Subject: Changes in the discharge head of
a centrifugal pump 1-1
We need to know the discharge head of
a centrifugal pump for several reasons:
- To select the correct size pump for a new application.
- To estimate stuffing box pressure in the event the discharge
valve is shut with the pump running.
- To determine if an installed pump is running at it's Best
Efficiency Point (B.E.P.). Any time we go off the B.E.P. the shaft
is subject to radial side loading and potential problems with
mechanical seals, packing, bearings and couplings.
- To be sure the pump will have enough capacity.
- To prevent cavitation problems, if the head is too low.
There are two methods of learning the head
- Look at the discharge pressure gauge that's been installed in
the system, or a chart recorder that's installed on the discharge
side of the pump. Subtract this reading from the reading obtained
from a gage installed on the suction side of the pump.
- Calculate the discharge and suction heads from losses in the
piping and associated hardware in the lines.
As easy as this sounds, we can still have many problems
determining accurate head information because of variations in the
discharge system. In the following paragraphs I'll describe how
design, operation and maintenance practices can affect this discharge
- An automatic or manual control valve is controlling:
- Level - Boiler feed systems are typical.
- Mixture - Many processes do this routinely.
- Pressure - Common with wash down pumps.
- Temperature - Typical for heat exchanger applications.
- Flow - Probably the most common application.
- Two or more pumps are running in parallel:
- A larger pump is closing the discharge check valve of a
- Only one pump is running and there's not enough head to stop
it from running on either side of it's B.E.P.
- The wrong size pump was originally installed. This is a common
problem in new applications because each individual involved has
added a safety margin to the original operating specifications.
Every one involved in the pump selection has a real fear of buying
a pump that's too small for the application. It's also a common
practice to purchase a pump that's known to be too large for the
application, because of anticipated future needs.
- A storage tank is being filled from the bottom and the
discharge head varies as the tank fills.
- The pump motor speed is changing:
- The packing was removed from a pump driven by an induction
motor. These motors have a slippage from 2% to 5% depending upon
their quality and load.
- A mechanical seal was installed and the motor has speeded up
changing the B.E.P..
- A variable speed motor is being used in a system with a high
static head. These motors work best where the head is mainly
caused by the piping and valves, such as in a hot or chill water
- A higher rpm. pump has been substituted because of system
demands, and the piping is offering too much resistance at this
increased flow. Keep in mind that the pipe resistance varies by
the square of the speed. In other words if you double the speed of
the pump you get four times the resistance in the piping.
- The system has been altered to accommodate a change in the
process. Another storage tank may have been added, more valves,
fitting or piping etc..
- A discharge valve is being closed to save amperage. This is a
common practice for starting centrifugal pumps with lower
"specific speed" impellers.
- The product viscosity has changed
- A different product is now going through the piping. This is a
common problem in pipe line applications.
- You are pumping a dilatant (Its viscosity increases with
- If the pump was sized for a higher viscosity product and the
temperature has increased, or the product viscosity lowers with
agitation (thixotropic products do this) you can also move off of
the best efficiency point towards the higher capacity side.
- The specific gravity of the product has changed, causing a
change in the rpm of the motor:
- A different product is now in the system
- The temperature of the product has changed. (Specific Gravity
changes with temperature)
- A discharge valve is being throttled.
- A common practice in an attempt to stop cavitation. (The
lower the capacity, the less N.P.S.H. Required)
- To compensate for impeller wear or worn wear rings
- To stop water hammer
- The impeller has been trimmed.
- The discharge of the pump is being restricted
- A foreign object was left in the pipe the last time the system
was opened, or maybe the swing check valve seat has come loose
- There is a buildup of material on the wall of the piping and
fittings. Hard water, depositing a layer of calcium on everything
is a good example. Many other fluids can also coat out on the pipe
- The pipe has collapsed (often not visible)
- A truck ran over a piece of the pump piping.
- A liner has come loose and is restricting the passage.
- A frozen water jacket has collapsed the pipe.
- A restricting fitting has been substituted or added to the
- A reducer has been installed
- A globe valve has been substituted for a gate valve. A globe
valve can add another 50 to 100 foot of head depending on the flow
and pipe size.
- A gasket is protruding into the piping.
To size a pump properly, or to do effective trouble shooting it's
important to know where the pump is running on its' curve. There are
low cost chart recorders available to help you anticipate some of
these problems and receive a clearer picture of what the system is
actually doing. I strongly recommend the use of one of these
recorders prior to specifying a replacement pump in an existing
This discussion was an attempt to show you how we sometimes make
an error in our calculations because of lack of knowledge about
varying discharge conditions. It should also demonstrate to you the
importance of the L3/D4 relationship to compensate for some of these unforeseen
For information about my CD
with over 600 Seal & Pump Subjects
Link to the Mc Nally home