The CO₂ concentration in a process line is to be determined using an NDIR measuring device. The gas contains, for example, 10 vol.% CO₂, is warm, nearly saturated with water vapour and is under several bar of pressure in the line. Can the gas sample simply be routed to the CO₂ sensor through a hose?
Usually not. With an NDIR CO₂ measurement, not only the actual CO₂ concentration in the process is relevant. The pressure, temperature and humidity conditions under which the gas reaches the measuring cell are also decisive.
If humid process gas cools down in a cold sampling line, water can condense. If the sample is dried before measurement, the reference basis of the stated CO₂ concentration changes. At the same time, a significantly different sample pressure can influence the infrared absorption or the concentration calculated from it.
Reliable NDIR CO₂ measurement therefore starts before the sensor: pressure, temperature, water vapour content, dew point, sampling line and the required wet or dry basis must all be consistent with one another.
How does NDIR CO₂ measurement work?
NDIR stands for Non-Dispersive Infrared. The measuring principle uses the property of CO₂ to absorb infrared radiation within specific wavelength ranges.
A typical NDIR measuring cell consists, in simplified form, of:
- an infrared source,
- a measuring path containing the sample gas,
- an optical filter or wavelength-selective evaluation,
- an infrared detector,
- reference and compensation methods.
The more CO₂ molecules are present in the optical measuring path, the more strongly the relevant infrared radiation is absorbed.
The electronics calculate the CO₂ concentration from this.
Modern NDIR sensors use additional reference measurements or compensation methods to account for factors such as ageing of the light source and changes in the optical system.
Why is water vapour relevant to CO₂ measurements?
Water vapour can influence a CO₂ measurement in different ways.
Three effects should be considered separately:
- Gas composition: Water vapour occupies part of the total gas volume and therefore changes the CO₂ concentration stated on a wet basis.
- Optical or sensor-specific influences: Depending on the NDIR technology, water vapour can be taken into account or compensated for in signal processing.
- Condensation: Liquid water in the line or measuring cell is a completely different condition from gaseous water vapour and should be avoided.
The statement “the sensor is suitable for high humidity” therefore does not automatically mean that liquid condensate may enter the measuring cell.
What is the difference between wet basis and dry basis?
With a humid gas mixture, the CO₂ concentration can either be referenced to the complete wet gas or to the dry gas remaining after the water vapour has been mathematically or physically removed.
These two values are commonly referred to as:
- Wet basis and
- Dry basis
.
Example:
A process gas contains on a wet basis:
10.0 vol.% CO₂
and:
5.0 vol.% H₂O
If the water vapour is completely excluded from the calculation, the simplified result is:
CO₂ dry = 10.0 % / (1 - 0.05)
which is approximately:
10.53 vol.% CO₂
Both values can be correct. They simply refer to different gas bases.
For process measurements, it must therefore be clearly defined whether the limit value, process specification and measuring device state CO₂ on a wet or dry basis.
Why is condensation in the sampling line a problem?
Warm process gas can contain large quantities of water vapour.
If this gas is routed through a cooler line and its temperature falls below the dew point, some of the water condenses.
This can cause several problems:
- water accumulates in the hose or pipe,
- filters can become blocked,
- the sample gas flow changes,
- optical surfaces can become contaminated or wetted,
- response time becomes longer,
- the composition of the remaining gas sample changes.
CO₂ is also soluble in water. When condensation occurs, it therefore cannot simply be assumed that only water vapour is removed from the sample while all other components remain unchanged.
Uncontrolled condensation is therefore not a suitable method of sample gas conditioning.
Why does gas pressure influence NDIR measurement?
Infrared absorption depends on the number of absorbing molecules within the optical measuring path.
If the pressure changes while the gas composition remains the same, the molecular density in the measuring volume also changes.
NDIR measuring devices can therefore exhibit pressure dependence.
High-quality process analysers can measure the current pressure and compensate for it mathematically. With other devices, however, a defined sample pressure is a prerequisite for the specified measurement accuracy.
