Home/6-axis sensors/Calibration interval

METROLOGY / CALIBRATION INTERVAL PLANNING

6-Axis Force Sensor Calibration Interval: Decision Guide & Worksheet

Estimate a review horizon from drift and error budget. Check overload triggers, then document the next action.

Calibration interval decision worksheet

JavaScript is disabled. The interactive worksheet requires JavaScript to calculate dynamic intervals. Review the ILAC-G24 standard methods and worked tables below or contact engineering for interval consultation.

Illustrative planning worksheet. All defaults and presets are invented examples. Enter values from a current verification on one axis and load point, then repeat for the other relevant points. Unknown drift or uncertainty must not be entered as zero. See the protocol below if evidence is missing.

Choose an industrial application scenario

1. Baseline & Historical Drift Rates

Current Absolute Error (% FS)0–20% FS; same axis and verification point. Document date and uncertainty coverage.

Expanded Uncertainty (% FS)0–20% FS; same axis and verification point. Document date and uncertainty coverage.

Owner-Approved Remaining Time (Months)0–60 months remaining to existing due date; default 12 is illustrative

Maximum Permissible Error (±% FS)0.05–20% FS; use the application requirement

Zero Drift Rate (% FS / Year)0% to 25% FS/yr (creep / aging)

Span Drift Rate (% FS / Year)0% to 25% FS/yr (sensitivity decay)

Planning Threshold (% of MPE)40–100%; 80% is an example, not a standard decision rule

2. Operating Duty & Dynamic Fatigue

Hours / Day1 to 24 hrs

Days / Week1 to 7 days

Dynamic Load (% FS)0% to 200% FS

3. Overload History & Robot Collisions

Reported Overload / Collision IncidentsWhole count 0–50 since last calibration; include unknown peaks

Max Recorded Peak (% FS)0–500% FS; any incident needs review. Enter 0 only for an unknown event peak.

4. Environmental Severity & Quality Risk

Thermal Swing ΔT (°C)Ambient fluctuation range

Vibration / Shock Level

Environmental Exposure

Out-of-Tolerance Consequence

In-Service Check Standard

Target TURTarget TUR and check frequency give no interval credit; actual uncertainty is entered above.

No result yet. Review the illustrative inputs, then calculate; use the static protocol if evidence is missing.

Core Metrological Decisions for 6-Axis Recalibration Intervals

Set an owner-approved review ceiling

Use the remaining time to the existing approved due date as the ceiling. A worksheet result cannot extend that date or release equipment.

Source basis: ILAC-G24:2022 / OIML D 10:2022. Metrological recommendations represent HexaForce engineering interpretations; cited standards do not endorse this specific worksheet.

Separate drift rate from present error

A rate in % FS/year does not show current conformity. Enter current absolute error and expanded uncertainty from a dated, accepted verification.

Source basis: BIPM JCGM 100:2008 — Measurement uncertainty. Metrological recommendations represent HexaForce engineering interpretations; cited standards do not endorse this specific worksheet.

Check every relevant channel

Use one axis and its own full-scale value per calculation. Repeat for Fx/Fy/Fz and Mx/My/Mz, load signs and points; keep the earliest candidate.

Source basis: ATI Force/Torque Sensor FAQ — published 2020-03-30. Metrological recommendations represent HexaForce engineering interpretations; cited standards do not endorse this specific worksheet.

Treat load incidents as review triggers

This worksheet withholds a calendar recommendation after any reported overload or above-range loading. Model-specific inspection determines what happened.

Use checks as evidence, not credits

Monthly or quarterly checking alone earns no extra months. Evaluate what each check can detect and preserve the readings and acceptance limits.

Source basis: ILAC-G24:2022 / OIML D 10:2022. Metrological recommendations represent HexaForce engineering interpretations; cited standards do not endorse this specific worksheet.

Agree the acceptance decision

The chosen MPE fraction is a planning limit. Actual uncertainty and an agreed decision rule are needed for a conformity statement.

Source basis: ILAC-G8:09/2019 — Decision rules. Metrological recommendations represent HexaForce engineering interpretations; cited standards do not endorse this specific worksheet.

