Surge tank detailed report - flow in tank and riser

Question

Hello, 
 I am modeling a differential surge tank. When looking at the detailed report, the results of "flow riser" is not as I would expect and I would like to make sure that I understand the definition of both flow columns in the report. Could you please confirm if my understanding is correct: 
 Flow riser: flow that goes through the top of the internal riser (when overspilling into the tank) 
 Flow tank: flow that goes through the orifice that connects the external riser to the tank. 
 If this is correct then I have an issue. I will share with you fictive dimensions of the surge tank in order to explain it better. 
 top of riser = 50 m; bottom of surge tank = 40 m; orifice elevation = 40 m 
 
 water elevation (WE) in the surge tank during simulation: 43 m < WE < 48 m 
 There is a valve downstream of my pipe network. The valve is closed at the begining of the simulation and opens after a few seconds so the level in the surge tank lowers. Using the "riser flow" data, I computed the volume of water that flows out of the riser. After a few seconds of simulation, this volume exceeds the volume of water that is stocked in the riser at time = 0 sec. I don't understand how this is possible since the riser can't be filled by the top (water elevation in surge tank lower than top of riser). 
 Can you help me to clarify this point? 
 Thanks, 
 
 Dominique

Jesse Dringoli · Answer

Dominique, 
 During a downsurge when water starts to leave the differential surge tank, water drains out of the riser, but at the same time water leaves the main part of the tank through the orifice. See diagram below. 
 The "Flow-Riser" shows the flow through the riser and the "Flow-tank" shows just the flow between the main part of the tank and the riser (from either the overflow above the top of the riser, and the flow through the orifice). 
 As an example in your model at timestep 21.0 seconds, the "flow-riser" shows as -14.4 CMS and the "flow-tank" shows as -7.7 CMS (negative values indicate outflow). From the Time history graph of flow in the adjacent pipes, you can see that 14.4 CMS leaves the tank at that time. So, this means that 7.7 CMS is going from the main part of the tank into the riser, and the remaining 14.4-7.7=6.7 CMS is draining down from the riser, with a total of 14.4 CMS leaving the tank. 
 So, the volume of water leaving the tank is coming partially from the riser and partially from the water in the main part of the tank. 
 I have made some updates to the Surge Tank Modeling Reference article to clarify these points.

Jesse Dringoli · Answer

Dominique Poissant said: at timestep 38 seconds the opposite happens: 52.056 cms leave the riser and 62.64 cms leave the tank ( flow in riser < flow in tank). These results are not consistent. Do you have an explination? 
 This is due to the balancing of the water level between the main tank and the riser, after the initial downsurge. 
 When the downsurge first occurs, water quickly leaves the surge tank's riser, so the water level in the riser drops faster than the water level in the main part of the tank. This can be seen in the diagram I posted in my last reply. 
 After things start to settle down a bit (by 38 seconds), the tank starts to balance itself because of the differential between the water level in the main part of the tank and the lower water level inside the riser. 62 CMS passes from the main part of the tank into the riser, due to the head difference, and 52 CMS leaves the riser into the pipeline (and down P-12). Since the flow into the riser from the tank is higher than the flow leaving the riser into the pipeline, the riser water level rises (as the main tank level drops) - you can see this by looking at the "head" column (water level in the riser) - note how this starts to increase when the "flow-tank" is greater than the "flow-riser". 
 By around 48 seconds, things balance out and the flow from the main tank to the riser (flow-tank) is about equal to the flow out of the riser and into the network (flow-riser).