Process Safety Time

  • Thread starter Edward M. Marszal, P.E.
  • Start date

Thread Starter

Edward M. Marszal, P.E.

I received a request for information about process safety time. I thought that some members of the list might be interested.

-----Original Message-----
From: Edward M. Marszal, P.E. [mailto:[email protected]]
Sent: Wednesday, January 03, 1990 11:32 PM
To: ####
Subject: RE: Request for advice

Dear Mr. ###,

I have received your request for information regarding the clarification of safety terms. I would be interested in knowing to which standard you are referring when you inquired about these
terms. The context of your questions may result in a slightly different answer.

I will give you a discussion of the terms as I employ them in my design and consulting, without specific regard to literal definitions as defined an any of the myriad of standards, books, and articles on the topic.

"process safety time" is the amount of time that is available to take action against the process to move it to a safe state after an out-of-control condition has occurred in the process. The key to this term is that it is completely PROCESS related, the SIS used has no bearing on the value. The "process safety time" can vary greatly between processes. For design purposes, I would calculate the "process safety time" as the time interval between the annunciation of an alarm identifying a particular out of control condition, and the occurrence of the unwanted event.

I will give you two examples. When a the flow rate through a compressor becomes very low, material can back through the compressor causing it to 'surge'. In this case, the "process safety time" is the time interval between when a low flow alarm is received and when the compressor actually surges. For this example, the "process safety time" is measured as a few milliseconds.

A second example involves a fired process heater. If the skin temperature of the heater tubes goes high the tubes can rupture causing the process fluid to be released into the heater firebox, potentially causing fire and explosion. In this example, the "process safety time" is the time interval between when a high tube skin temperature alarm and the point in time when the tube ruptures. In the worst case this time period will be minutes, in the best case, this time interval could be weeks or months.

"Process safety time" is important because it determines the maximum amount of time that can be allowed for the SIS to perform ALL of the
following actions.

1. Sense the out-of-control condition
2. Digital filtering of input signal
3. Input process scan time
4. PLC program scan time
5. Any trip delay timers set to remove process noise must time out
6. Output process scan time
7. Digital filtering of output signal
8. Fully actuate the output device (some valves can require up to 30 seconds to fully stroke!)

Depending on the hardware that is used to perform the shutdown, more or less actions than described in the list above will have to be performed.

The term "maximum controller output time" refers to the time period required for the instrumented system to perform. That time interval should consider items similar to the 8 items that I have listed above. This time interval is strictly dependent on the instrumentation that is used to perform the shutdown, and is mostly independent of the process that is being controlled.

A key rule of thumb when selecting controller scan times and input signal delays is:

"maximum controller output time" <= 1/2 ("process safety time")

I hope this reply completely answers your questions on the topic. If you have any other questions or need further clarification, do not
hesitate to contact me.

PS offers a monthly electronic newsletter which contains topics of interest to engineers involved with safety instrumented system design. I will have your e-mail address added to our list. If you wish to expedite the process, send an email to [email protected] with subscribe in the subject line. Perhaps future questions that have might be answered in our newsletters or
by the information contained in our web site.

Edward M. Marszal, P.E.
Principal Engineer,
Ph: 614-226-4263
Fax: 614-459-9764
e-mail: [email protected]

-----Original Message-----
From: ####
Sent: Thursday, June 15, 2000 8:29 PM
To: [email protected]
Subject: Request for advice

Date: 15th June, 2000

Dear Sir,

We appreciate the services provided by you by replying to many questions related to safety in List Management.

I am an instrumentation engineer. Recently I came across the terminology " process safety time" . In this regard I request you to kindly clarify the following:

1) What is the definition of "process safety time" ? What is its significance on the design & performance of safety system.
2) What is the definition of "maximum controller output time". What is its significance on the design & performance of safety system.

I appreciate your early response.

With kind regards,


Yesenia Khair

Please, could we conclude that the Process Safety Time is mainly the output process scan time?

Thanks in advance,
>Please, could we conclude that the Process Safety Time is
>mainly the output process scan time?

In accordance to the above question, i do have a question related Process safety Time. one of my client is asking me to calculate the Process safety time. Can we calculate the Process safety time?
"process safety time" is somewhat of a misnomer,

1. process response time or time constants
2. reaction rate and time,
3. measurement delays
4. control response delays or time constants,
5. safety threat response when dealing with toxic materials where the threat to life depends on the concentration and the given local environment

the list goes on and depends on the particular plant and the various systems and people involved...