Keep forecasts conditional

Zero measured slope, a quiet environment or a high target TUR cannot demonstrate future stability. The stress model is an illustrative sensitivity exercise.

Source basis: ILAC-G24:2022 / OIML D 10:2022. Metrological recommendations represent HexaForce engineering interpretations; cited standards do not endorse this specific worksheet.

1. Who approves the interval?

Separate published guidance from the local measurement plan. ILAC-G24:2022 describes review approaches; it does not certify this calculator. The responsible quality or metrology authority must approve the actual schedule, considering the manufacturer and any applicable contractual or regulatory requirements.

Approved ceiling → Current verification → Owner review. See the table for required evidence and limitations.

1. Who approves the interval? — engineering reference data
Decision item	Known / required evidence	Action / limitation
Schedule owner	Existing approved due date and responsible person	Enter remaining months, not a new full cycle from today
Laboratory	Scope, ranges, uncertainty and agreed deliverable	Confirm capability before booking; no service accreditation is claimed here
Guidance edition	OIML D 10 / ILAC-G24, 2022 edition	Use the linked edition; older 2007 references are superseded for this guide
Local procedure	Accepted records and review criteria	Record why a candidate was accepted, shortened or rejected

Source basis: ILAC-G24:2022 / OIML D 10:2022. Metrological recommendations represent HexaForce engineering interpretations; cited standards do not endorse this specific worksheet.

2. Build a channel-specific evidence record

Log matched readings under repeatable mounting, temperature, electronics and matrix settings. A zero change alone cannot identify material damage, moisture, mounting motion or electronics drift. The table gives diagnostic questions, not universal fault thresholds.

Match configuration → Check each channel → Investigate residuals. See the table for required evidence and limitations.

2. Build a channel-specific evidence record — engineering reference data
Decision item	Known / required evidence	Action / limitation
Zero	Unloaded readings before tare, with temperature	Investigate a jump; do not erase the only as-found evidence
Span	Known load, sign, point and channel full scale	Convert sensitivity drift to % FS/year on the same basis as zero
Cross-axis	Unloaded-channel residuals during reference loading	Check separately; no fixed matrix condition-number trigger is assumed
Configuration	Serial, matrix revision, amplifier, cable, filter and mounting	Compare only matched configurations; validate after changes

Source basis: ATI Force/Torque Sensor FAQ — published 2020-03-30 · BIPM JCGM 100:2008 — Measurement uncertainty. Metrological recommendations represent HexaForce engineering interpretations; cited standards do not endorse this specific worksheet.

3. Method 1: staircase review

Staircase review uses calibration outcomes to reconsider the next interval. Step sizes and acceptance bands belong in the approved local procedure. A single favorable result is weak evidence for extending a heterogeneous fleet.

As-found record → Local decision limits → Approve next review. See the table for required evidence and limitations.

3. Method 1: staircase review — engineering reference data
Decision item	Known / required evidence	Action / limitation
Clearly within local limits	As-found results and uncertainty	Consider retaining the approved schedule
Approaching a local limit	Trend and repeatability evidence	Review earlier and investigate the cause
Outside accepted limits	Affected channels, dates and uses	Contain affected work and define the impact review
Proposed extension	Comparable history and risk review	Separate owner approval; no automatic percentage increase

Source basis: ILAC-G24:2022 / OIML D 10:2022. Metrological recommendations represent HexaForce engineering interpretations; cited standards do not endorse this specific worksheet.

4. Method 2 context: control charts and the planning bound

A control chart uses repeat observations and scatter. This worksheet does not fit a chart or a confidence interval. Its original linear bound uses drift magnitudes at full scale on one axis: D = Z + S. It assumes the supplied slopes bound future drift; changes in environment can invalidate that assumption.

A − error − U → Divide by Z + S → Cap by owner ceiling. See the table for required evidence and limitations.

4. Method 2 context: control charts and the planning bound — engineering reference data
Decision item	Known / required evidence	Action / limitation
Planning limit A	MPE × selected fraction / 100, in % FS	The fraction is user selected; 80% is an example
Remaining margin B	A − current absolute error − expanded uncertainty	B ≤ 0 gives review now, not a diagnosis of OOT
Drift bound D	Zero magnitude + span magnitude, % FS/year	Not RSS; no independence assumption for systematic drift
Linear horizon	12 × B / D months when B > 0 and D > 0	Time from the current verification; zero D means no forecast

Source basis: BIPM JCGM 100:2008 — Measurement uncertainty · ILAC-G24:2022 / OIML D 10:2022. Metrological recommendations represent HexaForce engineering interpretations; cited standards do not endorse this specific worksheet.

