Torque Specifications in Manufacturing: How to Document Them Correctly

Undertorqued fasteners loosen. Overtorqued fasteners fail. Both are documentation problems as much as process problems — here is how to write torque callouts that prevent both.

Why Torque Specifications Are Safety-Critical Documentation

A torque specification in a work instruction must include five elements: the value, the tolerance, the units, the friction condition (dry vs. lubricated), and the tightening sequence for multi-fastener joints. Missing any one of these elements creates the conditions for systematic fastener failure — not a random event, but a repeatable deviation that accumulates across every part built against that procedure. A 2019 aerospace safety bulletin traced fastener undertorquing to a unit mismatch between the work instruction (foot-pounds) and the operator's calibrated wrench (Newton-meters). The error rate was 100%.

In June 2019, a commercial aircraft manufacturer issued a safety bulletin after discovering that a batch of engine cowling fasteners had been torqued to approximately 60 percent of their specified value. The bolts had not failed. They had not loosened catastrophically. But vibration data showed an anomalous signature that, in isolation, would not have been caught until a scheduled maintenance interval — weeks later, in flight.

The root cause was not an operator who did not know how to use a torque wrench. It was a work instruction that specified torque in units of foot-pounds when the operator's calibrated wrench was marked in Newton-meters. The operator converted in their head. The conversion was off by a factor of roughly 1.36. The number felt right. The bolts felt tight. Nothing in the procedure flagged the discrepancy.

This is what inadequate torque documentation produces: not always catastrophic failure, but systematic deviation that accumulates until it is discovered or until it is not.

Dry vs. Lubricated Torque Values: The Distinction That Gets Skipped

Most torque specifications are developed assuming a specific friction condition between the fastener and the mating surface. The two conditions are "dry" — no lubricant, no thread compound — and "wet" or "lubricated" — with a defined lubricant applied to threads, bearing surface, or both.

This distinction matters because torque is a proxy for clamp load. What you are actually trying to achieve is a specific tension in the fastener. Torque is how you measure your way to that tension through a relationship that depends heavily on friction. Change the friction coefficient and the same applied torque produces a different clamp load.

Applying a dry torque value to a lubricated fastener overtightens the joint. The reduced friction means more of the applied torque goes into stretching the fastener and less into overcoming friction. Applying a lubricated torque value to a dry fastener undertightens it for the same reason in reverse.

Every torque specification in a work instruction must state the friction condition it assumes.

If the specification calls for anti-seize compound, the torque value must be the anti-seize value — typically 65 to 75 percent of the dry value for most common anti-seize compounds. If the specification calls for thread-locking compound, the torque value must account for any torque contribution from the compound itself.

Procedures that list a torque value without specifying the friction condition are incomplete. They look complete. Auditors who do not know to look for this will not catch it. But any experienced fastener engineer will.

Torque Sequence and Patterns

For multi-fastener joints — flanges, covers, mounting plates, head bolts — the sequence in which fasteners are torqued affects the final clamp load distribution. Tightening one corner fully before moving to the opposite corner introduces distortion that changes the effective friction at adjacent fasteners.

Standard torque patterns for multi-fastener joints:

Cross-pattern tightening is the most common approach for rectangular or circular flanges with four or more fasteners. Fasteners are tightened in opposing pairs — diagonally across the pattern — to distribute load evenly as the joint comes together.

Concentric-outward is used for large circular flanges where the center fasteners should be snugged before outer fasteners to prevent buckling or distortion of the joint face.

Sequential with multiple passes is required when fastener yield load is close to the specification torque value, or when joint distortion could significantly change the effective torque at adjacent fasteners. Typically: 30 percent of final torque first pass, 70 percent second pass, 100 percent final pass.

Work instructions must specify not just the torque value but the complete tightening sequence. "Torque to 45 N-m" applied to an eight-bolt flange without a sequence specification is an invitation to eight operators to do it eight different ways.

