Réglementations, Technology • 23.01.2025
Refrigerant leaks : how to detect them early
The European refrigeration sector is experiencing sustained growth, with the commissioning of 600 million m2 of refrigerated warehouses and 5.6 billion air conditioners by 2050. At the same time, the European Commission has set itself the target of reducing greenhouse gas emissions by 55% by 2030. To meet this major environmental challenge, doing everything possible to detect and seal refrigerant leaks in good time is becoming a regulatory obligation, but above all a social one.
What leak detection methods are available, and what do French and European regulations say about them?
Fuites de fluide frigorigène : Le point sur la réglementation
Since 1987, the Montreal Protocol has regulated ozone-depleting substances such as chlorofluorocarbons (CFCs).
In 1997, the Kyoto Protocol tackled greenhouse gases such as hydrofluorocarbons (HFCs). Europe then introduced the first version of the F-Gas regulation in 2006 to regulate the sale and use of refrigerants.
Its second version, EU Regulation No. 517/2014, came into force on January 1, 2015. It sets out best practices for limiting refrigerant-related emissions. It also sets quotas for the quantity of fluids placed on the market.
This regulation governs leak detection, requiring the installation of a detector to alert the operator in the event of refrigerant leaks, as well as regular leak checks based on the tonnes of CO2 equivalent (T CO2 eq.) contained in the installations.
In 2016, a French decree went further than the European regulation and defined the authorized detection methods: direct and indirect detection.
In 2019, a new version of this decree defines the indirect measurement method as mandatory for installations containing quantities of refrigerants greater than 500 T CO2 eq. , or 128 kg of R-404A.
In March 2024, the new European “F-Gas III” regulation comes into force:
- HFOs are now subject to leak testing
- Installations containing quantities of unsaturated fluids (HFO) greater than or equal to 100 kg or HFC and HFC/HFO greater than 500 tonnes CO2 equivalent must be fitted with a permanent leak detection device such as the PolarBox or DNI. In the event of a leak, an alert must be sent out.
Frequency of leak checks according to refrigerant categories and the new F-Gas III regulation, and according to the French decree of July 2019:
Important point: HFC HFO blends are considered as HFCs.
The direct method: useful for finding local refrigerant leaks
Direct methods are not authorized in France for installations containing quantities of refrigerant in excess of 500 T CO2 eq.
They are nonetheless very useful in the search for local refrigerant leaks, although they do have certain limitations:
- They are used on an ad hoc basis, according to the periods defined by the regulations, and therefore do not ensure constant surveillance. If a leak occurs the day after a visit, it will not be detected until the next visit (3 months, 6 months…). The system will then operate in a degraded manner, and may even stop production. Remember: an underloaded system also consumes more energy.
- If they help locate a leak, there’s no guarantee that another one doesn’t exist. A complete search of all parts of the installation is required to ensure that no leaks remain.
Direct and indirect methods complement each other!
Please note that any alarm triggered by a leak detection system using indirect measurement methods requires a direct leak detection method.
The indirect measurement method: for early identification of refrigerant leaks
To facilitate early detection of leaks
The method for detecting refrigerant leaks by indirect measurement is based on continuous analysis of at least one of the following parameters:
- Pressure ;
- Temperatures ;
- Compressor current ;
- Liquid levels ;
- The volume of the refilled quantity.
In the event of a drift in any of these parameters, the system must trigger an alarm based on the following thresholds: 50g/h or 10% of plant volume.
This permanent monitoring, coupled with an alarm system, makes the indirect measurement method a monitoring tool particularly well-suited to the early detection of refrigerant leaks.
And for effective remote identification of refrigerant leaks
The indirect measurement method, by continuously analyzing several parameters, ensures constant plant monitoring.
That’s the whole point of this method of monitoring refrigerant leaks: to move from periodic leak checks to continuous monitoring.
It alerts you as soon as a leak occurs, enabling you to intervene as quickly as possible and avoid a plant shutdown leading to a potential break in the cold chain, and high financial losses.
Connected to the Internet, this type of method is particularly effective for remote monitoring of plant operation, thanks to Matelex’s PolarVisor web monitoring platform. As soon as a leak is detected, an alert is sent with an estimate of the leak’s flow rate and a deadline for action.
In brief: the different leak detection methods.
Direct methods | Indirect methods |
---|---|
Portable detectors | Low level switch |
Room controllers | Expert systems: P, T measurements, compressor current level, recharged qty volume |
Foaming solutions or soapy water | |
UV liquid Introduction or colorant | |
Vacuum packing | |
Pressurization |
Finding fluid leaks efficiently: do you know about expert systems (IoT)?
Low-level detection is one of the indirect methods available for measuring refrigerant leaks. How does it work? An alarm is triggered as soon as the fluid level falls below a specified threshold. This method has proved inadequate. It provides a late warning when the plant is almost empty and therefore soon to be shut down. It can also encourage systems to be underloaded, which has a major impact on performance and leads to higher energy consumption.
Expert systems, on the other hand, operate on a simple principle: the demand for cold is constantly changing. For example, the cold demand of large food retailers varies according to the frequency of store visits: high on Saturdays, when windows are constantly opened and closed, and low on Sundays, when the store is closed.
This involves differentiating between “normal” operation of a system with fluctuating cooling demand that causes the refrigerant level to vary, and abnormal operation with a drop in level due to a leak.
The emergence of the Internet of Things (IoT) and the introduction of algorithms simplify the identification of abnormal plant operation. After a learning period, a leak detection algorithm determines a reference level. Based on the refrigerant level in the liquid receiver (HP), this algorithm is then able to identify any deviation from normal operation.
Detect refrigerant leaks with the Matelex expert system
A test carried out by the French Environment and Energy Management Agency (Ademe) on large-scale food retail refrigeration systems highlighted the refrigerant savings achieved by the algorithm’s early detection.
The two detection methods used by Matelex were compared on a panel of 5 well-tuned positive and negative installations:
- Traditional method: direct leak detection in accordance with regulatory inspection periods based on CO2 equivalent tonnage.
- Indirect method with expert system: detection by level measurement (fluid level weighing system and detection using the reference level determined by the algorithm).
Over a 6-month period, the system saved the equivalent of 1,538 tonnes of CO2 in 5 stores with well-adjusted installations, by identifying leaks quickly (from 5 to 119 days earlier), representing a refrigerant saving of 79% compared with a traditional method.
In addition, Matelex has developed an Energy module that calculates the consumption of compressors, condenser fans and pumps (indirect circuits). Coupled installation of a DNI (Intelligent Level Detector) and this Energy module provides real-time monitoring of energy consumption and compressor operation, as well as early detection of refrigerant leaks. This dual approach makes it possible to identify direct and indirect greenhouse gas emissions, so as to reduce them more effectively.