Straightening Before Centerless Grinding: Runout, Hard Rolling and Grinding Stock

On a bolt line that ends in centerless grinding, straightening is not a separate quality check that happens somewhere upstream and stays there. It is a grinding-readiness input. The grinding step removes a limited radial stock. If post-straightening runout or surface damage is larger than that stock, the wheel cannot clean up the part all the way around.

That relationship is what ties the straightening target to the grinding allowance. On the documented project used here, the process chain is also explicit: heat treatment -> straightening -> hard rolling -> grinding -> surface treatment -> final inspection -> packaging. That sequence matters because each step changes what the next step has to absorb.

Table of Contents

• Straightening Is a Grinding-Readiness Input
• Work the Grinding Stock Backwards
• The Drawing Requirement and the Intermediate Target Are Not the Same Number
• Hardness and Heat Treatment Change the Straightening Window
• Why Hard Rolling Has to Stay in the Chain
• Length and Machine Boundaries Still Bind the Process
• Measure Runout the Way the Process Uses It
• Questions From the Shop Floor

Key Takeaways

• Post-straightening runout competes directly with the grinding stock; if it exceeds the stock, the wheel cannot clean up the surface fully.
• The documented process chain is heat treatment, straightening, hard rolling, grinding, surface treatment, final inspection, and packaging; hard rolling is not optional in this sequence.
• The drawing requirement of 0.060 mm runout over a 100 mm supported interval is not the same thing as the looser intermediate straightening target used before downstream finishing.
• Heat-treated hardness bands, part family, and length range all affect how wide the real straightening window is on the line.

Straightening Is a Grinding-Readiness Input

A centerless grinder uses the part’s own outside surface as the running datum. The work is supported by the grinding wheel, the regulating wheel, and the work blade. Nothing clamps the part into a perfectly straight axis before material removal starts. The machine grinds the envelope the part actually presents.

That is why post-straightening condition matters. If the residual runout stays inside the available radial stock, the grinding wheel can remove material all the way around and finish the part cleanly. If the runout is larger than the stock, the wheel will clean up the high side first and can leave black skin, under-clean-up, or local size loss on the low side. In that sense, straightening is not a separate pass/fail check. It sets whether grinding has enough allowance to finish the surface.

The documented project makes the process order visible as well: heat treatment -> straightening -> hard rolling -> grinding -> surface treatment -> final inspection -> packaging. This is a project-specific chain from the source archive, not a claim that every bolt line is arranged the same way.

Work the Grinding Stock Backwards

The grinding allowance on the documented line is recorded with explicit diameters:

Worked backwards, that gives the real stock budget:

The critical number for the straightening discussion is the minimum radial stock of 0.10 mm. That is the smallest cleanup depth available to the worst-case part in this project example. It is not a rated machine capability. It is a project-specific allowance budget taken from the recorded incoming and finished diameters.

The same source archive records intermediate post-straightening targets of 0.15-0.25 mm for the smaller family and 0.10-0.15 mm for the larger family. That overlap with the minimum radial stock is the reason the line has to treat straightening and grinding as one process decision.

The Drawing Requirement and the Intermediate Target Are Not the Same Number

The drawing wording in the source archive is more specific than the earlier article stated. It requires runout in the center area of the supported interval to stay within 0.060 mm when the measured distance is 100 mm. The archive records this as a TIR-based requirement.

That drawing requirement is not automatically the same as the intermediate straightening target. The intermediate values of 0.15-0.25 mm and 0.10-0.15 mm belong to a process stage before the later finishing chain is complete. The 0.060 mm requirement belongs to a specific measurement setup: supported interval, center area, 100 mm distance, and runout interpretation.

This distinction matters because the same nominal straightness number can mean different things when support span, datum, or reporting convention changes. A runout figure taken over 100 mm on a supported setup is not interchangeable with a looser in-process control value unless the process documentation says so. The safe reading is that the line uses an intermediate straightening target to keep the work inside the downstream finishing window, then evaluates the final condition against the drawing’s 100 mm / 0.060 mm runout requirement.

Hardness and Heat Treatment Change the Straightening Window

The source archive does record fastener grade ranges, but the stronger source-grounded way to describe straightening difficulty here is through heat-treated part condition. The documented hardness bands are 25-32 HRC, 27-34 HRC, and 32-38 HRC across the recorded part set.

That matters because straightening is applied after heat treatment in the project chain. Once the part is heat treated, the correction window is affected by how much elastic recovery remains in that hardness condition and by how sensitive the part is to over-correction or marking. A setting that works on one part family in a lower hardness band should not be assumed to transfer unchanged to another family in a higher hardness band.

For this reason, the safer process statement is not “higher property class parts are always harder to straighten” as a generic law. It is that the documented line contains several post-heat-treatment hardness conditions, and the straightening setup has to be validated by part family and hardness condition. The recorded grade span of 8.8-12.9 remains a project descriptor, but the published process explanation should stay anchored to the hardness values that are explicitly present in the source archive.

