SAC-SINGLAS Accredited ISO/IEC 17025 Acc. No.LA-2023-0845-C Traceable to Singapore's NMC View accreditation
Service Guide

Torque Wrench Calibration in Singapore: Standards, Intervals, and What an Accredited Certificate Must Include

Torque wrenches must be calibrated to ISO 6789 to confirm accuracy within their rated tolerance. Wrong torque causes structural failures and safety incidents. Here is what calibration involves, how often it is required, and what a compliant SAC-SINGLAS certificate must state.

Unitest Editorial10 min readWritten by an ISO/IEC 17025 accredited lab
Torque wrench calibration being performed on a calibration bench at Unitest Instruments, Singapore
The short answer Torque wrench calibration in Singapore verifies that a torque tool applies the correct torque within its rated accuracy tolerance. Typically ±4% for Type I (indicating) and ±4% for Type II (setting/click) wrenches under ISO 6789:2017. Calibration is performed on a torque calibration bench using a calibrated torque transducer and reference standard traceable to the National Metrology Centre of Singapore (NMC). A proper calibration certificate states the model, serial number, calibration points, measured deviations, expanded measurement uncertainty, and the accreditation body, not just a pass/fail sticker. Industries that require torque wrench calibration in Singapore include aerospace, automotive, oil & gas, construction, and pharmaceutical equipment maintenance.

Key takeaways

  • ISO 6789:2017 is the international standard for hand torque tools. It defines two types (Type I indicating; Type II setting) and two classes (Class I ±4%; Class II ±4%) and specifies the calibration procedure including the number of calibration points and pre-load cycles.
  • Torque wrenches should be calibrated at a minimum every 12 months or after 5,000 cycles, whichever comes first. More frequently for safety-critical applications in aerospace, pressure vessel assembly, and oil & gas.
  • A valid ISO 6789 calibration certificate must state the calibration points tested, the measured torque values, the deviation from nominal, and the expanded measurement uncertainty. A sticker or a simple pass/fail is not compliant.
  • Calibration should be performed at three or more torque values across the tool's range (typically 20%, 60%, and 100% of full scale), with five readings at each point after three pre-load cycles.
  • A dropped, overloaded, or repaired torque wrench should be recalibrated immediately. Impact events can alter the click mechanism or spring, even without visible damage.

Why torque wrench calibration is not optional. The consequences of wrong torque

Torque is a safety-critical measurement. It is not a best-efforts guideline. The function of a properly torqued fastener is to clamp two mating surfaces together with enough preload that the joint remains secure under the service loads it will experience. Vibration, thermal cycling, pressure, or dynamic stress. When that preload is wrong, the consequences range from nuisance to catastrophic depending on the application.

Under-torque leaves insufficient clamping force. Fasteners loosen progressively under vibration. The loosening may be gradual and undetected, particularly in enclosed assemblies where the fasteners are not routinely visible. The end result can be a structural connection that fails, a pressure boundary that leaks, or a flange that separates. In safety-critical systems, loosened fasteners have been the root cause of aviation incidents, pipeline failures, and pharmaceutical equipment contamination events.

Over-torque introduces a different failure mode. Applying torque beyond the fastener's yield point permanently stretches the bolt shank. A yielded bolt has lost its elastic recovery. It cannot maintain clamping force reliably and may fracture under subsequent service loading. Flanges can be distorted; threaded holes in soft materials can be stripped; gaskets can be crushed beyond their design limits, eliminating their sealing function rather than enhancing it.

In Singapore's industrial landscape, the sectors where wrong torque carries the most serious consequences are well defined. Aerospace MRO facilities at Seletar and Changi (operating under CAAS and EASA Part 145 approvals), must use calibrated torque tools for any fastener application where the manufacturer's maintenance manual specifies a torque value. Boeing and Airbus specifications are explicit: the torque tool must be calibrated, the certificate must be current, and the traceability must be documented. Oil and gas facilities on Jurong Island apply specified torque values to piping flange bolts on process lines carrying hydrocarbons under pressure; under-torqued flanges leak and over-torqued ones distort gaskets. Pharmaceutical facilities under HSA GMP requirements document torque application for critical equipment connections in clean rooms. Where GMP calls for recorded evidence of the measurement, a calibrated tool and a current certificate are part of the documentation chain. Construction contractors on BCA structural projects must verify torque application on high-strength bolted steel connections to SS EN 1090; auditors increasingly ask for calibration evidence for the torque tools used.

