The relief valves are another independient Protection layer (IPL) in your plant. This kind of valves do not depend of control system. For this reason you can not eliminate them.

It is mandatory to have different independient protection layer, for example:

1) Basic Control System
2) Emergency Shutdown System (alarms, shutdown valves, etc)
3) Mechanical Protections (relief valves)
4) Phisical Protections (dike design, distances between equipments, etc)
5) Hazard Plans for your plant in case of a emergency
6) Hazard Plans for the community


The different layer needed depend of the specific safety analysis for your plant.

Regards, Hianbo

   

Additionally the control system and shutdoen systems helps to reduce the frequency of the overpressure. the relief valve help to mitigate the severity of the overpressure. Each layer attack
Regards, Hianbo

Hianbo has described the different areas very well, the only thing I would to add is that with an ESD you have to evaluate the consequences for other equipment in the "train" as well. In general an ESD will generate multiple signals (or trips) to other equipment also, adding complexity to manage the system.  
In some cases as the situation gets out of control, a safeguard system can be needed to take over control when critical limits are reached.

Regards

I disagree with Hianbo's statement that

"The relief valves are another independent Protection layer (IPL) in your plant. This kind of valves do not depend of control system. For this reason you can not eliminate them."

A high integrity Safety Instrumented Function (SIF) (often called High Integrity Pressure Protection System (HIPPS))can sometimes be used i.s.o. a relief device, see for example NORSOK P-001, chapter 4 (http://www.standard.no/imaker.exe?id=1409), or API RP 521 section 2.2.

Note that a SIL classified Safety Instrumented Function also needs to be independent of the Basic Process Control System (BPCS).

I also don't agree with the statement "the control system and shutdown systems help to reduce the frequency of the overpressure, the relief valve helps to mitigate the severity of the overpressure."

A relief device also reduces the likelihood of overpressure. If a relief device operates on demand, it will prevent having overpressure (e.g. a pressure higher than 110% of the Maximum Allowable Working Pressure). If the relief device fails on demand (about once in every 1000 demands a conventional relief valve fails to open), it will no prevent having overpressure. In other words, the conventional relief valve reduces the likelihood of overpressure with a factor 1000. Same can be achieved with a SIL3 classified SIF.

Normally alternatives to conventional pressure relief devices are only considered when using a pressure relief device is considered unpractical (e.g. because of safety or environmental concerns or because of excessive costs).

Whether the code allows for this depends on the code that is applicable on your plant... When ASME code is applicable, requirements of ASME Code Case 2211 have to be met. In Code states in the USA, approval by the jurisdictional authority is required prior to the use of Code Case 2211.

Thank you all for your inputs
I agree with what Guidoo has stated regarding this subject.
But still not clear from your explanations if it is a mandatory of having both PSVs as well as HIPPS at the compressor discharge piping.
I am after a way of reducing the flare load in the plant during design stage with taking into account the protection provided by the shut down systems.

Brainstorming,

You cannot expect an answer to your question "if it is a mandatory of having both PSVs as well as HIPPS at the compressor discharge piping" from a forum such as this. We do not even know in which country your plant is located...

You should know yourself which codes, standards, recommended practices etc. are applicable on your plant. It may be required to discuss this with the authorities as well.

You must ensure that your plant is sufficiently protected against overpressure and that your plant is meeting all applicable codes, standards etc. and that it has all the necessary authority approvals.

Guidoo
If I know the answer, I wouldn't ask and post this question.
But I am expecting an input and feedback from the members who have an experience with such similar case as I have currently.
Regardless the standards and codes my company is applying for such case that might not be right or good compared to the best practices applied worldwide.
Thus, I raise this question again for those only have such case and they have experienced in projects.

Cheers


 

What is mandatory in Norway is not necessarily mandatory in the USA or in Russia...

What is good engineering practice in one application may not be good engineering practice in another application.

We don't know where your plant is located. We don't know any details of the process other than you have a compressor. We don't know how flexible the authorities in your country are.

etc. etc.

Are you getting the point?

brainstorming:

This is a valid and important safety point that has been brought out and discussed.  I’m adding this post because enough information can never be written or discussed when it comes to safety issues, and I feel there is an important result emanating from this thread.

You are not going to find as knowledgeable, hard-nosed, and experienced Chemical Engineer in safety issues and equipment as Guidoo that easily.  I believe I perfectly understand his point and the logical basis he employs.  I not only agree with him, but more importantly, I perceive just how he is trying to help you out of this dilemma.  He is pointing out that no matter how great, thorough, or “safe” a code or standard may be considered, it is of little or no importance if it is not backed up by a system that enforces, polices, and ensures that it is applied.  We still don’t know where you propose to install your HIPPS and much less what rules, laws, or codes apply there.  Therefore, we can’t comment on that.