5. Method 3 context: usage and sensitivity factors

In-use scheduling needs a justified usage budget. This demo applies unvalidated factors to the owner ceiling to explore sensitivity. They are disclosed here so the result is reproducible; they are not fatigue-life coefficients, OOT probabilities or values prescribed by a standard. Drift is not multiplied again by these factors.

Log usage → Demo stress factors → Candidate for review. See the table for required evidence and limitations.

5. Method 3 context: usage and sensitivity factors — engineering reference data
Decision item	Known / required evidence	Action / limitation
Duty	max(1, hours/day × days/week ÷ 40)	40 hours is an illustrative normalization, not a universal reference
Load factor	≤70% FS: 1; >70–90: 1.3; >90: 1.7	Thresholds are illustrative; above 100% also activates event review
Environment	Temperature: 1 + 0.035 × max(0, ΔT − 10); vibration: 1 / 1.1 / 1.5; exposure: 1 / 1.1 / 1.35	Multiply these demo factors; benign conditions earn no extension
Risk and synthesis	Risk weight: low/moderate 1, high 0.75, safety-critical 0.5; Tstress = ceiling × weight / product of factors	Candidate = min(remaining owner ceiling, Tdrift, Tstress); no one-month floor or 36-month standards cap

Source basis: ILAC-G24:2022 / OIML D 10:2022. Metrological recommendations represent HexaForce engineering interpretations; cited standards do not endorse this specific worksheet.

6. Method 4 context: checks and their blind spots

Design a check around the failure it should detect. Passing a partial check cannot validate unexcited load directions, dynamic response or every matrix coefficient. Check frequency changes the monitoring plan here, never the numerical horizon.

Traceable reference → Known load coverage → Log check residuals. See the table for required evidence and limitations.

6. Method 4 context: checks and their blind spots — engineering reference data
Decision item	Known / required evidence	Action / limitation
Unloaded check	Stable pose and temperature, raw readings	Catches some offsets; does not test span
Reference force	Traceable force or mass with local gravity and geometry	Verifies the loaded directions and points only
Reference torque	Known force and perpendicular lever arm	Include arm, alignment and parasitic-force uncertainty
Gravity orientations	Known payload and center of mass	Do not assume independent six-component excitation or complete matrix validation

Source basis: ILAC-G24:2022 / OIML D 10:2022 · ATI Force/Torque Sensor FAQ — published 2020-03-30. Metrological recommendations represent HexaForce engineering interpretations; cited standards do not endorse this specific worksheet.

7. Overload and unexpected events override the calendar

Measurement range, safe overload and breaking load are different specifications. No generic 125% or 150% FS damage boundary is used. A recorded incident causes review even when its peak was clipped or unknown; simultaneous forces and torques can violate a combined-load envelope below any single-axis full scale.

Incident or range issue → Withhold calendar → Inspect and verify. See the table for required evidence and limitations.

7. Overload and unexpected events override the calendar — engineering reference data
Decision item	Known / required evidence	Action / limitation
Any reported overload	Count > 0, even if peak is 0 / unknown	Withhold interval and preserve event evidence
Above-range peak or duty	Peak >100% FS or dynamic amplitude >100% FS	Review even when the recorded event count is zero
Shock, unexplained shift or saturation	Waveform and device status; peak may be unobserved	Stop relying on the forecast and seek competent inspection
Return to measurement	Model-specific limits, checks and disposition	Repair, recalibration or release follows evidence, not this tool

8. Planning margin and conformity decisions

The current absolute error and expanded uncertainty must refer to the same channel and point. Subtracting U here reserves planning margin; it does not implement a universal decision rule. Coverage, distribution and acceptance criteria need separate agreement.

Tolerance and error → Actual uncertainty → Agree decision rule. See the table for required evidence and limitations.