A torque wrench that has drifted 8% from its set value (well within the range of drift that can accumulate if a wrench is used heavily and not calibrated), is applying torque 8% outside the fastener manufacturer's specification on every application. Calibration finds and corrects that drift before it causes a failure.

ISO 6789:2017. The standard that governs torque wrench calibration

ISO 6789 is published by the International Organization for Standardization and is the globally accepted standard for hand torque tools. It covers the design requirements, accuracy classes, and (critically), the calibration procedure for indicating torque tools (Type I) and setting torque tools (Type II). Singapore's SAC-SINGLAS accredited laboratories that perform torque calibration follow this standard as the basis for their accredited scope.

The standard defines two types of hand torque tool. Type I tools indicate the torque being applied (through a scale and pointer, a dial gauge, or a digital display), while the technician applies force. Type II tools are pre-set to a specific torque value and produce a tactile signal (typically a click or a release) when that value is reached; the operator stops applying force at that signal. Both types are in wide use in Singapore's industrial base, though Type II click wrenches are the most common tool found in manufacturing and maintenance environments.

The standard defines two accuracy classes. Class I and Class II both specify a maximum permissible error of ±4% of the indicated or set value. This means a wrench set to 100 Nm that clicks between 96 Nm and 104 Nm is within the ISO 6789 Class II tolerance. A wrench that clicks at 108 Nm is out of tolerance and should be adjusted and recalibrated before further use.

The 2017 revision of the standard introduced a requirement that was absent from the 2003 edition: calibration certificates issued under ISO 6789:2017 must state the expanded measurement uncertainty. This is a significant change. A certificate produced under the older standard (or by a lab that has not updated its procedure to the 2017 requirements), may omit uncertainty entirely. That certificate does not comply with the current standard. For organisations whose audit framework references ISO 6789:2017 specifically (as most current quality and GMP frameworks do), a certificate without stated uncertainty is a gap.

The standard also specifies environmental conditions for calibration: the calibration should be performed at 23°C ±5°C. Torque wrenches are calibrated at ambient temperature because their click mechanisms and spring properties are temperature-dependent; a wrench calibrated at 10°C in a cold warehouse may behave differently when used at 30°C on a production floor. The calibration environment requirement exists to ensure that the certificate reflects performance under representative conditions.

Direction of application matters. A torque wrench designed for clockwise tightening must be calibrated in the clockwise direction. Calibrating it anticlockwise will produce different results because the internal mechanism is direction-specific. The certificate must state the direction of calibration, and calibration records should confirm it matches the wrench's intended service direction.

ISO 6789 Type I vs Type II. How calibration differs

The two types of torque wrench require slightly different calibration approaches because the way torque is applied and the signal that marks the calibration point differ between them.

Feature Type I. Indicating Torque Wrench Type II. Setting (Click) Torque Wrench
How it works Measures and displays torque as it is applied Produces a tactile "click" when preset torque is reached
Display type Scale, dial, or digital readout No display. Preset scale only
Accuracy (ISO 6789) ±4% of indicated value ±4% of set value
Calibration method Applied torque measured against indicated reading Torque at which click releases measured against set value
Calibration points Typically 20%, 60%, 100% of range Typically 20%, 60%, 100% of range
Pre-load cycles 5 cycles before 5 measurement readings 3 pre-load cycles then 5 trigger readings
Direction sensitivity Clockwise only (standard); dual-direction models exist Clockwise only (standard); left-hand models for anticlockwise
Applications Assembly, quality control, tightening verification High-volume fastening, production lines, field maintenance
Most common in SG? Less common Very common. Most torque wrenches in use are Type II

For Type II wrenches, the calibration reads the torque transducer value at the precise moment the click mechanism releases. This requires a smooth, slow loading rate (typically under 2 Nm/s), because dynamic overshoot from a fast snap-loading can produce a transducer reading higher than the actual release torque. Technicians are trained to apply force gradually and read the transducer at the click event, not the peak reading that follows it.

How torque wrench calibration is performed

The calibration procedure follows the ISO 6789:2017 method, adapted to the wrench type. The process requires equipment and technique that together produce results traceable to the SI unit of torque (the newton metre), via an unbroken chain of calibrations back to the NMC.

