The functions of the I2R 8 CNC machine are explained

• 1
  What an I2R 8 CNC machine actually is
• 2
  How the two rotary axes work together
• 3
  Accuracy and surface finish: what the numbers mean
• 4
  Which materials behave well and which fight back
• 5
  Where this setup stops being the right choice
• 6
  What to send us with your RFQ
• 7
  Five-axis layout versus simpler setups

Strip the marketing away and an I2R 8 CNC machine is a machining center built around two rotary axes that carry the workpiece, not the spindle. The table tilts and rotates, the tool stays on a fixed line, and the controller keeps the tool tip on the intended path while both rotary axes are moving. That single design choice drives everything else on this page.

The practical difference shows up in how the part is held. A trunnion-style machine has to swing the whole workpiece through the A axis. An I2R 8 CNC machine places the part on a Ø400 mm rotary table and lets the head reach in from a fixed direction. Deep pockets, tall bosses and thin ribs stop fighting the machine geometry, because the part no longer has to rotate past the spindle.

That does not make it universal. The rotary table carries a real mass limit, and the further the part sits from the table center, the more torque the C axis has to hold. Long shafts and heavy blocks often need a different setup. We will come back to those limits.

What you get in return is access. Five sides of a part can be reached in one fixturing, which removes re-datuming between operations. On a bracket with bores on four faces, that alone can remove two setups and the stack-up error that comes with them.

The controller solves a simple problem continuously: keep the tool tip on the commanded path while X, Y, Z, A and C all move at once. The rotary axes are not indexed between cuts. They run under interpolation, so the tool contact point stays normal to the surface as the table tilts.

Feed rate is the first thing that changes. On a flat face, a 3-axis machine moves the tool at the programmed feed. On a tilted face under simultaneous motion, the same programmed feed produces a different surface speed at the contact point. The controller compensates, but only if the post-processor and the CAM output agree on the machine model.

Tool axis control is the second. Ball nose cutters can be held at a fixed lead angle to the surface, which puts the cutting speed on the flank of the tool instead of the dead center. Dead-center cutting on a ball nose gives near-zero surface speed, and that is where smearing and poor finish start.

The third is collision handling. With the part on the table, the head can approach from a consistent direction, and the machine envelope is easier to check in CAM. This is why shops can program aggressive tool lengths on this layout without a simulator run for every operation.

None of these functions are new in principle. What matters is whether the post-processor, the CAM model and the physical machine agree. When they do not, you see it as chatter marks on a tilted face.

Positional tolerance on a well-maintained 5-axis center sits around ±0.005 mm (±0.0002 in). That figure covers the linear axes. Rotary axes add angular error, and angular error becomes linear error at a radius. A 10 arc-second error on a Ø400 mm table is roughly 0.019 mm at the rim. That is why part size matters more than the spec sheet number.

Surface finish follows the same logic. As-machined surfaces land around Ra 1.6–3.2 μm. With the right tool path and a light finishing pass, Ra 0.8–1.6 μm is normal. Fine finishing can reach Ra 0.2–0.8 μm on aluminum and some stainless grades, which usually removes the need for hand polishing on sealing faces.

Two things drive the finish more than the machine. One is tool stick-out: every extra millimeter of flute length adds deflection. Two is the finishing strategy: a constant-scallop path on a curved surface beats a parallel pass at fixed stepover, especially near the tangent points.

The rotary table itself is a heat source. On long roughing cycles the table drive warms up and the centerline drifts. Shops that hold tight tolerances run a warm-up cycle and re-check the datum before the finishing pass, not after the part is off the machine.

Inspection closes the loop. We check raw material on receipt, monitor in process, and inspect 100% before shipment, with reports on request. On a 5-axis part, that usually means CMM work on the critical features plus a surface roughness measurement on any sealing or sliding face.

Aluminum is the easy case. Grades 6061, 7075 and 6082 cut cleanly at high spindle speeds, and the rotary axes mostly need to handle the mass of the fixture rather than the cutting load. Thin-wall aluminum parts benefit most from the one-setup layout.

Stainless is a step up. Grades 303, 304, 316L and 17-4PH work harden if the tool rubs instead of cuts. Under simultaneous motion, a slight dwell on a tilted face is enough to start that. Keep the chip load up and avoid spring passes on 316L.

Titanium and nickel alloys are where the machine layout earns its keep. Ti-6Al-4V, Inconel and Hastelloy generate high cutting forces and poor heat conduction. Moving the part on a rigid table, instead of swinging it on a trunnion, shortens the force loop. That reduces chatter on deep cavities.

Brass and copper are straightforward but gummy. They need sharp edges and generous coolant. Magnesium AZ31B and AZ91D cut fast but demand strict chip control, since fine magnesium chips are a fire risk when they collect dry.

Plastics are a different problem. POM, PEEK and PA move with temperature, so the finish depends on coolant strategy more than on the rotary axes. Carbon fibre is abrasive and needs tooling that will survive it.

The honest rule: the harder the alloy, the more the layout matters. On aluminum, a 3-axis machine with two extra setups often costs less.

Mass and reach set the first boundary. A part that exceeds what the Ø400 mm rotary table can hold, or that sits far from the table center, will need a different machine. Large prismatic parts on a 4,000 mm bed belong on a gantry or a large 3-axis mill, not on a tilting table.

Simple parts set the second boundary. A flat plate with holes on one face does not benefit from five axes. You pay for programming time and machine hours, and you get nothing back. A 3-axis machine with two fixtures is faster and cheaper for that geometry.

Thin, long parts set the third. A shaft with a length-to-diameter ratio above roughly 10:1 will deflect under cutting force no matter how the table tilts. Turning on a mill-turn center or a lathe is the better route.

Very tight angular features set the fourth. If a bore needs a true position of ±0.005 mm relative to a datum 300 mm away, the angular error budget becomes tight. Sometimes the answer is to machine it in a single fixturing on a smaller machine, not to add axes.

We quote all of these routes. The point of the analysis is to pick the cheap one when the part allows it.

The 3D model is the starting point, but it is not enough on its own. Send a drawing that marks the datum features and the tolerances that actually matter. On a 5-axis part, the datum choice decides the fixturing, and fixturing decides whether the single-setup benefit is real.

Tell us the surface finish on functional faces only. Specifying Ra 0.2–0.8 μm across an entire part adds finishing time with no benefit on non-contact surfaces. Mark the sealing faces, the sliding fits and the bearing bores.

State the material grade. 6061 and 7075 machine very differently. So do 304 and 17-4PH. The grade changes the tool path, the feed rate and sometimes the machine choice.

Mention the quantity and the target date. We quote prototypes from one piece up to 10,000+ part runs with no minimum order quantity. If you need DFM feedback, we return the quotation and a free DFM analysis within 12 hours, and production can start within 24 hours.

If the part is confidential, send the NDA first. Uploads are secure, and we sign an NDA on request.

One more item: tell us how the part will be inspected at your end. If a gauge or a fixture is used, we can machine to the same reference and avoid an argument at incoming inspection.