Bump Test vs Calibration for Gas Detectors: Not the Same

Defining the Terms: What Each Procedure Actually Does

The industrial hygiene and gas-safety community uses "bump test" and "calibration" as if they were synonyms. They are not, and the distinction carries life-safety consequences. A bump test (also called a functional test or response check), applies a brief pulse of test gas to the sensor and records whether the detector's audible alarm, visual alarm, and display respond within the manufacturer's specified response time. The typical acceptance criterion is alarm activation within 30 seconds at a concentration above the instrument's alarm set point. No concentration value is measured, no correction is applied, and no certificate is issued.

Calibration, by contrast, is a metrological act defined by the International Vocabulary of Metrology (JCGM 200:2012, the VIM) as an operation that, under specified conditions, establishes a relation between the quantity values with measurement uncertainties provided by measurement standards and corresponding indications with associated measurement uncertainties. In practical terms: a calibration technician exposes the detector to a certified reference gas mixture (CRM) at one or more known concentrations, records what the instrument displays, adjusts the instrument's span and zero if the error exceeds tolerance, and documents the before- and after-adjustment readings in a calibration certificate that references the CRM's traceability chain back to a national measurement standard, in Singapore, the National Metrology Centre (NMC).

The key output differences are stark. A bump test produces a pass/fail entry in a maintenance log. A calibration produces a certificate with serial number, calibration date, reference gas concentrations, pre- and post-adjustment readings, expanded measurement uncertainty, and the accreditation body's endorsement. Only the certificate satisfies regulatory, audit, and insurance requirements.

The Underlying Physics: Why Gas Sensors Drift and Why It Matters

Understanding why both tests exist requires a brief look at sensor physics. The three dominant sensor technologies in portable gas detectors (electrochemical cells, catalytic bead (pellistor) sensors, and non-dispersive infrared (NDIR) sensors), all degrade over time through different mechanisms, and each type of degradation is invisible to a bump test.

Electrochemical Sensors (O₂, CO, H₂S, SO₂, NO₂)

Electrochemical sensors function through a redox reaction at working and counter electrodes immersed in an electrolyte. The electrolyte is consumed in the reaction, and over months of normal use the electrolyte volume decreases, changing the sensor's output at a given gas concentration. A sensor that read 20.9% oxygen at commissioning may read 19.8% two years later, a 1.1% absolute error that is clinically significant because Singapore's confined-space entry threshold for acceptable oxygen is 19.5% to 23.5% (per WSH Confined Spaces Regulations). A bump test at 20.9% ambient air would still pass: the alarm would not trigger because ambient is not a hazardous concentration. Only a calibration using a certified span gas at a defined concentration (say, 18.0% O₂), would reveal whether the sensor is correctly reading below the alarm threshold.

Catalytic Bead (Pellistor) Sensors (LEL/Combustible Gas)

Catalytic bead sensors detect combustible gases by measuring the heat of oxidation on a platinum coil. The active bead can be poisoned by silicone compounds, lead, sulfur, and halogen vapours. All common in Singapore's petrochemical, semiconductor fabrication, and marine-painting environments. Poisoning suppresses the sensor's output: a poisoned pellistor may read 20% LEL when the actual methane concentration is 40% LEL, a dangerously low reading. A bump test at a concentration above the alarm set point (typically 10% LEL) will still trigger the alarm, giving a false pass, because even a poisoned sensor produces some output. Only a calibration at a certified 50% LEL or 25% LEL span gas, where the technician compares the displayed value against the known concentration, will reveal the suppressed sensitivity.

Non-Dispersive Infrared (NDIR) Sensors (CO₂, HC gases)

NDIR sensors are generally more stable than electrochemical cells, but they are sensitive to optical contamination of the light path. Condensation, dust, and hydrocarbon films on the optical windows reduce transmission and introduce measurement bias. Temperature coefficients also vary, and high-humidity environments (common in Singapore's tropical climate) can cause condensation-related errors. Again, a bump test that simply asks "does it alarm?" will not reveal a systematic 8% low bias across the measuring range.

Singapore's Regulatory Framework: What MOM Actually Requires

The Workplace Safety and Health (Confined Spaces) Regulations 2009 (updated under the WSH Act Cap. 354A) require that any instrument used for atmospheric testing of a confined space must be "properly maintained" and "calibrated in accordance with the manufacturer's instructions." The regulations do not prescribe calibration intervals numerically but defer to SS 531 and manufacturer guidance, which typically specify 3–6 months for the gas types used in confined-space monitoring (O₂, LEL, CO, H₂S).

Singapore Standard SS 531:2014 (Code of Practice for the Selection, Use, Care and Maintenance of Respiratory Protective Equipment), provides the practical calibration framework. It requires that gas detectors used in conjunction with RPE selection and confined-space entry be calibrated at intervals not exceeding 6 months by a competent person using traceable reference gases. The standard explicitly distinguishes between a functional check (bump test) and a calibration, stating that bump tests do not replace calibration.

At the bizSAFE Level 4 and Level Star tiers, an organisation's safety management system must demonstrate control of monitoring equipment under a clause analogous to ISO 9001:2015 Clause 7.1.5 (Monitoring and Measuring Resources), which requires a calibration program with traceable certificates and documented calibration status. A bump-test log does not satisfy this requirement. To understand what information must appear on a valid calibration certificate, see our detailed guide: What Should a Calibration Certificate Contain?

