Requirements for gas quality and gas appliances

Introduction
The gas transmission network in the Netherlands transports two different qualities of gas, low-calorific gas known as G-gas or L-gas and, high calorific gas (H-gas). These two gas qualities are transported in separate networks, and are connected by means of five blending and conversion stations where high-calorific gas can either be blended with low-calorific gas or ballasted with nitrogen to produce gas that can be introduced into the low-calorific, G-gas, network.
The network was originally developed following the discovery of the large Groningen gas field. The Groningen field is low calorific gas. The low-calorific gas from the Groningen field became the standard for the consumers in the Netherlands. Later on, H-gas gas was produced from the so-called small fields in the Netherlands. High-calorific gas is also imported from Norway, Russia and through the LNG terminal in Rotterdam. The H-gas is supplied to industrial end users via approximately 80 connections to the high-calorific network and is also exported to other countries. The standard Wobbe Index bandwidth currently specified for exit points in the gas transmission network for H-gas is 47-55.7 MJ/m3 (25/0), with a number of regional variations as described in the Ministerial Ruling (MR) for gas quality.
The supply of low-calorific gas from the Groningen field is in decline, and ultimately the future supply to end users will be high-calorific gas. In March 2012, the Minister of Economic affairs declared that all appliances falling under the Gas Appliance Directive (the GAD) should be able to switch to high-calorific gas to prepare for a smooth transition at the end of the lifetime of the Groningen field. For this purpose, a formal notification has been given to the European Commission. However, at the moment it is unclear what the future range of high-calorific gas quality should be; that is, the range of gas quality that appliances falling under the GAD can accept is at present uncertain.
The question posed by the Dutch government is whether or not the government should change the existing notification of the requirements that GAD gas appliances should be able to meet to cope with high-calorific gas, and if so, how and when.
Analysis
The study comprised three major areas of investigation. In the first step, the limitations in appliance performance with varying gas quality were assessed. This assessment is based on a critical analysis of existing data from laboratory experiments on appliances from well-defined field tests and also from a critical analysis of non-experimental information (theoretical studies and evidence-based experience). In the second step, the future ranges of gas quality expected to be supplied in the Netherlands was inventoried, and the possibilities for, and costs of, treating the gases to limit the range of gas quality was assessed. In the third step, the potential of innovation for widening the range of gases acceptable for appliances was assessed. The recommendations summarized below are based on the synthesis of these three elements.
Limitations in appliances
We assessed the existing experimental, theoretical and practical/experiential evidence regarding which bands of Wobbe Index maintain the safety and reliability of the population of H-/E-band appliances installed in the field. In our opinion, the distribution practices in the UK, France, Denmark and Belgium give the best reflection of a practical range for Wobbe Index: the years-long practice in these countries shows that H-/E-band appliance performance with distribution limits in the range 4-5 MJ/m3 satisfies the national requirements and/or customs for safety and reliability in these countries. We also note that
DNV GL – Report No. 74106553 .01b – www.dnvgl.com Page 2
these countries have some form of active maintenance regime. Provided adjustable appliances are properly adjusted, the laboratory experiments assessed support a range of 4-5 MJ/m3, although these experiments require extra interpretation before being applied to the situation in practice. The theoretical analyses show that the approval regime does not safeguard the intended appliance performance in a number of situations. However, these analyses say nothing about the actual safe and reliable performance of appliances in the field. Therefore, we conclude that a bandwidth of in the range of 4-5 MJ/m3 can be realized for appliances approved under the GAD.
Future supply
Ultimately, when Dutch indigenous gas production becomes small, gas is expected to be supplied by pipeline imports from Norway and Russia, and as LNG from the worldwide market. The Netherlands is expected to become a net importer from 2025 onwards. The expectation is further that the imported gas will have a Wobbe Index between 51 and 55.7 MJ/m3 (25/0). The upper limit of 55.7 MJ/m3 is set by the Dutch government.
