Spray Chapter3

PIPENET VISION TRAINING MANUAL PAGE 1 OF 66
SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010
PIPENET SPRAY/SPRINKLER MODULE CHAPTER 3 FIRE PROTECTION SYSTEMS DESIGN - FIREWATER RINGMAIN SYSTEMS
1 Introduction
In this chapter, we build on the material covered in Chapter 1, and extend the principles to cover firewater ringmain systems. In order to model firewater ringmain systems well, we need to understand some of the capabilities of PIPENET VISION that were not covered in Chapter 2. This chapter is intended to cover both the design methodologies and the techniques for using PIPENET VISION itself. Chapter 2 of this document contains design tips and techniques, and Chapter 3 covers additional principles in more detail. Chapter 4 contains a discussion on how to model firewater ringmain systems, and Chapter 5 contains information on how to set up the desktop. The different phases of input (namely, initialisation and libraries) are discussed in Chapter 6. In Chapter 7, we describe in detail how to input the network. Systems based on clack shut type deluge valves are discussed in Chapter 8, and systems based on elastomeric type deluge valves are discussed in Chapter 9. In Chapter 10, we describe techniques for selecting pumps. We give some hints and tips for refining the design in Chapter 11, and conclude with an example of a real firewater ringmain in Chapter 12. The material in this document is partly for discussion and partly for actual input.
2 Design Tips and Techniques

In the previous chapter, we looked in detail at how deluge systems are modelled. We also saw why it is better to model deluge systems separately from firewater ringmains. In this chapter, we shall see how to model firewater ringmain systems in detail. The basic principles of setting up a network are the same as for deluge systems (and is covered in Chapter 1 of the Spray/Sprinkler Training Manual). However, in order to be able to fully utilise the capabilities of PIPENET VISION for modelling firewater ringmain systems, we have to learn the following new aspects. Inputting pump curves: Firewater systems often have pump curves, and it is necessary to learn how to input vendor-provided pump-performance curves. How to deal with systems using clack shut valves and elastomeric valves as deluge valves: The best technique for modelling systems using conventional clack shut valves is quite different from that for modelling elastomeric valves.
-1-
How to deal with multiple inputs and outputs: Firewater ringmain systems often have multiple inputs (for example, multiple pumps) and/or multiple outputs (for example hydrants, hose reels, etc.). We need to understand how to apply specifications to systems with multiple inputs/outputs. Dealing with monitors, hydrants and hose reels: Firewater ringmain systems contain all the above types of item. understand how to model these items.
We need to
Usually, there is not one unique way of modelling a firewater ringmain system. This chapter is intended mainly to cover the basic principles. Each individual engineer and each company may wish to develop its own method of applying these principles.
3 Additional Principles in More Detail

Let us discuss the above items one at a time. Please note that the dialog boxes shown in this section are for information only. Please do not input data at this stage. The paragraphs below simply contain a more-detailed discussion of the above matters.
3.1 Inputting Pump Curves

It is assumed that appropriate units have already been chosen. Pump curves can be input using Libraries | Pumps Coeffs. unknown (which, in PIPENET VISION, is used to denote choosing Pumps Coeffs. unknown from the Libraries Menu). As an example, let us use the data in the table below. Flow Rate (lit/min) 10000 20000 30000 Pressure (Bar G) 14.5 12.5 10
In addition, we use the following range for the flow rate. Minimum flow rate = 10000 lit/min, Maximum flow rate = 30000 lit/min. The dialog box for entering this data is shown below.
-2-
Sometimes the error message Gradient must be negative over the whole range... is issued, in which case, the fitted curve (and not necessarily the input data) has a peak between the minimum and maximum flow rates. Effectively, this error means that, for some values of pressure, there could be two corresponding flow rates. Under these circumstances, there may not be a unique solution, as two flow rates could give rise to the same pressure. This problem generally arises when the steep part of the pump curve is input together with the shallow part. There are perhaps three ways of dealing with this problem: Reduce the range between the minimum flow rate and the maximum flow rate (either by increasing the minimum value, decreasing the maximum value, or doing both). Input either the steep part of the pump curve or the shallow part, but not both parts together. Slightly modify the data points so that the peak does not occur.
One other important point that should be borne in mind, especially in offshore firewater systems, is the following. Often the pump curve refers to the flow rate and pressure at the discharge flange of the pump assembly. In other words, the static head loss and frictional
-3-
loss in the riser pipe have already been taken into account. If this were the case, the caisson riser pipe must not be input again.
3.2 How to Deal with Systems using Clack Shut Valves and Elastomeric Valves as Deluge Valves
Usually, the type of a deluge valve is either clack shut or elastomeric. Clack shut deluge valves are characterised by the fact that the flow rate depends on the inlet pressure. For this reason, if more than one system is in operation, the deluge systems will interact with each other. It is common to model a clack shut deluge valve using an equivalent nozzle. For example, if the system including the deluge valve requires a flowrate of 5697 lit/min at a pressure of 9 barg, an equivalent nozzle would have a K-factor of 5697/9 = 1899 (lit/min, bar). As each deluge system in the ringmain would have a different K-factor, the nozzles are not normally set up in the nozzle library. Instead, they are input as user-defined nozzles. The minimum and maximum pressures can be set to any reasonable values.
User defined
Equivalent nozzle
Elastomeric deluge valves, on the other hand, control the downstream pressure. Consequently, they control the flow rate entering the deluge system. As the flow rate is fixed during commissioning, this node is treated as an output with a known flow rate.
-4-
Output node
Design phase flowrate
Calculation phase flowrate
Please see the section below for a clarification of the meaning of design-phase and calculation-phase specifications.
3.3 How to Deal with Multiple Inputs and Outputs

