Velocity Limit at Centrifugal Suction Flange

[tmartin125]

I have seen pump suction velocity limit recommendations of between 5 - 7 ft/s. Assume this is for the suction pipe, however what are the velocity limits at the pump flange? Goulds model 3196 centrifugal pump appears to have very high velocities at the pump inlets. For example the 1.5x3-10 has a 3 inch suction while the pump curve extends beyond 360 gpm. This corresponds to velocities upwards of 17 ft/s at the pump inlet which appears improper. Is there a published guideline for velocity at pump inlets?

 

[bimr]

The pump manufacturer is setting the velocity criteria at the pump inlet through the use of the manufacturers published pump data. It is typical design practice to install a pipe reducer on the pump flange to increase the pump diameter at the inlet of the pump.

The velocity at the pump inlet is the pump manufacturer's responsibility.

 

[LittleInch]

Ultimately though why do you need / want to know? This is the size chosen by the pump vendor for your requirements so all you need to do is mate to it. 5m/sec is quite fast, but not unheard of.

Inlet piping is often, though not always a larger diameter than the discharge line / header due to issue over friction losses and NPSH, but you need to size for your conditions.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[tmartin125]

My supervisor is convinced that 5 m/sec at the inlet of the pump will deteriorate the system (I guess the inlet reducer?). I agree that the pump inlet velocity was designed by the manufacturer and therefore not in our scope of concern. However my supervisor is not convinced so I have been doing additional research to determine who is correct, ie find published standards if they exist, contact vendor, post on forum, etc. It should be noted that my supervisor is correct most of the time.. guess thats why he gets the big bucks.

 

[1gibson]

Ask for a reference list of that exact pump at that speed. Sometimes there is nothing to stop the sales force from saying "it works fine at 3,000 rpm, let's sell it at 3,600 rpm and see what happens." All they have to do is talk someone into creating the new curve.

 

[rconner]

I agree 17 fps sounds high; however, for at least a couple reasons I don't think one should really want to run pump/suction piping at the extreme "maximum" velocity. [The risk of cavitation is greater, and the pump is less efficient from the standpoint of energy input vs pumping output etc.,(e,g see

http://www.pump-magazine.com/pump_magazine/pump_articles/article_01/article_01.htm.)

]

 

[LittleInch]

Sometimes "recommendations" become more fixed in peoples heads as rules and limits than they were ever supposed to. The key start point is your pump vendor. After that it's up to you if you want to use a pump with a bigger inlet flange then you can, no one is stopping you.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[jgailla]

Why are you considering running the pump near the end of its curve?

 

[bimr]

What do you mean by "will deteriorate the system (I guess the inlet reducer?)".

Maybe what your supervisor is referring to is the practice of selecting pumps to operate at higher speeds than lower speeds. Higher speeds result in a less expensive pump but with a shorter lifespan. Pumps that operate at lower speeds tend to last longer.

 

[TenPenny]

The Goulds 3196 pumps have been built since the 1960s. Countless thousands of these size pumps have been installed and operated at 3600 rpm.

The inlet velocity in the suction nozzle to a pump is not really comparable to the velocity in pipe, it's an integral part of the hydraulic design. The inlet nozzle size is related to the eye area of the impeller which is related to the flow rate the pump is designed for.

Answer me this: if you think a 3" suction is too small for 360 gpm, what do you think about the 1.5" discharge on that pump? Doesn't the 360 gpm have to exit the pump as well?