This is a very common question when auditing “older plants”, say plants built prior to 1960 (not a specific date with any meaning). For example, during an audit it is discovered that a vessel was built to a RAGAGEP that no one had even heard of or maybe there is NO documentation as to what the vessel was built to. Is this a compliance issue and if it is what must the facility do to meet the PSM/RMP requirements?
First let me state that YES, this is a slam dunk for OSHA/EPA on at least either of these two requirements:
1) 1910.119(d)(3)(ii) The employer shall document that equipment complies with recognized and generally accepted good engineering practices.
or
2) 1910.119(d)(3)(iii) For existing equipment designed and constructed in accordance with codes, standards, or practices that are no longer in general use, the employer shall determine and document that the equipment is designed, maintained, inspected, tested, and operating in a safe manner.
In our auditing we always use (d)(3)(ii) as it is more clear cut when the facility does not have a construction drawing, U1A form, or some type of engineering analysis that shows this vessel is SAFE for it’s intended operation and that the current maintenance plan would ensure it’s ongoing integrity. And only when the facility claims, “how can we be held accountable for a code that is no longer in print” do we use (d)(3)(iii).
It is (d)(3)(iii) that I want to cover in this posting. 29 CFR 1910.119(d)(3)(iii) states:
For existing equipment designed and constructed in accordance with codes, standards, or practices that are no longer in general use, the employer shall determine and document that the equipment is designed, maintained, inspected, tested, and operating in a safe manner.
So if we have a process component (i.e. vessel, pipe, RV, etc.) that was designed and constructed to a RAGAGEP that is no longer in print/use (NOTE this is very DIFFERENT than one that has been “revised” – more on this later in article) we must perform an “engineering evaluation”, such as a fitness-for-service assessment per API 579, to ensure that the process component is still SUITABLE for our process operating conditions. We have to ensure that the process component is suitably designed, maintained, inspected, tested, and operated in a safe manner.
Here is what two (2) OSHA LOI’s state regarding this:
LOI #1 – dated 1998 and emphasis is added by me – NOT by OSHA (CLICK HERE for the actual LOI)
Question 1
Is a previous edition of a current code or standard (i.e. 1972 edition of the Section VIII of the ASME Code) considered to be “a code or standard no longer in general use”?
Reply 1
Yes. This question is generated from OSHA’s standard 29 CFR 1910.119(d)(3)(iii). The intent of this requirement is for the employer to determine and document that PSM covered equipment which was designed and constructed to codes, standards or practices which are no longer in general use can continue to operate in a safe manner. After the employer makes the determination required by the standard, it will be the basis for the decision to take the equipment out-of-service or to continue operations. If the equipment is to be kept in-service, the determination will be the baseline from which all future operation, inspection, testing and maintenance is conducted. In making the determination the standard requires the employer to evaluate the design, maintenance, inspection, testing and operation of the PSM covered equipment. When making the determination required by the standard, OSHA intends that the employer document that PSM covered equipment is consistent with the latest editions of codes and standards. If an employer determines that covered equipment is not in conformance with the latest editions of codes, standards or practices, the employer must document which codes and standards were used and that the design, construction, maintenance, inspection, testing and operation are still suitable for the intended use.
Question 2
Given a pressure vessel designed, constructed, inspected and Code-stamped in accordance with a previous edition of the ASME Code, and for which a manufacturers U-1 form exists, is it necessary to recalculate the vessel’s design characteristics (wall thickness, nozzle reinforcement, etc.)?
Reply 2
If the original design and construction is in accordance with the latest edition of design and construction codes and standards for equipment covered by the PSM standard, then OSHA’s standard 1910.119(d)(3)(iii) does not require the employer to recalculate design characteristics of PSM covered equipment. However, if there are differences between the original and the latest edition of design and construction codes and standards, then OSHA requires the employer to document those differences. The employer must also show how those differences are consistent with the latest editions of design and construction codes and standards to assure that the equipment can continue to function in a safe manner. A pressure vessel designed and constructed using any edition of the ASME Code and which posssses a manufacturers’ U-1 certification, may have had a change in-service at sometime during its operating history. To assure safe operation the employer when making its determination as required by this standard must also determine and document that the in-service operation, inspection, testing and maintenance of PSM covered equipment considers the change in-service and its impact is consistent with the latest codes, standards and practices so as to assure safe operation.
Question 3
When performing recalculation as a means of establishing suitability for intended service, and given a pressure vessel for which no documentation of material of construction, welding procedures or radiographic testing exists, is it acceptable to assume the lowest value for these variables (i.e. lowest weld joint efficiency factor, lowest allowable stress value for the class of materials involved, etc.) and to incorporate these values into the recalculation?
Reply 3
When an employer conducts an engineering analysis, including recalculation, when no documentation exists for the material of construction, welding procedures or radiographic testing , it is appropriate to assume the lowest value for the listed variables in the engineering analysis as a means of determining that the in-service condition of covered equipment is appropriate for its intended use. The engineering analysis must be conducted in conformance with the latest editions of codes and standards.
LOI #2 – dated 2001 and emphasis is added by me – NOT by OSHA (CLICK HERE for the actual LOI)
Question: If an employer determines that a vessel is suitable (vessel design, construction, maintenance, inspection, testing and operation) for PSM service to a particular American Society of Mechanical Engineers’ (ASME) Code, is it required that all of this information be updated periodically or only if there is a change of service?
