Basar Ozar, PhD and Jens Conzen

Link to SUNCENTER

Pressure build-up in a system may occur as a result of a process upset, instrument failure or fire. These upset scenarios can result in either, or both, vaporization or generation of gases; this is of particular interest in terms of system integrity and public safety. The extent of the potential hazards caused by such incidents are generally determined by the magnitude of the energy (heat from fire or exothermic reactions), the chemical mixture composition, the quantity and type (e.g. two-phase) of the effluent.

The pressure vessels or systems, which contain chemicals (e.g. chemical reactors or storage tanks), are required to be protected by a relief device (e.g. rupture disc and/or PRV). The relief device is part of an emergency relief system (ERS) that prevents the over-pressurization of the equipment as a consequence of the vaporization of chemicals and/or generation of gases. Some systems discharge straight to the atmosphere whereas others require a catch tank (or other retention mechanism) and a tall stack to reduce toxicity and flammability concerns via atmospheric dispersion.

The operation of the PRV includes three stages: opening, continuous discharge, and re-seating. The disk leaves the valve seat to reach sufficient valve lift, and its discharge capacity reaches the rated capacity, thus completing the opening process and transitioning to continuous discharge. The process, in which the disk returns to the seat from the fully open position, is the process of re-seating. PRVs can experience instabilities (valve flutter and chatter) under certain conditions. PRV instability occurs when the disk experiences a rapid oscillatory motion during any one of the stages. PRV instability has led to mechanical failure of the valve. The typical failure modes for a PRV were discussed by Grolmes [1]. Depending on their severity, failures may affect valve leak tightness and future demand actuations. If the valve leak tightness is affected, flammable fumes may escape from the systems and potentially create a combustible environment. On the other hand, if the actuation of the PRV is affected for a future demand, the valve may become inadequate to relieve the over-pressurization of the protected vessel in an upset scenario. This can lead to incidents with serious danger to human health and the environment. 

The following standards provide guidance for PRVs and design/construction of inlet and outlet lines to prevent valve instability:

-         International Standards Organization (ISO) -9 [2] and

-         American Petroleum Institute (API) 520 [3]

Both standards recommend the &#;3% rule&#; for the inlet piping. According to the &#;3% rule,&#; the pressure loss along the inlet piping should be kept below the 3% of the PRV set pressure (and at the rated flow capacity). Similarly, the &#;10% rule&#; was established for the maximum allowable backpressure. These rules allow for the inlet and outlet piping to be designed based on standard pressure loss calculations. However, recent experimental studies demonstrated that the &#;3% rule&#; and the &#;10% rule&#; did not adequately address the PRV instability [5]; PRV instability occurred even though the pressure drop requirements at the inlet and outlet piping were satisfied based on these rules under certain conditions. On the other hand, some experiments showed stable PRV behavior, although the inlet and outlet piping pressure losses did not meet the pressure requirements. Dannenmaier [4] provides an overview of various studies illustrating this point. Several of those are discussed in this paper.

Izuchi [6] performed an experimental and analytical study using 1E2 and 1.5F2 size API valves. PRVs were connected to a vessel via an inlet pipe. Izuchi varied the length of the inlet piping (up to 20 m). The tests showed that at very short inlet piping lengths, PRVs were stable. PRVs became unstable at intermediate piping lengths and stable again at longer lengths. He classified PRV instability as both dynamic instability and static instability. Dynamic instability was caused by the interaction between valve motion and pressure wave propagation at the inlet pipe. He concluded that the pressure waves attenuated at longer pipe lengths and resulted in stable condition. He also concluded that excessive inlet pressure drop caused static instability; however, he determined that the &#;3% rule&#; was too conservative for this purpose.

Hős et al. [7] studied three different types of API valves, 1E2, 2J3 and 3L4. The valves were connected to a reservoir with pipes. Hős et al. injected air into the reservoir at a constant flow rate. They varied the length of the inlet piping up to 1.8 m (72 inches) to study the PRV instability. They also concluded that PRV instability was attributed to the interaction between valve motion and pressure wave propagation at the inlet pipe. They showed that the pressure waves had a frequency corresponding to the quarter wave frequency. Quarter wave frequency corresponds to the quarter of the ratio of the speed of sound of the fluid inside the pipe over the inlet pipe length. Hős et al. concluded that the &#;3% rule&#; is not conservative.

Smith [8] tested eight different API 526 valves and 24 different piping configurations (a total of 396 tests) with air. He compared the test results with the detailed API methodology. The detailed API methodology consisted of a series of calculations and a flow chart, which were used to predict whether a PRV was stable or unstable. Detailed methodology predicted the stability/instability correctly only for 130 of the tests out of 396.