Dealing with multiple inputs and outputs is mainly a matter of learning how to assign appropriate specifications. The basic mathematical rules are: Number of inputs + Number of outputs Number of pressure specifications = Number of pressure specifications + Number of flow-rate specifications 1
Please note that nozzles are ignored in the above rules, as all of the specifications for nozzles are automatically assigned in PIPENET VISION. Please note the following point about applying the above rules. PIPENET VISION always performs the calculation twice, once in the Design Phase and once in the Calculation Phase. In PIPENET VISION, the terms Calculation Phase and Analysis Phase mean the same (and so are interchangeable). The way in which the above rules are applied is different between the Design Phase and Calculation Phase. The Design Phase The purpose of the Design Phase is to determine the sizes for pipes whose sizes have been left unset by the user. However, even if all of the pipe sizes have been specified (by the user), PIPENET VISION still performs the Design Phase. During the Design Phase, flow-rate specifications should be given to all input/output nodes except one. PIPENET VISION automatically assigns a pressure specification to one of the input/output nodes, thereby satisfying the basic rules shown above.
-5-
PIPENET VISION TRAINING MANUAL PAGE 6 OF 66 The Calculation (or Analysis) Phase
The purpose of the Calculation Phase is to determine the flow rates and pressures throughout the system. During this phase, the specifications for pressure and flow rate can be given in any combination, provided that the above rules are satisfied. Typically, the pressures at the inlet nodes and the flow rates at the outlet nodes would be known. However, this may not always be the case; for example, if PIPENET VISION is to be used for pump selection. It is essential to understand the manner in which specifications are given for firewater ringmains. For this reason, let us illustrate the basic principles by considering a few examples of simplified systems.
3.3.1 A System With Two Fire Pumps and Four Monitors
Input/output nodes are terminal points in the system, and so monitors that are modelled as nozzles do not count as input/output nodes. Therefore, there are two input nodes in the system. For the Design Phase, we need to provide a flow-rate specification for all but one input/output node. Therefore, we have to give one flow-rate specification, which can be applied to either of the input nodes, and can have any reasonable value.
-6-
In practice, we would know the inlet pressures of the pumps. So, for the Analysis Phase, we can give, say, a pressure of 0 barg to both input nodes.
3.3.2 A System With Two Fire Pumps and Four Outlets
The total number of input/output nodes in this system is six. In the Design Phase, we need to give a flow-rate specification at all but one input/output node. So, we have to supply five flow-rate specifications, which can be given to any five of the input/output nodes (and can have any reasonable values). We now consider the Calculation Phase. Again, we would know the inlet pressures of the pumps. So, we can give, say, a pressure of 0 barg to both input nodes. For the outlets that are working, we give appropriate flow rates; for the outlets that are not working, we give a flow rate of zero.
3.4 Dealing with Monitors, Hydrants and Hose Reels

These items have known required flow rates and pressures. There are two methods for dealing with them: Either (a) treat each item as an output, and apply the appropriate flowrate specification or (b) convert each item into an equivalent nozzle, using the formula
-7-
K=
q P
in which q is the flow rate and P is the pressure. We can then ensure that the pressure available at the output node is more than the required pressure. It is important to be aware of the weaknesses of the output-node approach, which just shows that the available pressure is more than the required minimum pressure. This approach, which would result in the flow rate becoming more than the minimum required, does not lend itself to studying the interaction between two items.
4 How to Model Firewater Ringmain Systems 4.1 The Objectives

Some of the objectives of modelling firewater ringmain systems are as follows. Adequacy of pump performance: The deluge-system calculations would indicate the system requirements by way of pressure and flow rate. We need to make sure that the fire pump arrangement can meet these requirements under a variety of conditions and a variety of demands. The skill in designing the firewater ringmain system well lies in making sure that it would work adequately under a wide variety of circumstances. Ability to generate sufficient pressure at the inlet to the deluge systems and other users: It is not just the sizing of the fire pumps that is important. The pipes have to be sized adequately too. They must be small enough to reduce the weight and cost, but large enough to offer acceptable pressure drops. Longest piping route by blocking pipes: Firewater ringmains have isolation valves for maintenance, strainers and so on. It is often necessary to ensure that the system would work adequately even if an isolation valve is left closed or a strainer is blocked. Integrity of the system if a pipe is broken: A broken pipe may be important if a firewater ringmain system is being analysed. For example, if the pipework near the helideck system is broken in a platform with several levels, it is possible that the deluge systems at the low levels might still work properly, albeit at a reduced efficiency. The reason is because the elevation difference might produce enough static head. Different fire scenarios: One of the interesting aspects in the design of a firewater ringmain system is the fact that the same piping system must work under a wide variety of conditions. Clearly, it is necessary to perform calculations for a wide variety of fire scenarios.
-8-
Different fire pump scenarios: Usually, the system is designed in such a way that one or more pumps would be required to meet the demand. It is important to ensure that this situation is possible.
4.2 The First Calculation in Analysing Firewater Ringmain Systems

