Shaft Runout & Straightness: TIR Measurement Guide | SHANGDA

Every straightening machine conversation eventually reaches the same sentence: "What TIR can you guarantee?" Buyers who can answer that question precisely — value, length basis, measurement points — get accurate quotes and machines that pass acceptance. Buyers who cannot get arguments after delivery. This guide explains runout and TIR in plain terms, how straightness is measured by hand and by automatic machines, where the measurement points go, and exactly how to write the requirement on your inquiry.

By SHANGDA Engineering TeamReading time: 9 minutesFor: quality & production engineers

The one rule behind this guide: a tolerance number without a measurement method is not a specification — it is a hope. "0.05 mm" means nothing until you define 0.05 mm of what, measured where, over what length, with the part supported how.

01Runout, TIR and straightness: what these words actually mean

When a shaft is supported on V-blocks or between centers and slowly rotated, a dial indicator touching its surface will swing up and down. That swing is radial run-out — the total movement of the needle through one full revolution. The difference between the highest and lowest reading is the Total Indicator Reading (TIR), sometimes called Total Indicated Runout.

TIR is not a single geometric error. On a real shaft it bundles together everything that makes the surface move under the probe: bending (the shaft is curved — the part straightening removes), out-of-roundness of the cross-section, and any eccentricity between the measured surface and the support journals.

Pure straightness is a slightly different geometric concept — how close the shaft's axis is to a perfect straight line, independent of roundness. In straightening practice, however, the industry measures and guarantees radial TIR because it captures exactly what matters in service: a bent shaft running in bearings will move, and that movement is what a dial indicator — or a machine — sees. When a drawing or a machine builder says "straightness of 0.05 mm", in nearly all production contexts it means 0.05 mm TIR measured on the part rotating in supports.

02Why the whole industry speaks TIR

TIR won out as the standard language for three practical reasons:

• It is measurable with cheap tools — V-blocks, a dial indicator and a surface plate — anywhere in the world, so buyer and seller can check the same thing.

• It mirrors service conditions — a shaft rotates in bearings or between centers; TIR measured that way predicts vibration, seal wear and bearing life directly.

• It is what an automatic straightening machine measures — the machine's probes read the same physical quantity as the inspector's dial gauge, so "guaranteed TIR" can be verified on the acceptance floor with familiar methods.

This last point is the key to a smooth machine purchase: negotiate the TIR value and the measurement method together, in writing, before ordering. (How that requirement feeds into machine selection is covered in our shaft straightening machine buyer's guide.)

03How TIR is measured manually

The classic inspection setup, unchanged for decades:

1. Support the part on two precision V-blocks resting on a surface plate, or between dead centers — always on machined journals or centers, never directly on a rough surface.

2. Place a dial indicator (or digital probe) perpendicular to the shaft surface at the first measurement point, with the plunger preloaded.

3. Rotate the shaft one full turn by hand, slowly. The needle's high and low readings give TIR at that point: TIR = max − min.

4. Move the indicator along the shaft and repeat at each measurement point; record the values per point.

5. The worst point governs — the part is only as straight as its largest TIR.

A skilled inspector also "walks" the probe along the surface while rotating, to catch the location of maximum bend — which is exactly where a straightening press must apply force.

04Why manual measurement breaks down in production

The V-block method is fine for incoming inspection of a few parts. It fails when straightening is a production process:

• It is slow: minutes per part at every check — and straightening needs measure → correct → re-measure cycles, multiplying the time. An automatic machine completes the full measure-correct-verify loop in 20–90 seconds per part.

• It is subjective: probe placement, rotation speed, reading the needle at the right instant — two inspectors on the same shaft can disagree by several microns.

• It measures too few points: a long shaft with a bend between inspection points passes the check and fails in service.

• It stops at the door: manual readings live on paper; there is no per-part data for traceability, SPC or customer claims.

05How an automatic straightening machine measures

A CNC straightening machine performs the same geometry as the inspector — support, rotate, probe — but with electronic probes and a control doing the arithmetic:

• The part rests on roller supports or between centers and is rotated automatically.

