In the first part of the series of experiments, we investigated the measurement behavior of a gas meter with increasing hydrogen content in natural gas and the effects on the sensors especially for the use case construction.

In this part we look at the measurement behavior in the case of “monitoring workspace”.

The use case “Workspace monitoring” is described as follows: Monitoring the working environment during work on gas pipelines and/or gas installations where explosion hazards have to be reckoned with. In addition, a warning of toxic gas concentrations such as carbon monoxide (CO) or oxygen deficiency may be given in the case of combination devices.

The following test gases were used to represent the different mixtures of H2 in the natural gas network:

The EX-TEC® PM 580 is set to the gas type methane

• CH4: The correct methane content is indicated by the IR sensor. H2 is not measured.
• CO2: The IR sensor for CO2 shows no cross-sensitivity to H2
• The CO sensor goes even at 5% H2 admixture up to the end of the measurement range at 300 ppm.
• The H2S sensor displays up to 15 ppm as cross sensitivity to 4400 ppm H2.
• The O2 sensor reacts with a slightly reduced display, even with a 2.2 vol.-% CH4 without H2 content

Conclusion:

If the proportions of the gas mixture are known, it is generally also to be used as a test gas for calibration/adjustment. However, in the case of hydrogen being fed into the gas network, the actual composition of the gas mixture is usually not known. Here methane and hydrogen content can vary within a predetermined range. In this case, a calibration/adjustment of the device with a “replacement test gas” is to be carried out according to leaflet T 023 of the BG RCI.

The replacement test gas used should contain as a measurement component the combustion gas of the gas mixture, which has the lowest sensitivity compared to the measurement principle of the gas sensors used. This refers to the respective value of the lower explosion limit (measurement signal / % of the UEG). However, since IR sensors have no sensitivity to hydrogen, the use of a substitute test gas is not possible here.

Rather, the undetected hydrogen content must be compensated by lowering the alarm thresholds, in this case by approximately 10 % of the alarm threshold specified in the explosion protection document.

The BG ETEM (Berufsgenossenschaft Energie Textil Elektro Medienerzudnisse) has already published the article "Energy Storage Hydrogen" in the issue 5, 2022 of its professional cooperative magazine. In a joint research project between BAM, DBI-GUT and BG ETEM, the effects of hydrogen additives in natural gas up to a concentration of 10 vol.-% tested for explosion protection. The safety parameters for these natural gas-hydrogen mixtures were also determined. Other interesting passages deal with gas propagation at blow-off and expansion lines, as well as the influence of hydrogen admixtures on gas alarm devices. The article can be found here: etem.bgetem.de/5.2022/etem/energy storage-hydrogen

A CO sensor can no longer be used because its end of measurement range is reached immediately.

With H2S, false alarms can occur from 10 % hydrogen admixture. A deployment must therefore be carefully considered.

In the third and final part of our series of experiments, the measurement behavior and the sensor properties in the case of gas alarms are used.