A bench Center looks like one of the simplest fixtures in a toolroom — two Centers on a rigid bed, holding a workpiece so it can be rotated. But it’s the standard tool for a whole family of checks on cylindrical parts, and getting reliable results from it depends on understanding what each check actually reveals. This guide covers mounting a workpiece correctly, then measuring radial runout, concentricity across multiple diameters, and tapering along a shaft’s length.
What a Bench Center Is For
A bench center supports a cylindrical workpiece between two opposed Centers so it can be rotated freely about its own axis. That rotation is the whole point: it lets an indicator sweep the full circumference of a feature and reveal how much the surface deviates as the part turns. It’s a distinctly different job from a comparator stand, which holds an indicator against a stationary part.
Typical applications include checking shafts, spindles, gear blanks, rollers, and turned components after machining, along with verifying that a part’s Center holes were drilled true. Because the fixture establishes the axis of rotation mechanically rather than relying on chucking, it gives a clean reference for exactly the kinds of checks that matter on turned work. The construction and material choices behind these fixtures are covered separately.
Mounting a Workpiece Between Centers
Start by cleaning both the workpiece’s Center holes and the fixture’s Center points. Debris or a burr in a Center hole shifts the axis of rotation slightly and produces runout readings that reflect the mounting rather than the part — one of the most common false results in this kind of inspection.
Adjust the tailstock so the part is held firmly enough that it can’t move axially, but not so tightly that it’s preloaded or deflected. A part squeezed between Centers under excessive force can bow measurably, and a slender shaft is particularly susceptible. The part should rotate smoothly by hand with no perceptible play and no binding.
A light film of oil on the Center points reduces friction and helps the part turn evenly, which makes indicator readings smoother and easier to interpret. Before taking any measurements, rotate the workpiece through a couple of full turns by hand to confirm it runs freely and to seat it properly in both Centers — parts occasionally settle slightly on first rotation, and it’s better for that to happen before you zero the indicator than midway through a set of readings.
Measuring Radial Runout
Position a dial indicator so its contact point rests on the surface you’re checking, with the stem perpendicular to the axis of rotation. Zero the indicator, then rotate the workpiece slowly through a full revolution while watching the needle. The total variation between the highest and lowest readings is the radial runout at that position.
Rotate through at least two full revolutions to confirm the pattern repeats — a reading that differs between revolutions usually means the part is shifting in the Centers or the indicator contact is slipping. Note that runout at a single position tells you about that cross-section only, so features spanning a length need checking at several points.
Checking Concentricity Across Multiple Diameters
Where a part has several turned diameters, the question is often whether they share a common axis. Set the indicator on each diameter in turn, taking a runout reading at each, without disturbing the part’s mounting between measurements. Comparing these readings shows whether one section is offset relative to the others.
It’s worth being precise about terminology here, because concentricity has a specific and fairly strict definition in geometric tolerancing that differs from what a simple runout sweep measures. In much practical shop work, a runout check is used as a proxy, which is usually adequate — but if you’re inspecting a drawing that specifically calls out concentricity, it’s worth understanding how runout, concentricity and circularity differ.
Detecting Taper Along a Shaft
Taper is a change in diameter along the length of a cylindrical feature, and a bench Center setup detects it by comparing measurements at different axial positions. Take a reading near one end, then move the indicator along the axis in steps and repeat, keeping the indicator stem perpendicular throughout.
A consistent progression in readings from one end to the other indicates taper; readings that rise and then fall suggest barrelling or waisting instead. Because you’re comparing positions rather than sweeping a single circumference, this check is sensitive to the indicator being repositioned squarely each time, so a stable mounting arrangement matters.
For longer shafts it’s worth distinguishing genuine machined taper from self-weight sag. A slender workpiece supported only at its ends deflects downward in the middle, and if you’re taking readings from above, that sag reads as though the Center of the shaft is undersized. Checking from the side rather than the top, or supporting the midpoint, separates the two effects.
Common Sources of Error
Most bad bench Center results trace back to a handful of causes. Contaminated or damaged center holes top the list, followed by excessive tailstock pressure deflecting the part, and indicator stems set at an angle rather than perpendicular to the axis.
Thermal effects also matter more than people expect. A shaft straight off a lathe is warm and will read differently once it stabilises, so parts should be allowed to reach ambient temperature before inspection. And the fixture itself needs to sit on a stable, level foundation — typically a certified plate within a properly organised inspection setup — because a bench Center resting on a surface that flexes or rocks undermines everything measured on it.
FAQs
Can a bench Center check a part without Center holes?
Not between Centers — the method depends on Center holes to establish the axis of rotation. Parts without them are usually checked using V-blocks or a rotary fixture instead.
How much tailstock pressure is correct?
Enough that the part cannot move axially and rotates without play, but no more. Excessive pressure can bow slender shafts and produce runout readings that reflect deflection rather than part geometry.
Does runout include out-of-roundness?
Yes — a circular runout sweep captures the combined effect of the feature being off-axis and being out of round, which is why it’s a useful screening check but not a substitute for a dedicated roundness measurement.
Should the indicator contact a machined or unmachined surface?
Always a machined surface where possible. An as-cast or as-forged surface has irregularities far larger than the deviations you’re trying to detect, making the reading meaningless.