Pump Decision

[kyhan]

I'm having a problem deciding on whether to keep the current centrifugal pump (which periodcally pumps nothing for unknown reasons, a guess is the excessive piping headloss, which is why we're moving the pump location to the bottom of the viscuos liq. column) or to replace it with a pump with half the speed, half the NPSHr and almost twice the impeller size or a PD (moyno) pump. Any suggestions? For any choice we want a 20gpm flow. Sometimes there might be solids in the stream, though rarely.

 

[TD2K]

You really need to identify the cause of your pump's problem or the new pump might suffer the same problem.

You say it could be excessive head-loss, that's fairly easy to check. Is there a suction strainer in the line that could be partially plugged? I'd suggest you stand back and start eliminating causes (which you may have done I realize).

For pump suggestions, you'll need to provide a lot more information than just the flow. What is the head required? What is the fluid? Temperature? Sealing requirements if that's critical wrt emissions? How much NPSHA do you estimate you have? What type of solids? Size range? How much?

 

[kyhan]

i'm not looking for a particular pump suggestion really, we already have quotes of the pumps of choice. the temperature npsha etc is the same for all three of the pumps.

the suggested centrifugal replacement for the current centrifugal is typically the same except for the minimum npshr, impeller diameter and speed.
the moyno was something someone else suggested.
I guess it really just boils down to whether its centrifugals or PD.s

 

[pmover]

to supplement TD2K comments and before obtaining quotes (otherwise the same problem(s) will likely occur again),

define the fluid and any solid concentrations,
define the process parameters (pressure, flow, temp, viscosity, etc.),
define any operational requirements or constraints,

then the appropriate pump can be choosen using sound principles.

good luck!
-pmover

 

[Longinthetooth]

All I can add is that I have heard at least 2 pump specialists say that over 90% of all centrifugal pump problems relate to the suction side of the pump. Suggest you start there.

 

[an0r0v]

Sound like you do not have a handle on the problem(discharge head requirements, suction head requirements, etc) and therefore not sure if the pump will work. Centrifugal pumps work well when properly sized and the 20 gpm range is also well within a typical gear pump range. If the problem is in the discharge area then I would be inclined to put in a gear pump. On the other hand, if the suction is starving due to undersize feed piping (perhaps caused by high viscosity, low temperature, etc), I am afraid you will have the problem no matter what kind of pump you use. The half speed is a trade off for doubling the impeller size.

 

[liberoSimulation]

Just to add some points to consider:
- Check the suction line including length, friction loss as you might find much friction loss at the suction line.
- Check the design basis vs. the operating conditions (e.g. solids content, vapor pressure/Temperature/static head..)that might be different and would cause different pump characteristics
- Check the head of the system such as elevation of both destinations, associated fittings and other friction loss piece.

Hope this would help

Cheers

 

[elguero]

Read the cover story of August "Chemical Engineering" magazine, written by yours truly.

 

[elguero]

Also see October issue of Chemical Engineering Progress, p. 14. Could be you have noncondensible degassing, causing a vapor lock.

What level of viscosity? Like kerosene? Like warm molasses? Like cold molasses? Have you calculated pressure drop in the suction line?