• Electronic probes (contact type on standard machines; high-resolution probes or laser systems on precision models) read run-out at every measurement point simultaneously or in sequence.

• The control builds a bend map of the shaft: where each high point is, and how large it is.

• It calculates the correction — press position and force — drives the servo-hydraulic ram, then re-measures in a closed loop until every point is in tolerance.

• Results are logged per part: run-out before and after, correction points, pass/fail — the traceability data automotive and tier-1 customers now expect.

The measurement tier is what separates machine performance levels: standard electronic probes hold the broad middle band of 0.02–0.05 mm TIR; high-resolution or laser configurations hold ≤0.02 mm for precision screws and piston rods. The machine's measurement system — not its press tonnage — is where tight tolerance is actually bought. (The automation and cost trade-offs are detailed in Hydraulic vs. CNC vs. Fully Automatic Straightening Machines.)

06Where to measure: points, supports and the sag problem

Measurement location is half the result. Three rules:

• Short parts: 2–3 measurement points (near each end and mid-span) are usually enough to characterize the bend.

• Long shafts and bars: 5 or more points along the length. A long shaft bends in curves a 3-point check misses; automatic machines commonly use 5–8+ probes on long-bed models.

• Support positions matter as much as probe positions. Measure at the wrong support spacing and the reading reflects the support layout, not the part.

The classic trap — gravity sag: a long, slender shaft laid horizontally on two end supports bends under its own weight, adding sag to the reading. Professional setups control this with proper support spacing (near the Airy-point positions), intermediate supports on long beds, or by referencing the machine's own fixture geometry. If a quoted tolerance on a 6-meter bar does not mention how sag is handled, ask.

07The three straightness tolerance tiers in practice

Note the "per meter" on the first tier: for long parts the requirement is often expressed as TIR over a length basis (e.g. 0.2 mm/m), because total deviation naturally accumulates with length. Always state whether your number is total TIR over the whole part or TIR per meter — confusing the two is a common source of quotation mismatches.

08Six measurement mistakes that distort every number

1. Dirty or damaged supports

A speck of swarf on a V-block or a burred journal shows up directly as run-out. Clean supports and journals before every check.

2. Confusing out-of-round with bend

TIR at one cross-section includes ovality. If the needle shows 2 high points per revolution, suspect roundness, not bending — the fix is grinding, not straightening.

3. Measuring hot parts

Shafts straight off the heat-treatment line are still moving thermally. Let parts reach ambient temperature, or readings change after they cool.

4. Too few measurement points

A mid-span bend between two probes is invisible. Long parts need 5+ points; specify the point count in your acceptance method.

5. Ignoring sag on long shafts

Self-weight deflection on horizontal supports inflates readings (see Section 6). Agree support spacing up front.

6. Vague acceptance language

"Straight within tolerance" without value, basis and method invites a stand-off at acceptance. Write the full method (Section 9).

09How to specify TIR on your straightening RFQ

A complete straightness requirement — the kind that gets you an accurate machine quote and a clean acceptance test — contains five elements:

The 5-part TIR specification

1. The value: e.g. 0.05 mm TIR (or 0.2 mm/m for long parts).

2. The length basis: total over full length, or per meter.

3. Measurement points: e.g. 3 points for shafts under 1 m, 6 points along the full length for long bars.

4. Support method: between centers, on V-blocks at the journals, or on machine rollers — and sag handling for long parts.

5. Incoming condition: measured run-out before straightening on your real heat-treated parts — this sets the machine's required tonnage and correction capacity.

Send that with your drawings and you should receive back a machine configuration with a guaranteed TIR in writing — and an offer to prove it. SHANGDA, for example, runs customer samples on its trial floor and delivers before/after run-out data with the machine, so acceptance is a formality rather than a negotiation.

10Frequently asked questions

Send your parts — see before/after TIR data before you decide

Every SHANGDA straightening machine is built around your workpieces and your TIR specification, then trial-run on your actual samples with full run-out reports. Since 2008, CE certified, closed-loop measurement and correction down to 0.02 mm.

Send Drawings & Tolerance RequirementsSee Automatic Straightening Machines