The accuracy specification of a CO₂ sensor should therefore always be considered together with the specified pressure conditions.
Why must process pressure be reduced in a controlled manner?
A portable gas detector or an analyser designed for ambient pressure must not simply be connected directly to a process line that is under several bar of pressure.
The gas sample requires suitable pressure reduction.
A typical setup may include, for example:
Process → sampling valve → pressure regulator → filter → condensate management → flow control → measuring device
The components actually required depend on the process.
The pressure reduction must be designed so that the permissible inlet pressure of the downstream measuring system is not exceeded.
It must also be taken into account that strong pressure reduction can thermally affect the gas. This can also alter the condensation conditions.
What role do gas temperature and dew point play?
Relative humidity alone is often not sufficient for assessing a process gas sample.
The more important question is:
How far is the temperature of the coldest point in the sampling system above or below the dew point?
As long as all gas-contacting areas remain sufficiently above the dew point, the water remains in the gaseous state.
If the temperature falls below the dew point, condensate forms.
Depending on the measurement strategy, there are therefore two basic options:
- keep the sampling line sufficiently warm and transport the humidity in gaseous form to the measuring device,
- remove water in a controlled manner using a defined sample gas conditioning system.
Which option is appropriate depends on whether the CO₂ concentration is required on a wet or dry basis.
What does a typical sample conditioning system look like?
Sample gas conditioning must be adapted to the process gas, pressure, temperature, particle loading and the required measured variable.
One possible setup consists of:
- Sampling point: Take a representative gas sample from the process.
- Isolation: Allow the sample to be safely isolated from the process.
- Pressure reduction: Reduce the process pressure to the permissible sample gas pressure.
- Particle filter: Protect the sensor and sample path from solids.
- Temperature or condensate management: Prevent condensation or handle it in a controlled manner.
- Flow restriction: Set a suitable sample gas flow.
- Measuring device: Only supply gas to the sensor under defined conditions.
For safety-relevant gases, safe discharge of the sample gas and possible explosion-protection requirements must also be taken into account.
Should the process gas be dried before CO₂ measurement?
This cannot be answered universally.
Drying can be technically useful if the analyser being used requires dry, non-condensing gas.
However, drying changes the reference basis of the measurement.
If an originally humid sample is dried, the CO₂ content in volume percent of the remaining dry gas mixture increases.
For correct evaluation, it must therefore be clear:
- what humidity was present before conditioning,
- how the drying is performed,
- whether CO₂ is required on a wet or dry basis,
- whether calibration and process measurement use the same basis.
A dry calibration gas and a humid process gas must not be compared without verification if the humidity represents a significant proportion of the gas mixture.
What influence does the sampling line have?
The connection between the process and the measuring device is also part of the measuring chain.
The following should be taken into account when selecting and routing the sampling line:
- length,
- internal diameter,
- material,
- temperature,
- low points,
- possible condensation,
- permissible pressure,
- chemical resistance.
An unnecessarily long line increases the internal gas volume and therefore extends the response time.
Low points can collect condensate and partially or completely block the gas flow.
With highly dynamic processes, a measuring device located far away can therefore indicate a condition that already occurred some time earlier in the process.
Why must the sample gas flow also be correct?
With a pumped gas detector, the gas is transported through the sampling line to the sensor at a defined volumetric flow rate.
If the flow is too low, the transport and response time increase.
Possible causes include:
- contaminated filter,
- condensate in the line,
- kinked hose,
- sampling line that is too long,
- unsuitable pressure conditions.
If, on the other hand, the gas flow is too high for the measuring system or the gas is under impermissible overpressure, the instrument can also be operating outside its specified conditions.
The sampling arrangement specified by the manufacturer should therefore be followed.
What does sample conditioning mean for alarm limits?
For a safety-relevant measurement, an alarm limit must correspond to the concentration basis actually being measured.
Example:
A process limit is based on:
CO₂ on a wet basis
but the sample gas is completely dried before reaching the sensor.