Equipment required: A torque calibration bench with a calibrated reference torque transducer rated for the wrench's range; a torque reference standard traceable to NMI; a data acquisition system to record transducer readings at the calibration events; and, where applicable, a fixture to hold the drive square square relative to the transducer axis. The transducer's own calibration certificate must be current and its expanded uncertainty must be appropriately small relative to the wrench's tolerance. The uncertainty hierarchy rule requires the reference to be at least four times more accurate than the item being calibrated.

The procedure for a Type II click wrench proceeds as follows. First, the wrench is visually inspected for damage, corrosion, or mechanical faults. If any are found, the calibration is paused and the condition is documented. Second, the wrench is set to the first calibration point, typically 20% of full scale. Three pre-load cycles are then applied: the wrench is loaded to the set value and released, three times in succession, without recording. This pre-load procedure conditions the click mechanism and ensures the spring is in a settled state before measurement. After the three pre-load cycles, five sequential measurements are taken: the wrench is loaded smoothly until the click event, the transducer reading at that instant is recorded, and the wrench is fully released before the next loading. The mean of the five readings and the deviation from the set value are calculated. The wrench is then set to the next calibration point (60% of full scale), and the pre-load and measurement sequence is repeated. Finally, the procedure is repeated at 100% of full scale. The result is fifteen measured torque values across three calibration points, from which mean deviations and expanded uncertainty are calculated and reported on the certificate.

For Type I indicating wrenches, the procedure differs slightly: the wrench is loaded to a specific torque as measured by the reference transducer, and the wrench's own indicator reading at that point is compared against the reference. This tests how accurately the wrench's display tracks the actual applied torque. Pre-load cycles are five rather than three for Type I, and the measurement is the wrench's indicated reading at the calibration point, not the transducer reading at a click event.

Throughout the procedure, the calibration temperature is recorded. The wrench should be at thermal equilibrium with the calibration environment before the procedure begins. Bringing a wrench from a hot vehicle or a cold storage area and calibrating it immediately may produce results that are not representative of the wrench's settled performance.

Calibration intervals. When to calibrate and when to recalibrate early

The standard calibration interval for torque wrenches is 12 months or 5,000 operating cycles, whichever comes first. These two criteria address different failure modes. The 12-month interval catches degradation from storage conditions, environmental exposure, and the gradual relaxation of spring pre-load that occurs even when a wrench is used infrequently. The 5,000-cycle interval catches wear-driven drift in high-usage applications. The click mechanism has moving parts that wear, and that wear changes the release torque over time.

Why cycles matter more than time in many Singapore environments: a torque wrench used on a production line assembling automotive components may complete 5,000 cycles in six to eight weeks. A wrench used once a month for annual flange bolt verification on a pressure vessel may not accumulate 5,000 cycles in ten years. The appropriate interval for each wrench depends on its use, not just the calendar. A calibration programme that treats all wrenches identically (regardless of usage), will over-calibrate some tools and under-calibrate others.

Early recalibration triggers must be defined in any competent calibration programme, alongside the routine interval. Several events mandate immediate recalibration regardless of when the wrench was last calibrated:

  • Impact event. A wrench dropped onto concrete from working height (even one that shows no visible damage), must be recalibrated. The click mechanism's spring and cam geometry are sensitive; a sufficient impact can alter the pre-load of the click spring without breaking or visibly deforming anything. The wrench may still feel normal in the hand and still click. Just at the wrong torque value. This is the failure mode that is most frequently underestimated on site.
  • Overload. A wrench used beyond its rated maximum torque (whether deliberately or accidentally), may have been mechanically stressed beyond the design envelope. The click spring or beam may have experienced permanent deformation. Recalibrate before further use.
  • After repair or service. Any mechanical intervention (cleaning, adjustment, lubrication), requires fresh calibration. The intervention may have altered the click mechanism, and the only way to confirm it has not is to measure it.
  • Observed inconsistency. If an operator notices that a wrench clicks inconsistently (sometimes clicking early, sometimes late), that is a symptom of click mechanism wear or spring fatigue. Send it for calibration and inspection; do not continue using a tool that behaves erratically on safety-critical fasteners.
  • Before critical use after extended storage. A wrench stored for an extended period (months in a case, or stored at the maximum torque setting), may have experienced spring relaxation. Calibrate before use rather than assuming the last certificate still reflects current performance.

