magnitude.
Cycle time, not accuracy, drives most of these decisions in high-volume work. If milling takes 90 seconds and drilling takes 8 seconds, putting them on one spindle wastes 82 seconds of spindle time per part. Two heads running in parallel bring the effective cycle closer to the slowest station.
Accuracy enters through the datum chain. Every unit references the same fixture or the same pallet locating feature. If the fixture repeats to ±0.01 mm and the units are aligned to that fixture, the stack stays tight. If each unit has its own datum, error accumulates station by station.
Thermal behavior is the quiet variable. A boring unit running at 8,000 rpm warms its spindle housing over the first hour. If a nearby drilling unit shares the same casting, that growth can move the drilling axis. Separate castings and separate coolant paths reduce the coupling.
Tool life is also easier to manage when functions are separated. A tapping head never has to fight a milling cut for the same spindle bearings, so tap breakage drops and feeds and speeds stay in a narrow, repeatable band.
- 1Rough and finish apartHeavy cuts and fine feeds rarely share a spindle well.
- 2Match cycle timesBalance stations so no unit idles for long.
- 3One datum for all unitsAlign every head to the same fixture feature.
- 4Watch thermal driftSeparate castings reduce cross-coupling.
Transfer, indexing, and how position is held between units
Two layouts dominate. In a linear transfer layout, parts advance along a line of stations, each station holding one or more units. In a rotary layout, parts index on a table, and fixed units around the perimeter cut as the table stops. Rotary indexing is compact; linear transfer scales to more stations.
Position between stations comes from the transfer mechanism. A rotary table with a Ø400 mm face and a curvic coupling can repeat to a few arc-seconds, which at 200 mm radius is roughly a few microns of tangential error. Ball-lock pallets offer similar repeatability with more flexibility.
Fixed units need alignment to the transfer datum, not to each other. Once the table or pallet repeats, each unit is dialed in against a master part or a laser tracker. Re-check after the first thermal cycle, then again after a week of production.
In-process gauging closes the loop when tolerances are tight. A probe measures a bore after boring, and the control offsets the boring unit for the next part. This handles tool wear without an operator touching offsets.
There is a limit. If the part needs five-axis contouring on a complex surface, a combined system with single-axis heads will struggle. Reach and orientation are fixed by the unit layout, so freeform geometry still belongs on a simultaneous five-axis machine.
- 1Linear or rotaryLinear scales in stations; rotary saves floor space.
- 2Repeat the transfer, not the headAlign units to one master datum.
- 3Probe to close the loopUse measured offsets for tool wear.
- 4Freeform surfacesKeep contoured work on a five-axis center.
When the combined layout pays off and when it does not
The math is straightforward. A combined system costs more than one machining center, so the part must carry enough volume to spread that cost. As a rough rule, a stable part family with annual demand in the thousands and a cycle that can be split across stations is a good candidate.
Part geometry decides as much as volume. Castings and forgings with features on several faces fit well, because the part can stay clamped while heads work each face. Thin-walled parts fit poorly, since clamping once and cutting hard on several faces loads the wall from multiple directions.
Changeover is the weak point. If the customer revises the part every few months, the unit layout becomes obsolete quickly. Mixed-model production with a shared fixture and quick-change heads is one answer, but it adds setup discipline that many shops underestimate.
Floor space and staffing usually improve. One operator can load a rotary system and monitor several units, and the footprint is smaller than the equivalent line of standalone machines. Power and coolant demand go up, not down.
Maintenance skill matters. Aligning and calibrating several independent units needs a different skill set from running one machining center, and spare heads should be stocked for the units that fail most often.
- 1Good fitStable casting or forging, features on several faces.
- 2Poor fitThin walls, frequent revisions, one-off jobs.
- 3Space and laborSmaller footprint, fewer operators.
- 4Stock spare headsRedundancy only works if a replacement is on hand.
Specifying units, fixtures, and tolerances for a combined setup
Start from the feature list and the tolerance on each feature. Group features that share a datum and a tool approach. Features needing ±0.005 mm and a fine finish should sit on a rigid unit with a fine feed, not on a roughing head that has already taken a heavy cut.
