Can a high peak and a low TWA both be correct?
A high H₂S peak and a low TWA are not necessarily contradictory. A peak is a reported maximum; a TWA averages readings over a defined period. Those numbers alone do not establish an instrument fault.
A low average does not cancel a recorded peak. Check the event against the applicable short-term or ceiling criterion, using the required measurement method. An event peak is not itself a short-term average.
Why can the two numbers be so different?
Imagine eight hours with a valid mean for every minute: one minute at 12 parts per million (ppm) and the other 479 minutes at zero.
Both bars use the same 0–12 ppm scale. The average bar is only 1/480 of the upper bar, so it is barely visible.
The highest saved one-minute mean is 12 ppm. This describes one minute; it does not tell you the highest instantaneous concentration.
All 480 minutes count in this average. The 479 zero-valued minutes bring the mean down to 0.025 ppm, while the 12 ppm minute remains in the record.
(12 ppm × 1 min + 0 ppm × 479 min) ÷ 480 min = 0.025 ppm
What should I check first?
First check where and when H₂S was sampled, which period the TWA covers, and whether the surrounding measurements are available. This establishes whether the two values describe the same monitoring task and period.
A remote sample from a tank and a reading from a monitor worn by a worker answer different questions. For personal gas monitoring, record who wore the monitor and where it was worn.
- Comparison of 3 methods characterizing H2S exposure in water and wastewater management work
- External manufacturer reference — Blackline Safety: G7c/G7x TWA and STEL guidance
- External manufacturer reference — Gas Clip Technologies: MGC Simple Plus manual, v1.04
- Personal gas monitoring: placement and recording requirements
Match the peak to the sampling task
Record the gas, units, peak and timestamp. Was the monitor worn by a worker, sampling remotely, or being exposed to test gas?
A remote sample describes air at the sampling point. It does not, by itself, establish the worker’s breathing-zone exposure.
Identify the TWA window and any resets
Check the model instructions and settings for the averaging period. Compare the alarm time with power-off, restart, manual-reset and mode-change records.
If the calculation window or reset history is unknown, do not label the displayed TWA as a complete-shift result.
Check the measurements around the event
Keep the original event and measurement exports. Note the software version and download scope; retain a timestamped screenshot if no export is available.
If rows are absent, use the checks below to investigate export scope, compression and overwrite. A peak cannot supply the missing concentrations.
Could a restart or setting change explain the TWA?
A restart, manual reset or setting change can alter an H₂S TWA, depending on the model, firmware and configuration. Check whether the calculation was cleared, retained or paused before treating the reading as a complete-shift result.
One published model-specific example allows TWA and short-term exposure limit (STEL) readings to be retained with firmware 3.422 or later and retention enabled. Both values survive shutdowns shorter than 15 minutes. For shutdowns longer than 15 minutes but shorter than eight hours, TWA and peak readings survive, but STEL resets.
Those conditions belong to the models identified in the source note. For an Eranntex monitor, send us its model, firmware and relevant restart times with the original records. Ask for the matching calculation and reset instructions; do not transfer another model’s retention settings to your instrument.
What if the peak is logged but readings are missing?
An H₂S alarm-event peak is a reported maximum, not a concentration history. If measurement intervals are absent, the peak cannot reconstruct their concentrations or a 15-minute average. The recorded alarm duration also does not establish total exposure duration.
Export scope, compression and storage overwrite are different possibilities to investigate. The examples below show what each can mean and which evidence to request. They do not diagnose the cause of a gap in your file.
For water and wastewater H₂S monitoring, retain the task and sampling-location record as well as the instrument files. The Norwegian study below illustrates why an alarm summary can remain available after its matching measurement history is no longer available.
- External manufacturer reference — Gas Clip Technologies: MGC Simple Plus manual, v1.04
- Comparison of 3 methods characterizing H2S exposure in water and wastewater management work
- Water and wastewater monitoring: tasks and sampling locations
| Possible mechanism | Documented example | What to check next |
|---|---|---|
| A restricted download scope | Download selections include events only, the last week, records since the previous download, or the full log.Example from the referenced manual and software; check the options for your instrument. | Record the selected scope and retrieve the measurement log covering the event, if still available. A broader download cannot establish that overwritten measurements are recoverable. |
| Compressed repeated records | The same manual describes one-second status logging with compression of redundant records.The manual does not establish a rule that every absent export row means zero. | Check how the software represents repeated values and elapsed time. Compression alone does not establish lost measurements or zero concentration. Do not turn unexplained blanks into zeros. |
| Separate circular storage | A 2024 Norwegian water and wastewater study reported 15-second logging and 16 hours of wraparound storage. Alarm records used separate storage, so a peak could be extracted without its corresponding measurement log.Study equipment and conditions; not an Eranntex storage specification or a universal detector rule. | Check the logging interval, storage capacity and download delay. Ask whether event and measurement records have separate retention rules. Circular storage overwrites older records when its space is reused. |
Which H₂S exposure limit should I compare with?
