Pump inlet velocity high, but too high?

[Mister T]

Hello guys,

I know that there are several topics dealing with the maximum pump inlet velocity but I still don't wether I face a problem or not.

There's a centrifugal pump (~1800rpm) connected to an atmospheric storage tank of hot water. The outlet of the tank is a 20"/DN500 line of approx 12m. The inlet nozzle of the pump (selected by the vendor) is 8"/DN200. Approx 1m before the pump inlet the diameter is reduced from 20" to 8". Pump flow is 550m³/h. That means that there is nearly 1m of suction pipeline with a velocity of < 5m/s. I know that it is common to have inlet nozzles with high velocity but here there is additional pipe lenght (<1m) in front of the nozzle.

Is this fact alone a reason to be concerned? Or only if the npsh_a value is too low?
The minimum npsh margin is 1.5m but this situation should only occur every now and then, if at all. For normal/ continuous operation the npsh margin is around 8m.

From this perspective I'd say it is fine. Looking at the maximum inlet speed I don't feel that comfortable. But it is more a feeling to be honest.

What do you guys think?

Thanks for your help!
T

 

[TiCl4]

I want to make sure I understand this correctly. You have:

[ul]
[li]A pump with an 8" NPS suction nozzle size.[/li]
[li]This pump is designed to run 550 m3/hr - i.e. that is on the normal operation range of the pump.[/li]
[li]Your inlet piping is reduced to 8" NPS before the pump.[/li]
[/ul]

Given these three things, I don't see an issue. These pumps are designed and tested by manufacturers to perform properly. I don't see why a pump manufacturer would design a pump with an 8" suction with a flow rate that would cause inherent issues due to high velocity.

 

[LittleInch]

NPSH margin of 1.5 is as low as you want to go.

Have you calculated that with the extra 1m of 8" pipe?

NPSH is measured from the centreline of the flange usually and if you're that close you need to get it right.

Remember NPSH isn't cavitation so when operating at what I assume is low level on the tank you could start to hear interesting tinkling noise coming from your pump....

But inlet velocities into pumps are often quite high.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Mister T]

A pump with an 8" NPS suction nozzle size.
This pump is designed to run 550 m3/hr - i.e. that is on the normal operation range of the pump.
Your inlet piping is reduced to 8" NPS before the pump.

3 times yes.
It's the normal flow rate of the pump and the 8" was selected/ given by the vendor.

My concern is not the nozzle itself. Inlet nozzles tend to be small. But normally (normally means what I saw in the past) the reducer was just before the pump and the inlet pipeline in front of the nozzles was very short (few inches).
But now it is different. There's a piece of pipe (approx 0.8m) with already 8" (=nozzle size) in front of the pump. Haven't seen such a "long" pipe with that small diameter upstream of the pump before. And I don't know if this is a problem or not.

 

[LittleInch]

It's not a problem unless you are really hurting for NPSH or inlet pressure.

550m3/hr in an 8" is about 5.3 m/sec.

That's nothing to write home about. 50 m/sec, yes, 5 No.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Mister T]

@LittleInch

I have measured from the top of the pipeline (not the center). So the additional height (~0.1m) should cover the additional pressure loss of that 1m lenght. Was never afraid of the pressure but because of the velocity.

I searched a lot at Google and for inlet velocites I always see values from <1 m/s up to 2m/s, in some cases up to 4m/s (but rarely) but never beyond.

 

[horacio Torres]

Just calculated the pressure drop of 0.8 mt suction side and NPSH Available , you will see.

Horacio

 

[LittleInch]

It's a little bit high, but not extraordinary.

It's up to the pump vendor / designer really. What's the outlet nozzle size?
What is the NPSHR figure?
assume you've allowed for the higher vapour pressure of the "hot" water(how "hot?") ? - Just checking.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Mister T]

Outlet nozzle is 6"
The NPSHr given by the vendor is 4,00m

I've considered the vapour pressure of 80°C water. The temperature should be below. However temperature will not exceed that temperature.

 

[Artisi]

Inlet pipe diameter is a function of NPSHr/a, if the pump is rated for 550m3h with an 8" inlet flange and without any further detail, we can only assume flow / inlet size is adequate, in your case with elevated temp. you simply need to undertake an NPSHa calculation to see if it's within NPSHr requirements as per the vendor information.
Because you haven't seen a pump with the same pipe diameter as the inlet flange doesn't translate into a problem.
A simple 101 hydraulic exercise.

Look at any reasonable hydraulic book on how to calculate NPSHa.

PS. I just checked on Google for NPSHa calc. here is even a plug-in the numbers calculater available.

It is a capital mistake to theorise before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts. (Sherlock Holmes - A Scandal in Bohemia.)

 

[LittleInch]

The pump model might be at the edge of its envelope, but should be ok.

At 1800 rpm it's going fairly slow so your NPSHR is ok and good to know you've checked.

We forgot to ask if this is just a question in general about an existing piece of kit or if there is something new / changed.

If the pump is new or is being proposed, make sure you check out its power and efficiency compared to alternatives - the CAPEX might be lower but you could save money money overall by reducing OPEX costs. Just a thought.

So are you happier now?

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.

 

[Mister T]

Thanks for the advice. CAPEX and OPEX has been considered. Regarding this the pump is fine.

So are you happier now?

More relaxed yes. The figures (NPSHa/r) were alright. Just my feelings weren't.

 

[Nutzman]

Typically a suction reducer should be flat topped (Eccentric) and should be 3 x diameter difference in length. Meaning 20-8 = 12.
12 x 3 = 36 inch long. The reducer would need to be fabricated and flanges fitted.
The flat top would prevent any gas's forming in the top of the pipe, causing turbulence and hydraulic instability at the eye of the impeller.
Suction isolation valve to be fitted on the 20 inch section.