The first calculation is almost always the worst-case scenario. In other words, it is the case that the system has been designed to cope with. This system could include one or more deluge systems, monitors, hydrants and so on. It could also include one or more fire pumps, depending on the operating philosophy. The objective of this calculation is to ensure that adequate levels of pressure are generated at the different points in the system. The system might include an overhead tank or may not contain deluge systems, but the principles of the calculation are the same. If the inlet pressure is higher than what is required, too much water may flow through the system. In an extreme case, this flow could lead to the need for a larger fire pump. Therefore, it may be necessary to restrict the flow rate into a deluge system.
4.3 Subsequent Calculations in Analysing Deluge Systems

Typically, the following calculations will also be performed as part of this exercise. The reason being that, in the real design environment, the supply side may not be exactly what is required, or the demand-side requirement may be less or even more than in the worstcase scenario.
More users than the worst-case scenario working, Fewer users than the worst-case scenario working, A fire pump (that is supposed to work) is stopped, The effect of a blocked pipe.
5 Training Example 5.1 The Network Schematic

The network that we shall be working with is shown below. This network schematic was developed in PIPENET VISION. It is being shown here for illustration purposes only. Please do not input the network at this stage.
-9-
In the above schematic, there is no indication of how the terminal nodes of the system are going to be treated. The reason is that the treatment of these nodes depends on whether clack shut deluge valves or elastomeric deluge valves are used. The steps involved in inputting the data are shown below.
5.2 How to Set Up the Desktop

Please note that much of this section has already been covered in Chapter 2 of the Spray/Sprinkler Training Manual, entitled Fire Protection Systems Design Deluge Systems. However, it is worthwhile to work through the steps again. Open PIPENET VISION, and set up the desktop using the steps that are shown below. The first step is to use the View Menu to make sure that all four windows are open.
- 10 -
Select all windows
Then, move the windows on the desktop so that the screen is similar to that in the diagram below. Note that you may also need to follow the steps described in the section entitled Display Options below.
Display Options From Options | Display options, choose the display options that are shown in the following dialog box.
- 11 -
The menu style that has been illustrated in this document so far is the Windows Menu Style (which is used in the diagram below). Those of you who are experienced users of PIPENET VISION might prefer to use the menu style that is similar to that of PIPENET Classic. This style may be selected as follows.
The Windows Menu Style will be used in the remainder of this document.
- 12 -
5.3 Initialisation and Libraries Stage 5.3.1 Initialisation Stage

For initialisation, please use the Options Menu and select Module options (i.e., Options | Module options, as denoted in the PIPENET VISION Training Manuals). Title
It is recommended that the title should relate to the calculation that is performed, especially for models that are to be kept. Therefore, choosing an appropriate title could potentially save a lot of time later when you wish to study the results.
- 13 -
PIPENET VISION TRAINING MANUAL PAGE 14 OF 66 Spray Options
The above dialog box shows a typical set up, which has been used for the calculations in this chapter of the training manual. The fluid properties depend on whether water or seawater is being used as the fire protection medium. NFPA 1996/2001 refers to the 1996 and 2001 editions of the NPFA rules. The option NFPA refers to earlier editions of the NFPA rules. The difference lies in the way that equivalent lengths for fittings are used. In earlier editions, it was assumed that Schedule 40 pipes were being used in calculating the values of the equivalent lengths of fittings. In the later editions of the NFPA rules, there is a correction for the actual inner diameters for pipes other than Schedule 40. The NFPA option is still retained for historical reasons; it is used to ensure that older data files produce the same results as before. For restriction orifices, which may be used in firewater systems, the Crane Option is generally used. BS 1042 is a flow-metering standard, and the Heriot-Watt Option is proprietary Sunrise Systems correlation.
- 14 -
With the Spray Option, all of the nozzles are kept open; with the Sprinkler Option, all of the nozzles are closed. The sprinkler heads may be opened individually afterwards, if required. Units A comprehensive range of units is available in PIPENET VISION. In this example, we shall use metric units, but with lit/min for the flow rate.
Please note the Save as defaults Button. If you click on this button, the default units on the specific computer will be set to the units currently shown on the screen. Then, PIPENET VISION will default to this unit selection every time a new model is created. The next step is to select the pipe schedule that is to be used. Before performing this step, we shall take a diversion into the Libraries Menu, and return to options later.
- 15 -
5.3.2 Libraries (Diversion from Initialisation)