For aerospace applications in Singapore, many CAAS and EASA Part 145 approved facilities operate on a 6-month interval for torque tools used on flight-critical fasteners, regardless of cycle count. This shorter interval reflects the consequence of an out-of-tolerance result in that context: a calibration finding on a wrench used on engine mounts or flight control systems has a different severity than a finding on a wrench used for non-critical ground support equipment.

What a valid ISO 6789 calibration certificate must state

A calibration certificate for a torque wrench is a technical document, not a sticker or a pass/fail report. The content required by ISO 6789:2017 and ISO/IEC 17025 for an accredited certificate is specific. Understanding what must appear (and why), helps quality managers, auditors, and procurement professionals assess whether a certificate they have received is valid.

A compliant certificate must include:

  • Instrument identification. Make, model, type (Type I or Type II), range (minimum to maximum Nm), drive size, and serial number. Without the serial number, the certificate cannot be definitively linked to a specific tool, and serial number matching is how auditors trace a certificate to the wrench it covers.
  • Accreditation reference. The accreditation body (SAC-SINGLAS for Singapore) and the accreditation number (e.g. LA-2023-0845-C). The scope must cover the torque parameter and the range of the wrench being calibrated. If torque is not within the lab's accredited scope, the accreditation reference on the certificate is misleading.
  • Calibration date. The date the measurements were taken. Not the date the certificate was issued or the date the instrument was received at the laboratory.
  • Standard used. ISO 6789:2017 (or the specific revision the calibration was performed to). A certificate referencing ISO 6789:2003 was produced under the older version of the standard, which does not require measurement uncertainty to be stated.
  • Results table. For each calibration point: the nominal (set) torque value, the mean measured torque from the five readings, the deviation in Nm and as a percentage of the set value, and whether the result is within the ±4% tolerance. The individual readings (not just the mean), should be available and may be required by some audit frameworks.
  • Expanded measurement uncertainty. Stated as a value in Nm and as a percentage (e.g. U = ±1.2% at k=2, ~95% confidence). This is the quantity that ISO 6789:2017 added as a mandatory requirement over the 2003 version. Without it, a quality engineer cannot assess whether the wrench's calibrated value is fit for the tolerances its application requires.
  • Environmental conditions. Temperature at the time of calibration. As noted earlier, torque calibration results are temperature-dependent; the recorded ambient temperature confirms that calibration was performed within the ISO 6789 requirement of 23°C ±5°C.
  • Traceability statement. Confirmation that the reference torque transducer used is traceable to a national metrology institute. Specifically, to Singapore's NMC for a SAC-SINGLAS accredited certificate.
  • Authorised signatory. The name and signature (or equivalent authentication) of the technically authorised person who reviewed and issued the certificate. The signatory's authorisation must be documented in the lab's quality management system.

What is not sufficient for a regulated application: a sticker printed with "CALIBRATED. Due: [date]" applied to the wrench body; a certificate that shows only a single pass or fail notation without calibration data; a certificate that states the wrench was compared to a reference but does not state the measured deviations; a certificate with no measurement uncertainty stated; or a certificate from a lab whose SAC-SINGLAS accreditation does not cover the torque parameter.

SAC-SINGLAS Accredited Torque Wrench Calibration

ISO 6789 torque calibration. Accredited certificates for aerospace, oil & gas, and construction

Unitest calibrates torque wrenches to ISO 6789:2017. Stated uncertainty, three calibration points, CAAS/GMP-ready certificates. Most tools returned within 3–5 working days.

Torque calibration in Singapore's regulated industries

Singapore's industrial and regulatory framework creates specific, documented calibration requirements across four sectors where torque measurement is safety-critical or compliance-critical. Understanding what each sector requires helps organisations build the right calibration programme from the outset rather than discovering gaps during an audit.

Aerospace. CAAS and EASA Part 145

Maintenance, repair, and overhaul facilities at Seletar Aerospace Park and Changi Airport hold approvals from the Civil Aviation Authority of Singapore (CAAS) and, where they service European-registered aircraft, the European Union Aviation Safety Agency (EASA). Part 145 approvals require maintenance organisations to control all tools and equipment used in maintenance operations, including a system for calibrating measurement tools. Torque wrenches are among the most scrutinised tools during Part 145 audits: inspectors check calibration certificates, verify that serial numbers on certificates match the tools in use, and confirm calibration intervals comply with the facility's own approved maintenance organisation exposition. Facilities that cannot produce a current, accredited certificate for every torque wrench used on a flight-critical application are at risk of an audit finding that can trigger suspension of the Part 145 approval for that tool type.

