How to Determine the Feed Rate During High-Speed Machining of a Double-Spindle Machining Center

• 1
  Key takeaways
• 2
  How to determine the feed rate: the variables that matter
• 3
  Match chip load to the workpiece material
• 4
  Tool geometry and radial engagement change the limit
• 5
  Check both spindles and the machine's dynamic limits
• 6
  Starting chip loads for common materials
• 7
  Mistakes that make the feed rate wrong
• 8
  Step by step: how to determine the feed rate on a double-spindle center
• 9
  Chip load and surface speed starting points

Feed rate is not a number you look up in a table. It is the result of four inputs: spindle speed, cutter diameter, flute count and the chip load the tool can carry. On a double-spindle machining center there is a fifth input, because the two heads run the same program on two parts at once. Any difference between them shows up as a dimensional mismatch, not as a crash.

The basic relationship is simple. Feed rate in mm/min equals RPM times number of flutes times chip load per tooth in mm. If a 12 mm three-flute end mill runs at 12,000 RPM with a 0.08 mm chip load, the feed is 12,000 x 3 x 0.08 = 2,880 mm/min. That figure is your starting point, not your final value.

Chip load itself comes from the tool supplier and from the material group. Aluminum 6061 tolerates a larger chip load than 17-4PH stainless at the same diameter. Titanium TC4 (Ti-6Al-4V) sits in the middle and punishes any tool that rubs instead of cutting. When in doubt, start conservative and raise the feed until the chip looks right.

• 1
  Spindle speedSet by surface speed and cutter diameter, then limited by the spindle's maximum RPM.
• 2
  Flute countMore flutes mean more feed at the same chip load, but less room for chip evacuation.
• 3
  Chip loadThe thickness of material each tooth removes; the single most important input.

Soft aluminum likes speed and feed. In 6061-T6 a 10 mm three-flute carbide cutter can run a 0.10 mm chip load at 15,000 RPM, giving 4,500 mm/min. Climb milling with a light radial step-over keeps the load steady. If the chip comes off as a fine powder instead of a curl, the feed is too low and the tool is rubbing.

Stainless 304 and 17-4PH work-harden. A chip load below roughly 0.03 mm per tooth will polish the surface and harden it, which destroys the next pass. Keep the feed up and the radial engagement moderate, use plenty of coolant, and never let the tool dwell in the cut. A 8 mm four-flute cutter in 304 typically runs 0.04–0.06 mm per tooth.

Titanium and Inconel generate heat at the cutting edge. Feed rates must stay high enough to keep the tool in the cut and low enough to control temperature. For TC4 with a 10 mm four-flute tool, 0.05–0.07 mm per tooth at 60–80 m/min surface speed is a workable band. Plastics such as POM and PEEK behave differently: they need high feed and sharp tools to avoid melting.

A tool's chip load rating assumes a specific radial engagement, usually around half the cutter diameter. When you reduce radial engagement to 10 percent of the diameter, as high-speed machining strategies do, the chip thins. The programmed feed must be increased to keep the actual chip thickness on target. This is the feed rate compensation that separates a stable HSM cut from a noisy one.

Corner radius matters too. A sharp corner on a square end mill concentrates stress and limits feed. A 0.5 mm corner radius lets you push the feed 20 to 30 percent higher in the same material. For deep pockets, a reduced neck or a long-reach tool lowers the safe feed, because deflection grows with the cube of the length-to-diameter ratio.

Helix angle and coating also shift the window. A 45-degree helix in aluminum clears chips fast and supports higher feed. A 38-degree variable helix in steel reduces chatter. Coatings such as AlTiN allow higher surface speed, but they do not allow a higher chip load by themselves. Feed limits come from the tool's core strength, not from the coating.

A double-spindle machining center is two machines sharing one bed. Each spindle has its own speed limit, tool length offset and thermal drift. The feed rate you program applies to both, so both heads must be capable of holding it. If spindle A is rated to 12,000 RPM and spindle B to 15,000 RPM, the program runs at the lower figure or the tools differ.

Acceleration and jerk limits matter more than top feed in small features. A machine that reaches 30,000 mm/min on a long straight cut may only average 6,000 mm/min in a 10 mm corner. The controller slows the feed to respect the machine's dynamics. Simulation software or the controller's own look-ahead shows the real average, and that is the number to check against the tool's chip load.

Spindle load and tool runout complete the picture. Keep spindle load below about 80 percent of the continuous rating to leave room for hard spots. Measure runout at the tool tip; beyond 0.01 mm TIR the effective chip load differs from flute to flute, and one edge wears out first. Both spindles should be measured separately.

The numbers below are starting points for solid carbide tools at moderate radial engagement, roughly 30 to 50 percent of cutter diameter. They assume rigid setups and good coolant delivery. Treat them as a first guess, then adjust from the chip and the sound of the cut.

Feed rate is the product of RPM, flute count and chip load. If you change the cutter diameter, the surface speed limit changes the RPM, and the feed moves with it. Do not copy a feed number from one tool to another without recalculating.

For finishing passes with a small radial step-over, raise the feed by the chip thinning factor. A 10 percent radial engagement roughly doubles the programmed chip load needed to keep the same chip thickness. This is the single biggest reason HSM feeds look high on paper.

The most common error is copying a feed rate from a different cutter diameter. Surface speed pins the RPM, and RPM scales inversely with diameter. A feed that works on a 16 mm cutter is far too high for a 6 mm cutter in the same material, because the smaller tool runs at more than double the RPM to keep the same surface speed.

The second error is ignoring chip thinning. HSM toolpaths use small radial engagement, and the programmed chip load must rise to compensate. Operators who keep the catalog chip load see the tool rub, heat up and fail early. The cut sounds fine for the first minute and then the edge goes.

The third error is treating the two spindles as identical. Thermal growth, runout and tool length offset all drift. Check both heads at the start of the shift and after long runs. If one spindle shows a different load, stop and find out why before the next part.