This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Negative pressure in pumps

Dear all, 

I am modeling a network in Hammer and I came across a strange situation. When I stop the pumps without protective devices (Hydropneumatic tank, air valves, ...), negative pressures occur only in the pipes after the pumps, as expected. But when I put on the protective device (in this case a hydropneymatic tank) and stop the pumps, I see negative pressure on the pump, which doesn't seem to make sense.

To solve the problem, I simulated a triple acting air valve just before the pump and the situation significantly worsened.

Without surge protection:

With surge protection:

Any suggestion? I will share the model to show better the situation. 

Parents
  • Joao,

    Often with a transient simulation you will see things that you did not expect.

    I would have to speculate here but it could be that without protection, the pressure drops upon pump shutdown, then stays low with no subsequent secondary effects from vapor/air pocket collapse. When introducing protection, it could be that the protection influences how the downsurge wave is reflected back toward the pump, possibly closing a check valve at the pump (if you are using one) because of the pressure staying or reflecting as higher than it would have, and the check valve closure may be causing the negative pressure on the upstream side of the pump.

    When you add an air valve, air may be entering, and then releasing too quickly, causing the adjacent water columns to collide and cause a worse transient than without the air valve (and these would quickly reflect in the typically small section of piping on the pump suction side). See: Transient pressure worse with air valve added

    Be sure to adjust your transient reporting options to enable you to see extra detail in the results, then animate the transient profile to better understand how things are interacting during a transient simulation.


    Regards,

    Jesse Dringoli
    Technical Support Manager, OpenFlows
    Bentley Communities Site Administrator
    Bentley Systems, Inc.

  • Hello Jesse,

    Thanks for the answer.

    Before sending the question, I had not simulated with a check valve but I already did it and the negative pressure in the suction pipe worsens... It is strange. I'll try to simulate different air valves and see if I can solve the issue.

    I have already sent the model. 

  • Hello Joao,

    You can try using delayed closure of check valve to compensate for the reflection of transient wave incase of a surge protection device (like a triple acting air valve). With a delayed closure, it is possible that the transient wave settles faster. Also, it is possible to use a pump without a check / control valve with a very high time to close (like 9999 seconds). This will allow reverse flow in the pump and can help if having a check valve close instantly is causing higher transients. See the below articles on this for more details;

    Modeling a Check valve or Control valve in a pump for transient analysis

    Modeling Reference - Check Valves in HAMMER


    Regards,

    Yashodhan Joshi

    Answer Verified By: Jesse Dringoli 

  • Hello Yashodhan,

    I followed your suggestion and modeled the pump with a control valve with a slower closing time and solved the problem! 

    Thank you!! 

Reply Children
No Data