Choose the H₂S exposure criterion for the workplace jurisdiction, sector and measurement period. A ceiling, an eight-hour TWA and a short-term average are different criteria. Satisfying one does not establish that the others are satisfied.
Ask the person responsible for your site’s exposure assessment which criterion and measurement method apply. The report’s language does not select the law. The five examples below are references, not alarm settings or permission to work.
WEL means workplace exposure limit; PEL means permissible exposure limit; STEL means short-term exposure limit. Australia’s WES label means workplace exposure standards. The table retains its 18 October 2026 source snapshot and jurisdiction-specific conditions.
| Jurisdiction and basis | H₂S reference | What changes in the review |
|---|---|---|
| United States — federal OSHA, general industry29 CFR 1910.1000, Table Z-2; check applicable State Plan rules and worker/sector coverage | 20 ppm ceiling; a 50 ppm peak exception has specific conditionsException: one period of no more than 10 minutes, only if there is no other measurable exposure during the 8-hour shift | This is not an 8-hour TWA row. State Plans may have different or more stringent requirements. A low displayed TWA does not answer the ceiling question. |
| Great Britain — EH40EH40/2005 Table 1, H2S entry | 5 ppm long-term WEL; 10 ppm short-term WEL8-hour and 15-minute reference periods | Keep the reference period and evaluation method with the number. An event maximum is not a 15-minute average. |
| Singapore — WSH General ProvisionsRegulation 2 definitions and First Schedule | 10 ppm long-term PEL; 15 ppm short-term PEL8-hour working day / 40-hour work week; 15-minute short-term period | Use Singapore's actual PEL definitions. Singapore is an example, not a substitute for the laws of other Southeast Asian countries. |
| Ontario, Canada — Regulation 833 tableCurrent Ontario occupational-exposure table, H2S row | 10 ppm TWA; 15 ppm STELUse the Regulation 833 reference-period and workplace requirements | Identify the applicable province or federal regime. Do not label this a Canada-wide limit. |
| Australia — national WES referenceNovember 2025 national WES reference; confirm applicable Commonwealth, state or territory law and effective dates | 10 ppm 8-hour TWA; 15 ppm 15-minute STEL8 hours and 15 minutes; confirm jurisdictional adoption and commencement before applying the national reference | The 24 June 2026 update says H2S levels remain unchanged at the 1 December 2026 transition. The proposed 1/5 ppm option is not treated here as adopted law. |
What should I send to technical support?
For an Eranntex H₂S record query, send us the model, firmware, event time, sampling task and original event and measurement files. Include known restarts or setting changes. This gives the query a specific instrument, period and set of records to investigate.
For a new instrument enquiry, specify the recording evidence you need before selecting a model. The BTYQ-MS104K-S2 example below separates documented capabilities from the file and firmware details that still need confirmation for your configuration.
Use the editable support message for an existing event, or the recording checklist for a new project. Unknown information can remain marked unknown. Do not include worker names or other personal details unless needed for the request.
Eranntex example: specify the records you need
The BTYQ-MS104K-S2 product documentation identifies these capabilities. Use them as a starting point for a configuration enquiry, then confirm the recording details needed for your task.
| Item | Documented capability | Confirm for your task |
|---|---|---|
| Sampling method | Diffusion | The intended sampling location and whether the task is personal monitoring or remote sampling. |
| Standard record capacity | More than 1,000,000 records | What counts as one record, the selected logging interval, and what happens when storage fills. Capacity alone does not establish retention time. |
| Stored-data functions | On-device viewing, deletion and data export; configurable storage interval | Supported interval settings, export fields, status definitions and the file format for the actual software and firmware. |
| Standard data interface | USB-C charging and data communication | The software and transfer procedure required for the proposed configuration. |
These documented capabilities do not specify a TWA reset algorithm, separate event storage, overwrite policy or a fixed-step CSV format. Confirm those items for the actual instrument and firmware.
Nothing is submitted. Notes are not saved; copy them before leaving or refreshing this page.
Planning new equipment? Confirm these recording requirements
Saved values and event peaks
What does each saved value represent: an instantaneous reading, an interval mean or a maximum? Over which period is an event peak recorded?
