PSV for Hydrotesting of Pipes

[CaracasEC]

Dear All,
What is your experience on hydro testing of pipes.
As far as safety is concern, is putting a pressure relief valve beneficial?
thanks in advance..

 

[BigInch]

No. It is a tightly controlled situation. Water, not product (I presume), in the line. Static, no flow. Test section is guarded and pressure is closely monitored. No transient pressures. Segments blocked off. Test area blocked off. There should be no chance that anyone would be hurt in this situation... at least not from bursting pipe.

If it ain't broke, don't fix it. If it's not safe ... make it that way.

 

[CaracasEC]

Hi Big Inch,
We will be using positive displacement pump during the hydro test of many test packages. To assure of human error and protect the pipe for overpressure is it not reasonable? On the world of pump if you used positive displacement on the line, psv is mandatory though the fluid is water. As part of the client side on the Ethylene Cracker Plant construction, safety is one of our utmost concern..thanks again..

 

[BigInch]

The PD pump needs a pressure relief itself, to prevent discharge pressures within the pump itself from exceeding it's own allowable pressure, or should be underdesigned to the extent that turning up to full power will raise the motor temperature to shut down range well before burnout, burst/rod damage pressure is reached.

The pipe probably has a range in which it will be tested. The minimum test pressure, to which every point on the pipe must be subjected, and a maximum test pressure, perhaps equal to the pipe's yield pressure. If the PD does have a PSV itself, you should set the PD pump's PSV within that range.

Alternatively, you can fill the pipe with a low pressure PD pump, then switch to quite a small capacity pump, with a discharge pressure rating over test, but under that of the yield pressure of the pipe, if you can. If you can't, the discharge from a small capacity pump, even though at a higher than pipe yield pressure, when running slowly will be disipated within the pipe with enough warning time in advance to stop the pump before sustained pressures higher than yield are reached.

If you only have a giant test pressure pump that you can't turn off, sure install another PSV.

If it ain't broke, don't fix it. If it's not safe ... make it that way.

 

[zdas04]

It is your test, do what you think best.

I've done a few dozen tests and not only do I not add a PSV, I've frequently had to remove PSV's set at MAWP when I was testing to a multiple of MAWP. During a test the protections typically provided by a PSV are replaced by human observation.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[BigInch]

Ya. As long as they arn't dummies, it works fine. Trying to design for dummies is a generally losing proposition. If you hire dummies, it seems that the only thing that they can routinely accomplish without a hitch is to defeat whatever safety system you build into it.

If it ain't broke, don't fix it. If it's not safe ... make it that way.

 

[Latexman]

True story. The owner of the main contractor for a project was watching his crew hydrotest the annular space on a jacketed pipe section they provided. I think it was 3" inside 4" and about 10 ft. long. Frustratecd with their slow, methodical, careful progress, he brushed them aside with language that would made a pirate blush. He grabbed the hand pump and rapidly pumped it 10-12 times, stopped and looked at the test guage - it was pegged to the max! The inside pipe had collapsed! He hung his head and shuffled away. The entire site burst into laughter and took turns not being able to look thru the inner pipe!

A new jacketed pipe came to site next day. The owner did not.

Good luck,
Latexman

 

[zdas04]

Every time I try to engineer something stupid-proof it ends up being: (1) too complex; (2) too expensive; and (3) not effective. I'm pretty sure that people's ability to be stupid will always exceed engineers ability to protect them.

David Simpson, PE
MuleShoe Engineering

"Belief" is the acceptance of an hypotheses in the absence of data.
"Prejudice" is having an opinion not supported by the preponderance of the data.
"Knowledge" is only found through the accumulation and analysis of data.

 

[Latexman]

Stupid is as stupid does.

Good luck,
Latexman

 

[BigInch]

Well, it takes a lot less external pressure to collapse a pipe then internal pressure to burst it.

If it ain't broke, don't fix it. If it's not safe ... make it that way.

 

[dicksewerrat]

You can't fix stupid. You can cure it but murder is a criminal issue.

Richard A. Cornelius, P.E.

http://WWW.amlinereast.com

 

[BigInch]

be sure to make it look like an accident. OMG!

If it ain't broke, don't fix it. If it's not safe ... make it that way.

 

[CaracasEC]

Hi All,
Based on your valuable comments the following will be done. Verify pumps maximum input power that will make the maximum pressure output and base on these data, we will be comparing the design pressure of the pipe. If the design pressure is less than the maximum pressure output then PSV will be required. However if the design pressure of the pipe is greater than the maximum pressure output of the pump based on the maximum input power then psv is not necessary.
thank you all...

 

[unclesyd]

I totally agree with the above post except I would have to call the dumb one "Hard Core Unemployable" that somehow got in the door.
In over 40 years of observing/witnessing Hydro Testing of piping both in service and newly fabricated I've never seen a pipe burst from the calculated test pressure. Testing pressure to 7200 psig to a routine 3500 psig for core pipe in polymer jacketed pipe and 175 psig in the jacket.
We use the following equipment and used the pressure relief/bypass valve on the higher volume requirements. While the Sprague is normally control by the operator. We normally use two gauges a dumb gauge and calibrate gauge for the actual test pressure. The dumb gauge will normally be located on a high point vent. We have a special valve that will allow the venting of air while keeping the gauge in line.

Having said all of the above we did do some burst test on SS and Ally 20 pipe that have lack of fusion and penetration in the long seam.

You can rent about any type pump you need.

http://www.hydratron.com/media/html/hydro_pumps.htm

http://sprague.cwfc.com/Products/spokes/Pumps.htm

 

[BigInch]

It's not pretty. Mud, dirt, water, rocks, chunks of pavement everywhere. Had one blow crossing under a parking lot back in the 80's.

If it ain't broke, don't fix it. If it's not safe ... make it that way.

 

[waterpipe]

I've seen hydrostatic pressures tests with pressure relief valves being used in addition to the one installed on the piston pump.

My suggestion is that you first look to your design code, standard and project specifications to see their requirements.
Then if you are a consultant, safety officer, or any other side than the Contractor, "ask" (or recommend based on your role)for the safety valve. Depending on the project contract, the Contractor must/may provide it or not but you have your request on the record. If you are on the Contractor side then add this valve as a safeguard against stupidity and to avoid any probable technical/contractual discussions in case of a possible over-pressurize of the pipe. These discussions could convert to legal charges in case of a pipe burst.

you might like not to specify this valve under circumstances that you find yourself.

PS: most of the hydrostatic test procedures ask for some sort of a step-rising pressures with holding times between each step before reaching to the maximum test pressure. This would give a room for finding "weak" points or realizing leakages before a burst under the maximum pressure. You can look to ASME B31 series for some guidelines. ASME B31.1 has no mandate for the relief valve for the hydrostatic test but makes it mandatory when it comes to Pneumatic leakage test "pressure relief device shall be provided, having a set pressure not higher than the test pressure plus the lesser of 345 kPa 50 psi) or 10% of the test pressure."

 

[CaracasEC]

Hi All, thank you very much for the intelligent comment and field experiences you have shared. It helps me a lot. AGain thanks...