Strainer or No Strainer?
by Stuart Asay, P.E., Ph.D.
January 3, 2008
Both schools of thought
have their strong points
For years, members of the
backflow prevention industry have debated the value of strainers at the
inlet of an assembly’s installation. Arguments on both sides of the issue have
merit and should be carefully considered before a strainer is installed.
A
strainer is essentially a screen placed in the flow path. It is intended to
catch sand and debris that may foul the integrity of a check valve. A fouled
inlet check valve in a reduced pressure principle assembly, for example, will
cause the relief valve to discharge intermittently with small sand and
potentially several gallons per minute with lunch buckets and two-by-four sized
debris.
If
a public distribution system has a reputation for water main breaks or silty
sand from shallow wells, a strainer will keep the backflow preventer
operational. Some backflow preventer assembly manufacturers may encourage
including a strainer in the contractor’s installation bid price, as the
strainer’s cost is far less than a return service call to clean a valve’s seat
after installation.
The
key concern rarely discussed, however, is the maintenance of the strainer.
Strainers will have ports to open and flush the debris free of the screen. This
must be performed periodically. If this is not done, the flow path will
eventually become plugged resulting in a pressure loss to the downstream use—or
worse.
Many
years ago, the director of a prominent evaluation laboratory expressed serious
concern over the use of strainers with backflow preventers. The basis of his
argument was the strainer’s lack of long-term maintenance.
As
the strainer catches debris in the piping system, the cross-sectional flow path
within the pipe is reduced. Because water is not a compressible fluid, a
Venturi is established, causing higher velocity flow through the restriction.
Keep in mind that a stainless steel strainer screen by its nature already
creates a higher velocity flow through the mesh openings, debris blocking the
screen openings compounds the effect and a still higher velocity is
established.
We
know that when sizing water supply piping, an important consideration is to
avoid high velocity flows. High velocity water causes erosion corrosion. The
upper velocity limit in a domestic system should not exceed 7.5 to 8 feet per
second.
Flow
velocities through a partially plugged strainer easily exceed that velocity.
With time, even a stainless steel screen will disintegrate. Because the
strainer has been retaining a mass of debris, the sudden release of that
material when the strainer’s integrity is lost will likely cause an immediate
failure of the closest check valve.
Perhaps
this argument from the evaluation laboratory had an impact on backflow
preventer designs for the last thirty years. One manufacturer of 1970s-era
backflow preventers included a wye strainer as part of the assembly. In fact,
the assembly’s number one test cock was located on the strainer. Conceivably,
part of this design was to encourage periodic flushing when the assembly was
field-tested.
It
should also be considered that the manufacturer was aware that the check
valve’s perpendicular orientation to the direction of flow might not allow
adequate flushing of debris from the check valve seats, thus resulting in increased
failures.
Backflow
preventers of the ’80s and early ’90s has a distinctive design change: check
valve components were designed at an angle. A few incorporated a characteristic
Y-pattern, with one check 45 degrees off the axis and the second check 90
degrees from the inlet check. The justification at the time was improved
flushing characteristics for the check valves. Water flow would lift the check
off the seats, and the flow’s direction would in turn sweep any debris through
the checks reducing the failure risk. Typically, strainers were not provided
for these assemblies.
The
industry continued to make demands for backflow preventer improvements.
Pressure or head loss has always been a concern, mainly for the fire-sprinkler
industry. Material costs have risen. Lay length is critical. In order to
accommodate these requested improvements, check valve orientations have
generally returned to being perpendicular to the direction of flow. However,
the newer designs have improved check valve flushing characteristics resisting
fouling from piping system debris.
Can
the strainer be installed between the backflow preventer’s inlet shut-off valve
and the inlet check valve? No, absolutely not. Neither the Foundation for
Cross-Connection Control and Hydraulic Research nor the American Society of
Sanitary Engineering will consider the assembly “approved.” In order to provide
for the strainer’s servicing, an additional shut-off valve must be installed
upstream of the device. One consideration should be a bypass with a control
valve around the strainer, so it can be maintained without service
interruption.
Please
remember this: An essential task prior to the installation of any backflow
preventer, especially on a service line, is to flush the pipeline with a
high-velocity flow. Most backflow preventer testers will report that a high
percentage of assembly failures occur with new installations. The reason for
the failures: construction debris in the check seats.
Please
consider all aspects of using a strainer. Hopefully these simple questions will
help lead to thoughtful decisions. If you have any questions regarding this
subject or any issue relating to backflow prevention, please feel free to contact me.
|