Tag: Management of Change

Direct Replacement, Replacement in Kind and Management of Change

Of all the PSM/RMP requirements, the Management of Change element is the most consistently problematic. Most of the difficulty is in answering two questions:

  • The compliance question: Does the change you considering count as a “change” per the PSM/RMP rule.
  • The safety question: Does the change you are considering have the potential to affect the safety of the process?

Note, it is quite possible that you answer NO to the first question and YES to the second question.

 

The text of the Rule

1910.119(l)(1) – The employer shall establish and implement written procedures to manage changes (except for “replacements in kind”) to process chemicals, technology, equipment, and procedures; and, changes to facilities that affect a covered process.

From a compliance perspective, how broadly you interpret the “…changes to facilities that affect a covered process” portion dictates how many changes will be subject to this element.

Of course, there’s also a little window that allows you to avoid the MOC element if you classify the change as a “replacement in kind.” The rule provides a fairly useless definition of Replacement in Kind:

1910.119(b) …Replacement in kind means a replacement which satisfies the design specification.

The “replacement in kind” exception is routinely abused to avoid MOC. To understand this element better, let’s consider a few scenarios: Replacing a Valve, replacing a Motor, replacing an Ammonia Detector, and replacing a Condenser.

 

Example: Replacing a Valve

In this example, we’re replacing a valve with the same model, size, etc. Is this a change? Some people would call this a Replacement in Kind, but I would not. I would call this a Direct Replacement. It’s not kind of like the valve we’re replacing, it’s exactly like it. Such a change is outside the scope of the MOC element entirely.

What if we were replacing the valve with a different brand or model? Then we don’t know if it is a Replacement in Kind until we ask enough questions to assure ourselves that it satisfies the design specification. Some questions we might ask are:

  • Is it made of the same materials?
  • Does it have the same flow ratings / capacity?
  • Does it have the same mode of operation in manual and automatic?
  • Does it have the same Mechanical Integrity requirements?
  • Does it affect the PHA section that this equipment belongs to?

It’s quite possible that you answer enough questions to assure yourself that the replacement valve satisfies the design specification making it a Replacement in Kind. While this means it is outside of the MOC element for compliance purposes, we’d still recommend you document the rationale you used to determine that it meets these design specifications. You could even take this documentation one step further and declare that in the future all replacements of Brand A Model X valve with Brand B Model Y valve can be considered a Direct Replacement in this application.

 

 

Example: Replacing a Motor

A motor might be considered by some (incorrectly) to be outside of the MOC element because it doesn’t (usually) contain ammonia, but this is short-sighted. Remember, the MOC element is about Changes to…equipment…that affect a covered process. A motor for equipment that is part of your covered process would certainly fall within the scope of the element for consideration.

Ideally, we’re replacing the motor with the exact some one – a Direct Replacement that would place it outside the MOC element. But, if we are replacing it with another motor, we will be looking to prove that it satisfies the design specification so we can consider it a Replacement in Kind. Again, we need to ask questions:

  • Does it have the same electrical requirements (phase, voltage, etc.)
  • Does it have the same frame size?
  • Does it have the same RPM, duty rating, capacity, etc?
  • Does it affect the PHA section that this equipment belongs to?

Just like earlier in our valve example, it’s quite possible that you answer enough questions to assure yourself that the replacement motor satisfies the design specification making it a Replacement in Kind. While this means it is outside of the MOC element for compliance purposes, we’d still recommend you document the rationale you used to determine that it meets these design specifications. You could even take this documentation one step further and declare that in the future all replacements of Brand A Model X motor with Brand B Model Y motor can be considered a Direct Replacement in this application.

 

 

Example: Replacing an Ammonia Detector

Like earlier, with the motor example, a change to an ammonia detector might be considered by some (incorrectly) to be outside of the MOC element because it doesn’t (usually) contain ammonia, but this is short-sighted. Remember, the MOC element is about Changes to…equipment…that affect a covered process. You definitely consider these detectors as safeguards in your PHA, so we need to exercise some caution on this change.

If we’re replacing the detector with the exact same one, then it’s a Direct Replacement. If we’re replacing it with a different detector, we need to assure that it satisfies the design specification so we can consider it a Replacement in Kind. Again, we need to ask questions:

  • Does it have the same electrical requirements?
  • Does it have the same output signal / alarm outputs?
  • Does it have the same sensitivity?
  • Does it have the same Mechanical Integrity requirements? The same calibration equipment, schedule and calibration procedure?
  • Does it affect the PHA section that this equipment belongs to?

