LUNA method (2017) (research report)

A CIPHE, CIBSE & Heriot Watt Joint Research Report

The LUNA (Loading Unit Normalisation Assessment) project was initiated in response to the growing concern within the building services and public health engineering sectors that the traditional loading unit method used in the UK results in an over-estimation of the design flow for domestic hot and cold water systems.

This report summarises the outcomes of the first phase of the LUNA project, and provides a recommendation for the development of a new methodological approach for assessment of simultaneous flow rate for greater accuracy when sizing pipework, and associated plant selection.

Although resourcing for this first phase of the project did not allow for data collection, data provided by a third party enabled a comparison against as installed designs, thereby providing an indication of the difference between estimated and actual values. Outcomes show that although the loading unit system within BS EN 806-3 produces lower design flow values compared to other UK loading unit methods, concern remains about the predicted combined hot and cold water flow, in particular regarding the inability of this method to account for low-flow outlets on a range of water efficient appliances. Comparisons with other methods also show how those that employ an empirically-based statistical framework offer a better fit with measured data. The report discusses the relative merits of different statistical models concluding, on the basis of both statistical validity and a comparison with measured data, that a move away from the use of traditional loading units is now timely.

Development of the empirical parameters required by this new model will be addressed in phase 2 of the LUNA project, thereby subsequently informing technical guidance documents and potentially forming the basis for a new national standard.

Contents:

Executive summary

1 Introduction

2 Project methodology

3 Statistical models

4 Review of relevant international standards
4.1 Establishing a ‘long list’ of relevant codes
4.2 Selection of approaches/standards for comparison

5 Application of codes and standards to case study buildings
5.1 Case studies and primary/secondary data
5.2 General assumptions made in the application of codes

6 Results and discussion

7 Statistical validity of techniques
7.1 Probability models
7.2 Stochastic models
7.3 Empirical models

8 Alternative options and recent code changes

9 Conclusion

References

Appendix A: additional tables and graphs

Acknowledgements
Heriot-Watt Research Team:
Project lead: Professor Lynne Jack
Co-lead: Dr Sandhya Patidar
Lead researcher: Ms Achala Wickramasinghe

LUNA Group Members:
LUNA Honorary Chair: Brian Whorlow
Memebers: Mike Carter (Arup); Paul Harmer (CIPHE); Steve Ingle (consultant); Lynne Jack (Heriot Watt University); Gary Mann (Cadline); Anastasia Mylona (CIBSE); Will Pitt (NG Bailey); Laurence Seymour (Imtech) Martin Shouler (Arup); David Taylor (AquaTech Pressmain); Mark Taylor (AquaTech Pressmain); Steve Vaughan (AECOM)