Why Hard Rolling Has to Stay in the Chain

The earlier draft skipped hard rolling and treated straightening as if it fed grinding directly. The source archive does not support that simplification. The documented route is heat treatment -> straightening -> hard rolling -> grinding.

That extra step changes how the process should be discussed. Hard rolling sits between shape correction and grinding, so the line is not only managing residual runout. It is also managing what surface condition reaches the grinder after hard rolling has done its work. Any marks, deformation, or geometry changes that survive hard rolling become part of what the grinding stock must absorb on the surfaces that will be ground.

This also keeps the surface discussion honest. Surfaces that will later be ground can absorb a limited amount of prior-process marking, but only if the mark depth stays inside the available stock. Surfaces that are not cleaned up in grinding do not get that second chance. That is why mark acceptability is a specification question tied to surface location and downstream removal, not a universal yes/no statement.

Length and Machine Boundaries Still Bind the Process

The source archive records two different straightening ranges: phi6-10 mm x 100-200 mm with stated precision to 0.1 mm, and phi10-14 mm x 140-330 mm with stated precision of 0.1-0.15 mm. It also records that the grinder originally handled 250 mm length while the project required 300 mm, and that the machine had to be modified to close that gap.

Those facts show two separate boundaries at once. First, part family and length range affect what straightening precision is being targeted on the line. Second, grinding length capability can become a physical machine boundary rather than a tuning issue. A process plan can therefore look acceptable on diameter arithmetic and still fail on work length.

The practical lesson is to review straightening range, grinding range, and final drawing requirement together. A line that can straighten a family is not automatically able to grind the same family to the required finished condition over the required length.

Measure Runout the Way the Process Uses It

The documented project includes two straightness gauges with dial indicators. That gives the line a measurement tool, but the useful part is the measurement definition around it.

Three points matter most:

1. Support the part the documented way. The drawing requirement is tied to a supported interval and a 100 mm distance, so support geometry is part of the requirement.
2. Treat the reading as runout, not a single-point offset. The source record frames the requirement as runout and interprets it as TIR.
3. Keep sample evidence in its lane. The recorded straightened outcomes of 0.05 mm and 0.1 mm are trial signals against a 0.10-0.15 mm requirement, not proof of throughput, Cpk, long-run repeatability, or final acceptance.

That distinction is important for publishing. The project source is strong enough to explain how the line thinks about straightening before grinding. It is not strong enough to claim long-run production capability beyond the recorded trial and acceptance boundaries. Gauge calibration and inspection records should remain traceable to the measurement system used for the drawing requirement; the NIST dimensional metrology programme is a public reference for that traceability context.

flowchart LR
    A[Heat treatment] --> B[Straightening]
    B --> C[Hard rolling]
    C --> D[Grinding input 9.40-9.50 mm]
    D --> E[Finish 9.17-9.20 mm]
    E --> F[Surface treatment]
    F --> G[Final inspection]
    G --> H[Packaging]

Process chain and allowance logic for the documented project: post-heat-treatment straightening and hard rolling feed a grinding step with 0.10-0.165 mm radial stock in the recorded example.

Questions From the Shop Floor

Can centerless grinding fix a bent bolt by itself?

No. Centerless grinding can remove surface stock from a part that is already inside the available cleanup window, but it does not create a new straight datum independent of the part. If residual runout is larger than the available stock, the part will not clean up fully.

How should I compare straightening and grinding requirements on a bolt line?

Start with the grinding input and finished diameter, convert the diameter stock to radial stock, and compare that with the in-process straightening target. Then separate that comparison from the final drawing requirement. On this project, the drawing uses a 100 mm supported interval and a 0.060 mm runout limit, while the intermediate straightening targets are looser process controls.

Why is hard rolling important in this discussion?

Because the documented project does not go straight from straightening to grinding. Hard rolling is part of the chain, so the grinder sees the condition left after both straightening and hard rolling. That affects whether the available stock can still clean up the surface.

What do the recorded 0.05 mm and 0.1 mm results prove?

They show that sampled trial parts reached the required in-process band on the documented line. They do not, by themselves, prove formal acceptance, long-run repeatability, throughput, or statistical process capability.

Conclusion

Straightening before centerless grinding is a downstream-process input because runout and surface condition determine whether the recorded grinding stock can finish the part cleanly. On the documented project, that decision sits inside a longer chain of heat treatment, straightening, hard rolling, grinding, and later finishing steps. The useful engineering check is to read the drawing requirement, the in-process target, and the grinding stock together instead of treating them as separate numbers.

References

• Dimensional Metrology | NIST Sensor Science Division — Supports the measurement-traceability context for gauge calibration and inspection records.