control valves and piping hydrotest 1

[elylrc]

In the past we have hydrotested systems prior to installing control valves by installing a short spool in lieu of the valve. I have a contractor who is insisting that this is unnecessary. The system is ANSI 600#, Class Location 3, per B31.8 2007. The design pressure is 1440 (not 1480)and the test prtessure is 2016. The valve is by Mokveld.

Any advice will be appreciated.

 

[DLite30]

Our company standard is to remove control valves prior to hydrotest...and the definition of a control valve is still open to some interpretation.

So in my opinion it depends on the specific valve you're using...I would remove it if:

--it has internal noise trim or any ports that can be plugged by sand, grit or welding slag. (in this case it should be left out until your gas purge is complete or install an upstream strainer)

or

--it is a globe valve type control valve (i.e. with a rising stem).

There are probalby some more instances where it would be prudent to remove it, but these are the two main instances I've had to deal with.

Now we have some folks in our company who consider any valve that has an automated pneumatic actuator on it, a control valve, and require them to be taken out of the line for testing, which is ridiculous.

 

[elylrc]

Many thanks for the input. It reinforces my experience. By "control valve", I am talking about the traditional type of valve with a sliding mechanism. So I agree, it's not necessary to remove other types of valves not prone to getting jamed by particulate matter.

 

[sheiko]

Hi,
I would also ask advice to the valve vendor.

"We don't believe things because they are true, things are true because we believe them."

 

[vpl]

I'll second sheiko's advice. If it's a new system and new valve, you want to ensure you don't void the valve warranty.

Unless, of course, the contractor is willing to take full responsibility (in writing) for procurring and installing a new control valve if you follow his advice.

Patricia Lougheed

******

Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of the Eng-Tips Forums.

 

[elylrc]

Thanks Patricia,

That was the first thing we did, unfortunately the vendor wasn't any help. Hence my question in this forum.

Industry standard seems to be to replace them with a spool during pressure testing, which jives with my experience.

Chris

 

[sheiko]

The vendor wasn't any help? What do you mean?

"We don't believe things because they are true, things are true because we believe them."

 

[vzeos]

Hydro tests are required by code to strength test line pipe. Components and fittings are usually exempted from the test if they meet the requirements outlined in the code. Below is the appropriate section of CFR 192. Make sure you keep a record of the certification documents for the components you use.

[tt]
§ 192.505 Strength test requirements for steel pipeline to operate at a hoop stress of 30 percent or more of SMYS.
(a) Except for service lines, each segment of a steel pipeline that is to operate at a hoop stress of 30 percent or more of SMYS must be strength tested in accordance with this section to substantiate the proposed maximum allowable operating pressure. In addition, in a Class 1 or Class 2 location, if there is a building intended for human occupancy within 300 feet (91 meters) of a pipeline, a hydrostatic test must be conducted to a test pressure of at least 125 percent of maximum operating pressure on that segment of the pipeline within 300 feet (91 meters) of such a building, but in no event may the test section be less than 600 feet (183 meters) unless the length of the newly installed or relocated pipe is less than 600 feet (183 meters). However, if the buildings are evacuated while the hoop stress exceeds 50 percent of SMYS, air or inert gas may be used as the test medium.
(b) In a Class 1 or Class 2 location, each compressor station regulator station, and measuring station, must be tested to at least Class 3 location test requirements.
(c) Except as provided in paragraph (e) of this section, the strength test must be conducted by maintaining the pressure at or above the test pressure for at least 8 hours.

(d) If a component other than pipe is the only item being replaced or added to a pipeline, a strength test after installation is not required, if the manufacturer of the component certifies that—
(1) The component was tested to at least the pressure required for the pipeline to which it is being added;
(2) The component was manufactured under a quality control system that ensures that each item manufactured is at least equal in strength to a prototype and that the prototype was tested to at least the pressure required for the pipeline to which it is being added; or
(3) The component carries a pressure rating established through applicable ASME/ANSI, MSS specifications, or by unit strength calculations as described in §192.143.