The displayed CO₂ value on a dry basis can therefore be higher even though the absolute amount of CO₂ in the original gas has not changed.
An alarm limit must therefore not simply be adopted without clarifying which reference basis actually applies at the sensor.
In addition, every sampling line increases the response time of the overall system. This must also be considered in an alarm strategy.
Calibration and process measurement under comparable conditions
During calibration, a gas of known concentration is supplied to the sensor.
For meaningful calibration, the conditions under which the calibration gas is supplied should correspond as closely as possible to the intended measuring method.
The following should be considered in particular:
- gas pressure at the sensor,
- gas flow,
- temperature,
- humidity or wet/dry basis,
- sample path.
A functional test or bump test is not the same as a complete calibration.
During the functional test, the main points checked are whether:
- the sensor responds to the test gas,
- the indication increases plausibly,
- the alarm functions correctly,
- the pump or sampling path is operating.
During calibration, by contrast, the quantitative relationship between gas concentration and measured value is checked or adjusted.
Practical example: humid CO₂ process gas under pressure
A process line carries a gas mixture with:
8 vol.% CO₂
at:
4 bar process pressure
and with a high water vapour content.
A sample is to be routed to a portable NDIR measuring device for checking.
On the first attempt, a long unheated hose is used. Visible condensate forms in the cooler workshop area.
At the same time, the gas flow fluctuates and the CO₂ indication responds unusually slowly.
The sampling system is therefore rebuilt:
- Sampling directly at a representative process point.
- Controlled pressure reduction.
- Defined condensate or humidity management.
- Short, suitable sampling line.
- Constant gas flow to the measuring device.
The indication then stabilizes much more quickly.
When evaluating the result, it is also taken into account whether the process value of 8 vol.% is defined on a wet or dry basis.
This example shows that a fluctuating NDIR CO₂ value does not necessarily indicate a defective infrared sensor. The sampling system itself can already alter pressure, humidity and gas composition to such an extent that the conditions at the sensor are no longer the same as those in the process.
Systematically diagnosing implausible CO₂ values
- Clarify the process CO₂ range and expected concentration.
- Check whether the value is specified on a wet or dry basis.
- Determine the process pressure.
- Check the permissible inlet pressure of the measuring device.
- Assess gas temperature and dew point.
- Check the sampling line for condensate.
- Check the filter for contamination or water.
- Check the sample gas flow.
- Keep the sampling line as short as possible and free of low points.
- Check whether pressure compensation is required or available.
- Use zero and test gas and check sensor response.
- For alarm applications, test the complete alarm chain.
Common mistakes
- Routing process gas directly from a pressurized line to a portable gas detector: The permissible sample gas pressure can be exceeded.
- Equating high humidity with condensate: Gaseous water vapour and liquid water are different measuring conditions.
- Confusing wet and dry basis: After drying, the same gas mixture has a different CO₂ volume fraction.
- Ignoring condensate in the sampling line: It can affect gas flow, composition and response time.
- Considering only relative humidity: The relationship between gas temperature and dew point is decisive for the risk of condensation.
- Ignoring pressure influence: NDIR measurements can be pressure-dependent or require defined pressure conditions.
- Supplying dry calibration gas and comparing it directly with humid process gas: The reference conditions may be different.
- Using very long sampling lines: This increases the dead time of the measuring system.
- Checking filters only for particles: A filter filled with condensate can also severely reduce the sample gas flow.
- Equating functional testing with calibration: A sensor response alone does not automatically prove quantitative measurement accuracy.
Gas-Pro IR for mobile CO₂ measurements
A specific portable gas detector with infrared measurement for carbon dioxide is the Crowcon Gas-Pro IR.
Depending on the configuration, the device can monitor several gases simultaneously. For CO₂, an infrared sensor with a measuring range of:
0 ... 5 vol.% CO₂
is available.
For remote sampling, the Gas-Pro can optionally be equipped with an internal pump. This allows gas to be drawn through a sampling line, for example before entering an area.
The device also features:
- audible alarm,
- visual alarm,
- vibration alarm,
- data logging,
- event logging,
- bump-test and calibration reminders.