8. Planning margin and conformity decisions — engineering reference data
Decision item	Known / required evidence	Action / limitation
Current error	Absolute as-found error at the evaluation point	A rate cannot substitute for this value
Expanded uncertainty U	Document value, units and coverage from the measurement	Unknown U must not be entered as zero
Selected fraction	40–100% of MPE is the tool input range	This is a UI range, not a standards recommendation
Target TUR	Requested ratio, not measured uncertainty	No time credit and no zero-risk claim; actual U remains required

Source basis: ILAC-G8:09/2019 — Decision rules · BIPM JCGM 100:2008 — Measurement uncertainty. Metrological recommendations represent HexaForce engineering interpretations; cited standards do not endorse this specific worksheet.

9. Environment and configuration change limits

A stable historical trend applies only while its measurement conditions remain relevant. The calculator cannot infer adhesive chemistry, flexure material, seal integrity or thermal lifetime from a duty schedule. Use a recheck to distinguish a changed setup from changed sensor performance.

Conditions change → Recheck setup → Reassess history. See the table for required evidence and limitations.

9. Environment and configuration change limits — engineering reference data
Decision item	Known / required evidence	Action / limitation
Temperature change	Record sensor temperature and stabilization	Recheck outside the conditions represented by the history
Washdown or ingress	Protection rating, exposure and observed symptoms	Inspect against the specific device instructions
Mounting or tooling change	Bolt procedure, adapter, payload and reference frame	Repeat installation-specific verification
Electronics or matrix change	Version, scale, filter, cable and amplifier settings	Preserve provenance and validate the measurement chain

10. Balance calibration cost and measurement risk

A useful planning comparison includes removal, installation verification and the cost of discovering bad measurements late. Prices and failure probabilities are unknown for this application; the page therefore gives cost inputs rather than an invented optimum curve.

Lab and downtime cost → Detection coverage → Owner trade-off. See the table for required evidence and limitations.

10. Balance calibration cost and measurement risk — engineering reference data
Decision item	Known / required evidence	Action / limitation
Shorter schedule	Lab fee, freight, downtime and reinstallation work	Lower time between reviews can cost more; it does not guarantee conformity
Longer schedule	Comparable stability data and accepted risk	Requires separate approval and monitoring; tool never grants extension
Spare rotation	Spare cost and interchangeability verification	Avoid treating a spare matrix or cable as interchangeable without evidence
In-situ checks	Fixture, labor, uncertainty and detectable failure modes	May reduce disruption; preserve lab coverage for untested behavior

Source basis: ILAC-G24:2022 / OIML D 10:2022. Metrological recommendations represent HexaForce engineering interpretations; cited standards do not endorse this specific worksheet.

11. Repeatable interval-review protocol

Use the worksheet to prepare a decision record. All defaults and named scenarios are invented examples, not product performance or customer case histories. If the inputs are unknown, retain the existing approved schedule and collect a baseline rather than creating zero values.

Verify inputs → Evaluate constraints → Record decision. See the table for required evidence and limitations.

11. Repeatable interval-review protocol — engineering reference data
Decision item	Known / required evidence	Action / limitation
1. Identify	Serial, configuration, axis, point, due date and evaluator	Enter remaining approved horizon and current verified error budget
2. Review history	Matched dated results, slopes and scatter	Use conservative justified drift magnitudes; inspect anomalies
3. Evaluate events	Overload, shock, saturation and configuration changes	Resolve event review before relying on a calendar candidate
4. Approve and retain	Inputs, model version, memo, sources and owner decision	Set actual due/check dates in the controlled system; re-evaluate on change

Source basis: ILAC-G24:2022 / OIML D 10:2022. Metrological recommendations represent HexaForce engineering interpretations; cited standards do not endorse this specific worksheet.

12. Choose an evidence approach

The five guidance methods support different evidence situations. A hybrid plan may combine calendar review and specific checks. Fleet statistical modeling needs comparable populations and an appropriate model; this page does not prescribe a minimum sample count or calculate confidence.

Available evidence → Suitable method → Validate local plan. See the table for required evidence and limitations.