Oil and gas. Jurong Island process industries

Jurong Island hosts the majority of Singapore's petrochemical and specialty chemical plants, operating under the Ministry of Manpower's Workplace Safety and Health (WSH) framework and the Environmental Protection and Management Act. Piping system integrity on Jurong Island depends in part on correctly torqued flange bolted connections. The joints between pipe flanges carrying hydrocarbons, chlorine, acids, and other process chemicals under pressure. Industry standards such as ASME PCC-1 (guidelines for pressure boundary bolted flange joint assembly) specify torque values and sequences for different flange sizes and gasket types. Plant owners are required to maintain calibrated measurement equipment for safety-critical operations; during WSH inspections and process safety management audits, calibration records for torque tools used in pressure equipment maintenance are reviewed. An out-of-calibration torque wrench used to assemble a flanged joint carrying a toxic or flammable medium is a process safety risk, not merely a quality management gap.

Pharmaceutical. HSA GMP

Pharmaceutical manufacturing and packaging facilities in Singapore operate under the Health Sciences Authority's Good Manufacturing Practice requirements, which align with WHO GMP, PIC/S GMP, and in some cases EU GMP Annex guidelines. GMP requires documented control of critical measurement equipment, and in clean room environments the torque applied to critical connections (equipment flanges, vessel closures, and sanitary fittings), is often specified in standard operating procedures. When a torque value is specified and documented in a GMP-controlled SOP, the wrench used to achieve that torque must be calibrated, the certificate must be current, and the calibration record must be retained as part of the batch or equipment qualification documentation. HSA inspectors conducting GMP facility assessments look for gaps between the SOPs that specify torque values and the calibration records for the tools used to apply them.

Construction. BCA structural steel

High-strength bolted connections in structural steel frameworks are specified to Singapore Standard SS EN 1090, the execution standard for steel and aluminium structures. This standard references EN 14399 for high-strength structural bolts and EN 13670 for concrete structures, both of which require torque-controlled tightening of structural bolts to specified values. Main contractors and structural steelwork subcontractors on BCA major project sites (particularly those operating under the BCA Quality Mark scheme or requiring structural engineer sign-off), may be required to produce calibration evidence for torque tools used on structural connections. This requirement has become more common as main contractors' quality management systems have matured and as structural engineers have become more specific about documentation of the tightening process. SAC-SINGLAS accredited torque calibration certificates are the appropriate evidence.

All four sectors share a common requirement: the calibration certificate must come from a lab whose SAC-SINGLAS accredited scope covers the torque parameter and the torque range of the specific wrenches being used. Verify the scope before submitting instruments, accreditation is scope-specific.

Common torque wrench mistakes that affect calibration and accuracy

Many torque wrenches arrive at calibration laboratories with avoidable damage or drift that could have been prevented by correct storage and use practices. Understanding the most common mistakes helps reduce the frequency of out-of-tolerance findings and the cost of recalibration or replacement.

1. Storing the wrench at its maximum setting. Click wrenches have an internal spring that maintains tension on the click mechanism. Storing the wrench at a high torque setting (or at its maximum), compresses the spring continuously. Over weeks or months, the spring can take a partial set, reducing its elastic recovery and causing the click to occur at a lower torque than the set value. The correct storage practice is to return the wrench to its minimum scale setting (typically 20% of full scale or the lowest marked increment), before storing. This releases the spring and preserves its calibrated tension.

2. Using the wrench to loosen fasteners. Click wrenches are designed for tightening in their rated direction. Using them to loosen fasteners applies torque in the reverse direction, which stresses the click mechanism in a way it was not designed for. The result can be click mechanism damage or drift. For loosening fasteners, use a standard ratchet or breaker bar; the torque wrench is for the final tightening step only.

3. Applying torque too quickly. A rapid snap-loading technique (pulling the wrench handle sharply until it clicks), adds dynamic inertia above and beyond the measured torque at the click event. The fastener may receive more torque than the wrench's setting, because momentum carries force into the fastener after the click. Slow, steady application (increasing load smoothly at roughly 2 Nm per second), allows the click mechanism to release at the correct value and gives the technician time to stop at the click signal.