At this point, we take a diversion and set up our library. The Libraries Menu is used to store items that are likely to be used several times. Typical examples are user-defined pipe schedules, nozzle data and pump curves. We are considering the modelling of firewater ringmain systems in this section. As these systems do not contain nozzles (other than user-defined nozzles), we shall not input nozzle data into the library. However, we shall input a pipe schedule and a pump performance curve. In order to access the Libraries Menu, simply click on Libraries in the Menu Bar.
Pipe Schedules As an example, enter the data below for the following pipe schedule (called NEW Schedule). Nominal Size (mm) 100 150 200 250 300 350 400 Inner Diameter (mm) 102 154 203 255 303 350 400
Use the following value for the pipe roughness. Pipe roughness = 0.0457 mm.
The pipe roughness must be input even if it is not used. It is not used with the HazenWilliams Option, but it is used with the Darcy Option. Choose Libraries | Schedules, and click on New Button. Then enter the data from the above table, as illustrated below.
- 16 -
Click on the Apply Button, and then the OK Button. Pump Data While we are in the Libraries Menu, we shall enter the following data for the fire pumps. Fire pump name = Weir type 120/45, Minimum flow rate = 10000 lit/min, Maximum flow rate = 30000 lit/min. Flow Rate (lit/min) 10000 20000 30000 Pressure (Bar G) 14.5 12.5 10
Enter this data by choosing Libraries | Pumps Coeffs. unknown, clicking on the New Button, and then typing in the data, as indicated below.
- 17 -
Click on the Apply Button and then the OK Button. When the Apply Button is clicked, the dialog box will display the pump curve, as shown below.
- 18 -
5.3.3 Initialisation (Returning from Libraries)

We now return to the initialisation phase. The next step is to select the pipe type. However, it is worth considering some discussion points beforehand. The difference between a pipe schedule and a pipe type is as follows. A pipe schedule is standard pipe data that is entered into a library. A pipe type, on the other hand, is a pipe schedule (from the library) that is to be used in the current network. It, also, has associated maximum velocities (or maximum pressure drops per unit length), C-factor and pipe lining, if applicable. For our network model, we use the following data. C-factor = Maximum velocity = Lining = 140, 6 m/sec, none.
To enter this data, choose Options | Pipe types, click on the New Button, and type in the values. A short cut for entering the maximum velocity is (a) enter 6 m/sec in one cell, (b) click on another cell, and (c) click and hold the left button on the cell that has the 6 m/sec already
- 19 -
entered, and drag the cursor down over the other cells in the maximum velocity column. Of course the maximum velocity could also be entered individually in each cell.
We have already completed the Display Tab. We shall return to the Calculation Tab again later, but, for the time being, the initialisation and library phases are complete. We can now return the input phase for the network.
5.4 Network Input 5.4.1 Input of Schematic Drawing

From the Tools Bar, choose the Pipe Tool, which is indicated below.
Pipe Tool
- 20 -
For the data-input stage, it might be preferable to display the grid (by either (a) using the Grid On/Off Button on the Display Toolbar or (b) ticking the Display grid box from Options | Display options). In your schematic diagram, try to use the pipe and node labels that are shown in the diagrams in this section, as it will then be easier to input the attributes later. It may be easier if the nodes and links are displayed (using either the Display Tools or Options | Display options).
Display node labels
Display link labels
In the drawing area, click on a grid point, preferably near the lower left-hand part of the drawing area.
Drag the tool in the Northeast direction, until you reach the point at which you wish to change the direction of the pipe, and then press and hold down the Shift Key (on the keyboard). Click on the left-hand mouse button every time you want to change direction. (In the network schematic that we are drawing, there are two changes of direction.) When you want to create the end node, release the Shift Key and click on the left-hand mouse button.
- 21 -
Then, to complete the pipe network, draw the other pipes as shown below.
- 22 -
5.4.2 Input of Pipe Data