How to Write Torque Specifications in Work Instructions

A compliant torque specification in a work instruction contains five elements:

1. The value and tolerance — "45 plus or minus 2 N-m" not "approximately 45 N-m" and not "45 N-m minimum"

2. The units — stated explicitly, not assumed from context

3. The friction condition — "dry threads, bare mating surface" or "MIL-PRF-907 anti-seize applied to threads"

4. The tightening sequence — for any joint with more than one fastener

5. The torque wrench type — "calibrated click-type torque wrench" or "torque-angle wrench per engineering specification ES-4421"

The sequence notation should be visual when possible. A diagram showing fastener positions numbered in the correct tightening sequence eliminates interpretation error entirely. An operator should be able to follow the diagram without reading the surrounding text.

Common notation mistakes: using "ft-lbs" when the specification is in "in-lbs" (a factor of 12 difference), writing "N-m" when the engineering source document is in "N-cm," specifying a minimum torque without specifying a maximum (undertorque protection only, no overtorque protection), and omitting the tolerance entirely ("torque to 45 N-m" leaves no upper bound).

What Auditors Look For

AS9100 Rev D and IATF 16949 both require work instructions to include sufficient detail to ensure consistent process execution. For torque specifications, auditors will look for:

Value completeness. Does the work instruction contain a specific torque value, tolerance, and units for every fastened joint that requires controlled torque? References to "per drawing" without the value included in the work instruction are acceptable only if the drawing is present at the workstation and the operator does not need to locate it to execute the step.

Traceability to engineering source. Where did the torque value come from? Auditors may ask for the engineering analysis, test data, or specification that establishes the torque requirement. A torque value that cannot be traced to an authoritative source is a finding.

Calibration status of equipment. The torque wrench used to apply the specified torque must be calibrated. The calibration must be current. Auditors will ask to see the calibration records for equipment used in critical assembly operations.

Sequence documentation. For any joint where the sequence matters — and it matters for most multi-fastener applications — the sequence must be documented.

Torque Wrench Calibration Requirements

A torque specification is only as good as the instrument used to apply it. Torque wrenches are measuring instruments subject to calibration requirements under AS9100 Clause 7.1.5 and IATF 16949 equivalent clauses.

Minimum calibration requirements for torque wrenches in certified manufacturing environments:

Calibration interval — typically annual for infrequently used tools, more frequent for tools in daily production use. Some customers specify calibration intervals contractually.

Calibration traceability — calibration must be traceable to national measurement standards (NIST in the U.S., NPL in the U.K., or equivalent). A calibration sticker from an unaccredited service shop does not meet this requirement.

Accuracy class — most industrial torque wrenches are specified at plus or minus 4 percent accuracy. For tight-tolerance applications, higher-accuracy instruments (plus or minus 2 percent) may be required by the engineering specification.

Pre-use check — for critical assembly, requiring a visual check of calibration currency before each use is a documented best practice that satisfies auditors who look for evidence of verification at the point of use.

Torque Verification Methods

Applying a torque specification is not the same as verifying that the specification was met. For critical fastener applications, verification is a separate documented step.

Breakaway torque verification — after the fastener is torqued and the wrench released, re-apply the torque wrench to confirm the breakaway (initial movement) torque meets the specification. This confirms the torque was applied correctly and that the fastener was not undertorqued.

Torque-angle verification — used when both torque and angle of rotation are specified. After reaching the torque value, verify that the angle of rotation falls within the specified range. Angle that is too low indicates undertorque; angle that is too high may indicate thread damage or thread compound issues.

Stripe marking — a visual indication applied across the fastener head and mating surface after final torque. If the stripe breaks (fastener rotates), the assembly requires re-verification. Simple, inexpensive, and effective as a tamper-evidence measure.

Torque audit — a quality hold point in the work instruction requiring a QC verification of torque before the assembly proceeds. Required for safety-critical joints where the torque specification is the single control preventing a failure mode with serious consequence.

The appropriate verification method depends on the consequence of improper torque application. For safety-critical joints, the control plan must specify the verification method, frequency, and responsible party.

Coplain's Torque Spec Extractor identifies every torque callout in your engineering documentation and generates correctly formatted work instruction steps with value, tolerance, units, friction condition, and sequence. Try it free at coplain.com.

Stop reading about better documentation. Start creating it.

Coplain turns any work instruction into a print-ready, audit-proof job aid in minutes.