The Special Value of Vertical Grinding Tools

• 1
  The special value of vertical grinding tools: how the layout works
• 2
  Why clamping force drives roundness on a vertical spindle
• 3
  One chucking for bore, face and outside diameter
• 4
  Where vertical grinding stops being the right answer
• 5
  Wheel choice and material behavior above the part
• 6
  How to verify a vertically ground bore
• 7
  Vertical or horizontal grinding: which part goes where

On a vertical grinding tool the workpiece sits on a rotary chuck near the base of the column, and the grinding spindle comes down from above. The wheel head can stroke along the vertical axis, index across the part face, and in many machines swivel so that a single spindle can reach an outside diameter, a bore and a shoulder face. On a horizontal cylindrical grinder the part is held between centers or in a chuck and the wheel approaches from the side.

The layout changes where the forces go. Gravity acts along the same axis as the wheel feed, so the part is pressed onto its locating face rather than hanging off a chuck or between centers. Radial grinding force still pushes the part sideways, but the chuck only has to resist that force plus the friction needed to keep the part from rotating. There is no long overhang to sag under its own weight.

That matters most on parts with a large diameter-to-length ratio. A thin disc, a ring gear blank or a bearing housing that would deflect between centers can sit flat on a face chuck and keep its shape. The chuck grips a face, not a diameter, so the clamping load spreads over a wide annulus instead of concentrating at three jaw contact points.

The trade is reach. A vertical spindle cannot pass a wheel between two closely spaced shoulders the way a small horizontal wheel can. Deep bores with small diameters, long slender shafts and parts that need support at both ends are still horizontal work. The vertical tool wins where the part is short, wide, heavy, or needs several features ground in one chucking.

A three-jaw chuck on a horizontal grinder has to hold the part against a tangential cutting force while the part spins. The jaws are driven inward until the friction is high enough, which means the part sees a three-point radial squeeze before the wheel ever touches it. On thin-wall rings that squeeze alone can turn a round blank into a triangle. After grinding, the jaws release and the part springs back.

On a vertical grinding tool the part rests on an axial face and is held down, not squeezed inward. Clamping force is mostly axial, so the radial distortion that shows up as lobing is much smaller. The wheel still pushes the part radially, but a face clamp resists that through friction on a large area rather than through point contact.

Roundness on the machine is not roundness in the application, though. If the part is clamped on a face that is not flat, or on a burr left by a previous operation, the error transfers into the ground bore. We check the locating face for flatness and clean it before the part goes on the chuck. A 0.01 mm burr under a 150 mm diameter ring can show up as several micrometres of roundness error.

Springback also depends on wall thickness. A ring with a 2 mm wall will move far more under clamping than a solid block. For thin rings we lower the clamping pressure to the minimum that still holds the part, and we grind in a light pass first to see whether the part is moving. If the spark pattern is uneven across the face, the part is not seated.

The clearest argument for a vertical grinding tool is setup count. A part that needs a bore, a shoulder face and an outside diameter can often be finished in one chucking because the wheel head indexes and swivels above the part. Each extra chucking adds a re-clamp error, and that error usually lands on the relationship between the bore and the face.

Consider a hydraulic manifold block with a large central bore and a flat mounting face. The perpendicularity between the bore axis and the face is a functional requirement, not a cosmetic one. Grind the bore on a horizontal machine, then move the part to a surface grinder for the face, and the perpendicularity depends on how well the part sits in the second fixture. On a vertical spindle the wheel grinds both without the part ever leaving the chuck.

The same applies to a bearing housing where the bore and the locating shoulder must stay concentric. Every time the part is released and re-clamped, the new grip defines a new axis. Keeping the part on one chuck keeps one axis.

This is also where in-process gauging pays off. On a vertical machine the probe or gauge head can reach the bore from above while the part is still clamped. We take a size reading after the spark-out pass, adjust the offset, and take a second pass. The part is not released until the size is in tolerance, which removes the guesswork that comes with measuring after unclamping.

Vertical grinding is not a general replacement for cylindrical grinding. The spindle hangs over the part, so the wheel diameter and the bore depth set a hard limit on what the head can reach. A bore that is 200 mm deep with a 40 mm diameter leaves little room for a wheel large enough to hold its form. The wheel wears faster and the finish drifts down the bore.

Long slender shafts are another poor fit. A shaft with a 10:1 length-to-diameter ratio needs support at both ends or a steady rest. On a vertical machine the lower end is in the chuck and the upper end is unsupported, so the part can deflect away from the wheel in the middle. Grinding a shaft this way usually produces a barrel-shaped profile rather than a straight cylinder.

Part weight matters in the other direction. Heavy parts are easier to load on a vertical machine because the crane or hoist can lower them straight into the chuck. Horizontal machines need the part lifted and aligned sideways, which is slower and riskier for a 300 kg casting. When the part is heavy and short, the vertical layout is the safer choice.

There is also a floor-space and foundation question. A vertical grinding tool with a Ø400 mm rotary table takes a compact footprint for the size of part it can hold, but the column and the swivel head need a rigid foundation. If the machine is installed on a weak floor, the column movement shows up as taper in the bore. We check the level and the anchor bolts before accepting any grinding job.

The wheel above the part behaves differently from a wheel approaching from the side. Coolant has a straight path down into the contact zone, so it reaches the grinding arc more easily and flushes the chips out. That helps on materials that burn or load the wheel, such as 17-4PH stainless and tool steel. The risk of heat checking on a hardened bore is lower when the coolant arrives from directly above.

Hardened steel above 45 HRC is common work for a vertical grinder. We use aluminum oxide wheels for 4140 and 4340 in the 45–55 HRC range, and cubic boron nitride when the batch is large enough to justify the wheel cost. CBN holds form longer on a deep bore, so the size stays stable across hundreds of parts.

Aluminium behaves differently. It loads a conventional wheel quickly, so we switch to a coarser, more open wheel and increase the coolant flow. Surface finish targets of Ra 0.8–1.6 μm are realistic on aluminium bores with the right wheel and a spark-out pass. Pushing for Ra 0.2–0.8 μm on aluminium usually means a second operation with a finer wheel.

Titanium and Inconel need the lowest wheel speed and the heaviest coolant flow. The vertical layout helps because the coolant floods the arc, but the wheel still has to be dressed often. If the spark stream turns from white to orange, the wheel is glazing and the part is heating. We stop and dress rather than push through the pass.

A ground bore is only as good as the measurement behind it. We check size with a bore gauge or an air gauge, roundness with a roundness tester, and perpendicularity to the mounting face with a dial indicator on a height stand. Size alone tells you nothing about lobing, and lobing is exactly what a face-clamped part can hide.

Roundness readings should be taken at more than one height in the bore. A bore can be round at the top and lobed at the bottom if the wheel is not dressed square or the chuck face is not flat. We record readings at three heights and compare them. A spread above a few micrometres means something in the setup is moving.

For parts that will run at speed, we also look at the surface finish direction. A vertical grinder leaves a crosshatch pattern that depends on the wheel feed and the chuck speed. That pattern affects oil retention in a bearing bore. If the print calls for a specific lay, we set the chuck speed and the wheel feed to match it rather than grinding to size and hoping.

All of this assumes the previous operation left enough stock. Grinding removes a small amount of material, and it cannot correct a bore that is out of position by 0.2 mm. We ask for a pre-grind dimensional report on first articles so the grinding stock is known before the wheel touches the part.