How to witness commissioning results

By Chris Parsloe:

Commissioning is an essential requirement for ensuring that systems operate as intended. For a piped heating or cooling system, the main activity of commissioning is the balancing of flow rates. By ensuring that each system branch receives its allocated design flow rate, this should ensure that the correct amount of heating or cooling is delivered to each zone.

Witnessing of commissioning is an activity often undertaken by the designer or client representative to ensure that the flow values achieved are acceptable. The usual method of judging acceptability is to verify that the flows measured and recorded lie within the acceptable tolerance bands indicated in CIBSE Code W. Depending on the size of the system and the criticality of its function, a percentage of flow measurement results are witnessed by the designer/client representative. All witnessed results should lie within the stated plus or minus percentage tolerance band.

All is fine when the results witnessed are within the stated tolerance limits. When manual balancing valves were the norm, the achievement of repeatable flow measurement values within CIBSE tolerances was relatively easy to achieve. All of the valves in the system were manually operated and once set, their resistances never changed. Therefore, any variations in flow rate could be attributed to issues unrelated to the valves such as air or dirt blockages.

The move to variable flow systems has resulted in the adoption of self-acting valves such as differential pressure control valves (DPCVs) and pressure independent control valves (PICVs). These valves are not fixed resistance valves and to maintain set flow rates under varying system operating conditions, their resistances must vary. This variation in resistance inevitably affects the repeatability of flow measurements.

Most self-acting valves are mechanically operated relying on the action of a spring. Basic mechanics tell us that all springs experience hysteresis i.e. the compression of the spring in response to a force may follow a different pattern relative its extension when the force is reduced. Hence, there will inevitably be some variation in the set flow rate through the valve as system pressure conditions change. The degree of flow variation will depend on the stiffness of the spring, the design of the valve and the flow rate it is trying to maintain. The degree to which flows vary from their set values is sometimes referred to as the valve’s “proportional band”.

In general, larger DPCVs with stiffer springs controlling larger pressure differentials maintain flow rates more accurately than small DPCVs with light springs controlling smaller differentials. It is just such DPCVs that are incorporated in PICVs to deal with flows to individual terminals. Despite the potential repeatability issue, I would argue that the potential benefits of PICVs in terms of simplified design, easy selection and optimisation of energy savings necessitate that they are not excluded from consideration.  

Nevertheless, variation from set flow values can make witnessing tricky. If the flow witnessed has changed from that set and recorded to a degree that it lies outside of the tolerances indicated in CIBSE Code W, then its acceptability may be challenged.

There are differing views on how to approach such a situation and in this respect I can only offer my own views on the subject.

First of all, it should be noted that ultra low flow rates (i.e. less than 0.015 l/s) are now excluded from the tolerance tables. Due to the difficulty of accurately measuring such low flow rates, a tight tolerance band should not be expected, regardless of the choice of valves. Alternative approaches to balancing ultra low flow rates are described in a separate article within this edition of BSE, and the adoption of one of these should overcome the issue of non-repeatability.

For larger flows that prove to be non-repeatable within CIBSE tolerances an appropriate course of action must be decided. Artificially increasing specified flow values to allow for negative variations should be resisted unless deemed essential. Such an action is likely to increase pump energy consumption and reduce system operating temperature differentials.

Opinions vary, but my own view is that for many applications, some degree of negative flow variation outside of the current CIBSE specified limits for both sensible heating and cooling coils can often be accepted. BSRIA Guide 2/2010 Commissioning water systems explains the factors that might be considered when making such a judgement.

Ultimately, the application is critical to the decision. Cooling coils performing dehumidification or which serve single zones requiring close control should be treated with priority and the performance of the proposed DPCV solution or PICV should be checked in advance to ensure that the “proportional band” is acceptable.   For heating coils or smaller cooling coils serving open plan spaces, I would say that some negative variation in flow outside of CIBSE tolerances can be considered especially when a margin has been added to the heating/cooling capacity of the selected coil.

Chris Parsloe (Parsloe Consulting Ltd) is the author of CIBSE Commissioning Code W 2010, BSRIA BG2/2010 Commissioning Water Systems, and BSRIA BG 29/2012 Pre-commission Cleaning of Pipework Systems.