12. Choose an evidence approach — engineering reference data
Decision item	Known / required evidence	Action / limitation
Staircase	Comparable as-found pass/fail history	Simple review; steps must be locally justified
Control chart	Repeated values, conditions and scatter	Trend assessment; worksheet linear bound is not a fitted chart
In-use time	Logged hours and validated usage relationship	Retain calendar/event constraints for aging and changes
In-service checking	Traceable reference and known coverage	Detects only failure modes the check actually exercises
Other statistics	Comparable cohort and model diagnostics	Avoid pooling unrelated loads, configurations or environments

Source basis: ILAC-G24:2022 / OIML D 10:2022. Metrological recommendations represent HexaForce engineering interpretations; cited standards do not endorse this specific worksheet.

Reproduce the worksheet example

Actual calibration interval worksheet result for the illustrative electronics assembly preset — Original browser capture, November 4, 2026: the default Electronics Assembly Pick-and-Place Cobot preset after Calculate. Inputs are invented; the complete parameter set appears in the worked example below. This demonstrates software execution, not measured sensor stability, customer results or laboratory accreditation. Only the result banner is captured; site navigation is outside this image.

Run the Free Interval Worksheet

13. Five Illustrative Worked Examples

These are invented scenarios calculated by the worksheet, not customer histories or physical sensor tests. Full inputs and decision flows let you reproduce each example and compare the planning assumptions.

24/7 Robotic Grinding & Deburring

Operational Premise: Illustrative inputs, not measured product data: owner remaining ceiling 12 months; MPE 1% FS; fraction 80%; current error 0.1% FS; expanded U 0.1% FS. Zero/span drift 0.85/0.45% FS/year. Duty 24 h/day × 7 d/week; load 85% FS. Events 2, peak 120% FS; ΔT 15°C; vibration severe; exposure washdown; risk high; checks none; target TUR tur_4_1.

Calculation & Mathematical Weighting: D = Z + S = 1.3% FS/year. A = 0.8% FS. Linear horizon: 5.54 months. Demonstration stress product 12.99; stress/risk horizon 0.69 months. Candidate takes the minimum with the owner ceiling, unless an event or exhausted margin requires review now.

Engineering Action Verdict & Protocol: Review Now — No Calendar Recommendation. A reported overload or above-range load requires the model-specific single-axis and combined-load limits, inspection and verification before return to measurement. Damage is not diagnosed by this worksheet. Preserve as-found readings, event loads and configuration; ask the manufacturer or competent laboratory for inspection and a verification / recalibration plan.

Aerospace Wind Tunnel 6-Component Balance

Operational Premise: Illustrative inputs, not measured product data: owner remaining ceiling 12 months; MPE 0.3% FS; fraction 85%; current error 0.1% FS; expanded U 0.1% FS. Zero/span drift 0.08/0.05% FS/year. Duty 8 h/day × 4 d/week; load 60% FS. Events 0, peak 90% FS; ΔT 2°C; vibration benign; exposure clean; risk high; checks monthly; target TUR tur_10_1.

Calculation & Mathematical Weighting: D = Z + S = 0.13% FS/year. A = 0.255% FS. Linear horizon: 5.08 months. Demonstration stress product 1; stress/risk horizon 9.00 months. Candidate takes the minimum with the owner ceiling, unless an event or exhausted margin requires review now.

Engineering Action Verdict & Protocol: Shorten Interval: Review Candidate 5.08 Months. This is an illustrative review horizon from the current verified check, capped by your approved remaining time. It is not a calibration due date, release decision or authorization to extend an interval. Document monthly checks with traceable loads, uncertainty and channel-specific limits. A passing partial check does not validate the full 6×6 matrix.

Medical Teleoperation & Surgical Research

Operational Premise: Illustrative inputs, not measured product data: owner remaining ceiling 12 months; MPE 0.5% FS; fraction 75%; current error 0.1% FS; expanded U 0.1% FS. Zero/span drift 0.12/0.08% FS/year. Duty 6 h/day × 5 d/week; load 35% FS. Events 0, peak 70% FS; ΔT 2°C; vibration benign; exposure clean; risk safety_critical; checks quarterly; target TUR tur_4_1.

Calculation & Mathematical Weighting: D = Z + S = 0.2% FS/year. A = 0.375% FS. Linear horizon: 10.50 months. Demonstration stress product 1; stress/risk horizon 6.00 months. Candidate takes the minimum with the owner ceiling, unless an event or exhausted margin requires review now.