The specified humidity range is 10 ... 95 % RH. However, this does not mean that liquid water or condensate should enter the sensor system.
When gas is sampled from a pressurized, hot or highly humid process line, suitable sample conditioning must therefore be installed upstream.
In particular, the portable Gas-Pro IR must not be treated like a directly pressurized inline process analyser.
Further information can be found under Crowcon Gas-Pro IR and under gas measuring devices / gas detectors at ICS Schneider.
Conclusion
NDIR CO₂ measurement in humid process gas is not determined by the infrared sensor alone. The complete sampling chain is part of the measurement.
Water vapour changes the proportion of the other gas components stated on a wet basis. If the sample is dried, the resulting CO₂ value on a dry basis changes accordingly.
Uncontrolled condensation is even more critical. Liquid water in the sampling line, filter or measuring cell can influence gas flow, gas composition and response behaviour.
Pressure must also not be ignored. The NDIR principle responds to the number of absorbing molecules in the optical measuring path. A measuring device therefore requires either suitable pressure compensation or defined pressure conditions.
With pressurized process gas, the sample pressure must be reduced in a controlled manner. Temperature and dew point must also be taken into account because pressure and temperature changes can create new condensation conditions.
For reliable CO₂ measurement, the following therefore applies: first define whether the value is required on a wet or dry basis, then check sample pressure, gas temperature and dew point, prevent condensation or manage it in a defined manner, and only then supply the conditioned gas to the NDIR sensor under the specified conditions.
FAQ: NDIR CO₂ measurement in humid process gas
What does NDIR mean in CO₂ measurement?
NDIR stands for Non-Dispersive Infrared. CO₂ absorbs infrared radiation within characteristic wavelength ranges. The gas concentration is determined from the measured absorption.
Does water vapour influence NDIR CO₂ measurement?
Yes. Water vapour influences the gas composition and therefore the distinction between wet and dry basis, among other factors. Depending on the sensor, sensor-specific optical influences may also be taken into account.
What is the difference between CO₂ on a wet basis and on a dry basis?
On a wet basis, water vapour is included as part of the total gas mixture. On a dry basis, the CO₂ content is referenced to the water-free remaining gas. The numerical value is therefore higher on a dry basis.
Why is condensate more problematic than high humidity?
Condensate is present as a liquid and can affect lines, filters and optical measuring paths. High humidity, by contrast, initially only means that a large amount of water is present in gaseous form.
Can an NDIR CO₂ sensor be connected directly to a pressurized gas line?
Only if the specific measuring system is expressly designed for that process pressure. With portable devices or devices designed for approximately ambient pressure, the sample pressure must first be reduced in a controlled manner.
Why does pressure influence CO₂ measurement?
Gas pressure changes the molecular density in the optical measuring volume. Depending on the design, NDIR devices therefore require pressure compensation or defined pressure conditions.
Why can pressure reduction cause condensation?
A pressure change can also alter the gas temperature. At the same time, the conditions under which water vapour condenses change. Pressure reduction and dew point should therefore be considered together.
Should humid process gas always be dried before CO₂ measurement?
No. This depends on the measuring device and on whether the process value is required on a wet or dry basis. Drying changes the reference basis of the CO₂ value.
How can condensation in the sampling line be prevented?
Depending on the application, the sampling line can be kept above the dew point or the humidity can be removed in a controlled sample gas conditioning system.
Why should the sampling line be as short as possible?
A long line contains a larger gas volume and therefore increases transport and response time. It also increases the risk of temperature changes and condensation.
What is the difference between a bump test and calibration?
A bump test primarily checks whether the sensor responds to gas and whether the alarm functions operate. Calibration checks or adjusts the quantitative relationship between a known gas concentration and the measured value.
Which specific device can measure CO₂ using infrared technology?
One example is the Crowcon Gas-Pro IR. It can be equipped with a CO₂ infrared sensor for the range 0 to 5 vol.% and is optionally available with an internal pump for remote sampling.