NPSHA affected by position of pipe irregularity?

[Keith1885]

Hi, we have a pump that's cavitating. Right before the suction inlet the piping has 2 45 degree bends welded together to gain about 2" of elevation. I know that there should be around 3-5 pipe diameters of pipe in a straight line before the suction inlet, but in this case it's about 0.5 pipe diameters. I would think moving this farther away from the pump would alleviate the cavitation, but I don't see anything in the calculation for NPSH that accounts for the position of this bend in the piping. Ideally, I know it'd just be best to get rid of the bend, but cutting a new outlet from the tank would be a lot more expensive. Thanks for any advice on this.

 

[ClicketyClack]

The location of the elbows is more for laminar flow reasons than NPSH. It allows the fluid to "straighten out" so it enters the eye of the impeller at the center, with minimized turbulence, rather than overloading one side of the pump inlet.

The turbulence and friction loss associated with elbows will reduce the NPSHa regardless of where they are located in the suction. But you want as close as you can get to laminar flow at the pump suction. This is why you want a straight run into the pump suction.

 

[BigInch]

Not to mention the flow loss of the bends themselves, you should not have any fittings within a preferred 10xD of the pump inlet.

http://virtualpipeline.spaces.msn.com

 

[Zapster]

By the way, what is the NPSHA and NPSHR for your pump?

 

[dickon17]

What is the NPSH required by the pump? What is the calculated NPSH available? Increased friction loss from 2 45 degree bends should be negligible. Are you sure the pump is actually cavitating? If NPSHR is significantly less than NPSHA, I would look elsewhere.

 

[BigInch]

Is anything negligible when facing cavitation?

http://virtualpipeline.spaces.msn.com

 

[25362]

Suction pipe vortices and pre-rotation increase the NPSHR.

 

[ClicketyClack]

NPSHr is a determined by pump variables (rating, RPM, impeller diameter, etc).

NPSHa is determined by system variables (friction loss, tank level, fluid temp, etc) Therefore vortices and pre-rotation are said to reduce NPSHa, not increase NPSHr.

 

[dickon17]

If this is a double suction impeller design then a bend close to the inlet could cause cavitation in one impeller eye only. I ask again, are you sure this is cavitation? Did you run the numbers?

 

[25362]

Experts tell us that prerotation effects may also depend on the impeller geometry and specific speed.

There are cases where at low (partial) flow rates the NPSHR undergoes a drastic increase, an effect attributed to prerotation setting in.

In extreme cases, with low NPSHA, the system may develop noise and vibration, as if subjected to hammer blows.

 

[Artisi]

What type of pump are you talking about - end suction or horizontal split case - as the positioning of bends at the inlet has different effects in each case.

 

[BigInch]

None of them good. Its best to live with the 10D rule or live with the consequences.

http://virtualpipeline.spaces.msn.com

 

[Artisi]

BigInch - not so for a horizontal split case pump unit, a 45 or 90 degree bend is ok fitted directly to the inlet flange provided the flow direction is in the same plain as the inlet branch leading to the impeller.

 

[BigInch]

Easier to remember the rule, rather than an exception or two.

http://virtualpipeline.spaces.msn.com