The attribute data for the network is shown below. Pipe Label 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 Diameter (mm) 200 200 200 200 200 200 200 200 200 100 100 100 100 100 100 100 100 Length (m) 20 20 20 20 20 20 20 20 20 5 5 5 5 5 5 5 5 Elevation (m) 5 5 0 0 0 0 0 0 0 5 0 -5 0 0 0 0 0 Fittings 2 x 90 2 x 90 1 x Tee 1 x Tee 1 x Tee 1 x Tee 1 x Tee 1 x Tee 1 x Tee 1 x Tee, 1 x 90 1 x Tee 1 x Tee, 1 x 90 1 x Tee 1 x Tee 1 x Tee 1 x Tee 1 x Tee
As this example is created for training purposes, the data for pipes and fittings is not realistic. It is geared towards practising fast data entry using capabilities such as global edit, copy/paste, etc. This data can, of course, be entered using either the Properties Window or a Data Window. We shall now revise how to enter the diameters quickly. Firstly, under the Data Tab of a Data Window, choose 200 from the pull-down menu in the Diameter Column for Pipe 1, as shown below.
- 23 -
Next, click on any other cell in the Data Window, and then click on the cell with the diameter of 200 mm again. Right click on that cell, and choose Paste in column from the context-sensitive menu. The Data Window will now appear as shown below.
- 24 -
According to the pipe data in the table above, Pipes 10 to 17 inclusive have diameters of 100 mm (and not 200 mm, as we have set). Use the following steps to obtain the 200mm diameters quickly. Firstly, choose a diameter of 100 mm for Pipe 10. Next, click on any other cell, and then click on the diameter cell for Pipe 10 again. Next, use the right-hand mouse button and select Copy.
- 25 -
Then, highlight the pipes to which this diameter is to be pasted, right-click, and select Paste.
- 26 -
The Data Window should then look as follows.
- 27 -
Enter the other data for the pipes in the same way. The Data Window should then look as shown below.
5.4.3 Input of Fittings

We gain access to the fittings by clicking on the Fittings Tab of the Properties Window.
Fittings Tab
- 28 -
To add the two 90-degree-elbow fittings to Pipe 1, begin by clicking on Pipe 1 in the Schematic Window. Next, in the Fittings Window, select 90 deg elbow and then click on the Add Button twice.
In a similar manner, we can input the pipe fittings (which are shown in the table of pipe data above).
- 29 -
5.4.4 Input of Pumps

The next step is to input the pumps. Select the Pump Tool from the Tools Toolbar and draw the two pumps.
Then, select each pump in turn, and choose Fire Pump 1 from the pull-down menu in the Properties Window, as illustrated below.
- 30 -
Choose the fire pump
The initial stage of inputting the network is now complete. From this point, the clack-shut valve-based system and elastomeric-valve-based system become different. In Section 5.5 below, we consider aspects that are specific to clack-shut-valve-based systems, and, in Section 5.6, we consider elastomeric-valve-based systems.
5.5 Systems Based on Clack Shut Type Deluge Valves

The requirements for the clack-shut valve-based system are described in this section.
5.5.1 Monitors
Monitors are attached to Nodes 12 and 13, and the data for these monitors is as follows. K factor Required flow rate Minimum pressure Maximum pressure = = = = 2500 (lit/min, bar), 5000 lit/min, 1 barg, 20 barg.
5.5.2 Deluge Systems

Nodes 14, 15 and 16 are attached to deluge systems, for which the following data is applicable.
- 31 -
PIPENET VISION TRAINING MANUAL PAGE 32 OF 66 Node Number Flow Rate (lit/min) 14 10000 15 12000 16 8000 Pressure (bar) 7 6.5 7.3
The above requirements at nodes 14, 15 and 16 can be converted into equivalent nozzles by using the formula in Section 3.4 (namely, that the equivalent K-factor is equal to the flow rate divided by the square root of the pressure). Node Number 14 15 16 Equivalent K-factor (lit/min, bar) 3779.6 4706.8 2960.9
As the nozzles in the system are not real nozzles, the minimum and maximum pressure for the equivalent nozzles can be arbitrarily set to 1 bar and 20 bar respectively. All of above data for these five items should be input as user-defined nozzle data. We add nozzles to the nodes, and assign user-defined attributes to these nozzles as shown below.
- 32 -
PIPENET VISION TRAINING MANUAL PAGE 33 OF 66 The Properties Window for Nozzle 1 is as follows.
After all of the equivalent nozzles have been input, the Data Window for the nozzles is as shown below.
- 33 -
5.5.3 Assignment of Specifications

In this system, there are two input/output nodes, both of which are inputs. As both nodes are inputs to the pumps, a pressure specification of 0 barg is applied at each node. We now consider the specifications for the Design Phase. As we saw earlier, we must assign a flow rate to all but one input/output node. So, we need to give one design-phase flow rate. The actual value is not important, as long as it is reasonable. The total required output flow rate from the system is 40,000 lit/min. Therefore, we can assign 20,000 lit/min to one input (as there are two inputs). This value is also consistent with the pump curve data being halfway between the minimum flow rate and maximum flow rate. The dialog boxes for the two input nodes are shown below.
Note that Node 17 has only an analysis specification, whereas Node 18 has a design specification and an analysis specification.
- 34 -
Finally, we must make sure that the type of calculation to be performed has been set to user-defined (since specifications for the inlet pressures have been given). Typically, the option for firewater ringmains is user-defined specifications.
The other two options (namely, the most remote nozzle and mass balance options) are typically used with deluge systems. The only exception is the pump selection case with firewater ringmain systems, in which case, the most remote nozzle option would typically be used.
- 35 -
5.5.4 The Simulations

