Have very short suction piping length. 2

[Lovison]

We have a customer who did not follow Hydraulic Institute Standards for having a suitable length of horizontal straight run pipe prior to the suction inlet of the pump.

Therefore we are getting vortexing or a non-streamlined entry into the pump which sounds like cavitation or a crackling sound.

I considered a vortex breaker but I'm wondering what other things I could do. The breaker worked somewhat but the noise is still present and they (we) are loosing mechanical seals like crazy and these aren't cheap. Help and ideas would be appreciated. Wayne E. Lovison
service-parts@naglepumps.com

 

[BobPE]

Lovison:

Sounds like there is something else going on like cavitation. Give us some more info on flows, diameters and heads. A short run does not necessarily attribute itself to immediate damage, rather the damage usually occurs over long periods of time.

If you have a vortex going into the suction from the surface of the suction liquid, you should be able to see it occuring. have you seen it? If not, a vortex breaker will not do anything and will just be another minor loss on the suction decreasing the NPSHa and increasing the noise you are hearing.

BobPE

 

[PUMPDESIGNER]

You never stated what the customer was complaining about.
What kind of pump?
PUMPDESIGNER

 

[ccfowler]

What about some creative pipe installation? Can some larger diameter pipe be routed around a bit so that a decent length of proper-size pipe can be installed at the pump suction port? What about extending the suction pipe into (or farther into) the source reservoir?

How does NPSHa compare to NPSHr, and how does the flow rate compare to BEP flow? If the flow rate is much less than BEP flow (a likely possibility), the NPSHa may need to be very much greater than the "3%" NPSHr.

 

[25362]

To avoid pre-rotation straightening vanes are sometimes recommended. But before anything else all the questions above must be answered by Lovison.

 

[Lovison]

It's a 10" centrifugal pump single suction with an enclosed impeller design. It's rated 5000 USGPM @ 180' TDH. The product is waster water with a 1.05 gravity and 100 Deg. F pumping temperature. The motor is an inline 400HP @ 1200 RPM.

The suction is flooded and the liquid is in a tank which does not allow you to visually witness what's going on. The system operates in a closed loop arrangement. The feed side back to the tank is gravity drained. The liquid goes into centrifuges where the solids are screened out.

The suction line run is only about 27" (max.) There is a gate valve and then the connection to the tank itself. NSPHr is not an issue that I've been told since we are flooded. The suction line is a 12" ips was a 10" earlier but to date we are still loosing seals.

The impeller is balanced and is a 4-vane design. Some system variables can take place such as creative valving by the operators we've since advised them that this creates excessive line velocities swings and could be a factor relating to the noise being heard from the casing areas.

I thought I had been exposed to most everything but I'm always learning something every day but this one has really got me stumped. I can not figure out what to do or try next?

Wayne E. Lovison
service-parts@naglepumps.com

 

[BobPE]

Lovinson:

You stated "rated flow," have you looked at actual flows in the field. I would lean towards suction cavitation based upon my experience with centrifuge applications in the field and the operators ambitions to process more solides to get home earlier. This means that they open the supply valve to the centrifuges a which are open referneces to atmosphere nd subsequently run the supply pumps off their curves. But then I come to your statement about the suction pipe diameter of 12 inches. At your stated flow rate, I would design a 24 inch suction. With this in mind, mabe you can answer some questions now. With a short 12 inch suction, one 90 degree bend and entrance and exit losses on the suction, you already have over 16 feet of headloss to the impeller. I am sure the NPSHr is quite high for a pump in this application.

Also, what is the percent solids of the fluid you are sending to the centrifuges? You could wuite possible be dealing with fluid transition to non-newtonian, which will have the direct result of multiplying the head losses encountered in the system amoungst other nasty hydraulic things.

I would get an engineer experienced in these things to look it over simply because of the size of the system. You can't affor to be wrong with a 5000 gpm system because too many people with notice that financially.

BobPE

 

[BRT549]

You could extend the suction pipe into the tank itself. That would effectively increase the straight run of pipe to the pump inlet (CCFowler mentioned it too). Is there a 90 ell in the suction line? I see that in the BobPE post, but not in your posts.

I've had success with the suction pipe extension trick. In addition, putting multiple inlets on the suction pipe inside the tank (cut some extra holes in the pipe near the end) minimizes vortexing from a single inlet.

Increasing the inlet pipe size from 10" to 12" was a step in the right direction, but I would go up to at least 16" for 5000 gpm.

 

[2081940]

Keep the velocity of the fluid at the suction pipe, between 1 to 2m/sec

 

[25362]

I repeat: Is there a chance of inserting straightening vanes ?