Pick spindle power from the cut, not from the catalog. A 12 mm carbide end mill in 6061 aluminium at 3,000 rpm and 0.15 mm per tooth needs far less power than the same tool in 4140 steel. Under-powering a roughing unit causes chatter that shows up later as poor finish on a finish unit.
Fixtures carry the accuracy. A hydraulic or pneumatic clamp with a repeatable locating pin gives better station-to-station consistency than a manual strap clamp. Plan a master part for alignment and keep it for the life of the system.
Coolant and chip evacuation deserve early attention. Drilling units produce long chips, boring units produce fines. If both share one conveyor, chips back up. Route high-volume drilling to a dedicated chip path.
Finally, agree the acceptance test before the build. Run a capability study on the critical features over a full shift, and include thermal drift in the run. A system that holds tolerance cold and drifts hot is not finished.
- 1Group by datumFeatures sharing a datum belong on one unit.
- 2Size power to the cutSteel needs more spindle power than aluminium.
- 3Fixture sets the stackRepeatable clamping beats manual straps.
- 4Plan chip pathsDrilling and boring produce different chips.
Combined units versus a standard machining center
Use this to check the part family against the two routes.
| Factor | Special combined machine tools | Standard machining center |
|---|---|---|
| Best part family | Stable casting or forging, high volume | Mixed low-volume work, frequent revisions |
| Faces per clamp | Several, via multiple units | One to three, via table rotation |
| Cycle time split | Parallel stations, balanced cycle | One spindle does all operations |
| Freeform surfaces | Limited by fixed unit orientation | Simultaneous five-axis contouring |
| Changeover | Slow, layout is part-specific | Fast, reprogram and re-fixture |
| Floor space | Compact, units share one base | One machine per operation |
| Redundancy | Spare head keeps line running | Machine down stops the job |
| Best volume band | Thousands of parts per year | One prototype to a few thousand |
The trade in one line
Choose special combined machine tools when the part family is stable, the volume runs into the thousands, and features sit on several faces of one casting or forging. Choose a standard machining center when geometry changes often, walls are thin, or you need simultaneous five-axis contouring.
Common questions
Can a combined system hold ±0.005 mm across stations?
It can, but the fixture decides. Every unit must be aligned to the same locating feature, and the transfer mechanism must repeat tightly enough. A rotary table with a curvic coupling or a ball-lock pallet system is the usual answer.
Plan a capability study over a full shift, including the warm-up hour, because thermal drift is the most common cause of a station falling out of tolerance.
How many independent units make sense on one base?
There is no fixed number. The practical limit comes from chip evacuation, coolant routing, and how much access an operator has for tool changes and inspection. Four to eight units on one base is common in production work.
Beyond that, a linear transfer layout with separate stations is easier to service than one crowded base.
Do combined systems replace five-axis machining?
No. They solve a different problem. A combined system cuts fixed features from fixed directions, fast and in parallel. A five-axis center orients a tool freely, which is what contoured and undercut surfaces need.
Many shops run both: five-axis for the contoured features, a combined line for the rest.
What drives the cost of a combined setup?
Unit count, spindle specifications, the transfer mechanism, and the fixture. Controls add cost too, since several spindles may need coordinated logic and in-process gauging.
The build also takes longer than a standard machine order, so the volume has to justify the lead time as well as the price.
How do you handle tool wear on a boring unit?
Use in-process gauging. Probe the bore after the cut, compare it to the nominal, and offset the unit for the next part. This keeps the size inside the band without an operator editing offsets.
For lower volumes, a scheduled tool change based on counted parts and a first-piece check is enough.
Is a combined system suitable for prototypes?
Usually not. The fixture and alignment work only pays back over volume, and a prototype part often changes before the second run.
For prototypes, a three-axis or five-axis machining center is faster to set up and easier to change. Move to a combined layout once the design is frozen.
Send the drawing and we will check the route
Tell us the part family and volume, and we will say whether a combined layout makes sense or whether a five-axis center is the better call. DFM feedback and a quote come back within 12 hours.
12-hour quote100% inspectionNDA on request