Engineering Action Verdict & Protocol: Shorten Interval: Review Candidate 6.00 Months. This is an illustrative review horizon from the current verified check, capped by your approved remaining time. It is not a calibration due date, release decision or authorization to extend an interval. Document quarterly checks with traceable loads, uncertainty and channel-specific limits. A passing partial check does not validate the full 6×6 matrix.

Electronics Assembly Pick-and-Place Cobot

Operational Premise: Illustrative inputs, not measured product data: owner remaining ceiling 12 months; MPE 1.5% FS; fraction 80%; current error 0.1% FS; expanded U 0.1% FS. Zero/span drift 0.15/0.1% FS/year. Duty 8 h/day × 5 d/week; load 30% FS. Events 0, peak 65% FS; ΔT 5°C; vibration moderate; exposure clean; risk moderate; checks quarterly; target TUR tur_4_1.

Calculation & Mathematical Weighting: D = Z + S = 0.25% FS/year. A = 1.2% FS. Linear horizon: 48.00 months. Demonstration stress product 1.1; stress/risk horizon 10.91 months. Candidate takes the minimum with the owner ceiling, unless an event or exhausted margin requires review now.

Engineering Action Verdict & Protocol: Shorten Interval: Review Candidate 10.91 Months. This is an illustrative review horizon from the current verified check, capped by your approved remaining time. It is not a calibration due date, release decision or authorization to extend an interval. Document quarterly checks with traceable loads, uncertainty and channel-specific limits. A passing partial check does not validate the full 6×6 matrix.

Zero reported slope — no extension

Operational Premise: Illustrative inputs, not measured product data: owner remaining ceiling 12 months; MPE 1.5% FS; fraction 80%; current error 0.1% FS; expanded U 0.1% FS. Zero/span drift 0/0% FS/year. Duty 8 h/day × 5 d/week; load 30% FS. Events 0, peak 65% FS; ΔT 5°C; vibration moderate; exposure clean; risk moderate; checks quarterly; target TUR tur_4_1.

Calculation & Mathematical Weighting: D = Z + S = 0% FS/year. A = 1.2% FS. Linear horizon: not established (zero reported slope). Demonstration stress product 1.1; stress/risk horizon 10.91 months. Candidate takes the minimum with the owner ceiling, unless an event or exhausted margin requires review now.

14. Frequently Asked Questions

Metrology Standards & Authority

Is there a universal 12-month interval?

No. Start with the applicable manufacturer, quality-system and contractual requirements. Review actual records with the responsible metrology authority.

Is this calculator an ILAC-approved method?

No. ILAC-G24:2022 supplies review context. The original linear bound and stress factors are disclosed planning assumptions, not endorsed or validated coefficients.

Can a lab help set the schedule?

Yes. Seek competent technical advice and agree deliverables. Do not interpret a certificate or this webpage as overriding the owner-approved schedule or applicable requirements.

Does ISO require a 36-month maximum?

This guide makes no such claim. The worksheet is capped by the owner-approved remaining time; its 0–60 month input range is a software boundary.

Does HexaForce offer accredited calibration here?

No accreditation or specific service capability is established by this page. Ask for the proposed laboratory, scope, ranges and uncertainty before procurement.

Operational Triggers & Drift Limits

Is annual drift the current measurement error?

No. A rate needs elapsed time and an initial state to predict error. Enter current absolute error separately from a dated verification.

Why add zero and span drift instead of RSS?

For this full-scale bound they are systematic magnitudes on one axis. Adding avoids assuming independent uncertainty contributions; it is not a complete uncertainty model.

What if the drift rate is zero?

No drift forecast is established. The tool retains owner and stress constraints, with no extension credit. Unknown or unresolved drift must not be entered as zero.

What happens after a 120% FS event?

The tool returns review now. Actual damage and permitted return to use depend on the specific single-axis and combined-load limits, inspection and verification.

Can loading below 100% FS still be excessive?

Yes, simultaneous forces and torques can exceed a device-specific combined-load envelope. This tool does not calculate that envelope.

What if the result is less than one day?