Scenario 1: All Three Deluge Systems, the Two Monitors and the Two Pumps are Working The network has now been completely specified, and all the systems are working. Therefore, the data that we have already input is sufficient, and so the calculation can be carried out.
If we look at the results for this calculation, we can see that the monitors are discharging too much water, and the deluge systems are under-performing. We need to investigate this behaviour by looking at the velocities in the pipes.
- 36 -
It is clear that the velocities in most of the pipes are too high. We can, therefore, consider increasing (a) the diameters of the 200 mm pipes to 300 mm and (b) the diameters of the 100 mm pipes to 200mm. We can easily carry out this change by copying and pasting in the Data Window.
- 37 -
PIPENET VISION TRAINING MANUAL PAGE 38 OF 66 The new results are shown below.
The results are now satisfactory, and so we shall use the increased diameters for all subsequent calculations.
- 38 -
Scenario 2: All Three Deluge Systems and the Two Pumps are Working We turn off the two monitors simply by switching off the nozzles. This change can be made using either a Data Window or the Properties Window.
Nozzle is turned off
The flow rates in the nozzles are shown below.
Scenario 3: All Three Deluge Systems are Working, but Only One Pump is Running The data for this scenario is the same as that for Scenario 2, except that the pressure specification (of zero barg) at the inlet of one of the pumps is changed to a flow-rate specification of zero litres per minute.
- 39 -
In particular, the analysis specification on Node 18 is changed from 0 barg to 0 lit/min.
The volumetric flow rates throughout the system are shown below.
The nozzle deviations are shown below.
- 40 -
Scenario 4: All Three Deluge Systems are Working, but Only One Pump is Running, and Pipe 8 is Blocked We use the Properties Window to change the status of Pipe 8, as illustrated below.
Blocked pipe
- 41 -
PIPENET VISION TRAINING MANUAL PAGE 42 OF 66 Pipe 8 now appears as a dashed line in the Schematic Window.
Blocked pipe
- 42 -
PIPENET VISION TRAINING MANUAL PAGE 43 OF 66 The pipe volumetric flow rates are as follows.
The nozzle deviations and node pressures are shown in the diagram below.
- 43 -
5.6 Systems Based on Elastomeric Type Deluge Valves

The system demands are the same as above. However, the manner in which the deluge systems are handled is quite different.
5.6.1 Monitors
Nodes 12 and 13 have monitors attached to them, and the data is as follows. K-factor Required flow rate Minimum pressure Maximum pressure = = = = 2500 (lit/min, bar), 5000 lit/min, 1 barg, 20 barg.
The monitors do not normally have elastomeric valves, and so they are modelled in the same way as before (namely, as nozzles).
5.6.2 Deluge Systems

Nodes 14, 15 and 16 are attached to deluge systems, with the following data. Node Number Flow Rate (lit/min) 14 10000 15 12000 16 8000 Pressure (bar) 7 6.5 7.3
As the deluge valves are of the elastomeric type, the flow rates are fixed, and so we cannot model them as nozzles. What we need to do is the following. We can set the flow rates at the deluge systems. The pressure generated by the pumps at the outputs should be at least equal to the system-demand pressure plus the minimum pressure drop in the elastomeric valve. As long as the pressure generated by the pump is higher than this threshold pressure, the deluge system can be expected to work satisfactorily.
- 44 -
PIPENET VISION TRAINING MANUAL PAGE 45 OF 66 The schematic is shown below.
5.6.3 Assignment of Specifications

In this system, there are two input/output nodes, both of which are inputs. As before, these nodes operate at 0 barg (as they are the inputs to the pumps). In the Design Phase, we must assign flow rates to all but one of the input/output nodes. In this case, as there are five input/output nodes, we need to specify four flow rates for the Design Phase. As usual, the actual value is not important, as long as it is reasonable. The total required output flow rate from the system is 40,000 lit/min. So, we can assign 20,000 lit/min to one input (because there are two inputs). The three outputs can simply be assigned their required flow rates.
- 45 -
PIPENET VISION TRAINING MANUAL PAGE 46 OF 66 The dialog boxes for the two input nodes are shown below.
Note that, as before, Node 17 has only an analysis specification, and Node 18 has both design and analysis specifications. The following specifications apply to the three output nodes.
- 46 -
5.6.4 The Simulations

