Product(s): | HAMMER |
Version(s): | 08.11.XX.XX and higher |
Area: | Modeling |
This article explains how the Discharge to Atmosphere (D2A) element works and its typical application in transient simulations in HAMMER. The example model file at the end of the TechNote can also be found the Samples folder of the HAMMER installation file (C:\Program Files (x86)\Bentley\HAMMER\Samples).
Note: for details on how the D2A element works in a steady state or EPS in WaterGEMS and WaterCAD, see: How does WaterGEMS/WaterCAD treat the discharge to atmosphere element?
The Discharge to Atmosphere element can be used to model a valve discharging to atmosphere, orifice discharging to atmosphere, and head vs. flow rating tables. It is used to model an opening or orifice that allows flow to leave the pipe network (similar to a junction with a demand) and discharge to the atmosphere. You can model it as a fixed orifice that is always open, or a valve that is either initially open or closed, then opens or closes during the transient simulation. It can be placed in series with the main water line or at a "Tee", as shown below.
Note: It is important to understand that this element discharges to the atmosphere, not between the adjacent pipes. So in the above case of an in-line orientation, flow still passes through the pipeline beneath the valve, regardless of whether the valve is open or closed.
In the numerical solver, it is modeled as a demand (outflow) point located a hydraulically short distance from its node coordinates. The initial pressure and flow entered by the user ("Pressure Drop (typical)" and "Flow (typical)") are used to automatically calculate an emitter coefficient to represent the pressure vs. outflow characteristics of the orifice. This is the term CA in the orifice equation below. In turn, this will be used during the simulation to calculate outflow as the pressure changes. This applies to both the initial conditions solver as well as the transient solver. HAMMER uses the term CA to calculate other flows and their corresponding pressure drop. The "drop" in this case ("pressure drop (typical)" field) refers to the difference between pipeline pressure (inside the D2A) and atmospheric pressure (outside the D2A). So, it is not referring to any headloss across the node element but rather the pressure reported at the D2A element (since the pressure results in HAMMER refer to the gauge pressure inside of the pipeline with atmospheric pressure as the datum)
For example if you were to set the "typical pressure" to 50 psi and the "typical flow" to 750 gpm, that means if the pressure at the node is 50 psi, the outflow will be 750. If the pressure at the node drops below 50 psi, the outflow will drop accordingly based on the equation mentioned below. Basically this element represents a pressure-dependent outflow.
Here is the orifice equation used for the D2A element in HAMMER:
Q = C A (2 g P)^0.5
Q - Discharge (cfs, cms)
C - The 'discharge coefficient' (distinct from discharge coefficient Cv used elsewhere in HAMMER) which will be computed based on the typical flow/pressure
A - The cross-sectional area of the opening (ft, m) - this is combined with the "C" as a single term
g - gravitational acceleration
P - Pressure head (ft, m)
As you can see, once the "CA" is calculated based on the initial head and flow, HAMMER can solve for other flows, as the pressure head changes during the simulation.
Here is an example, where the "Pressure Drop (Typical)" is set to 1 m and the "Flow (Typical)" is set to 0.5 m^3/s :
Q = CA (2*g*P)^0.5
0.5 = CA (19.6)^0.5
0.5 = CA * 4.4272
CA = 0.1129
If the pressure at the D2A is 1 m, then the outflow is exactly 0.5 m^3/s:
Now, say the pressure changes from 1 m to 10 m:
Q = CA (2*g*P)^0.5
Q = (0.1129) * (2*g*10)^0.5
Q = (0.1129) * 196^0.5
Q = 1.58 m^3/s
This is the flow you see in HAMMER when the pressure in the model is 10 m:
Note: There is a known issue in HAMMER 2023 (version 23.00.00.19) where the pressure results in the initial conditions calculation are inaccurate if the D2A is in a closed system and the upstream pump is off. This could occur in a pump start transient event with no downstream source, such as a reservoir. This issue will be fixed in a future release of HAMMER. The reference for this is 1423356. As a workaround, you can add a very small diameter pipe (such as 0.1 mm) that bypasses the closed pump or valve.
A sensitivity analysis wherein the cross sectional area, A, is varied would illustrate the consequences of a range of breaks, with an upper limit to A being the diameter of the incoming pipe(s). The analysis should also consider different locations of the break(s). Depending on the pipe network's topology, a sudden break can lead to the formation of vapor pockets with ensuing collapses and pressure spikes.
As an example, assume a case where you know the flow but not the pressure or head. In this case, the flow is 10 cfs through a 12 inch opening. The velocity in this case would be: V = Q/A = 10 / (pi*0.5^2) = 12.73 ft/s. If you assume C = 0.6, you could calculate the corresponding appropriate headloss using the orifice equation relationship: = (V/C)^2 / 2g = (12.73 / 0.6)^2 / (2 * 32.174) = 7 ft. In this example, the "Pressure Drop (Typical)" would be set as 7 ft H2O. HAMMER would then use the orifice equation to compute a change in outflow as the pressure changes. If you already know the flow at a corresponding head, you would simply enter them in the D2A properties and HAMMER will calculate the discharge coefficient internally and use that during the transient simulation to vary outflow with pressure.