4. Not accounting for extension bars or adapters. When an extension bar or crow's foot adapter is attached to the wrench drive, the effective torque arm changes. If the extension is not in line with the wrench handle, the applied torque to the fastener is different from the indicated value. The correction formula (adjusted torque = (wrench setting × wrench length) ÷ (wrench length + extension length)), must be applied when using off-axis extensions. Failing to account for this can result in significant over- or under-torque even with a correctly calibrated wrench.

5. Using a wrench near its minimum range. Torque wrenches are less accurate near the bottom of their operating range. The ISO 6789 calibration points of 20%, 60%, and 100% reflect this: 20% of full scale is the practical lower limit of reliable use. A 340 Nm wrench used at 10 Nm (less than 3% of full scale), is operating far outside its accuracy zone; a small, separate lower-range wrench is the correct choice for that application.

6. Calibrating in the wrong direction. A wrench calibrated clockwise must be used clockwise. The internal mechanism produces different click-release characteristics in each direction, and a clockwise calibration does not validate anticlockwise performance. Left-hand thread applications require a wrench rated and calibrated for anticlockwise operation, or a separate verification of the clockwise wrench's reverse-direction performance.

7. Skipping pre-load cycles in service. Before applying a calibration-critical torque in service (particularly on high-value assemblies), applying two or three pre-load cycles at the target torque value conditions the wrench's mechanism and gives more consistent results on the measurement cycle. The pre-load requirement in the calibration procedure exists for this reason; the same principle applies in critical service use.

8. Using a torque multiplier without recalculating effective torque. Torque multipliers apply a mechanical ratio to the input torque from the driving wrench. A 3:1 multiplier, for example, produces 300 Nm at the output from a 100 Nm input wrench setting. The driving wrench must be set to the target output torque divided by the multiplier ratio. Connecting a calibrated torque wrench to a multiplier and setting it to the target output torque will produce three times the intended torque at the fastener.

Choosing the right torque wrench for your application

Selecting a torque wrench with the right range for the application is a prerequisite for accurate torquing. A wrench used at an inappropriate point in its range (whether too close to minimum or operating at maximum), will produce less consistent results regardless of how recently it was calibrated.

Drive size and torque range are closely correlated. 1/4-inch drive wrenches are designed for low-torque applications in the 2–25 Nm range. Electronics assembly, small component fastening, and precision instrument work where torque values are in single-digit newton metres. 3/8-inch drive wrenches cover the 10–80 Nm range that encompasses most automotive, light engineering, and general workshop applications. 1/2-inch drive is the most common size for general engineering and structural applications, spanning roughly 20–340 Nm depending on the specific model. 3/4-inch drive covers the heavy engineering range of 100–700 Nm. Large flanges, heavy machinery, and industrial equipment. 1-inch drive wrenches extend to 1500 Nm and above for large pressure vessels, ship equipment, and heavy plant.

The range-matching rule: use the wrench in the middle third of its calibrated range whenever possible. For a 100 Nm wrench, the reliable working zone is approximately 33–67 Nm. If your application requires torque at the extremes of the range, a wrench specifically sized for that value will give more consistent and accurate results.

For Singapore's industrial applications, several brands are commonly specified or accepted by quality systems and aerospace customers. Norbar (UK) is the standard reference in aerospace and precision engineering. Norbar's TruTorque and Professional ranges are cited in many OEM maintenance manuals. Gedore (Germany) is widely used in automotive and general engineering. Snap-on and Proto/Stanley are established in professional automotive and industrial settings. Bahco covers general workshop use. All these brands manufacture tools that can be calibrated to ISO 6789:2017; the key is ensuring the specific model you purchase has stated accuracy performance consistent with its ISO type and class.

Beyond click wrenches, a note on digital torque wrenches: electronic torque wrenches with digital displays offer data logging capability, audible and visual alerts at the target torque, and often Bluetooth connectivity for records transfer. They are increasingly used in aerospace and pharmaceutical assembly where a digital record of each fastener tightening is required. Digital torque wrenches must still be calibrated to ISO 6789:2017. The electronic transducer inside requires the same traceable verification as any other torque-measuring system, and the same deterioration modes (wear, impact, battery-induced drift in some models) apply.

Managing a torque wrench calibration programme

A calibration programme for torque wrenches is not complicated, but it requires consistent discipline to be effective. The following practical steps cover the core of a functional programme suitable for ISO 9001, GMP, or CAAS Part 145 compliance contexts.