Scenario 5: All Three Deluge Systems, the Two Monitors and the Two Pumps are Working The specifications that are described above are sufficient for running the calculation, and the results are as follows.
- 47 -
Scenario 6: All Three Deluge Systems and the Two Pumps are Working We turn off the two monitors by simply switching off the nozzles (which is carried out using either a Data Window or the Properties Window). The results are as shown below.
Scenario 7: All Three Deluge Systems are Working, but Only One Pump is Running The data is the same as that for Scenario 2, except that the pressure specification (of zero bar G) at the inlet to one of the pumps is changed to a zero flow-rate specification. In particular, the analysis specification on Node 18 is changed from 0 barg to 0 lit/min, as illustrated below.
- 48 -
The results for this scenario are shown below.
- 49 -
Scenario 8: All Three Deluge Systems are Working, but Only One Pump is Running, and Pipe 8 is Blocked The following results are obtained for this scenario.
6 Pump Selection Techniques 6.1 Clack Shut Deluge Valve Systems (Scenario 9)
Naturally, this system will have no pumps. For a system with one pump, we use the most remote nozzle option. The inlet node will have no operating condition associated with it. The logic of using the most remote nozzle option is that the smallest pump would be the one that meets the requirement of the most remote nozzle exactly.
- 50 -
- 51 -
PIPENET VISION TRAINING MANUAL PAGE 52 OF 66 The resulting nozzle deviations are as follows.
We note that the smallest pump that would meet the demand exactly would have the following operating point. Flow rate = 45120.9 lit/min, Pressure = 19.513 bar. This information can be used to choose the pump. We note, in passing, that one nozzle has a deviation of exactly 0%, thereby confirming that this is the smallest pump that would satisfy the requirement.
6.2 Elastomeric Deluge Valve Systems (Scenario 10)

We have to guess which output node is the one where it is most difficult to meet the requirement, and set both the pressure and flow rate at that node. At the other output nodes, we simply set the flow rate. As before, the inlet node will have no operating condition associated with it. Node Number Flow Rate (lit/min) 14 10000 15 12000 16 8000 Pressure (bar) 7 6.5 7.3
- 52 -
We guess that Node 16, with required flowrate of 8000 lit/min and pressure of 7.3 barg, is probably the most difficult to satisfy.
First of all, we confirm that our guess of the worst output node is correct. As can be seen from the following schematic, the pressures at all of the output nodes are more than is required.
We then note that the required inlet conditions are as follows.
- 53 -
PIPENET VISION TRAINING MANUAL PAGE 54 OF 66 Flow rate = 43071.7 lit/min, Pressure = 9.09 barg. These conditions enable us to determine the required pump size.
7 Some Hints and Tips for Refining the Design

Good design is not a matter of ensuring that the system works. Good design is about making sure that the system works in the best possible way. For example, if a fire pump is larger than required, it is not necessarily true that the pump is good. The firewater ringmain system may be supplied from a water tank, in which case the tank may drain more quickly than necessary, thereby requiring a bigger thank. Even if the firewater pump is not supplied from a tank, if it generates a pressure that is higher than necessary, a deluge system may discharge at a greater flow rate than is necessary, leading to greater power consumption. In an extreme case, this might lead to the need for a bigger pump. For these reasons, it is better to perform a small number of additional calculations to ensure that the system works in the optimal manner. In this section, we look at some techniques for refining the system. For the sake of convenience, we add pipes to the inlet of the pumps. The system now looks as shown below.
- 54 -
Please note that the use of equivalent nozzles for modelling clack valves is useful as a simulation tool for studying the interaction between systems. However, for design and design refinement, it is often better to treat the connections to the deluge systems as outlets. In the section, we shall adopt the method of using outlets. At all of the outlets, the flow rates are set to their minimum required values. From the results of the initial calculation (with no additional restriction), which are shown below, we see the following levels of over pressurisation.
We note that the following are the requirements for the pressures at the output nodes. Node Number 12 13 14 15 16 Minimum Pressure (barg) 4 {= (5000/2500)2} 4 {= (5000/2500)2} 7 6.5 7.3 Actual Pressure (barg) 10.772 10.772 11.097 10.804 11.160 Deviation (bar) 6.772 6.772 4.097 4.304 3.860
Note that the minimum pressures for Nodes 12 and 13 are obtained using the formula in Section 3.4, with a flow rate of 5000 lit/min and a K-factor of 2500 (lit/min, bar).
- 55 -
7.1 Use of Overboard Dump Valves (Scenario 11)

In PIPENET VISION, this term could represent any device that fixes the pressure by bypassing water flowing out of the fire pump. It could also be used for an overboard dump valve on an offshore platform or an FPSO, or it could be used for a PCV, and so on. One of the most important questions is How do we select the overboard dump valve, and what should its set point be? We note from the above table that the critical node from a pressure point of view is Node 16, and it has an excess pressure of 3.86 bar. So, the maximum reduction that we can have in the pressure of the system is 3.86 bar. In other words, if we dropped the pressure in the system by more than 3.86 bar, Node 16 would not receive its required pressure. On the supply side, we note that the fire pumps have an average discharge pressure of approximately 12.286 barg. So, this pressure must be reduced by 3.86 bar to 8.426 barg. We can achieve this reduction by placing overboard dump valves on Nodes 1 and 3, and setting their pressures to 8.426 barg.
- 56 -
The results of the calculation can be seen below. Please note that Node 16 satisfies its required pressure of 7.3 barg, but the pressures at all of the other output nodes are above their required values.
The pipe volumetric flow rates are illustrated below.
- 57 -
The following procedure could be adopted for selecting the overboard dump valves. The fire pump inlet and outlet pipes have flow rates of around 33582 lit/min and 20000 lit/min respectively. Therefore, we conclude that the overboard dump valve should dump about 13682 lit/min at a pressure of 8.426 barg. It can be seen that the overboard dump valve can be used for the dual role of ensuring that the pressure in the system does not exceed the design pressure, as well as restricting excess flow rates.
7.2 Use of Orifice Plates (Scenario 12)

