What is the purpose of the "Valve Type" field in the properties of a TCV? When is this used and what are the assumptions behind it? Do I need to use this dropdown field when I need to model a Globe, Butterfly, Gate, Ball or Needle valve?
The selection made from the "Valve Type" field for a TCV (butterfly, globe, etc) allows the program to relate "relative closure" (%) in a pattern to an internally used discharge coefficient to represent the loss through the valve for that position, based on the entered "fully open discharge coefficient". So, this enables you to accurately model the closure characteristics.
The "User defined" option enables you to model any type of valve, which is useful if your valve type is not one of the other default options, or if the closure characteristics are different. Some examples of when you might use the User Defined option are: Plug valves, Diaphragm Valves, Pinch Valves, Iris valves.
When the Coefficient Type field for the TCV is set to Headloss Coefficient or Discharge Coefficient, headloss in the initial conditions will always be based on that value compared to the fully open conditions (modeled either with the "minor loss coefficient" or "fully open discharge coefficient" fields, respectively), and the reported "Relative Closure (Calculated)" will then be based on the Valve Type. When using the "Pattern (Valve Settings)" for an EPS (or logical controls to change the valve setting), or when using a transient Operating Rule for a transient simulation (in HAMMER), the program will use the Valve type to determine the change in discharge coefficient as the relative closure in the pattern is changing.
If you're only modeling a Steady State simulation in WaterCAD or WaterGEMS, or an EPS with a fixed pattern on the valve, then the Valve Type will not have an impact on the headloss results, as they will be based on the entered headloss coefficient or discharge coefficient.
The Valve Characteristics Curve option for the Coefficient Type of a TCV allows you to specify initial conditions by way of a relative closure instead of directly entering the headloss coefficient or discharge coefficient. In this case, the Valve Type will be used more directly, to translate the initial relative closure that you enter, into a discharge coefficient that is used to determine the headloss through the valve (both in steady state/EPS and in a transient simulation). The program accomplishes this by looking at the initial relative closure that you enter, along with the "full open discharge coefficient" that you enter, along with the characteristic curve defined by the selection of Valve Type (see graphic below)
With HAMMER, you can configure a transient Operating Rule to change the position (relative closure) over the course of the transient simulation. You would set the initial relative closure equal to the position in the initial conditions, as seen in either the "Relative closure (initial)" field (with the Valve Type set to Valve Characteristics Curve) or the calculated initial relative closure as seen in the Results section of the TCV properties.
Below is the relationship between relative closure (the X axis) and relative discharge coefficient (the Y axis - relative to the "fully open discharge coefficient") used when selecting one of the predefined "Valve Type" options.
This chart is based on Fok, A.T.K., “A Contribution to the Analysis of Energy Losses in Transient Pipe Flow”, Ph.D. Thesis, University of Ottawa, 1987.
T/Tc is time over time to fully close (so 0.5 would mean 50% relative closure, or "stroke") and A/Ao is area over full area (which can be correlated to discharge coefficient, so it is essentially relative discharge coefficient). Basically it shows the relationship between stroke and discharge coefficient. Meaning, one particular valve will have a different flow control characteristic when closed half way versus another type of valve.
If your particular valve doesn't align with these relationships, you always have the option to choose "user defined" as the valve type, then enter your own relationship between %closed and %discharge coefficient. Or, you could of course enter valve positions the traditional way using headloss coefficient or Discharge coefficient instead of %closed.
The curves in the chart above have the functional form:
1 – Y^k ...
where needle valves have k = 2.0; circular gate valves, k = 1.35; and globe valves k = 1.0;
(1 – Y )^k ...
where for ball valves, k = 1.35; and butterfly valves, k = 1.85.
More information can be found at the following paper: Fok, A.T.K., “A Contribution to the Analysis of Energy Losses in Transient Pipe Flow”, Ph.D. Thesis, University of Ottawa, 1987.
Note: most valve manufacturers can provide the discharge coefficient(s). The closure vs discharge coefficient relationship could potentially be significantly different from the above, in which case the "User defined" option should be used, to enter the manufacturer data.
A spreadsheet can be downloaded from the below link (you must be signed in first), as a reference to see the actual closure vs discharge coefficient relationships used with the default "valve types" in HAMMER.
HAMMER Valve Type Comparison Tool
In HAMMER, changing the Valve Type of a valve that is closing, can have a large impact on the transient response (the magnitude of the resulting surge/transient spike in pressure). Consider the Globe valve Type compared to the Butterfly Valve Type as an example.
From the valve opening area figure shown above, you can see for example the Butterfly valve type's discharge coefficient changes more gradually than other valves, near the critical last bit of closure. In almost all cases, the last bit of closure of a valve will govern the transient response. Hence, it is best to close a valve slowly right before it is fully closed. Accordingly, in the HAMMER model, the maximum pressure caused by closing this type of valve is smaller than that of valves of other types and the surge/peak occurs earlier.
Notice in the below graphic, a butterfly valve is on the extreme end of the spectrum with a "decelerating" trend - note the part of the curve near the bottom-right side, just before closure. So, in the example case shown further below of a valve closure, the momentum change happens slow enough that a significant transient response does not occur, and you see the system settle on it's final steady state condition without a notable "upsurge" (spike in pressure). It can also help to view a profile anile animation to visualize what happens as the valves are closing.
Conversely, a globe valve has a linear closure rate with respect to discharge coefficient, so the last bit of closure occurs faster than that of the butterfly valve. Because momentum changes faster in this case, it causes a slight "upsurge" which you can observe as the spike in pressure in the graph.
Modeling Reference - Valves
How are headlosses determined for TCV's with the different initial status settings?