Step 1: Inventory and asset numbering. Assign a unique asset number to every torque wrench in use. Mark the number on the wrench using an engraver, a durable label, or a stamped tag. The asset number is the link between the physical tool and its calibration certificate. Without it, a certificate cannot be definitively matched to a specific wrench during an audit.

Step 2: Register the wrench details. Record each wrench in your calibration register: asset number, make, model, type (Type I or II), range (min–max Nm), drive size, serial number, calibration interval, last calibration date, calibration due date, and the name of the lab that holds the current certificate. A spreadsheet is adequate for small toolrooms; a dedicated metrology management system provides better visibility for larger organisations.

Step 3: Label each wrench with its calibration due date. Attach a durable calibration status label to each wrench showing the due date and the asset number. Green labels for in-calibration, red for overdue. This allows anyone picking up the wrench to see at a glance whether it is available for use on calibration-controlled applications.

Step 4: Store correctly. All wrenches should be stored at their minimum torque setting, in their original cases or in a dedicated, clean toolroom location, away from moisture and extremes of temperature. Wrenches stored correctly degrade significantly less between calibration events than wrenches stored carelessly in a shared toolbox at whatever setting they were last used at.

Step 5: Define and communicate early-calibration triggers. Document the events that mandate recalibration before the scheduled date (drop, overload, repair, observed inconsistency), in your maintenance SOP or quality procedure. Ensure all technicians who use torque wrenches know these triggers and understand that a wrench involved in a trigger event must be removed from service and submitted for calibration before further use.

Step 6: Retain certificates for the life of the wrench. Calibration certificates for torque wrenches used on safety-critical applications should be retained for the lifetime of the wrench, not merely until the next calibration. For aerospace and pharmaceutical applications, the retention requirement may extend beyond the tool's operational life. To the life of the products it was used to assemble. Define your record retention requirement in line with your quality management system and regulatory framework, not as a default calendar period.

Step 7: Retire wrenches that repeatedly fail calibration. A wrench that fails calibration, is adjusted, and fails again at the following calibration has a worn or damaged click mechanism. Continued use of a repeatedly non-conforming tool is a quality risk. Retire it, document the disposal, and ensure it cannot be returned to service accidentally. Some organisations mark retired instruments permanently (cut the drive square, paint the handle red, or physically destroy the mechanism), before disposal.

For aerospace and pharmaceutical facilities, traceability records for torque tools must be retained for the lifetime of the products they were used on, consistent with the relevant regulatory framework. This requirement makes the long-term certificate retention practice not just good management, but a regulatory obligation.

Unitest's torque calibration capability

Unitest Instruments Pte. Ltd. calibrates torque wrenches under its SAC-SINGLAS accreditation (no. LA-2023-0845-C), following the ISO 6789:2017 procedure for both Type I indicating and Type II setting torque tools. Calibrations are performed on a torque calibration bench with a reference transducer traceable to Singapore's National Metrology Centre, covering the range and drive sizes used across Singapore's industrial, aerospace, and regulated manufacturing sectors.

Every Unitest torque calibration certificate states the calibration points tested (minimum three points: 20%, 60%, and 100% of full scale), the individual and mean measured torque values at each point, the deviation from nominal in Nm and percentage, whether each point is within the ±4% ISO 6789 Class I/II tolerance, and the expanded measurement uncertainty. Stated as a percentage and newton metre value at k=2 (~95% confidence). The certificate includes the ambient temperature at calibration, the standard referenced (ISO 6789:2017), the accreditation number, and the authorised signatory. Certificates are formatted to satisfy CAAS Part 145 audits, HSA GMP documentation requirements, and ISO 9001 clause 7.1.5 compliance.

To submit torque wrenches for calibration, clients provide: the wrench itself (in its original case where possible), the asset or serial number, and the torque values at which calibration is required (or a request to calibrate at the three standard ISO 6789 points). Wrenches should be packaged to protect the click mechanism from impact during transit. A drop in shipping has the same effect as a drop on site and may require recalibration on receipt before meaningful calibration results can be obtained. Most standard torque wrenches are returned within three to five working days; urgent turnaround can be arranged on request. Contact Unitest to confirm range coverage, current turnaround times, and submission logistics before sending instruments.

Frequently asked questions

How often should a torque wrench be calibrated in Singapore?