What I can comment on is the point being made: You, and any other engineer(s) involved in this application are the responsible persons for seeing to it that the relevant and locally prescribed safety procedures are installed and operated correctly.  This is a minimal requirement.  Additionally, from a professional and moral standpoint an engineer has a responsibility to analyze and critique local safety requirements as they pertain to his specific application.  No amount of bureaucratic or political jurisdiction can substitute for sound engineering judgment.  And it is this essential engineering judgment that Guidoo is stressing.  We can all easily recline back and let the “codes” handle it; if something tragic happens, we can always turn away and state we did it according to the “code” and let it go at that.  Most codes are very thorough and have been extremely well up-graded to reflect good sound engineering practice.  However, there are still areas – especially in relatively new technology – which require careful and valued consideration before going forth.  Substituting a PSV on a compressor discharge with a HIPPS application is certainly a case to consider carefully.  We all know exactly what hardware, how it works, and the details of a PSV; we don’t know anything quantitative or qualitative about your HIPPS substitute except that it is considered as HIPPS.  This, logically, cannot be a basis for a decision or a recommendation.  Hence, Guidoo’s experienced comments.

A decision as important as the one that you present is one that should be decided logically based on experienced knowledge, careful scrutiny, legal and local codes, company policy, and sound engineering.  You, and those close to the application, are the most qualified to undertake that decision.  Certainly, detailed and candid discussion on this issue is a good start towards that decision and how it can be implemented safely.

Needless to say, I wish you luck and success in your decision.

It sounds good, but can you give the code which is asking about such requirements of having both HIPPS and PSVs.
The standard we have in my company calls for the need of having both but it doesn't mean that it is 100% true, and it might be changed if there are other codes worldwide allow for having only HIPPS without PSVs in the compressor discharge piping.
I would appreciate if you can give those codes which allow for HIPPS and not ask for another safeguard as PSVs.

Cheers

brainstorming:

You state:  “The standard we have in my company…. might be changed if there are other codes worldwide…”  I believe you are confusing the issue by equating the meaning of codes, standards, recommended practices, guidelines, etc., etc.  API 520 – parts I & II - as well as API 521 are clearly described as “Recommended Practice”.  This was noted by Guidoo in his post.

A standard, recommended practice, guideline, etc. is not a code.  What is meant by a code is a legally mandated (& enforced) practice, clearly outlining what is legally permitted.  The ASME “code” is not a code in Germany; there, DIN is the recognized code.  Anybody and everybody and their brothers can come up with a “code” whenever they so please.  It means nothing as a code if it isn’t legally mandated.  Otherwise it is a recommended practice.

What you are asking is next to impossible.  You ask us to cite all the “codes” that permit PSV substitution with HIPPS.  I can assure you no one on this forum will take the time – months or years – to compile such a list.  You’ve got a long wait ahead of you.  I also wish you good luck on this.

Regardless of which codes/standards/practices your particular area is under, I doubt they will say a PSV can be replaced with a shutdown system.  Although there may be some special cases, cost savings will probably not be one of them.

In General:
Equipment will require a relief device for overpressure protection.  
Equipment overpressure protection is separate from the IPL’s (Independent Protection Layers) associated with safety instrumented systems.  A relief device can count as an IPL, but sufficient layers provided by other instrumented systems will not take the place of a relief device.

I would imagine that limiting the flow to a flare using high integrity shut down systems would be possible, but no I don’t know where this would be spelled out in a code. (eg. In a plantwide power failure scenario, where multiple PSV’s from multiple units are relieving; tripping a furnace to reduce heat input, thereby removing flow from one PSV.  In this case, the header and flare would still need to handle this PSV and all its scenarios, but not necessarily this PSV plus all the others.  Also note that this example PSV would still need to be sized for the full ‘furnace is still on’ scenario.)  

Hope this helps.  
Again, note the ‘In General’ above.

In some countries, flaring is heavily restricted and HIPPS are used to greatly reduce environmental releases.  HIPPS are also used to reduce load on undersized flare systems.  It is okay to say that you want to replace a PSV with a HIPPS.  This must be considered very carefully and a process hazards assessment (i.e. HAZOP) is required.  All of the layer of protection must be considered, as many have already said.  Most companies consider HIPPS to meet a SIL 3 or SIL 4 design criteria.  SIL 3 is tough to meet as it will probably require 2 ESD valves installed in series with 2 transmitters voting 1oo2 (or triplicated).  It is also advised that the HIPPS be controlled by a separate safety rated logic solver.  A SIL 4 system will be quite a bit more expensive, but possible.

I would recommend following IEC 61511 or IEC 61508 standards as they outline the processes and steps required to design a system such as this.

In countries where flaring is heavily penalized, HIPPS is cost effective but in countries where the environmental restrictions are not as severe, HIPPS will prove to be too expensive.

Just my two cents.

Ok Brainstorming,

just to be on the same page as you; Are you looking for a standard or code that will support the argument that you can replace a PSV with HIPPS?

If YES, then Guidoo has provided you with the answer: ASME VIII Division I code case 2211, approved in 1996, sets the conditions under which over-pressure protection may be provided by an instrumented system instead of a PRV.

However, this substitition of the HIPPS for the PRV should provide a safer installation.  

The over-pressure protection can be provided by a SIS in lieu of a pressure pressure relieving under the following conditions.

1. The vessel is not exclusively in air, water, or steam service.

2. The decision to utilize overpressure protection of a vessel by a system design is the responsibility of the user

3.The user must ensure that the MAWP of the vessel is higher than the highest pressure that can reasonably be expected to be encountered by the system.

4. A quantitative or qualitative risk analysis of the proposed system must be made addressing all credible over pressutre scenarios.

5. The analysis in (c) and (d) must be documented.

So if you can satisfy the above stringent conditions, i believe you can confidently argure that you dont have to use both PRV and a HIPPS.

Buchi

Thank you Buchi for your understanding to the issue posted here
I will look for this standard along with your and the others recommendations, then justifications of having only HIPPS could be found and used to eliminate PSVs and reduce flare load and size as well.

Cheers

See also article by Angela E. Summers: "Consider an instrumented system for overpressure protection", Chemical Engineering Progress, November 2000, pages 65-68. IT can be downloaded from http://www.iceweb.com.au/sis/overpressure.htm

Some additional remarks:
- Code Case 2212 and the above mentioned interpretation may not be accepted by authorities all around the world.

- Specific regulatory and enforcement jurisdiction requirements must be determined. In some instances, approval of local authorities is required.

- Note that Code Case 2211 doesn't simply tell you that it is allowed to use a SIS i.s.o. a PSV. This is an interpretation of this Code Case.

- Case 2211 is not a standard or a code but an official guideline to the use of the ASME BPVC.

I think a key point that needs to be stated clearly is that you don't have to remove the PSV to reduce the design flare load. sdl inferred this but I thought it was worth making it explicit.

HAZOP at www.curryhydrocarbons.ca

I am aware of a few projects that eliminated a PSV related to a reboiler application in a refinery or petrochemical plant.  If I had been in the PHA I would have argued against this.  Even if the PHA could eliminate the PSV thus lower the flare load, couldn't the analysis consider keeping the PSV and disregarding its flare load contribution - as it should never open?

An operator can see the open/closed status of the PSV inlet block valve.  The maintenance bypass status may be apparent but the operator cannot see the program.  As a control systems engineer I advocate the segregated systems and assorted redundancy levels.  However, I would prefer to keep the PSV - even if it is mechanical and subject to leakage once lifted off the seat.

John

John:

You do not have to eliminate the PSV.  HIPPS are installed where flaring is an initiating event i.e. undersized flare system or severe economic penalties in the event of an environmental release.  Some HIPPS valve manufacturers insist on a PSV downstream to protect against any possible leakage through the valve (additional protection).

In some countries, flaring is heavily penalized and leaky PSVs can be quite costly.  A HIPPS system is quite expensive to design, install and maintain, but the life cycle cost is much lower than any potential penalties levied by regulatory bodies.

By eliminating the PSV, one independent layer of protection is removed, thus the SIF will have to be designed to meet an additional SIL.  Most HIPPS are designed to SIL 3 and are separate and independent systems from the plant SIS.

SDL

Guidoo,

I agree with everything you write but I have one question concerning the reliability data of safety valves.

I know that some companies use a PFD (probability of failure on demand) factor of 0.01 (which comes up to a failure of 1 per 100 demands compared to 1 per 1000 demands in your message) for PSVs.

This would result in a different SIL classification for the described safety instrumented function. Are there any reliability data for PSVs public available (besides OREDA)?