This gas quality bandwidth is significantly smaller than the current standard bandwidth permitted in the H-network, from 47 to 55.7 MJ/m3. The ‘small fields’ contribute predominantly to the lower half of this range. At the moment an annual volume of 25 bcm is produced from the small fields. By 2032 the total capacity including so-called futures is expected to be almost three times lower than the 2014 capacities. These futures are however uncertain. When the futures are not taken into account, the total capacity will be 15 times lower compared to 2014. In 2032 supply from the small fields will then only be 3% of demand (20% if the futures fully materialize). Excluding futures, the production volume of the small fields decreases to 1 bcm in 2030.
Three cases with different Wobbe Index bandwidths were evaluated: a band of 2 MJ/m3, significantly narrower than the expected range of import qualities, a wide band of 8 MJ/m3 and a band coincident with the expected future import band of 4.7 MJ/m3. The results show that the narrow band option requires the most gas treatment, particularly nitrogen ballasting, which given the expected import will be required indefinitely. The costs for gas treatment (especially nitrogen ballasting) will thus recur every year. The widest band allows the widest accommodation of both import gases and residual small-field gases. Referring to the appliance limitations analysis above, not all appliances can handle a band of 8 MJ/m3 without further gas treatment measures. We note that the potential measures that support the intake of ‘off-spec’ gas with a Wobbe Index lower than the minimum Wobbe Index of H-gas imports are expected to be temporary. The volume of gas from the small fields is expected to become very small after 2030 as mentioned above. The use of the gas quality management options described here is seen more as a transition measure and not as permanent. In light of the expected gas supply, the widest band gives therefore only a temporary advantage.
For the intermediate band of 4.7 MJ/m3, the import gases can be easily accepted, but relatively more small-field gases become ‘off spec’, requiring blending with the H-gas import. The forecast of H-gas imports volumes indicate that sufficient H-gas is available for blending the ‘off spec’ gas in this case. Ballasting with nitrogen will not be required if the upper Wobbe limit is set at 55.7 MJ/m3.
Concluding, choosing the intermediate band of 4.7 MJ/m3 corresponds to the bandwidth of the expected import of future H-gases of 51-55.7 MJ/m3, and is also within the range of 4-5 MJ/m3 to which the GAD appliances have been exposed in practice. In this option there are no extra costs for nitrogen ballasting. This range may provide an optimum between appliance performance and expected gas supplies.Innovation
An inventory has been made to determine the status of the development of innovative products aimed at extending the fuel flexibility of GAD end-use equipment, e.g. by means of active control systems. The inventory is based on interviews with different stakeholders and collecting existing information available, including progress made in existing innovation programs such as SBIR.
It became clear that commercially available (premixed) appliances having active control systems are suited to operate across the entire E-band and for handling abrupt Wobbe fluctuations. We also observed promising developments for (inexpensive) control systems for premixed appliances to extend the fuel flexibility for both new domestic appliances and suitable appliances already installed in the field. Also, a sensor-based hob burner (cooker) is under development to guarantee high performance and capacity while using variable gas quality (L+H band). DNV GL is not aware of any existing developments or innovations to make other type of partially premixed domestic appliances, such as flow-through hot water heaters, suitable for a wide range of gas compositions. To our knowledge, no innovation regarding the development of fuel-adaptive control systems for non-domestic burners is currently being undertaken. However, several control strategies are possible and economically feasible.
Recommendation
Based on the analysis, DNV GL recommends setting the long term quality bandwidth for H-gas at 51 – 55.7 MJ/m3 (25/0). In this choice, the bandwidth (4.7 MJ/m3) is within the 4-5 MJ/m3 with which millions of H-/E-band appliances function in other EU countries, aligning the distribution practice with those in other countries, and complex and costly gas quality management measures are limited. A consistent policy for appliance adjustment using known gas quality is essential for maintaining this band. We recommend updating the existing notification and suggest modifications to the notification as formulated in Appendix A. We further note that in terms of Wobbe Index this band allows room for the accommodation of a reasonable bandwidth for renewable gases.