Jump to content

Featured Articles

Check out the latest featured articles.

File Library

Check out the latest downloads available in the File Library.

New Article

Product Viscosity vs. Shear

Featured File

Vertical Tank Selection

New Blog Entry

Low Flow in Pipes- posted in Ankur's blog


Cooling Water Failure Of Refrigeration Package

5 replies to this topic
Share this topic:
| More

#1 MacProcess


    Junior Member

  • Members
  • 14 posts

Posted 29 July 2019 - 01:47 AM



I have a proplylene refrigeration package with screw compressor.


It is an old vendor pacakge and there is lack of data and vendor closed down.


When cooling water fails, condenser will fail and there is no more condensing and compressor is still running.


And finally PSV at discharge will pop up.


However, vendor did not consider this scenario and there is no flare load for refrigeration package in cooling water failure.


Is there any good reason or just vendor mistake?


Is there any mechanical reason which there is no flare load in cooling water failure such as compressor trip?



#2 Bobby Strain

Bobby Strain

    Gold Member

  • Members
  • 2,825 posts

Posted 29 July 2019 - 10:43 AM

It's not the vendor responsibility to determine relief loads and relief scenarios.

#3 Art Montemayor

Art Montemayor

    Gold Member

  • Admin
  • 5,721 posts

Posted 29 July 2019 - 11:13 AM



I’m going to reword your post to check out that I fully understand what your questions are based on:

You have a propylene refrigeration package employing a screw compressor that doesn’t shut down when the cooling water to the refrigerant condenser fails to flow.  Consequently, the compressor’s discharge pressure continues to rise due to lack of condensation of the refrigerant and the discharge PSV is actuated (probably depleting the inventory of the propylene refrigerant within the refrigeration cycle).
Is this a correct understanding?


You further ask: “Is there any mechanical reason which there is no flare load in cooling water failure such as compressor trip?”
I don’t understand what you mean to state.  Do you mean to say: “Is it proper to have the propylene compressor shutdown upon a condenser water supply failure without having to vent the propylene refrigerant to flare?”
My answer to this is: Yes.


When you state you have “a propylene refrigeration package”, I assume you mean that you are using a propylene refrigerant to cool something (probably a process stream) - and you do not mean that you are using a refrigeration package to cool down a propylene stream.


You also state the supplier of the refrigeration package is out of business and did not supply a safety relief valve actuation upon condenser cooling water failure.  You also don’t have any plant design or operational data on this refrigeration package in your plant files.

I would state that you find yourself in a sorry state of engineering affairs.  A failure in maintaining local engineering files on all process equipment is a bad situation to find yourself in.  All process information - especially detail specifications and operational history and as-built P&IDs - should be kept in plant engineering files and up-dated.  I can never understand how anyone can operate a serious process plant without this organization of needed information.  This information is vital for local and government regulations and inspections, as well as insurance policies.  In the event of an accident or environmental spill, your local civil agencies would certainly be knocking on your door asking for all that information to ensure that you are operating a safe process and represent no safety threat or hazard to the local community.


I strongly urge you to initiate a serious and detailed hazard review of your process and document all information gathered as a result.  The fact that your refrigeration package supplier is out of business is no excuse.  You have inherited a bad situation and it is now your problem to resolve.  You have to gather all field basic data and calculate loads, develop detailed P&IDs and all other normally required engineering data to operate your process.

I advise you of this because I have, in my younger days, also gone through such situations.  I inherited badly operated plants that had no engineering files or basic data and had to re-organize and develop engineering files, calculations, standards, and other basic information.  It was not a picnic, but it has to be done to protect the operators who run the process, the community that surrounds you, and the stakeholders that invested in your plant operation.  If you can't do it, have someone else do it.

#4 PhilippM


    Gold Member

  • Members
  • 98 posts

Posted 29 July 2019 - 04:05 PM

Regarding the general state of available documents and data in your plant, I don't think there is much to add to Art Montemayor's description.


If I remember correctly however, the engineering company I worked for, did a safety upgrade project on a propylene refrigeration package a couple of years ago, and one very important task was to find out whether or not a relief load (and replacement of the available relief valve with a much larger one) was necessary in case of cooling water failure and compressor trip. We did lots of calculations and had to check the lengths of the lines on-site (as-built isometrics were not available) in order to determine the volume of the system. As it turned out the settle-out pressure was well below the lowest MAWP of any piece of equipment or line in the system, therefore there was no need for a relief load in case of cooling water failure/compressor trip.

Edited by PhilippM, 30 July 2019 - 12:25 AM.

#5 MacProcess


    Junior Member

  • Members
  • 14 posts

Posted 29 July 2019 - 07:27 PM

Mr. Bobby Strain


It was vendor package including equipment, piping and instrument and vendor also provided PSVs. 


Mr. Montemayor


Your understanding is correct. propylene is refrigerant. 

compressor shut down logic already exist in cooling water flow Low,

but because it is a conventional logic, I can not assume  compressor stop when PSV sizing according to API.


So, I wonder if there is a mechanical reason of compressor stop when cooling water failure, not by instrumentation.

Because vendor did not consider that scenario.


Thanks, anyway I think there was an vendor's mistake and it is correct that PSV should be considered for cooling water failure without compressor stop.

#6 Bobby Strain

Bobby Strain

    Gold Member

  • Members
  • 2,825 posts

Posted 30 July 2019 - 02:47 PM

The discharge PSV should be sized for cooling water failure at least. If it's not you need to correct this. I'll repeat my earlier statement. Don't rely on vendor to properly size safety valves. I have corrected even more dangerous shortcomings by vendors.



Similar Topics