PDA

View Full Version : Negative Flows in Forcemain



Healz
January 29, 2016, 10:15 AM
Hi,

I've noticed a few Forcemains in my model that show negative flows and not sure why? Checked to make sure MH's along the main had a surcharge depth and elevations seem correct as well.

Any help would be appreciated.

Thank you.
Shawn

Robert Dickinson
January 30, 2016, 01:11 PM
Hello Shawn,
It is negative because when the pump turns off the HGL at the downstream end of the force main is higher than the HGL at the upstream end of the force main. You can have a flap gate in InfoSWMM to present backflow
Regards,
Robert Dickinson

The key problem discussed here is to keep the Force Mains full or d/D equal to 1 when the pumps turn on, it is sometimes difficult to get it to work like ICM or IWCS due to the fact that there is only one Q in the link of the SWMM 5 solution compared to the 4 or more flow points in the IWCS solution. I have suggested in past that the user add a break node in the force main at the end so that the downstream end of the Force Main is full – however does not always work if there is a gravity main at the end of the rising force main. The gravity main takes all of the flow coming down the long force main and the downstream node depth stays small so the Force Main is never full – this annoys the customers. My suggestion is to increase the gravity main roughness to simulate the effect of the force main transitioning to the gravity main – this keeps the depth higher and the force main full most of the time and full when the pump turns on.
Here are eight suggestions:
1. Use a Flap Gate for the rising main with HW Force Main Coefficients,
2. Add a Break Node at the end of your longer Force Mains with a Surcharge Depth in the node using the Insert Manhole Tool,
3. The reason d/D values for the force main were normally less than 1 was that the downstream node of the Force Main had a low depth, as the link downstream was taking the Force Main flow almost immediately, adding a Break Node keeps the downstream end fuller,
4. Change the link AFTER the Break Node to a Gravity Main and make the roughness n value higher to simulate losses from the Force Main to the gravity main (2 to 3 times rougher),
5. This will increase the depth of the node at the downstream end of the Force Main and force it to be full most of the time.
6. As mentioned there is one Q in the force main link and three depths, the reported d in the d/D graph is from the midpoint depth or the average of the upstream and downstream link depths,
7. In reality (model reality) the force main is always full at the upstream end of the link but due to the low downstream depth
8. Adding the increased roughness in the gravity main makes the results look more like the what the users expect for the d/D value and have the advantage of somewhat matching reality

Robert Dickinson
January 30, 2016, 01:15 PM
More due to Forum image limitations


http://swmm5.files.wordpress.com/2015/02/wlemoticon-lightning1.png Use a Flap Gate for the rising main with HW Force Main Coefficients,


http://swmm5.files.wordpress.com/2015/02/wlemoticon-lightning1.png Add a Break Node at the end of your longer Force Mains with a Surcharge Depth in the node using the Insert Manhole Tool,


http://swmm5.files.wordpress.com/2015/02/wlemoticon-lightning1.png The reason d/D values for the force main were normally less than 1 was that the downstream node of the Force Main had a low depth, as the link downstream was taking the Force Main flow almost immediately, adding a Break Node keeps the downstream end fuller,


http://swmm5.files.wordpress.com/2015/02/wlemoticon-lightning1.png Change the link AFTER the Break Node to a Gravity Main and make the roughness n value higher to simulate losses from the Force Main to the gravity main (2 to 3 times rougher),

Robert Dickinson
January 30, 2016, 01:16 PM
http://swmm5.files.wordpress.com/2015/02/wlemoticon-lightning1.png This will increase the depth of the node at the downstream end of the Force Main and force it to be full most of the time.


http://swmm5.files.wordpress.com/2015/02/wlemoticon-lightning1.png As mentioned, there is one Q in the force main link and three depths, the reported d in the d/D graph is from the midpoint depth or the average of the upstream and downstream link depths,


http://swmm5.files.wordpress.com/2015/02/wlemoticon-lightning1.png In reality (model reality) the force main is always full at the upstream end of the link but due to the low downstream depth


http://swmm5.files.wordpress.com/2015/02/wlemoticon-lightning1.pngAdding the increased roughness in the gravity main makes the results look more like the what the users expect for the d/D value and have the advantage of somewhat matching reality

Robert Dickinson
January 30, 2016, 01:18 PM
http://swmm5.files.wordpress.com/2015/02/clip_image002_thumb.jpg (http://swmm5.files.wordpress.com/2015/02/clip_image002.jpg)Break Node Placement in InfoSWMM