Evidence to request: An annotated export with value definitions, units and the event-field description.
Timestamps and coverage
Do timestamps identify interval starts or ends? How are the time zone, clock changes and recording interruptions documented?
Evidence to request: Example timestamps, clock settings and records identifying the monitoring start, stop and gaps.
Logging intervals and status flags
Which logging intervals are available? How do missing, invalid, warm-up, fault and over-range states appear in the export?
Evidence to request: The supported interval settings and a field or status-code guide for the proposed configuration.
TWA calculation and resets
Which averaging period and readings are used? What happens to the calculation after power-off, a reset or a mode change?
Evidence to request: Model- and firmware-specific calculation instructions, configuration settings and a reset record where available.
Download scope
Which export options are available, and which records does each include? How must any compressed data be interpreted?
Evidence to request: An export made with the intended software settings and its matching file-format documentation.
Storage and retention
Do event and measurement logs share storage? What happens when storage fills, and how long are records retained at the selected logging interval?
Evidence to request: Documented record capacities, storage-full behaviour and retention information for that configuration.
Instrument, wearer and sample location
How will the records identify the instrument, the wearer where relevant, and the location from which air was sampled?
Evidence to request: An instrument assignment or activity record that distinguishes personal monitoring from fixed-point or remote sampling.
Related questions
Optional: follow a fictional 20-minute example
In this fictional H₂S case, a peak falls within two minutes of missing measurements. The files cannot establish the complete 20-minute average or explain the separate reported TWA. The example contains no real worker or detector data.
The example data pack below contains the files and values used in the exercise. Download it if you want to inspect the rows. Read the four-file guide before comparing the event and summary with the measurements.
Start with these four case files. The ZIP also includes a worksheet, supporting notes and the separate eight-hour arithmetic example.
- metadata.json
- Read first. Defines units and what each value means. Timestamps use Coordinated Universal Time (UTC).
- measurements.csv
- Contains twenty one-minute intervals labelled M01–M20. NR on this page means no measurement was supplied; it is not zero.
- events.csv
- Contains one alarm event, labelled E01, with its start, end, recorded peak and alarm-event duration.
- summary.json
- Contains a reported TWA with intentionally missing calculation details. It is a separate summary, not a result calculated from the measurement CSV.
The fictional record covers 09:00–09:20 UTC. Each numeric measurement is defined as a mean for one complete minute. Eighteen minutes have values; the two minutes from 09:04 to 09:06 have none.
Event E01 runs from 09:04:15 to 09:04:45 UTC, inside that gap. It reports a 12 ppm peak and a 30-second alarm duration. The separate summary reports a TWA of 0.030 ppm.
The 0.030 ppm value was chosen for this exercise; its calculation period, inputs and reset history are deliberately unspecified. The files use a generic format, not an Eranntex or other manufacturer’s export.
Use the findings below to separate what the files establish from what is still unknown. You do not need this exercise to use the checks above.
Fictional example · one-minute interval means · UTC
Each cell represents one minute. NR means no measurement was supplied. No line connects the values across the gap.
- E01 · events.csv
- 09:04:15–09:04:45 UTC · 12.000 ppm peak · 30-second alarm event
- summary.json
- Reported TWA 0.030 ppm · calculation window and inputs not supplied
Peak without the surrounding curve
Referenced records: E01, M05, M06
- What the file tells us
- Alarm event E01 reports a 12.000 ppm peak and a 30-second alarm duration.
- Still unknown
- The concentrations during missing intervals M05 and M06, the total exposure duration and the true atmospheric maximum remain unknown.
- Evidence to request
- Obtain the associated measurements, event-field definition and export or storage description.
Zero and missing are different
Referenced records: M01, M05, M06
- What the file tells us
- Measurement M01 is a valid 0.000 ppm one-minute mean. M05 and M06 have no values. Eighteen of twenty minutes have numeric values.
- Still unknown
- The files do not explain why two intervals are missing or what concentrations occurred then.
- Evidence to request
- Check the original download scope, status information and relevant activity records.
A reported TWA without its calculation history
Referenced records: summary.json
- What the file tells us
- The summary file labels 0.030 ppm as a reported TWA.
- Still unknown
- Its calculation period, input readings and reset history are unspecified. The wearer and sample location are also unknown.
- Evidence to request
- Obtain the actual model and firmware instructions, settings, monitoring times and instrument assignment.
For the 18 valid minutes, the covered-interval mean is 1.500 ppm·min ÷ 18 min = 0.0833 ppm. This is not the complete 20-minute average. The separate 0.030 ppm TWA remains unexplained because its calculation period and inputs are unspecified.