Parallel Pump Operation and Cavitation

[MatAI]

We have three boiler feed pumps operating in parallel and suspect cavitation in the #1 & #3 Pumps. The pumps supply water to a boiler, and 90% of the time two pumps must operate to satisfied the steam demand. The original pumps were installed in 1970, and the #2 pump was replace last year due to failure. Since that time we've experience what we suspect as cavitation, #1 Pump failed from outboard bearing demise and the #3 Pump failed from the 1st stage impeller, #1 of six, being excessivley eroded to the point holes existed. We suspect cavitation has been present for quite some time because operational records indicate similar problems. Do you feel this is a correct assumption? The recirc line was removed years ago, and as of this writting, there is no ability to balance flow by recirculation, which is what we'd like to achieve as we feel this is the resolution to our problem. I am new at this, so please understand when I ask how is this accomplished. I would think installation of a recirc line would be somewhat simple, but have been insuccessful in finding any plans or what instrumention and equipment one should use to correct this type problem. I would appreciate any help one could provide, and will be looking forward to whatever responses I receive. Thank you.

 

[vanstoja]

Finding holes in the 1st stage impeller of your #3 pump certainly supports your presumption of cavitation damage involvement. However bearing damage to #1 pump and a previous unspecified failure mode of #2 pump suggests that some or all of these pumps are operating far enough off-design for long enough time periods to cause significant damage. Your statement that 2-pump operation occurs for 90% suggests that the 1-pump or 3-pump operating modes which together account for only 10% of operating time could be the culprit modes. Damage can occur in a relatively short time in some off-design conditions, particularly cavitation bubble implosion damage on hydraulic components of high-energy pumps(>850(?)ft or 600(?)HP per stage). If 2-pump operation is at or near best efficiency flowrate for these pumps then 1-pump operation is at a higher off-design flowrate and 3-pump operation is at a lower off-design flowrate. Generally, off-design flows, either high or low, can cause radial hydraulic thrust overload or cavitation damage to bearings and impellers/diffusers. More details are needed about the actual vs design (bep) flowrates for your three operating pump modes and the energy level of your pumps to further assess the possible causes of the damage that has been occurring.

 

[MatAI]

In explaining the above, we've concluded the problem is cavitation due to problems evolving from the elimination of the recirculation line years ago, which was a poor design, no pressure or flow devices and no check valves. Although the pumps are supposedly the same manufacturer, design, flow, discharge, etc., the strongest pump rules, and the other pump in service experiences cavitation. My problem is how does one design or what are the devices used in a typical recirculation system required to circumvent this problem, and what is the system layout. We've developed a design utilizing a "governor" type effluent regulator on each pump, and I think our design will work, but its too sophisticated for a normal recirculation piping system, and the cost to make the governor device is high. Does anyone have pumps operating in parallel that use a recirculation system to avoid this type problem? If so, what is the recirc system like and what equipment do you use? I'd appreciate any help I can get, and thank you for your concern. Happy Holidays!!

 

[ccfowler]

MatAI,

Other than use of a recirculation line to avoid overheating at startup, there can be substantial energy usage penalties resulting from a recirculation flow. It is likely that the added energy costs can significantly exceed the pump damage and repair costs due to operation without a recirculation system. Your careful study of your system and possible options may lead to unexpected indications for actions to mitigate the problem.

You may want to consider improving the suction piping system to increase NPSHa. The suction and discharge conditions that the pumps actually experience are what really matter. Parallel pump operation does not necessarily imply greater risks of damaging conditions. Careful troubleshooting may disclose that damaging conditions are present in your system even when only one pump is operating.

 

[WebWizz]

To explore the suggestions of the other replies online calculation tools are handy. I recommend to use EngineeringPage.com. The influences of a larger suction pipe, lower installation of the pump etc can be easily calculated using the tools on this website.

for reading:

http://www.engineeringpage.com/info/pump/pumpinfo.html

then use the links to the Pump Conditions and Centrifugal Pump calculator found at

http://www.engineeringpage.com/nav/pump.html

 

[checman]

MatAI: It is very possible you are experiancing low flow cavitaion. The damage looks just like that of insufficient NPSHA. The corresponding vibration from cavitation can and will take out your outboard bearings. The possibility of ballancing your pump flows depends on the pump curves. If the curves are flat it will be tough. Also you need to keep a minimum flow through the pumps for cooling and to avoid low flow cavitation. Regards Checman

 

[quadtracker]

Is it possible that when you put on two pumps that the suction pressure drops below the pressure required to keep the steam condensate from flashing into steam in the first stage of the pump causing cavitation? Is it possible to increase suction pressure to the pumps by increasing your condensate flash tank pressure to bring up your supply pressure to your pumps? To reinstall the recirculation line might be a good idea to reduce high head or high pump discharge pressures when feedwater demands are much lower than two pump design output flow rates. Those (dead head) conditions can also cause high thrust loads on the bearings and unnecessary hydraulic forces (churning)of the steam condensate which maybe trapped within each stage and show up as cavitation in first stage being at a lower pressure.

 

[lilliput1]

You need to calculate the available NPSH and compare it with the required minimum. The available NPSH must be greater than the minimum required. You may have to raise the feedwater heater or put in a recirculation orifice to prevent the pumps from developing too much suction.
For positive (flooded) suction:
NPSH = ha-hvpa+hst-hfs

where:
ha=ft of liquid absolute pressure on the surface of the liquid supply tank at suction (this will be barometric if from opent tank)
hvpa=head in ft of liquid corresponding to the vapor pressure of the liquid at the temperature being pumped
hst=static ht in ft the liquid supply level is above the pump impeller eye.
hfs=all suction line loss if ft of liquid, including entrance losses & friction losses through pipe, valves & fittings.