PRV instability has also been a focus of several analytical studies. Izuchi, Melhem [9], and Hős et al. developed detailed analytical methods to characterize the PRV instability. All the analytical studies highlighted the importance of capturing the reservoir dynamics, pipeline dynamics and the PRV dynamics in their studies. The analytical models predicted PRV behavior, which was consistent with experimental data.

The authors of this paper with the process safety team at Jensen Hughes performed a separate analysis of the PRV chatter phenomenon utilizing a software solution. The software model captured the key physical phenomena of a vessel, a straight pipe and a PRV. Three different transient calculations were performed by varying the pipe the length. Model parameters (e.g. vessel volume, pipe diameter and length, PRV parameters) were consistent with Hős et al.&#;s and Izuchi&#;s experiments. This also allowed for the software model to be benchmarked with the experimental data. The results of the software model and the test data were interpreted. One of the key conclusions is that the &#;3% rule&#; is not a reliable criterion for PRV chatter. Detailed analytical models may be more reliable tools for predicting PRV instability and designing emergency relief piping for systems that especially contain highly toxic or flammable fluids since failure of these systems have significant consequences.  The process safety team continues to perform research related to PRV chatter. Questions and inquiries are welcome on LinkedIn or at .

References

[1]           Grolmes, M., A., (), &#;Odds and Ends, Relief Valve Stability-part 4: Welcome to the Morgue,&#; presented at the DIERS Users Group Meeting, October, , Burr Ridge, IL.

[2]           ISO Standard -9, () Safety Devices for Protection Against Excessive Pressure-Part 9: Application and Installation of Safety Devices Excluding Stand-alone Bursting Disc Safety Devices.

[3]           API Standard 520, (), &#;Sizing, Selection, and Installation of Pressure-relieving Devices in Refineries,&#; 8th edition.

[4]           Dannenmaier, T., Schmidt, J., Denecke, J., Odenwald, O., (), &#;European Program on Evaluation of Saefty Valve Stability,&#; Chemical Engineering Transactions, Vol. 48, pp. 625-630.

[5]           Izuchi, H., (), &#;PRV Stability Study for Control Valve Outlet Piping&#;, presented at Joint DIERS-EDUG Spring Meeting, May 22-24, , Karlsruhe, Germany.

[6]           Izuchi, H., (), &#;Chatter of Safety Valve,&#; presented at the API Fall Meeting, November, , Los Angeles, CA.

[7]           Hős, C.J., Champneys, A.R., Paul, K., McNeeley, M. (), &#;Dynamic Behavior of Direct Spring Loaded Pressure Relief Valves in Gas Service: Model Development, Measurements and Instability,&#; Journal of Loss Prevention in the Process Industries, Vol. 31, pp. 70-81.

[8]           Smith, D., (), &#;Considerations for RV Stability,&#; presented at Joint DIERS-EDUG Spring Meeting, May 22-24, , Karlsruhe, Germany.

[9]           Melhem, G. A., (), &#;Pressure Relief Valve Stability with SuperChems,&#; presented at the API Fall Meeting, November, , Los Angeles, CA.

INTELLIGENT WORK FORUMS
FOR ENGINEERING PROFESSIONALS

Log In

Come Join Us!

Are you an
Engineering professional?
Join Eng-Tips Forums!

• Talk With Other Members
• Be Notified Of Responses
  To Your Posts
• Keyword Search
• One-Click Access To Your
  Favorite Forums
• Automated Signatures
  On Your Posts
• Best Of All, It's Free!

Join Us!

*Eng-Tips's functionality depends on members receiving . By joining you are opting in to receive .

Posting Guidelines

Students Click Here

Promoting, selling, recruiting, coursework and thesis posting is forbidden.

Eng-Tips Posting Policies

Contact US

The 3% rule

2

thread- Forum Search FAQs Links MVPs

• Forum

• Search

• FAQs

• Links

• MVPs

  For more information, please visit relief valve test bench.

The 3% rule

The 3% rule

SrChemE

(Chemical)

(OP)

17 Aug 17 18:02

The company where I work has several relief valves identified as having inlet line losses greater than 3%.
- What kind of engineering analysis can be done to show that the installation will not chatter?
- What guidelines can be used to ensure the installations will not chatter?
Thank You!

Replies continue below

Recommended for you

RE: The 3% rule

Latexman

(Chemical)

17 Aug 17 19:52

Did you look at this one.

Did you search the database?Did you look at this one. thread-: Relief Device Inlet Piping - Beyond the 3% Rule

Good luck,
Latexman

To a ChE, the glass is always full - 1/2 air and 1/2 water.

RE: The 3% rule

mk

(Mechanical)

17 Aug 17 20:08
http://oilandgasprocessing.blogspot.com//01/ps...

I just looked into the same topic. Here is to share what I found,

RE: The 3% rule

don

(Chemical)

17 Aug 17 23:35

Engineering analysis (alternative to the 3% rule) is explained in API 520 Pt. II, 6th ed., section 7.3.6. Included is a high-level explanation of the suggested steps and suggested analyses. There are two analyses. One is a force balance and other is an acoustic analysis. I suggest you get a copy of the ioMosaic papers in references 31 and 32.