Comparing the Two Procedures: A Technical Summary

Calibration Procedure in Detail: What Happens at an Accredited Lab

When a portable gas detector arrives at an ISO/IEC 17025-accredited laboratory for calibration, the procedure follows a defined sequence that is audited by SAC-SINGLAS against the requirements of ISO/IEC 17025:2017. The process begins with incoming inspection: the technician records the instrument make, model, serial number, firmware version, and physical condition, noting any contamination or mechanical damage that could affect measurements.

The instrument is then conditioned in the calibration environment (temperature between 20°C and 25°C, relative humidity 45–55%), for at least 30 minutes to allow thermal stabilisation. Gas flow is applied via a calibrated mass flow controller at the manufacturer's specified flow rate (typically 0.5–1.0 L/min). The zero point is checked first using zero-grade air or pure nitrogen, and the zero offset is recorded. The span gas (a certified reference mixture with a stated concentration and uncertainty), is then applied, and the instrument's displayed value is recorded as the pre-adjustment reading.

If the pre-adjustment error exceeds the manufacturer's tolerance (commonly ±5% of reading or ±2% of full scale, whichever is greater), the instrument is adjusted to read correctly, and a post-adjustment reading is recorded. The calibration certificate documents both values, allowing the user to see how far the instrument had drifted. For multi-gas instruments, this process is repeated for each sensor channel using the appropriate CRM. The expanded measurement uncertainty of the calibration result (typically ±2–4% of reading at k=2 for electrochemical gas sensors), is calculated and reported in the certificate.

For a deeper understanding of traceability requirements and why it matters that your CRM links to a national measurement standard, refer to our article on What Is Metrological Traceability in Calibration?

Common Mistakes and How to Avoid Them

Mistake 1: Using non-certified bump-test gas for calibration

Some operators use the same non-certified or "nominal concentration" gas cylinders for both bump tests and calibration. Non-certified gas has no traceable concentration value. The label says "50 ppm CO" but there is no CRM certificate, no lot number, and no accredited laboratory has measured that cylinder's actual concentration. Using such gas for calibration is equivalent to using an uncalibrated ruler to calibrate another ruler. The resulting "calibration" has no metrological validity and will not satisfy an MOM inspector or ISO 9001 auditor.

Mistake 2: Accepting a calibration certificate without stated uncertainty

ISO/IEC 17025:2017 Clause 7.8.2 requires that calibration certificates report measurement uncertainty. A certificate that states only "Pass" or lists only a tolerance without an uncertainty value does not comply with ISO/IEC 17025 and is not issued by an accredited laboratory. Always verify the issuing laboratory's accreditation status and scope by checking SAC's online directory, where Unitest Instruments is listed under accreditation number LA-2023-0845-C.

Mistake 3: Extending calibration intervals without documented justification

Calibration intervals are not arbitrary. They are based on empirical drift data for the sensor type, operating environment, and usage intensity. SS 531:2014 and OSHA's confined-space standard (29 CFR 1910.146) both link calibration interval to the instrument's service environment. A gas detector used in a Singapore semiconductor fab with solvent vapours every day degrades far faster than one used weekly in a cleanroom. Extending the interval because "it always passes the bump test" is not a risk-based justification, it is confirmation bias. The bump test, as established above, cannot detect gradual drift.

Mistake 4: Calibrating only the span point and ignoring zero

Single-point span calibration is common in field practice but insufficient for instruments used in safety-critical applications. A sensor can have correct span sensitivity but a non-zero baseline offset. Particularly common in electrochemical CO and H₂S sensors that are cross-sensitive to hydrogen. A two-point calibration (zero check + span check) is the minimum acceptable procedure for confined-space gas detectors. Laboratory calibration typically includes a zero-gas check, a mid-scale span point, and optionally a linearity check at a second concentration.

Mistake 5: Failing to record the calibration gas CRM lot number

Traceability requires not just that you used a CRM, but that you can link the calibration result to a specific certified cylinder. If an incident occurs and the calibration is challenged, the investigation will ask for the CRM certificate of analysis showing the certified concentration, expanded uncertainty, and expiry date. Without the CRM lot number on the calibration record, the traceability chain is broken and the calibration evidence is inadmissible. Accredited laboratories maintain this documentation as part of their ISO/IEC 17025 quality management system.

Practical Compliance Programme: Integrating Both Tests Correctly

The most effective gas-detector maintenance programme treats bump tests and calibration as complementary layers of assurance, not alternatives. A typical programme for a Singapore construction or petrochemical operator looks like this: bump test every instrument before each day's entry into confined spaces, document the result in the confined-space entry permit log, and reject any instrument that fails. Send each instrument to an ISO/IEC 17025 accredited laboratory for formal calibration every 6 months (or 3 months for instruments used in harsh environments). Maintain calibration certificates in the instrument's asset file and present them on demand during MOM inspections.

For organisations running ISO 9001:2015 quality systems, the calibration certificate satisfies Clause 7.1.5.2 requirements for monitoring and measuring equipment when issued by an accredited laboratory. Understanding how calibration intervals should be set and reviewed is covered in detail in our article on How Often Should You Calibrate Your Instruments?

An important administrative detail: calibration status must be visible on the instrument. Most accredited laboratories affix a calibration sticker showing the calibration date and next-due date. If this sticker is missing or expired, the instrument must be withheld from safety-critical use until recalibrated, regardless of its bump-test history.