Analysis of Basic Technologies and Advantages of the CNC Double Head Tour
A double-head CNC lathe carries two spindles on one bed, so a part can be cut from both ends without being unloaded. This page breaks down the basic technologies behind that layout and the specific part families where it saves cycle time. Written for process engineers and buyers who need to decide whether the setup fits their drawing.
What the double-head layout actually changes
Two spindles on one bed turn a single-setup lathe into a two-setup machine that runs in one cycle.
The basic technologies inside a double-head lathe
A double-head lathe is not two lathes bolted together. It is one bed carrying a main spindle and a second spindle, usually opposed and coaxial, with independent Z travel. Each head has its own tool turret or gang slide, and the two sides share a single CNC that coordinates them. That shared control is the whole point: the controller knows where both spindles are, so a part can be cut on face one, transferred, and finished on face two inside one program.
The transfer step is where the engineering lives. The subspindle must grip the part while the main spindle is still holding it, then the main collet opens and the subspindle pulls the part clear. Any mismatch between the two spindle axes shows up as runout on the second operation. Builders hold spindle-to-spindle concentricity to a few microns and let the control compensate with a work offset, which is why a bar-fed double-head machine can hold ±0.005 mm across a handoff.
- 1Opposed spindlesMain and secondary head face each other on one bed, cutting both ends without re-chucking.
- 2Subspindle handoffThe second head grips and pulls the part; the main collet opens under program control.
- 3Live toolingC-axis indexing plus driven tools add cross-drilling, milling and slotting to a turning cycle.
- 4Bar feederContinuous bar stock feeds the main spindle, so the machine runs unattended between loads.
How the two heads split the work
The simplest use is balanced cutting: both turrets work the same part at once, one on the outside diameter and one on a bore or a second diameter. Cycle time drops because two tools are in cut at the same moment, not because the machine runs faster. On a part with a 90-second single-spindle cycle, splitting the operations across two heads often lands between 55 and 70 seconds, depending on how well the two tool paths can be balanced.
The second use is sequential: the main spindle cuts face one, the subspindle takes the part and cuts face two, then ejects it. Here the gain is not simultaneity but the removal of a manual flip. On a shaft with a thread on each end, a single-spindle lathe needs two operations and an operator between them. The double-head machine does it in one cycle, and the concentricity between the two threads comes from the machine, not from a re-chuck.
Single-spindle lathe vs double-head lathe
Where the second head pays back and where it does not.
| Factor | Single spindle | Double head |
|---|---|---|
| Setups per part | Two, with a manual flip | One, with an in-machine transfer |
| Concentricity across ends | Depends on re-chuck accuracy | Set by spindle alignment |
| Cycle time on two-ended parts | Two cycles plus handling | One cycle, often 25-40% shorter |
| Best batch size | Low volume, simple geometry | Medium to high volume, both ends cut |
| Tooling cost | Lower | Higher: second turret and driven tools |
| Programming effort | Straightforward | More work on handoff and balance |
| Setup time for a new part | Shorter | Longer, needs proving on the handoff |
| Weak fit | One-off prototypes | Parts with no second-end feature |
Which parts suit a double-head machine
The layout favors parts that are turned from bar or a small billet and have features on both ends: fittings, hydraulic adapters, valve bodies, motor shafts, sensor housings, connector pins, and threaded studs. If the part has a bore that must stay coaxial with an outside diameter on the other end, the single-setup transfer is a real quality gain, not just a speed gain. It removes the stack-up that comes from chucking twice.
It is a poor fit for a part with only one machined end. You pay for a second spindle and never use it. Long, slender shafts are also a problem: the subspindle adds support, but bar whip and deflection still limit how far the part can extend. Deep internal features that need a long boring bar may not fit the second turret envelope. And a part that starts as a casting with irregular stock is usually better on a mill-turn or a 5-axis center, where the first operation is facing and fixturing rather than bar feed.
- 1Good fitBar-fed parts with threads, bores or diameters on both ends.
- 2Good fitParts needing coaxial features across a handoff.
- 3Poor fitOne-ended parts, one-off prototypes, irregular castings.
- 4WatchSlender shafts and deep bores that exceed the second turret reach.
Tolerances, tooling and inspection on the floor
Two heads mean two sources of error. Thermal growth on the main spindle shifts the part during a long run, so warm-up cycles and in-process probing matter more than they do on a single-spindle lathe. Tool wear on the second turret is easy to miss because the operator is watching the first. We keep a wear offset schedule per tool and check the handoff dimension at set intervals rather than only at the end of the run.
On the materials side, aluminum and brass run cleanly on a double-head setup because chips break and clear. Stainless 303 and 316 need more attention to chip control on the subspindle side, where coolant access is tighter. Titanium and Inconel run hotter and slower, and the second spindle rarely justifies its cost unless the part has a critical second-end feature that would otherwise need a separate operation.
For inspection, the handoff dimension is the one to watch. Measure concentricity between the two machined ends, not just each end on its own. A part can pass every diameter check and still fail assembly because the two ends are not coaxial. We inspect 100% of parts before shipment and can supply reports on request, covering raw material, in-process checks and final inspection.
Questions engineers ask about double-head turning
How accurate is the transfer between the two spindles?
It depends on spindle alignment and how well the work offset is set. On a maintained machine, concentricity across the handoff can be held in the low micron range, which supports a ±0.005 mm part tolerance on the transferred features.
The number to verify is not each end separately but the relationship between them. Ask for a concentricity check across the handoff, not just diameter reports.
When is a double-head lathe worse than a single-spindle lathe?
When the part has only one machined end, or when the batch is a handful of pieces and the setup cannot be amortized. Programming a balanced two-turret cycle and proving the handoff takes longer than a simple single-spindle setup.
Irregular castings are also a poor fit. The first operation is about establishing a datum, and bar feed does not help there.
Does the second head replace a second operation on a mill?
Partly. With C-axis indexing and live tooling, cross-holes, flats and slots can be cut in the same cycle. Deep pockets or features needing five-sided access still belong on a machining center.
The decision is usually about how many features can be reached from the two turret positions without a re-fixture.
Do I need a bar feeder to run one?
No, but it is where the economics improve. Without a bar feeder the machine still removes the manual flip, which helps on medium batches. With continuous bar feed it can run unattended between loads.
For short bar or billet work, a magazine or robot loader does the same job.
What materials run well on a double-head setup?
Aluminum 6061 and 7075, brass C36000, and stainless 303 and 316 are the common choices. They break chips predictably and hold size across the transfer.
Titanium and Inconel are workable but slower, and the second spindle only pays off when a second-end feature is truly critical.
Can the machine hold a fine surface finish on both ends?
Yes. Finish is set by the tool and the parameters, not by the number of heads. We typically run Ra 0.8–1.6 μm on turned diameters and can reach Ra 0.2–0.8 μm where the drawing calls for it.
The second-end surface depends on how rigid the subspindle grip is. A short, well-supported grip holds finish better than a long overhang.
Send a drawing and we will tell you if the second head helps
Upload the part and we return a quotation plus a DFM analysis within 12 hours, with a straight answer on whether double-head turning is the right process.
Quotation and DFM in 12 hoursNo minimum order quantityNDA on request100% inspection before shipment