It is clear that using overboard dump valves does not restrict the flow to individual demand points. Orifice plates can be used for this purpose. Node Number 12 13 14 15 16 Minimum Pressure (barg) 4 {= (5000/2500)2} 4 {= (5000/2500)2} 7 6.5 7.3 Actual Pressure (barg) 10.772 10.772 11.097 10.804 11.160 Deviation (bar) 6.772 6.772 4.097 4.304 3.860
The orifice plates, therefore, should produce the following restrictive pressure drops.
- 58 -
PIPENET VISION TRAINING MANUAL PAGE 59 OF 66 Pipe Label 13 14 15 16 17 Required Pressure Drop (bar) 6.772 6.772 4.093 4.304 3.860
To achieve these pressure drops, place orifice plates on all the relevant pipes, and specify the required pressure drops.
This scenario results in virtually the exact required performance from the system. The orifice plate details are shown below.
- 59 -
7.3 Use of Bypass Valves (Scenario 13)

The basic objective might be to ensure that a pump (which might be bigger than the minimum requirement) does not cause a water tank to drain too quickly. Attach two bypass lines to the model, as shown in the diagram below.
- 60 -
PIPENET VISION TRAINING MANUAL PAGE 61 OF 66 Use the following data for Pipes 20 and 21.
Use the Equipment Item Tool to place a restriction on the bypass line.
Equipment item
Set the equivalent length of both equipment items to be 20 m.
- 61 -
PIPENET VISION TRAINING MANUAL PAGE 62 OF 66 The pressures at the nodes are as follows.
The calculated results for the bypass line are shown below. The network can now be fine tuned as required.
- 62 -
8 Example of a Real Firewater Ringmain

In this section, we look at an example of a real system, which is included only as a matter of interest (and is not a training example that is to be input by the user). Please note that the screenshots in this section were not obtained using the most recent version of the PIPENET VISION Spray/Sprinkler Module.
- 63 -
- 64 -
- 65 -
- 66 -

Spray Chapter3

Uploaded by

Copyright:

Available Formats

Spray Chapter3

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Spray Chapter3

Uploaded by

Copyright:

Available Formats

What are the main components of a firewater ringmain system?

What are the main components of a firewater ringmain system?

How are pump curves input in PIPENET VISION?

How are pump curves input in PIPENET VISION?

PIPENET VISION TRAINING MANUAL PAGE 1 OF 66

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

2 Design Tips and Techniques

PIPENET VISION TRAINING MANUAL PAGE 2 OF 66

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

3 Additional Principles in More Detail

3.1 Inputting Pump Curves

PIPENET VISION TRAINING MANUAL PAGE 3 OF 66

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

PIPENET VISION TRAINING MANUAL PAGE 4 OF 66

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

PIPENET VISION TRAINING MANUAL PAGE 5 OF 66

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

Design phase flowrate

Calculation phase flowrate

3.3 How to Deal with Multiple Inputs and Outputs

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

3.3.1 A System With Two Fire Pumps and Four Monitors

PIPENET VISION TRAINING MANUAL PAGE 7 OF 66

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

3.3.2 A System With Two Fire Pumps and Four Outlets

3.4 Dealing with Monitors, Hydrants and Hose Reels

PIPENET VISION TRAINING MANUAL PAGE 8 OF 66

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

4 How to Model Firewater Ringmain Systems 4.1 The Objectives

PIPENET VISION TRAINING MANUAL PAGE 9 OF 66

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

4.2 The First Calculation in Analysing Firewater Ringmain Systems

4.3 Subsequent Calculations in Analysing Deluge Systems

5 Training Example 5.1 The Network Schematic

PIPENET VISION TRAINING MANUAL PAGE 10 OF 66

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

5.2 How to Set Up the Desktop

PIPENET VISION TRAINING MANUAL PAGE 11 OF 66

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

Select all windows

PIPENET VISION TRAINING MANUAL PAGE 12 OF 66

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

PIPENET VISION TRAINING MANUAL PAGE 13 OF 66

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

5.3 Initialisation and Libraries Stage 5.3.1 Initialisation Stage

PIPENET VISION TRAINING MANUAL PAGE 14 OF 66 Spray Options

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

PIPENET VISION TRAINING MANUAL PAGE 15 OF 66

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

PIPENET VISION TRAINING MANUAL PAGE 16 OF 66

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

5.3.2 Libraries (Diversion from Initialisation)

PIPENET VISION TRAINING MANUAL PAGE 17 OF 66

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

PIPENET VISION TRAINING MANUAL PAGE 18 OF 66

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

PIPENET VISION TRAINING MANUAL PAGE 19 OF 66

SPRAY: CHAPTER 3 REVISION 2.1 SEP 2010

5.3.3 Initialisation (Returning from Libraries)