Note: The "Gas Volume (Initial)" will impact the timing of the release of the air. The value you enter will be up to your engineering judgment, but a good starting point may be the volume of the empty pipe. A larger volume of air for the same size orifice will take longer to be expelled from the D2A. This, in turn, will impact the head increase at the source. The most important impact on the system will occur with the air is fully expelled, which is when the transient would occur. So while a large air volume will take longer to expel, the setup and size of the D2A may prove to the be most important part of the transient event.
Note: the "Rating Curve" discharge element type is used when the discharge out of your orifice does not follow a typical orifice-equation relationship. It allows you to explicitly define the flow released out of the system for certain pressures at the discharge location.
The following attributes are available when the "Discharge Element Type" is set to "Valve":
"Valve Initial Status" - This specifies whether the valve is initially open or initially closed.
"Time to Start Operating" - The valve starts to operate after this time, either opening or closing based on the initial status selection. It is measured from the start of the simulation. So a value of 5 seconds means that the valve remains in a fixed position for the first 5 seconds, and then starts to operate.
"Time to Fully Open or Close" - This is the time it takes for the valve to either fully open (if the initial status is closed) or fully close (if the initial status is open). It is measured from the "Time to Start Operating" value. Meaning, if the "Time to Start Operating" is set to 5 seconds and the "Time to Fully Open or Close" is set to 10 seconds, then the valve closes linearly between time t=5 and t=15, and the valve is fully closed 10 seconds after it starts operating. If you need to model more advanced closure characteristics or closure patterns, set the D2A as an Orifice, and use a TCV on the immediate upstream side. See: Modeling Reference - Valve Closure
"Flow (Typical)" - This is the typical discharge out of the valve when it is open. It is used along with the "Pressure Drop (typical)" to determine the pressure vs. outflow relationship (see equation further above).
"Pressure Drop (Typical)" - This is the pressure corresponding to the typical flow through the valve. It is referred to as the "drop" because the pressure beyond the orifice is zero (atmosphere). As mentioned further above, the "drop" in this case is not referring to any headloss across the node element but rather the pressure reported at the D2A element (since the pressure results in HAMMER refer to the gauge pressure inside of the pipeline with atmospheric pressure as the datum) at the "typical flow". The pressure and flow computed in the initial conditions will not necessarily be equal to these values, so you only need to enter any known pair. For example, if modeling a hydrant closure, you might enter the typical pressure and flow as the flow and pressure observed in a field test when the hydrant was opened.
You are basically defining an orifice size by way of the "typical" flow and pressure drop fields. By supplying one pair of pressure and flow, HAMMER can figure out the relationship based on the orifice equation that gives the pressure drop for any flow value. So, if unsure, you can use the orifice equation along with the size of your opening and an estimate of the "head" (pressure head drop) to solve for the typical flow. Selecting a pressure head drop close to a typical value you might see under normal operating conditions will yield the most accurate pressure/flow relationship during both the initial conditions and transient simulation. See further above under "How it works" for more information.
Note: a standard 2.5 in. (100 mm) hydrant outlet would have a pressure drop of roughly 10 psi at 500 gpm.
When the Discharge Element Type is set to "Orifice," only the typical pressure drop and typical flow are available. When set to Rating Curve, only a rating curve table is available, where you would enter the table of head versus flow for your discharge. Initial conditions and transient head/flow is computed based on the values in this rating table.
In the properties of the D2A element, results such as pressure and calculated discharge represent the results from the initial conditions. To see the outflow and other results during the transient simulation, use the Transient Results Viewer. Ensure that the D2A and adjacent pipe are included as a report points in the transient calculation options, then look at the Time History tab of the Transient Results Viewer to view a graph of flow, pressure, and other results during the transient simulation. Select the pipe endpoint adjacent to the D2A if it is at a tee (lateral pipe). If the D2A is in-series (two connected pipes), you would need to view the flow in both pipes and the difference would be the outflow.
If the pressure drops below zero (sub-atmospheric) at the D2A element during the transient simulation, the D2A allows air into the system. When this happens, the air enters the pipeline freely on the assumption that the opening for the liquid is infinite for air. In this case, the node acts like a reservoir at zero pressure and the air pocket respectively expands or contracts accordingly as the liquid flows away from or towards the node, but the air remains at the branch end point(s) located at the D2A. The rate of change of the air volume is based on the water flow rate of the adjacent water column in the pipe. If a check valve further upstream is closed, such as in an upstream pump station at a lower elevation, you may not see any air enter because the water column will not be able to move backwards (or may only be able to move slightly due to the elasticity of the water column). Similar to an air valve, the admittance of air from a D2A will only be able to help protect the system in the immediate vicinity of the D2A; pressure can still drop further upstream. See below for details on the application of a D2A for simulating an "inrush" or pipe filling/emptying.
For more information on the assumptions related to tracking air/vapor pockets in a transient simulation, see the following link: Assumptions and limitations of tracking air or vapor pockets in HAMMER.
The below model is an example of the use of the D2A element in HAMMER and has several scenarios for different configurations. Note:
How do do WaterCAD/WaterGEMS treat the discharge to atmosphere element?
OpenFlows Methods Product Tech Notes And FAQs