Torque wrenches should be calibrated at a minimum every 12 months or after 5,000 operating cycles, whichever comes first. For safety-critical applications (aerospace MRO, pressure vessel assembly, oil and gas piping), many facilities require 6-month intervals. Any wrench that has been dropped, overloaded, or repaired must be recalibrated before further use, regardless of how recently it was last calibrated. High-usage production line wrenches may require calibration checks more frequently than the 5,000-cycle rule if operational experience shows earlier drift.

What standard does torque wrench calibration follow?

Torque wrench calibration follows ISO 6789:2017. The international standard for hand torque tools. It defines two types (Type I indicating; Type II setting/click) and two accuracy classes (Class I and Class II, both ±4%), specifies the calibration procedure including pre-load cycles, calibration points, and number of measurements, and (critically), requires measurement uncertainty to be stated on the certificate. A certificate without stated uncertainty does not comply with the 2017 revision of the standard, which is the version current quality and GMP frameworks reference.

Does a calibration sticker count as a valid torque wrench calibration certificate?

No. A sticker (or any certificate showing only a pass/fail result without calibration data), does not constitute a valid ISO 6789 calibration certificate. A compliant certificate must state the calibration points tested, the measured torque values at each point, the deviation from nominal, the expanded measurement uncertainty, the standard used (ISO 6789:2017), the accreditation body and number, and the technician's signature. Stickers are useful as a visual reminder of the calibration due date, but they carry no evidentiary value for an ISO 9001, GMP, or regulatory audit on their own.

Should I recalibrate a torque wrench after dropping it?

Yes, immediately and before further use. A drop onto a hard surface from working height can shift the click mechanism's spring pre-load or alter the beam geometry, even without any visible damage to the wrench. The change may be subtle (perhaps a fraction of a percent), but for safety-critical applications that is enough to cause a non-conformance or, in the worst case, a structural failure. The cost of recalibration is negligible compared to the liability. Treat any impact event as a mandatory recalibration trigger without exception.

At what torque values should calibration be performed?

ISO 6789:2017 requires calibration at a minimum of three points across the wrench's working range. Typically 20%, 60%, and 100% of the full scale value. For example, a 100 Nm wrench would be calibrated at 20 Nm, 60 Nm, and 100 Nm. Five measurement readings are taken at each calibration point, preceded by three pre-load cycles to the calibration value. This gives 15 data points in total across the range, from which mean deviations and expanded uncertainty are calculated. Some clients request additional calibration points at application-specific torque values, this is available on request.

Can Unitest calibrate torque multipliers and torque screwdrivers as well?

Yes. Unitest's torque calibration capability covers hand torque tools across the range from small torque screwdrivers used in electronics assembly through to large torque wrenches used in heavy engineering. Torque multipliers can also be calibrated. Note that when a torque multiplier is used in service, the input torque setting on the driving wrench must account for the multiplier ratio. Contact Unitest with the make, model, and torque range of the specific instruments to confirm coverage and confirm the correct calibration procedure before sending them in for calibration.

What is the acceptable accuracy tolerance for a click-type torque wrench?

Under ISO 6789:2017, both Type I (indicating) and Type II (setting/click) wrenches must achieve accuracy within ±4% of the set or indicated value. This is the Class I and Class II tolerance. For a wrench set to 100 Nm, the acceptable click-release range is 96–104 Nm. Some high-precision applications specify tighter tolerances through their own engineering standards; in that case, the torque wrench selected must have a rated accuracy that is a fraction of the process tolerance. A wrench used at the ±4% limit in a process requiring ±2% accuracy will not meet the process requirement even when the wrench itself is within its rated class.

SAC-SINGLAS accredited laboratory mark
Written by Unitest Instruments

Unitest Instruments Pte. Ltd. is a SAC-SINGLAS accredited calibration laboratory (ISO/IEC 17025, no. LA-2023-0845-C) based in Singapore. We calibrate torque wrenches, electrical instruments, temperature sensors, pressure gauges, and related measurement equipment for aerospace, oil & gas, pharmaceutical, and general manufacturing industries across Singapore and the region.

Torque wrench calibration in Singapore. ISO 6789, SAC-SINGLAS accredited

Unitest calibrates torque wrenches to ISO 6789:2017 with stated measurement uncertainty. Accredited certificates accepted by CAAS Part 145, HSA GMP, and ISO 9001 auditors.

Verifiable at sac.gov.sg · Acc. No. LA-2023-0845-C · ISO 6789:2017