Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations waross on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Safety device required after positive displacement pump before shutoff valve? 1

Status
Not open for further replies.

USAeng

Mechanical
Jun 6, 2010
419
In our plant we have a positive displacement pump with PVC piping from the outlet to the equipment. Before and after the pump there is a pvc shutoff valve. There is no pressure relief, pressure switch, etc in the piping.

Is there a requirement somewhere that says we need a safety device before the valve on the outlet side? If so can someone please provide the location of a written requirement? Thanks for any help!
 
Replies continue below

Recommended for you

Hi

I would have thought that the process safety engineer would provide that answer.
From my experience, they would weigh against the criticality of the chemical or fluid in used, and pressure of the system.
 
The pump manufacturer will probably state it so you do not destroy the pump. Here is Warren Pump's statement:

Relief valves are required in Systems with PD pumps. A centrifugal pump can operate briefly without damage against a closed discharge valve and only generate a pressure equal to its shutoff head. Positive displacement pumps can not operate against a closed discharge valve or plugged line. PD pumps create flow and if that flow is blocked, pressure in the system downstream of the pump builds rapidly. If there is no relief valve, the peak pressure generated will be a function of driver horsepower. Pressures several times the designed operating limit can be attained in seconds. Relief valves need to be sized for the full flow of the pump. Relief valves for centrifugal pumps need only be sized for partial flow; as pressure increases, flow decreases so at elevated pressure, there is less flow to relieve, PD pumps have a constant flow so relief valves must be able to pass the rated flow without raising the pressure above a set design point. The bypass flow from the relief valve should be piped back to the suction source, rather than the pump suction. Returning the flow to the pump suction creates a short loop for the recirculating fluid allowing
temperature to rise too quickly. Rupture disc type relief valves work well for rotary PD pumps, but may fatigue due to flow pulsations from reciprocating PD pumps. Spring actuated relief valves have adjustable set points, but can become clogged if the fluid has a high viscosity.

 
API-521, Section 5.5, Closed outlets.

Good luck,
Latexman

Technically, the glass is always full - 1/2 air and 1/2 water.
 
Apart from common sense, a general duty to not design something beyond its design limits, the following appears in ASME B31.3 and I'm sure in other codes as well as a general requirement. I assume you need this so that you can persuade someone that this set-up needs to be modified. As bimr says also look nin the pump suppliers installation instructions as they often warn about this issue as well.

301.2.2 Required Pressure Containment or Relief
(a) Provision shall be made to safely contain or relieve
(see para. 322.6.3) any expected pressure to which the
piping may be subjected. Piping not protected by a pressure
relieving device, or that can be isolated from a
pressure relieving device, shall be designed for at least
the highest pressure that can be developed.
(b) Sources of pressure to be considered include ambient
influences, pressure oscillations and surges,
improper operation, decomposition of unstable fluids,
static head, and failure of control devices.

in your case the pump might well be able to create a pressure higher than the allowable design pressure of your PVC pipework.

You might find that there is an internal pressure relief valve in the pump itself which has saved you so far, but normally these are not trusted as they are small, difficult to test / repair and pressure is set at the maximum for the pump, not the pipework.

PVC pipework (see recent thread when it fails can fail like a fragmentation grenade.... Therefore treating it like this is a huge risk every time.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way
 
I think it depends upon the kind of PC pump. If it is an oscillating PC pump like a piston pump it will probably have a pulsation dampener. This is a pressure vessel and the pressure vessel codes require a pressure relief valve (or a rupture disc). If it is a rotating PC pump without pulsation dampener there won't probably be any legal code/standard that requires the pressure relief valve but of course the technical knowledge of the PC pumps functions makes it more than only recommended to install one in any case.
 
Thank you all for your input. I did not design this system. I work in a large plant and once in a while I wonder who put things like this together. Now that we are aware of this we will have to inspect all old outlets of rotary lobe pumps and make changes. All of our new setups have both pressure switches and pressure transmitters for safety.

FYI they are rotary lobe pumps with no built in safety. The valves were put in only to shut if changing the pump so I am guessing this is why whoever put it in did not consider safety devices

Thanks again
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor