Q. When installing a horizontal pump, is it better to grout the pump baseplate or let it float free?   Different sources provide opposing views on this subject.

A. Whether to grout depends on the specific application and the design of the baseplate. Applications that undergo wide temperature variation may do well with floating baseplates to allow for movement of the pump and to minimize pipe strain caused by thermal expansion of the pipe. However, the baseplate in such cases must be stiff enough to avoid misalignment of the shaft coupling as the baseplate moves.

The grouted baseplate is designed to allow grout to be poured underneath and inside the base. The grout placed inside the base contributes to the baseplate's installed rigidity and damping. The cross members in this type of base are normally designed to lock into the grout and further resist deflection or vibration of the baseplate. Typically, the cross member geometry chosen to achieve this is an L-section, T-section or I-section.

If the baseplate is closed on top, then grout holes must be provided to allow the grout to be placed inside the base.

Figure 1.3.8.2.1a shows a typical grouted baseplate design.

[[{"type":"media","view_mode":"media_large","fid":"210","attributes":{"alt":"Grouted baseplate, cast iron","class":"media-image","id":"1","style":"float: left;","typeof":"foaf:Image"}}]]A freestanding baseplate is intended to be elevated off the floor and supported by stilts, shims or springs. See Figure 1.3.8.2.5a. This type of baseplate must be designed to provide its own rigidity, as there is no grout for additional support.

The features of this type of baseplate are:

  •     The need for concrete foundation pads is eliminated
  •     An allowance for height adjustment can easily be included
  •     Spring-mounted bases provide vertical displacement under applied loads associated with thermal expansion of the piping
  •     Optional addition of slide bearing plates may also be provided below the stilts or springs to allow horizontal displacement associated with thermal expansion of the piping

More detail on baseplates can be found in the recently revised HI Standard ANSI/HI 1.3-2007 Rotodynamic (Centrifugal) Pumps for Design and Application.
 

Q. I understand that the efficiency of variable speed drives is reduced as operating speed and power is reduced. Does this negate the power saving from the accompanying reduction in pump speed?

A. The difference in the power consumption with and without the variable speed drive (VSD) must be evaluated. The power savings from the reduction in speed of the pump is typically greater than the loss in efficiency of the driver. Assuming a change in demand results in a reduction in pump speed of 10 percent, the power required by a 100-hp pump is reduced by the cube of the speed-so at 90 percent speed the power required is 0.90 cubed times 100-hp, or 0.729 times 100, which equals 72.9-hp.

At 90 percent of full speed, the efficiency of the VSD may be reduced by about 10 percent of that at full speed and power. This depends on the design of the drive.

If we assume the full load efficiency of the driver is 92 percent, then:

Full load electrical power = pump power / driver efficiency

= 100 / 0.92

= 108.7-hp

At 90 percent full speed and 90 percent reduction in efficiency:

Variable speed electrical power = pump power / VSD efficiency

= 72.9 / (0.92 X 0.90)

= 88.0-hp

The result is nearly 20 percent reduction in electrical power.
 
Q. After installing a new 20-hp centrifugal pump, we checked its performance and found that the total head was 10 percent below the guaranteed point. The system-head curve was checked and determined to be as predicted. Before we contact the supplier, is there something we should check?

A. Accurately verifying pump performance in the field is difficult because of the care necessary in the placement and use of instruments. To begin, all terms in the following total head equation must be considered.

[[{"type":"media","view_mode":"media_large","fid":"211","attributes":{"class":"media-image","id":"1","typeof":"foaf:Image"}}]]

Where:

H = Pump total head in feet

p =  Gauge pressure in pounds per square feet or 144 times psi

γ =  Pounds per cubic feet

Z =  Elevation of measurement point in feet

v =   Liquid velocity in feet per second

g =   Gravity constant = 32.2 feet per second squared

hf= Friction loss between points 1 and 2 in feet

Point 1 refers to a point on the suction side of the pump, and point 2 is a point on the discharge side.

The most common problems during tests are:

  • Inaccuracy of instruments. They must be calibrated at the recommended intervals.
  • Improper location or installation of pressure gauges. They should be preceded by 5 diameters of straight pipe and have no sharp edges at the inside wall of the pipe.
  • Ignoring or incorrect gauge height measurements. If Bourdon gauges are used, Z is measured at the center of the gauge. If the pump takes suction from an open surface of water, point 1 is usually taken on the surface of the liquid.
  • Velocity measurements are often ignored. This is most important when testing low head pumps.

For a detailed procedure for testing pumps, see the series of Hydraulic Institute Test Standards ANSI/HI 1.6 Centrifugal Pump Tests, ANSI/HI 2.6 Vertical Pump Tests and ANSI/HI 11.6 Submersible Pump Tests.

Pumps & Systems, March 2008