Response: Based on your letter, your question is related to the PSI portion of OSHA’s PSM standard, specifically 29 CFR 1910.119(d)(3)(iii). The intent of this requirement is for the employer to determine and document that PSM-covered equipment that was designed and constructed to comply with codes, standards or practices no longer in general use can continue to operate in a safe manner. After the employer makes the determination required by the standard, it will be the basis for the decision to take the equipment out-of-service or continue operations. If the equipment is to be kept in-service, the determination will be the baseline from which all future operation, inspection, testing and maintenance is conducted. Generally speaking, OSHA intended for the employer determination and documentation required by 29 CFR 1910.119(d)(3)(iii) to be completed prior to the implementation of the original PHA or startup of a PSM-covered process. Therefore, once an employer is in compliance with this requirement, there is no additional requirement per 29 CFR 1910.119(d)(3)(iii) for future determinations/documentation simply because a code or standard related to the covered equipment has been revised. After the employer has made this baseline determination and documentation, other PSM elements such as management of change, mechanical integrity, PHA-revalidation, pre-startup safety review, etc., are intended to address on-going safe operation and maintenance of PSM-covered equipment.
So as we can see, OSHA does expect the employer to make the baseline determination that the process component found to be designed and constructed to a RAGAGEP that is no longer in print/general use that it can continue to operate safely and that the maintenance plan for the component will ensure it’s on-going integrity.
However, OSHA does NOT require us to do this type of review every time a RAGAGEP is UPDATED. So here is the difference that many get confused on, including me!
Example: We find a vessel that we have no documentation on. We have no as-built drawings, we have no U1A, we have no fit for service inspection(s), etc. This is a MAJOR problem! We cannot simply state that the vessel was built before any of us were born and therefore there was no RAGAGEP at the time and therefore we have no responsibility to ensure it is safe to operate! Yes there are some that use this excuse! What we must do, is have a QUALIFIED engineer perform an engineering evaluation to basically develop our Process Safety Information for this vessel. For example, the engineering evaluation MUST establish the maximum allowable working pressure (MAWP) for the vessel and then we must ensure that our operating Safe Upper Limits are BELOW this newly created MAWP. Now please keep in mind I used MAWP as just ONE (1) example of the data that we would need. This same concept applies to materials of construction and ALL operating parameters. In essence, the engineering evaluation will create a sudo U1A form and/or as-built drawing for us.
Once we have this engineering evaluation completed and we have ensured that the component can SAFELY operate within our process conditions and operating parameters, we must also use this engineering evaluation to establish a suitable mechanical integrity plan for this component. Often times this is the most difficult since most of the time we can just follow the manufacturer’s recommendations; however, with respect to some process components we often times find the manufacturer is NO LONGER in business and we can not find an “owners manual” for the component. We use the engineering evaluation to establish the inspections/testing and the frequency of this inspection/testing.
Here is an example of how this has impacted me when I was a Safety/PSM Manager:
We had an old PSM covered process that used process piping made of “XYZ” materials of construction. The piping was in horrible shape and there were far too many “minor leaks” occurring for the business. So as PSM came out we began to put our PSI together and when it came to this system within the process we discovered that the vendor who supplied us with our PSM covered chemical, I’ll call it “Chemical A”, they clearly stated in their literature that piping made of “ABC” or “DEF” was to be used. We could not determine from our records review why this process used pipe made of “XYZ” materials of construction and in consultation with the supplier they found that “way back” “XYZ” was stated to be “acceptable” which is quite different from “preferred”. To be clearer: we used carbon steel which was “acceptable” and the supplier of this chemical “preferred” the pipe to be “lined with a special liner” that prevented accelerated corrosion/erosion. However, the supplier at the time of the evalaution “prohibited” the use of unlined pipe with their chemical. As many of us have experienced, “acceptable” turns out to be MUCH cheaper than the “preferred” and this is what we end up with. So the business needed to perform an engineering evaluation on our current piping for this system within the process. The evaluation pretty much stated what the supplier said; “it is OK, but not the preferred materials of construction for the application/chemical”. What nailed us was the engineering evaluation stated that since we used “XYZ” that the service life of the pipe was 15 years, unless we had historically data to show that our operating experience proved a lesser erosion/corrosion loss that what the evaluation predicted. We did some fit for service testing (NDE) and found the pipe to be beyond its minimum thickness for it’s material of construction. Essentially, the age of the pipe and the loss of thickness meant that we could keep the pipe in service, but we would have to do NDE annually till its retirement. It was cheaper and SAFER to just replace the pipe with the pipe the supplier of “Chemical A” recommended and would you know it – the leaks stopped!
So to close this discussion out, remember there are TWO (2) objectives that need to be met when we find a process component that we have NO data on and the supposed RAGAGEP it was built to is no longer in use/print:
1) engineering evaluation to ensure it can currently operate SAFELY within our current operating parameters, and
2) develop a maintenance plan to ensure the on-going integrity of this process component based on the data from the engineering evaluation.
I STRONGLY encourage all my clients to use API 579 – Fitness for Service code to ensure their engineering analysis is done properly. This code is NOT a requirement by OSHA, but it fits perfectly, albeit not in all situations, into our engineering evaluation process. And if OSHA or EPA ever ask “how did you determine this component is suitable for your process” you have a very well recognized RAGAGEP to fall back on; otherwise the engineer who conducted the evaluation will have one of his/her worse days in their career as they are interviewed by the inspecting agency.
We have to DOCUMENT that our equipment complies with recognized and generally accepted good engineering practices. This is EVERYTHING that could lead to a release/accident involving our PSM/RMP covered process. And when we find a piece of equipment that was built to a RAGAGEP that is no longer in use/print, we must conduct an engineering evaluation, much like in API 579, to DOCUMENT that the process component is SUITABLE for our process.
I would like to hear from anyone who has something to add or challenge. You can e-mail me privately or post your comments publically below in the comment sections. All comments and challenges are welcome.