Notice that API 520 doesn't prescriptively define a particular engineering analysis methodology. But this is a complex topic, so I don't suggest going off on your own and creating an engineering analysis. Over the last decade a lot of research has been done on PRV stability analysis, and there are process safety engineering companies that are practicing analysis methodologies based on that research. ioMosaic's website advertises a seminar in which they teach how to do these engineering analysis calculations. Alternatively, you can pay one of these process safety companies to do this work for you.

This is still an evolving technology. There's no guaranteed way to be certain that a PRV installation will not cause the valve to chatter. However, what is known based on research and testing is that the legacy 3% rule isn't sufficient for predicting chatter. This research has shown that a PRV with 2% inlet loss may chatter, and a PRV with 10% inlet loss may not chatter.

RE: The 3% rule

georgeverghese

(Chemical)

18 Aug 17 00:44

Also note that while the API publishes obligatory recommended practices, the ASME BPVC design codes are mandatory in jurisdictions where these are enforced, and this 3% inlet dp loss rule is stated here too - so you wont be able to wriggle your way out of the ASME BPVC, no matter what your engineering analyses may conclude.

RE: The 3% rule

The Obturator

(Mechanical)

18 Aug 17 07:38

Your PSV solution is to install a pilot operated SRV, but with remote sensing line. Instead of the sensing line pick up being located on the inlet neck of the SRV. engineer it such that it will pick up directly from the vessel. This works and is an option from all SRV manufacturers.

Per ISO, only the term Safety Valve is used regardless of application or design.

RE: The 3% rule

don

(Chemical)

18 Aug 17 13:53

George - Possibly that is true in Australia - I'm not knowledgeable about Australia's legal requirements. But in the US, the legal requirement (OSHA) is that relief design practices must be consistent with "RAGAGEP" (Recognized And Generally Accepted Good Engineering Practices), and the practices in API 520 Pt. II are cited by OSHA as RAGAGEP. API standards and ASME Sec VIII are complementary rather than competing standards. Engineering analysis, as an alternative to the 3% rule, is an accepted and legal practice in the US. The practices published by API are also accepted (and legal) in most other countries too.

RE: The 3% rule

mk

(Mechanical)

18 Aug 17 15:18

Thank you, don and georgeveghese for sharing.

RE: The 3% rule

SrChemE

(Chemical)

(OP)

18 Aug 17 20:04

I want to thank everybody for your comments. I have since looked at articles in this database and taken various notes based on your feedback.
Many thanks!

RE: The 3% rule

georgeverghese

(Chemical)

19 Aug 17 04:38

Don, The last time I saw this clause in one of the appendices in ASME Sec 8 Div 1, there is nothing stated about an engineering analysis or otherwise to allow for inlet dp to exceed 3%.

From the title of this RAGAPEP, I gather these are also recommended good practices, and do not carry the same mandatory / regulatory compliance obligations as with the ASME BPVC.

To summarise, if you wish to exceed 3% inlet dp loss by following some route stated in the API, the risk is all yours - if anything goes wrong, you may not be able to take umbrage by the API at an audit / inquiry (depending on who runs the audit) since it is a ASME Code violation.

RE: The 3% rule

don

(Chemical)

21 Aug 17 14:15

Quote (georgeverghese)

To summarise, if you wish to exceed 3% inlet dp loss by following some route stated in the API, the risk is all yours....

OK, specifically what is that risk? For US facilities it is the risk that OSHA can successfully claim that API 520 Pt II is not a RAGAGEP standard. Can they claim such a thing? Yes - anything can be claimed. Would such a claim be upheld in a court of law? Again, anything is possible, but it's very unlikely that a globally used standard, such as API 520, could be judged to be contrary to RAGAGEP.

OK, specifically what is that risk? For US facilities it is the risk that OSHA can successfully claim that API 520 Pt II is not a RAGAGEP standard. Can they claim such a thing? Yes - anything can be claimed. Would such a claim be upheld in a court of law? Again, anything is possible, but it's very unlikely that a globally used standard, such as API 520, could be judged to be contrary to RAGAGEP.

Red Flag This Post

Please let us know here why this post is inappropriate. Reasons such as off-topic, duplicates, flames, illegal, vulgar, or students posting their homework.

Red Flag Submitted

Thank you for helping keep Eng-Tips Forums free from inappropriate posts.
The Eng-Tips staff will check this out and take appropriate action.

Reply To This Thread

Posting in the Eng-Tips forums is a member-only feature.

Click Here to join Eng-Tips and talk with other members! Already a Member? Login

News

If you want to learn more, please visit our website electric tube expansion system machine.