In my experience, unless you are dealing with a Direct Replacement, no detector meets the requirements for a Replacement in Kind because they almost all fail the last question on calibration equipment, schedule and procedure. That means such a change would require the implementation of a Management of Change procedure.

Here’s where we can get a little clever. The PSM/RMP rules require that we “establish and implement written procedures to manage changes” but they don’t require that we use the same procedure for every change! If we sit down and think through all we need to do to successfully change from Brand A Model X NH3 detector to Brand B Model Y NH3 detector, we could establish a standard procedure for doing so. That means that in a facility with, say, 45 detectors that you are changing over a period of time, you have a Single MOC (to establish the new procedure) and then simply implement the new NH3 Detector Change SOP 45 times as the changes occur.

 

 

Example: Replacing a Condenser

In this example we’re replacing a brand A evaporative condenser with 500 tons of capacity with a brand B evaporative condenser with 500 tons of capacity. Note: this would work the same with if you were replacing it with a different model of brand A as well. Also, if you were replacing a condenser with an exact duplicate, then theoretically you may be able to get by with a PSSR, but that assumes you don’t need any non-standard operating modes during the change-out.

First question: Is it a change to equipment which satisfies the design specification? Answer: Well, I don’t know because there is a lot that goes into that determination. But every part of a condenser changeout has the potential to affect the safety of your system. Questions you should ask include:

  • Does it have the same electrical requirements (phase, voltage, amp draw, etc.)
  • Is it have the same size and weight?
  • Is it made of the same materials?
  • Does it have the same flow ratings / capacity?
  • Does it have the same mode of operation in manual and automatic?
  • Does it have the same Mechanical Integrity requirements?
  • Does the P&ID need to be updated?
  • Does the SOP need to be updated?
  • Does it affect the PHA section that this equipment belongs to?
  • …and the list goes on.

While it is technically possible that you could ask these (and 100 other) questions concerning a condenser replacement in such detail that you ensure it satisfies the design specification, you are going to want to document all that work. I’ve personally never seen it happen unless it was the same make and model. In our opinion, the best way to document all that work is by following the Management of Change procedure.

 

Closing Thoughts

Management of Change is a difficult element. But by working this element, you can find and address hazards before they are introduced to your process. There’s very little that can be said about it better than this advice from the Petroleum NEP:

OSHA’s MOC requirement is prospective.

The standard requires that an MOC procedure be completed, regardless of whether any safety and health impacts will actually be realized by the change. The intent is, in part, to have the employer analyze any potential safety and health impacts of a change prior to its implementation. Even if the employer rightly concludes there would be no safety and health impacts related to a change, 1910.119(l)(1) still requires the employer to conduct the MOC procedure.

Composite Fan Blades – A Replacement in Kind?

Imagine we’ve decided to replace the existing fans of an air unit (probably a blast cell) with composite blades. Should you consider that a Replacement in Kind?

Before we go into the change, you might be asking why people would change from a metal fan to a composite. The answer is usually three-fold:

  • Composites allow for some really elaborate / efficient shapes that are difficult to pull off with traditional metal blades. This means that the fan blades can be designed for low noise or to maximize throughput / throw. It’s not uncommon to see a composite blade manufacturer claim a 5% energy savings by using these types of fans.
  • Corrosion Resistance. The commonly used FRP (Fiber reinforced polymer) blades are more resistant to common cleaning chemicals and water than traditional metal fans.
  • Composite fans usually provide a far better weight/strength ratio than even cast aluminum. This means higher efficiency, lower motor wear, etc.

Should you consider the replacement of traditional metal fans (in an air unit for example) as a Replacement in Kind? My thinking on this is the same as the answer to every other “Is it a Replacement in Kind” question: It’s a change that falls under the MOC procedure(s) until you prove that it isn’t.

To meet the lowest possible compliance threshold of OSHA and the EPA we have to know if it “satisfies the design specification.” In this case, that pretty much comes down to asking if it has the same (or better) performance characteristics as the fan it’s replacing.

Assuming you do those checks, and you find the new fan can operate at the same temperatures, static pressure, speed, etc. as the old fan, OSHA and the EPA would allow us to qualify this as a Replacement in Kind. As always, I would suggest you document that work so you can prove you’ve correctly qualified the change as a Replacement in Kind. That’s the minimal compliance answer. The process safety answer is different and it’s the answer you should care about!

For the process safety answer we also ask a few more questions. For example:

  • Does the fan have the same MI requirements?
  • Does the fan have the same failure modes?
  • Are our PHA answers concerning fans (evaporators, condensers, etc.) still the same based on these new fans?

I would like to focus on the PHA question because the others are part of our PHA answer. If you are using the IIAR’s What-If checklist (or something like it) you have a question that looks like this:

EV1.22  What If…a fan fails catastrophically? (air unit cooler only)

It’s likely you answered that for metal fans by saying there were appropriate safeguards to control this hazard such as your MI program, daily walk-through’s, 5yr independent inspections, etc. It’s also likely that if you had those safeguards in place you made “no recommendation” on that question.

Now, let me ask: What if the new fans have different failure modes / MI requirements? Wouldn’t that change your analysis?

Recently I’ve been getting reports of fan failures in the field. While not unheard of with traditional metal fans, these reports lately seem to all be about composite fan blades.

What we seem to be seeing in these composite fan failures is that they appear to be caused by:

  • Turbulent uneven air flow through the coil (usually caused by debris on the coil but could also be caused by partially blocking the airflow out of a coil)
  • The turbulent air flow causes the fan to oscillate, sometimes at a resonant frequency
  • Over time this resonant frequency causes small stress cracks to start on the fan blade at the hub
  • If the condition lasts long enough, the crack propagates, and the fan catastrophically fails

Now, that’s not a failure mode / sequence of events we are used to seeing on traditional metal fan blades. Does that mean we shouldn’t use composite blades? No, of course not – it means Identifying new hazards like this requires us to institute new / altered controls. Some good examples of appropriate changes:

  • LOWER our acceptable threshold for debris on air unit intakes. This could mean more walk-through’s or just more debris removal. We’re not talking about dust buildup here, we’re talking about the removal of large debris: cardboard chunks, labels, etc. The things that are likely to cause turbulent and uneven flow through the unit.
  • LOWER our acceptable threshold for product blocking airflow out of a coil.
  • Specifically look for signs of cracks during annual / 5yr MI inspections.

Making changes to our program like this show us that the change from a traditional fan blades to composite fan blades should not be treated as a Replacement in Kind. It is, however, a fairly simple Management of Change. I want to re-emphasize here, had we chosen to view this change as a Replacement in Kind, we would have missed these opportunities to identify and control these new hazards introduced by the change.

Why does my PSM program make me review the PHA during any equipment/facility MOC?

Nearly two years ago, I changed the Management of Change Written Plan template in my PSM programs to make a few changes:

  • Removed the “minor change” and “major change” categories.
  • Explicitly require a Process Hazard Analysis review during all equipment/facility MOC’s.
  • Explicitly require a Pre-Startup Safety Review (PSSR) before startup for all equipment/facility MOC’s and suggesting that it be conducted by someone other than the person that performed the MOC procedure.

These changes are related and they were made for a few reasons, which broadly break into three categories:

  1. PSM Guidance History
  2. Lessons Learned from Audits, Citations, and Incidents
  3. Plan Simplification

PSM Guidance History: There are two main ongoing issues with the NH3 refrigeration industry’s understanding of MOC. The first is the fiction that there are “minor” and “major” categories of changes in the PSM/RMP rules. It seems that the example written plan provided in the first edition of the IIAR Compliance Guidelines has been so internalized into our industry that people have stopped looking at the actual regulations or guidance documents from OSHA & the EPA. Essentially, the IIAR guidance splits changes into either “minor” or “major” based on whether they invalidate the PHA. If they do invalidate the PHA, they are “major” changes, and if they don’t, they are “minor” changes. This (arguably useful) fiction is NOT present in the PSM/RMP MOC rules.

Here’s the relevant IIAR guidance on “minor” changes:

“When processes undergo minor changes (e.g., minor rerouting of a piping run), information is typically added to a PHA file to reflect the change, even though the validity of the PHA is not affected by the modification.”

Here’s the relevant IIAR guidance on “major” changes:

“A major change is a modification which has significant impact on process conditions or system parameters and was not addressed by a previous Process Hazard Analysis (PHA) study.”

If that’s your policy, then the IIAR is already telling you that the PHA should be reviewed during the change. Why? How could you possibly know if the change “was not addressed by a previous PHA study” or whether the “Validity of the PHA is not affected by the modification” without reviewing the PHA?

Since the IIAR guidance (and experience) tell us we have to review the PHA as part of the change, there is little benefit to the IIAR scheme of “minor” and “major” changes. You can’t justifiably determine which category the change falls into until after you’ve reviewed the PHA. Unfortunately, that’s not what was happening…

Lessons Learned: PSM covered facilities have a LONG history of failing to properly manage changes. You would be stunned to see some of the changes that occur in covered processes without undergoing an adequate MOC procedure – or without any MOC at all. We’ve seen entire machine rooms added with no PHA review because the facility felt that the change was a “minor” change. The reality of implementing the IIARs guidance was that people were deciding if a change was “minor” or “major” based on their gut feeling of the size/complexity of the change – They weren’t actually reviewing the PHA to make this determination. The IIAR scheme acted as a “shortcut” for many facilities as they tried to jam ever larger & more complex changes into the “minor” category.

Furthermore, we found that almost none of these poorly performed MOC procedures were being caught by the facility. Since one person was in charge of (and administering) the entire program, there was essentially NO oversight. Adding the PSSR at the end of the equipment / facility MOC procedure provided a degree of oversight.

Simplification: Removing the “minor” and “major” categories meant that all equipment / facility changes followed the same procedural steps. Each equipment / facility change now requires the “Responsible Person” to review the relevant PHA section(s). Please note that this is not a full-scale team-based exercise – it’s the “Responsible Person” reading through the relevant section(s) and making a determination based on their expertise. Obviously, as in all other cases, if the “Responsible Person” is encouraged to seek any operational or engineering expertise they might lack. If the PHA section(s) appears unaffected by the change, then that’s the end of the issue until the regularly scheduled PHA revalidation. If, however, the PHA section(s) appears to be affected, then they need to be revalidated by a team meeting the requirements of 1910.119(e) & 40CFR§68.75.

Also, added at this time were individual component PSSR sheets. That is, for each common type of equipment in an NH3 refrigeration system, a sheet was created that allowed you to quickly document the Pre-Startup Safety Review. These sheets covered various required PSI items as well as IIAR Bulletin 109 items that have been commonly requested and IIAR Bulletin 110 inspection & maintenance items. Since, this PSSR is required (by the Written Plan) for every equipment / facility change, they provide a oversight function as well as meeting the regulatory requirements.

All of these changes are made to improve the MOC & PSSR process.

To meet the requirements of the MOC element 1910.119(l)(2)(ii) you have to “assure that the following considerations are addressed prior to the change: …Impact of change on safety and health.” If you haven’t read this OSHA MOC guidance recently, I highly commend it to you:

An MOC procedure is required anytime a change per the requirements of 1910.119(l) is considered. An MOC procedure is a proactive management system tool used in part to determine if a change might result in safety and health impacts. OSHA’s MOC requirement is prospective.

The standard requires that an MOC procedure be completed, regardless of whether any safety and health impacts will actually be realized by the change. The intent is, in part, to have the employer analyze any potential safety and health impacts of a change prior to its implementation. Even if the employer rightly concludes there would be no safety and health impacts related to a change, 1910.119(l)(1) still requires the employer to conduct the MOC procedure.

The MOC requirements are important because many large incidents have occurred in the past when changes have been made and the employer either did not consider the safety and health impacts of the change, or did not appreciate (wrongly concluded) the potential consequences of the change before it was too late. Therefore, it is not only required, but important that the employer conducts an MOC procedure on each change, even those changes the employer believes will have no safety or health impacts. (OSHA, Refinery PSM NEP, 2007)

It is the word prospective in the above text that is most important. You can think of the dictionary-like definition of “a study that starts with the current condition and follows it into the future.” You could also think of those men and women who panned for gold in the western territories of the US during the 19th century.  Like them, you are looking through a lot of mundane things to find the rare oddity – that unique nugget – that can cause trouble.  In an MOC the nugget is a hazard that needs a safeguard – either a new one, or one you already have but haven’t yet applied to the discovered hazard.

Where do you document the hazards you’ve found and the safeguards you’ve put in place? The Process Hazard Analysis!