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!

Recip PD Pump Piping Question

Status
Not open for further replies.

Krausen

Mechanical
Jan 1, 2013
289
Gentlemen,

For Recip. PD pump piping design, a well known best practice is to install a bypass or recirculation line from the pump discharge back to the pump suction/inlet. This aids in pump start-up where a high amount of mechanical inertia requires high starting torques & power from the motor.

One resource I've read states that this bypass/recirculation line should "take liquid back to the liquid source and not into inlet line."

I've seen it done both ways. Does anyone know the technical basis behind this specific requirement? The only advantage I can see to route the bypass line back to a tank would be to eliminate any air or gas in the pump or inlet line at start-up. I'll also note that these pump motors will be driven by VFDs, which will also aid in easing the start-ups.

Thanks
 
Replies continue below

Recommended for you

The main disadvantage of returning the bypass to the suction line is the possibility of overheating. If the pump is operated in pure bypass (rightly or wrongly), the product will get hotter and hotter until it boils. If it goes back to the supply tank, there is likely a much larger volume available and it would take much longer to overheat. Even if the intention is that the bypass would only be used for a few seconds at start-up, it would be better to spill back to the tank. Mistakes happen. At some point, the pump may be left in full spill-back for longer than anticipated.

Johnny Pellin
 
JJPellin - Great, thank you for the response. That makes a lot of sense.
 
The other reason is that if the bypass system is a simple pressure relief valve it can be affected by the pump inlet pressure and may not open at the correct pressure. If the inlet pressure can vary because of things upstream, then it makes more sense to go back to the tank which is at a more reliable and steady pressure.

The main issue as JJ says is heat accumulation.

Some pumps include this internally within the pump, but most people prefer to have their own system which is specific to your requirement and can be repaired and maintained much easier.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.
 
Little Inch - Thanks for the confirmation. This bypass will be controlled by a motor-operated block valve with the unit relief valves separate. The downstream side of this bypass valve will tie into the relief valve disposal line going to tankage. The bypass operation will take liquid volumes back to the source (tank) while simultaneously flushing the relief disposal system. Why not kill two birds with one stone?

My only problem now is my meters upstream of these pumps will be measuring the same volumes twice during this bypass operation (start-up). Time to get the PLC programmer geeks involved ...

 
A sketch would be helpful. Your comment about relief disposal system is confusing. Systems should be designed so that relief valves function infrequently.
 
Krausen,

The devil is in the detail. If your relief disposal system is a multi-user system and quite small or long, you could raise the pressure in it to affect operation of some other relief systems.

Metering upstream is now not a good place to meter anything that occurs d/s of the pump. metering, especially fiscal metering, needs to be as close to the end point as you can and not have multiple paths.

Only if you have 100% of flow either in the bypass or the main line will a software switch be able to stop / start flow metering. Seems like a poor location to me for the meters.

Remember - More details = better answers
Also: If you get a response it's polite to respond to it.
 
Metering the flow from a PD pump is also problematic since the flow is not steady, it pulses.
 
Bimr - Re: relief valves - The relief system is designed to only relieve in upset conditions. The problem with relief disposal lines, especially with hydrocarbons, is maintaining the "dead leg" piping that normally has no flow. These kind of lines can be nightmares for operators & companies long term. An operating bypass line that ties into this relief disposal line would flush out an otherwise "dead leg" line during pump startups.

Re: metering with PD pumps - meter types are bidirectional coriolis meters. The suction & discharge stabilizers on each pump will also smooth out the pulsed flow. Not to mention there are also check valves between u/s meters & d/s pumps.

Little Inch - Code explicitly allows for designing for only one upset condition at a time in any shared relief system. Anything beyond that would be considered double jeopardy. Also, the leak detection/allocation meters u/s of pumps should never be measuring multi-path (split) flow. Prior to bypass mode start, the outgoing station valves will be closed. As an origination station, we are required to have leak detection/allocation meters tracking the outgoing volumes.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor