4 replies to this topic

[riven]

I am recommissioning a pilot plant that has been out of service for 3 years.

Short description
In 01-D1, organic feed liquid of 35 L is used (e.g. butanol). This liquid (35L) has a maximum operating temperature of 150C and 25 barg. The feed is heated by heat tracing around 01-D1. There are several temperature trips (01TIA4, 01TICA1) and pressure trips (01-PIA2 01-PIA4/5) which should keep the operation within these limits.
This liquid is fed to pump 01-P1 where it is eventually returned to the 01-D1. This circulation loop has a connection to a nitrogen header for in-erting/flushing (after 01-V15).

There are several ways to supply pressure

• 01-P1
• Nitrogen header
• External pumps (not shown)

In an overpressure event, there are 3 relief valves than can potentially relive; 01-SV2, 02-SV2 and 01-SV1. The first is specifically for overpressure after 01-P1, 02-SV2 is for overpressure after the external pumps and 01-SV1 is for over-heating on the feed tank (02-SV2 will also relieve in this event).

These relief valves on walk down are actually found to be check valves i.e. one direction valves. I have never seen check valves used as relief valves. I would prefer relief valves as the relieving rate is higher.
Can check valves be used in this type of service?

Attached Files

•  Close upx.pdf   32.18KB   40 downloads

[Art Montemayor]

Riven:

I would tread very carefully when commissioning this pilot plant. I would apply a lot of caution and suspicion. It has been my personal experience that pilot plants that have been left in “moth balls” (storage) usually have little or no technical or engineering documents regarding their design, operation, or scope of work. Once the experimenter obtains the results he is looking for from a pilot plant, the equipment is set aside and forgotten. The specific purpose and design of the equipment is also “forgotten”. The next experimenter that follows must be cautious because it is highly probable that his process will be very different in scope and purpose. Let me give you some examples based on your diagram information:

• You don’t tell us if vessel 01-D1 is a PRESSURE VESSEL or what is the mechanical design – MAWP and MAWV; the maximum OPERATING pressure and temperature are of minor interest – the most important pressure and temperatures are the MAXIMUM ALLOWABLE WORKING values.
• Is the vessel filled with 100% (35 liters) Butanol LIQUID and then heated?
• The system heats up the liquid by heat tracing – which is inefficient and slow; this will work by pump recirculation – but a reasonable vapor space MUST be available for liquid expansion upon heat up.
• What is “01-V15”? – a manual vent valve? It is shown as plugged.
• What is YOUR scope of work with respect to this pilot plant? In other words, what process and purpose are you preparing it for?
• I have never seen your symbol for a Pressure Relief Valve (PSV), but I have to presume it is a PSV. If so, then IT MUST REPRESENT A CONVENTIONAL PSV, AND NOT A CHECK VALVE.
• To prove my point above about the PSV symbol, look at your diagram and note that the check valve on the discharge of pump (?) 01-P2 is clearly depictd in the conventional way. It is far different than the PSV symbol. Therefore there must have existed a PSV there before. I would suspect that someone replaced the original PSVs with check valves - either because the process changed or it was done in ignorance.
• Why don’t you furnish your flow diagram symbology? That way, we all would know what equipment is being described.

According to what you describe and your diagram, I would STRONGLY ADVISE YOU TO REMOVE THE CHECK VALVES AND INSTALL APPROPRIATE PSVs – as depicted on the diagram. Under no circumstances can a check valve serve as a relief device – at least not on this system as depicted.

You must allow for a reasonable vapor expansion space in any liquid vessel that you are heating! To do otherwise is to invite a disaster from a hydraulic overpressure and subsequent vessel rupture. Like most pilot plant systems, this one is probably “low cost” and has no HazOp history on it. I strongly recommend that you draw a correct and detailed P&ID for your scope of work and HazOp the process to ensure that it is safe to operate. From what you describe, you do not have a safe process and I would not commission it for startup.

I hope this experience assists you in taking the appropriate safe engineering steps and actions.

[riven]

I will your reply at each point. It is a pleasure to read your response by the way. At the end is a bit of history.

I would tread very carefully when commissioning this pilot plant. I would apply a lot of caution and suspicion.
That is my starting point

It has been my personal experience that pilot plants that have been left in “moth balls” (storage) usually have little or no technical or engineering documents regarding their design, operation, or scope of work.
As we speak I am rebuilding the documents (they were missing) and am currently in the middle of re-doing a Hazop.

Once the experimenter obtains the results he is looking for from a pilot plant, the equipment is set aside and forgotten.
If I had dime for every time I have experienced that. I work currently in an energy research institute (not university) where there are many scientists with little knowledge or interest in basic maintenance procedures.


You don’t tell us if vessel 01-D1 is a PRESSURE VESSEL or what is the mechanical design – MAWP and MAWV; the maximum OPERATING pressure and temperature are of minor interest – the most important pressure and temperatures are the MAXIMUM ALLOWABLE WORKING values.
01-D1 is a pressure vessel with a MAWP of 30 barg. MAWV (volume?) is 60L. Design temperature is 300C with the maximum allowable temperature being set at 200C. I have it retested by Llyods (external) who have completed the pressure testing based on acids. It is passed and registered.

Is the vessel filled with 100% (35 liters) Butanol LIQUID and then heated?
Yes but the vessel from above, is not 100% full.


The system heats up the liquid by heat tracing – which is inefficient and slow; this will work by pump recirculation – but a reasonable vapor space MUST be available for liquid expansion upon heat up.
Agreed. I was scratching my head as to why trace heating was used when I first encountered it. The equipment is proposed for long term experiments covering a period of months. The heating ramp to set point temperature is adequate for such experiments (ramp is approx 12 hours).

What is “01-V15”? – a manual vent valve? It is shown as plugged.
It is a manual ball valve. I should have added the caveat that the valves in the PnID may not be correctly drawn. I have a new version being drawn up as we speak.


What is YOUR scope of work with respect to this pilot plant? In other words, what process and purpose are you preparing it for?
The purpose is to dehydrate wet solvents. These streams are laboratory grade so no solids or acidic/basic streams are considered.


I have never seen your symbol for a Pressure Relief Valve (PSV), but I have to presume it is a PSV. If so, then IT MUST REPRESENT A CONVENTIONAL PSV, AND NOT A CHECK VALVE.
It is not correct. I am approaching the project as if this plant were just delivered. In the walk down I am checking everything against the available documentation and serial numbers.
That symbol for the PSV for me at least is easily recognizable (experience in Netherlands, Ireland).


To prove my point above about the PSV symbol, look at your diagram and note that the check valve on the discharge of pump (?) 01-P2 is clearly depictd in the conventional way. It is far different than the PSV symbol. Therefore there must have existed a PSV there before. I would suspect that someone replaced the original PSVs with check valves - either because the process changed or it was done in ignorance.
01-P2 is a pump. The depiction there is correct. However the PSV's (01-SV2 etc) are in-correctly drawn. From the valve serial numbers printed on them, I have a date and this corresponds to the original design/commission date. I have that they are indeed check valves visually, by taking them out of the system (their relieving pressure would have to be tested anyway) and talking with the manufacturer.
In other words people in the past have been using this equipment based on using check valves are relief devices. I have never seen this before and would not consider it normal or correct, hence the question.

Why don’t you furnish your flow diagram symbology? That way, we all would know what equipment is being described.
Do not have access to it at this time. In our company we have a separate fabrication department. i will ask them to supply me with a digital list that is easy to read and post it here (if they have one).

According to what you describe and your diagram, I would STRONGLY ADVISE YOU TO REMOVE THE CHECK VALVES AND INSTALL APPROPRIATE PSVs – as depicted on the diagram. Under no circumstances can a check valve serve as a relief device – at least not on this system as depicted.
That is in tandem with my thoughts on the issue.