Review now rather than waiting a minimum month. Resolve the error budget and verification status with the responsible metrologist.

Does a zero shift prove plastic deformation?

No. Preserve as-found readings and investigate mounting, temperature, wiring and mechanics. A single symptom cannot establish the physical cause.

Intermediate Checks & Verification Protocols

Do monthly checks add 50% to the interval?

No. Frequency alone proves neither stability nor coverage. The calculator applies no check-frequency credit; any extension requires a separate documented review.

Does a gravity tumble check verify all six components?

Do not assume that. Loads and moments from a fixed payload are coupled; assess independent excitation and untested matrix terms before claiming complete coverage.

Can a shunt check replace mechanical loading?

An electrical check does not establish the complete mechanical force/torque response. Confirm exactly which portions of the measurement chain it tests.

What uncertainty goes into the worksheet?

Use actual expanded uncertainty for the same verification point in % FS, with coverage documented. A desired TUR cannot replace measured or evaluated U.

How should I handle six different full scales?

Convert each channel using its own full-scale force or torque. Evaluate relevant points and signs separately and retain the earliest review horizon.

Can the result approve safety-critical use?

No. That requires application-specific validated measurement controls and safety analysis. A numerical candidate is not release authorization.

Cost, Downtime & Scheduling Strategy

Can I use the candidate as a new due date?

Only after the responsible authority reviews the evidence and records an actual decision. The tool never extends the approved remaining time.

What should a recalibration request include?

Serial, matrix and electronics revisions, mounting, loading ranges, current errors, uncertainty, drift history, incidents, due date and required laboratory scope.

Does a spare remove reinstallation checks?

No. Verify configuration, matrix identity, installation and application performance before accepting the spare into service.

What can I do without reliable history?

Retain the approved schedule, collect a traceable baseline and planned interim checks, and request a competent review. The static protocol remains usable without JavaScript.

15. Sources, Review Date and Limitations

Authored and reviewed by HexaForce Metrology & Robotics. Technical guide formally established November 4, 2026; sources checked and illustrative model reviewed November 4, 2026.

Primary sources were checked on November 4, 2026. This guide uses the 2022 edition of ILAC-G24 / OIML D 10. The original model, sensitivity coefficients and examples are not prescribed or validated by those sources. No material lifetime, universal overload boundary or accredited service capability is established here. Source scope is stated individually below. Review this guide at least every six months, or sooner after a source, worksheet or configuration change.

ILAC-G24:2022 / OIML D 10:2022
Sections 4–6: interval selection, records and five review methods. Guidance, not a source for the worksheet coefficients or a universal interval.
ILAC-G8:09/2019 — Decision rules
Sections 3–7: uncertainty, guard bands and agreed conformity decisions. A guard band does not eliminate decision risk.
BIPM JCGM 100:2008 — Measurement uncertainty
Sections 5.1–5.2 distinguish uncorrelated and correlated inputs. RSS is not a general rule for adding systematic drift magnitudes.
ATI Force/Torque Sensor FAQ — published 2020-03-30
Sections 2.1 and 3.1 discuss measurement range, overload and load monitoring. Manufacturer-specific evidence; its ratings do not transfer to HexaForce configurations.

Transducer drift rates, thermal aging coefficients, and overload resilience depend on specific robot tooling dynamics, mounting rigidity, and environmental protection. Recalibration intervals must be formally approved by the responsible metrology or quality authority. HexaForce does not guarantee that any transducer will remain in-tolerance for the estimated duration without routine in-situ intermediate checks.

Return to the calibration interval decision worksheet. To audit an incoming calibration certificate or verify matrix uncertainty, visit 6-axis force sensor calibration certificate review. For loading arm design and deadweight rig error budgets, see 6-axis calibration fixture design. To qualify a candidate calibration facility and screen against the "uniaxial scope trap", see 6-axis calibration laboratory selection. For general load cell calibration theory, see 6-axis load cell design and calibration.

Prepare an Interval Review or Laboratory Request

Share your installed sensor model, operating schedule, recorded overload history, and target tolerance envelope. Request review of the evidence and laboratory requirements. Confirm service availability, scope, ranges and uncertainty before arranging recalibration.

Request an Interval Evidence Review ↗