The difference between circular runout and total runout
On the drawing of an aerospace drive shaft, the runout symbol directly decides which inspection set-up to build. Choosing between circular runout and total runout changes the probe path, the inspection time and whether axial form errors are caught. Misreading this symbol leads either to accepting bent parts or to scrapping good bearing seats for nothing.
Circular runout and total runout under ISO 1101
The difference between circular runout and total runout lies in how the probe moves while the part turns about its datum axis. Circular runout assesses the error on independent circular sections, without moving the indicator along the part, whereas total runout sweeps the entire surface by combining rotation with continuous axial or radial travel.
| Characteristic | Circular runout (↗) | Total runout (⌰) |
|---|---|---|
| Indicator movement | Fixed on each section | Continuous travel along the surface |
| Form errors controlled | Ovality, local eccentricity | Ovality, eccentricity, taper, overall bending |
| Relative movement | Pure 360° rotation | 360° rotation + longitudinal or radial sweep |
| Severity of the tolerance | Local, per cross-section | Overall, across the whole enveloping surface |
Circular runout (a single slanted arrow, ↗) calls for measuring radius variations section by section. Total runout (two parallel slanted arrows, ⌰) covers at once coaxiality, roundness, cylindricity and straightness of the generating lines.
Inspecting radial or axial circular runout
Radial circular runout is checked by placing the stylus of a dial indicator or an inductive probe perpendicular to the cylinder surface. The part is mounted between centres or on two V-blocks that define the datum axis specified on the drawing.
While the part turns through a full 360 degrees, the probe stays still along the length. The difference between the highest and lowest dial readings gives the circular runout for that section.
To approve a whole bearing seat, the inspector has to repeat this on several separate sections spread along the cylinder. If the tolerance reads ↗ 0.015 A-B, each section on its own must show an error of 15 microns or less, without adding up the variations from one section to the next.
Axial circular runout follows the same logic, but the probe sits parallel to the axis of rotation, on a flat shoulder. The measurement checks the local perpendicularity error on a given circular trace.
How total runout is inspected across the whole seat
Total radial runout needs a mechanical slide or a CNC spindle able to move the probe continuously parallel to the datum axis while the part keeps turning.
The theoretical path of the contact point is a tight helix over the part surface. The tolerance ⌰ 0.020 A-B means that the absolute difference between the highest reading recorded along the whole seat and the lowest one must not exceed 20 microns.
This method catches at once the taper of a shaft or a longitudinal waviness. A part with a conforming circular runout of 10 microns on each individual section can show a total runout of 35 microns if its mean diameter flares from one end to the other.
The analysis of ISO 1101 geometric symbols sets out how these specifications apply to features of revolution.
A worked inspection on a turbine shaft
An inspector has to check a ⌀40 h6 bearing seat, 90 mm long, on an Inconel 718 turbine shaft. The assembly drawing calls for a total radial runout tolerance of ⌰ 0.018 A-B relative to datum seats A and B.
The part sits on a ground inspection bench fitted with two carbide V-blocks. The datum axis A-B is the line joining the centres of the two supporting seats. The LVDT probe is connected to a high-resolution acquisition unit calibrated at 20 °C.
- Set the probe at the left end of the 90 mm seat.
- Zero the digital indicator on the starting point.
- Turn the shaft at a steady 15 revolutions per minute.
- Feed the probe automatically along the seat at 1 mm per second.
- Record the extreme readings continuously over the 90 mm of travel.
During the 90-second sweep, the lowest point recorded comes out at -0.005 mm and the highest at +0.016 mm.
Total runout found = 0.016 - (-0.005) = 0.021 mm
The measured total runout is 0.021 mm (21 microns). With the tolerance set at 0.018 mm (18 microns), the error exceeds the limit by 3 microns. The verdict on the inspection report is non-conforming.
Had the operator only checked circular runout on three sections (left, middle, right), the errors would have been 0.012 mm, 0.014 mm and 0.011 mm respectively. The shaft would have been wrongly accepted, although it had an overall taper unacceptable for bearings running at high speed.
Practical pitfalls in form metrology
The first pitfall is the choice of datum supports. V-blocks that are too short or badly aligned introduce false runout. If datum seats A and B are themselves out of round, that error carries straight into the measurement of the seat being inspected.
The second comes from the probe’s feed rate in total runout. If the probe moves too fast for the part’s rotation speed, the pitch of the helix becomes too wide. Whole areas of the surface escape inspection, which amounts to an incomplete circular runout check.
The third concerns surface finish. A pronounced roughness (Ra above 1.6 µm) makes the ball stylus oscillate slightly, which adds to the macro-geometry error. A probe ball of a suitable diameter (usually 2 mm or 3 mm) mechanically filters out the machining marks without spoiling the runout measurement.
To bring these complex geometric tolerances into your inspection packages correctly, CoteCote handles automatic GD&T symbol detection and carries them faithfully into your dimensional inspection tables.