(e) For fabricated units and short sections of pipe, for which a post installation test is impractical, a preinstallation strength test must be conducted by maintaining the pressure at or above the test pressure for at least 4 hours.
[35 FR 13257, Aug. 19, 1970, as amended by Amdt. 192–85, 63 FR 37504, July 13, 1998; Amdt. 192–94, 69 FR 32895, June 14, 2004; Amdt. 195–94, 69 FR 54592, Sept. 9, 2004]


[/tt]

 

[elylrc]

Thanks Vzeos,

This plant is overseas and built to B31.8 2007 ed. All valves, flanges and fittings are normally all hydrotested together, in this case to 1.4 times MAOP. However, control valves are not covered by API 6D, so we normally install a dummy spool prior to FGT operations and install the control valve later with a witch's hat.

C.

 

[Duwe6]

Assuming that you don't have any vendor warranty, hydro with the valves in. The tradition [and the term 'spool'] come from the old cases where the last thing to arrive at a plant being built was the control valves, hence they were 'spooled thru'. The short pieces with a flange on each end look like an oversize thread spool.

The only problem other than warranty will be packing leaks. Bring the test pressure up to Operating, and adjust the packing to be leaktight. Don't touch the packing during the overpressure portion of the Hydro, otherwise you will have the packing WAY too tight. Identified leaks of this nature are acceptable in a Code hydro, even at nuclear plants.

 

[elylrc]

Thanks Duwe6,

We are going with the spool piece since it's not practical to gauge a pipeline with a control valve installed !!

Thanks anyway.
C

 

[BigInch]

And, if you ever test with a valve "in", at least make sure its open.

 

[BigInch]

Leaving a control valve in for hydro is like driving crash test dummies around in a Ferrari.

 

[Kiwi2671]

elylrc,
Just doing up a hydrostatic test procedure for piping on the $50 billion Gorgon LNG Project in Australia.
The client is Chevron and this is an excerpt from the clients project specifications.

a. Tests through control valves will be permitted provided there is no internal cage assembly or possibility of damage. A list shall be created identifying those not suitable for testing through; these will be tagged and removed.
b. Pressure balanced type control valves and their pressure connecting leads shall be isolated from the test pressure. The valves shall be bypassed and blocked off with the drain or vent open and the pressure connection to the diaphragm closed or
disconnected.

Hope that helps,
Regards,
Kiwi

 

[ione]

From Lees' Loss Prevention in the Process Industries Hazard Identification, Assessment and Control – third edition - vol. 1 - table 19.5

"Pre-test preparation
.... Remove or spade off any control valve or instrument liable to damage under test pressure, replace with spool or make good joint and open any by pass vale"

 

[sheiko]

I agree with ione, morever is it possible that the valve has already been hydrotested in factory?

"We don't believe things because they are true, things are true because we believe them."

 

[elylrc]

Thanks all,

The pipeline system was successfully flooded, gauged and hydrotestedand the contraol valves were reinstalled this morning.

Regards,
Elijah C

 

[BigInch]

Kiwi, that's the same as basically just telling you not to do it.

 

[davefitz]

The valve design code max allowable overpressure is 10% above design pressure, while most hydro tests are 30-50% above design pressure. If the 110% design pressure of the valve is exceeded during the hydrotest, and it fails, then you own the valve.

 

[elylrc]

Dave, you maybe confusing design pressure with test pressure. The B31 series allows 10% over design to account for transients. Afterall the Barlow Hoop stress is overly conservative being based on the OD when the actual hoop stress is based on the ID.

Remember the hoop stress for design purposes is governed by the class locations and 80% SMYS is the limit for gas. However, when it comes to hydrotesting, as long as one does not yield the steel, its ok up to 100% SMYS. So the non steel internals are what one must be careful with, like seals etc. and thois is why valves are tested in the half open position.