Do Weighted Check Valves Significantly Affect Pump Station Calculations? 1

[CivilEngineer401]

Does anyone know a good way to figure out the head loss through a weighted check valve in a sewage pump station?

Most of the pump stations I work on have weighted check valves on the discharge line. We typically install them because operators like the visual confirmation when the pump is running, and they provide a way to hold the check valve open if needed to empty the forcemain back into the wetwell.

The problem I’ve encountered over the years is that there always seems to be a discrepancy between the actual flow of the pump station and my calculations. My theory is that the weighted check valves might be the cause.

Typically, I use a K value of 2 for the head loss coefficient for a 6” weighted check valve. However, I’m starting to believe that the “K” value might be much higher than typical when the valve is barely open (e.g., at flow velocities of 1 or 2 ft/s), potentially exceeding 10. When we turn the VFD up to full speed and the valve opens more, resulting in higher velocities (e.g., ~7 ft/s), the K value seems to drop back to a more typical value of around 2.

Since I don’t typically account for different K values at varying flow velocities, this variation may not be reflected in my calculations. What do you think? Could this explain the discrepancies I’m seeing?

 

[Artisi]

Your calculations are probably the problem, can't see 1 correctly sized valve in the overall system making enough difference to even be considered.

 

[pointsman]

Partially open check valves will have higher K values. This can happen when flow velocity is insufficient to force the disk fully open. You can check with the valve supplier to find the minimum velocity required to fully open the valve. They might even be able to provide you with a chart showing the K value as a function of valve opening percentage.

There are many other potential causes of unexplained headloss in a piping system. I can't say whether or not this fully explains your calculations' discrepancies.

 

[bimr]

Agree with Artisi on this one. The headloss through the check valve is considered a minor loss and should only be a few feet of headloss. Issues are created as a result of the Weight, Lever & Hinge assembly:

• Firstly, the disc is pulled directly into the flow often allowing it to only open about half way in typical flow rates, needing approximately 2.5m/s to fully open. This restricts flow to sometimes less than half and increases head loss.

The chart from Mueller shows the headloss is minimal:

 

[CivilEngineer401]

Agree with Artisi on this one. The headloss through the check valve is considered a minor loss and should only be a few feet of headloss. Issues are created as a result of the Weight, Lever & Hinge assembly:

• Firstly, the disc is pulled directly into the flow often allowing it to only open about half way in typical flow rates, needing approximately 2.5m/s to fully open. This restricts flow to sometimes less than half and increases head loss.

The chart from Mueller shows the headloss is minimal:

View attachment 2055

Thank you! That’s the chart I’ve been looking for. I was theorizing that the head loss at 1 or 2 fps might be unusually high because the check valve was barely open, and then it would stabilize at higher velocities. This chart clearly shows that nothing unusual is happening at low velocities