Like most pilot plant systems, this one is probably “low cost” and has no HazOp history on it. I strongly recommend that you draw a correct and detailed P&ID for your scope of work and HazOp the process to ensure that it is safe to operate. From what you describe, you do not have a safe process and I would not commission it for startup.
I agree. It has had a Hazop history but the Hazop was so inadequate that I threw it out (it was based on a conceptual design diagram and not on the existing PnID).


*The plant was built in 1999 and was used for several years and then mothballed. However, as I would find out the original plant was different to the PniD currently in existence. Extensive changes had been made to the plant and had not been documented. Recently there was a drive from a manager (no technical experience with experimental equipment) to restart the plant using water. The basis was the we have used it before and it was fine. I was away but got wind of the situation and rang the tech and told him under no circumstances to start it up. I am currently dragging the situation kicking and screaming away from commissioning. My document list on my desk at this moment is Hazop, control logic diagram, risk and ATEX evaluation, technical information package etc. It has not been a pleasant experience but I refuse to let unsafe equipment to be operated on. I will probably leave this company when I am finished with this item.*

Edited by riven, 03 December 2010 - 04:42 AM.

[Art Montemayor]

Riven:

Thank you for your timely response to my comments. You have made me feel more comfortable about this pilot plant passing into experimenters hands with a professional engineering criterion and safety design input. I commend you on the work you are doing and the professional attitude you have taken towards it. Instead of acting emotionally upon coming across such a potential disaster, you have obviously acted seriously and professionally, knowing exactly what you must do.

I took the time and effort to comment on this topic because I also have gone through this type experience and I felt then – and still do – that there is something of worth to be acquired from this. Engineers are trained and bred to solve problems. That is our claim to fame. Most times the problems belong to individuals who – because of a lack of sufficient brain cells, education, and common sense – are unable to understand the problem and much less resolve it. But there are other times when the problems are brought about not by educationally-challenged people, but by what presumably are the “crème de la crème” in science and learning. And, incredibly, these problems can be the most potentially disastrous. It is our obligation and calling to help them as well – although they can be very difficult and impossible to cooperate.

You are doing exactly what is expected of you as a professional engineer by attacking the obvious basics of any process:

• Document and check out all hardware;
• Certify all pressure vessels and protect them with the appropriate PSVs according to Code;
• Develop a detailed and accurate P and ID, with which you can generate a HaZop.

You are making it possible for some erudite and learned scientists to work in a safe and productive environment – in spite of themselves. They may not take a liking to you because you are an engineer, but I am sure they secretly are aware that you are safeguarding them from their own shortcomings.

Keep up the good work.

[gel]

As per ASME Section 8,Division 1 and API, we can not use check valve even in inlet and outlet piping of relief valve. So check valve can not be used in place of relief valve.
quote name='riven' timestamp='1291303023' post='44658']
I am recommissioning a pilot plant that has been out of service for 3 years.

Short description
In 01-D1, organic feed liquid of 35 L is used (e.g. butanol). This liquid (35L) has a maximum operating temperature of 150C and 25 barg. The feed is heated by heat tracing around 01-D1. There are several temperature trips (01TIA4, 01TICA1) and pressure trips (01-PIA2 01-PIA4/5) which should keep the operation within these limits.
This liquid is fed to pump 01-P1 where it is eventually returned to the 01-D1. This circulation loop has a connection to a nitrogen header for in-erting/flushing (after 01-V15).

There are several ways to supply pressure

• 01-P1
• Nitrogen header
• External pumps (not shown)

In an overpressure event, there are 3 relief valves than can potentially relive; 01-SV2, 02-SV2 and 01-SV1. The first is specifically for overpressure after 01-P1, 02-SV2 is for overpressure after the external pumps and 01-SV1 is for over-heating on the feed tank (02-SV2 will also relieve in this event).

These relief valves on walk down are actually found to be check valves i.e. one direction valves. I have never seen check valves used as relief valves. I would prefer relief valves as the relieving rate is higher.
Can check valves be used in this type of service?
[/quote]