Sorry, this is not exactly matching to this thread but I thought it would be a good opportunity to ask...

Thanks a lot and Best Regards

The CCPS (Center of Chemical Process Safety) database shows for spring loaded safety relief valve, failure mode "Fails to Open on Demand" a mean value of 0.212 failures per 1000 demands. This falls in the SIL3 range.

For Pilot Operated Safety Relief Valves they give a mean value of 4.15 failures per 1000 demands. This falls in the SIL2 range.

From my previous experience I always used three PSHH (dublicated) at cenrifugal compressor's discharge line.
Code for this  - B31.8 par. 843.441
For reciprocating compressors PSV should be used.

API 14C (1998), API 521 (1997), and Code Case 2211 (1995) of ASME Section VIII, Division 1 and 2, provide alternatives in the design of overpressure protection systems.  These alternatives revolve around the use of an instrumented system that achieves a level of safety that meets or exceeds the protection provided by a pressure relief valve and flare system.
Any instrumented system used to provide over-pressure protection is a safety-related system, since its failure would result in the rupture of the pipeline/vessel or in overloading the flare.  As a safety-related system, the instrumented system must meet either the United States domestic ANSI/ISA S84.01-1996 (1996) or the international standard Draft IEC 61508 (1998, 1999).  Due to the high likelihood that the instrumented system would be needed and the high severity of the consequence should these fail, the SIL assigned per the standards is often 3 (or simply as high as achievable with redundant architecture, high availability devices, and frequent proof testing).  Due to the high availability requirements, these over-pressure protection systems are often called “high integrity protection systems” or HIPS.
Industry is increasingly moving towards utilizing HIPS to reduce flare loading.  They are becoming the option of choice to help alleviate the need to replace major portions of the flare systems in existing facilities when adding new equipment or units.  The relatively low capital cost of HIPS compared to flare system piping upgrades and the ability to install HIPS without incurring significant additional downtime during a turnaround, makes these systems an extremely attractive option.
However, prior to making the choice to install the HIPS, the regulatory and industrial standards pertaining to their design must be well understood.  Due to the unique nature of the HIPS application, certain design aspects must be carefully evaluated.  Any company considering HIPS is cautioned to do a thorough hazard evaluation prior to the implementation of HIPS.
    HIPS do not differ greatly from other trip systems.  The systems are composed of field-input devices, a logic solver and final elements.  The necessity for high availability and reliability is where the differences truly begin.  Redundancy in field devices is utilized to provide a high level of availability while, at the same time, increasing reliability.  Typically, the inputs are configured in a two-out-of-three voting basis, the logic solver should have high availability, and the final elements are configured one-out-of-two.  The design of any HIPS should be quantitatively verified to ensure it meets the required availability.
    Care must be taken in any decision to implement HIPS.  The use of HIPS should be generally restricted to the reduction of relief and flare loading in existing facilities.  The use of HIPS should not be a justification for reducing the pressure relieving requirements on individual pieces of equipment.  The pressure relieving of vessels should be sized for the worst credible scenario for each piece or groups of equipment irrespective of the HIPS design.
    
    Advantages of HIPS:
•    Low capital costs compared to upgrading flare systems
•    Can be installed without incurring additional downtime during a turnaround
    
    Disadvantages of HIPS:
•    HIPS require that many different components work as designed.
•    Effectiveness of system is highly dependent on the field design, device testing, and maintenance program.
•    Limit of knowledge in the identification of all over-pressure scenarios
•    HIPS becomes the “last line of defense”.  Failure results in potentially over-stressing of vessel.

Hi brainstorming,
After giving due consideration to all of the above, pl consider this:
If it is a positive displacement compressor, provide PSV and if it is centrifugal, axial, etc, go for HIPPS. This agrees with JETEE, I suppose.
My two bits.

HIPPS is a substitute for those cases when the risk cannot be practically reduced to a tolerable level.  The Safety Integrity Level is assigned based on a risk assessment of the identified hazards.

As mentioned earlier, some countries place heavy restrictions on flaring so in these cases, the environmental and/or financial risk assessments may over-ride the safety assessment.  There is no hard and fast rules on when to use PSVs vs. HIPPS.  Each situation needs to be addressed individually because individual corporate risk tolerances vary.  In existing facilities where the flare size is fixed, HIPPS would be a method to reduce the risk where to increase the size of the flare system would be cost prohibitive.

The best method we have today for designing and maintaining safety systems is IEC 61508/61511.  These standards outline a good practical methods.

sdl