January 27, 2012

Chlorine tonner incident

Thanks to Mr Harbhajan Singh Seghal for sending this incident:
In one of the chlorine consuming industry two persons were affected with chlorine while the operator disconnected the tonner from the process due to hard valve operation of the tonner.
· The consumer withdraws liquid chlorine from the chlorine tonner and consumes gas after evaporation.
· As per practice, the consumer keeps the tonner in line to withdraw maximum chlorine from the tonner.
· On the specific day of the incident, the operator tried to isolate the tonner at 1.0kg/cm2 pressure when about 40-50 kgs liquid chlorine was left in the tonner and there was ice formation at the bottom of the tonner.
· He could not close the valve fully. The spindle of the valve damaged due to excessive force.
· The operator decided to cut off the tonner by wearing SCBA.
· This action resulted in heavy gas leakage and affected two persons in the surrounding area.
· The tonner brought to the works.(of the chlorine supplier)
· It was depressurized and the valve was dismantled
· Iron chloride rust and greenish color sludge was observed in the threading of the valve.
· Damaged valve replaced with new valve.
· Liquid chlorine withdrawal rate is 180 kgs/hr. Maximum liquid chlorine is used up in 4-5 hrs operation at this rate. Some quantity (40-50) kgs remains in the tonner at the bottom.
· Some consumers try to recover this 40-50 kg liquid chlorine as gas by keeping the tonner in line for more time.
· The left over liquid chlorine evaporates at 5-6 kgs/hr as gas and lowers the temperature of the tonner/pipe lines due to fall in pressure till the remaining liquid chlorine is exhausted.
· 1.0 kg/cm2 pressure can lead to about -20 degree C and takes 8-10 hrs to completely empty the tonner.
· Normally chlorine in the tonner is dry. But under such conditions (-200) the same chlorine becomes wet.
· Water in the chlorine separates out and freezes in the spindle of the chlorine valve. It makes the valve hard to operate at that time.
· After attaining normal temperature, chlorine evaporates first and water later. This chlorinated water reacts with the sprindle and makes the chlorine spindle greenish.
· The evaporators which do not have backflow prevention system (from evaporator to chlorine tonner) results carry over of iron chloride rust to valve spindle and makes the valve hard in operation.
· The tonners containing some quantity of liquid chlorine (40-50 Kg) are not to be cut off at 1.0kg/cm2. The tonner is to be depressurized by releasing the chlorine to neutralization system through header or evaporator.
· After depressurizing, check that no chlorine gas comes from the upper valve of the tonner, and also check that no ice formation appears on tonner or pipe lines before the tonner is disconnected.
· Chlorine header and evaporator must have chlorine release facility connected to neutralization system.
· Chlorine evaporator must have liquid chlorine flow control interlocked with temperature and outlet pressure to avoid the back flow of chlorine
· Evaporator should have emergency release system with rupture disc and safety valve.
· Temperature of evaporator should be maintained between 80-85 degree C to avoid formation of rust as Fecl3 in the evaporator.
· Dry air (-40 degree) dew point is to be utilized for evaporator maintenance.
· Glass wool filter is to be utilized in gas line to avoid carry over of Fecl3 to main products and choking in chlorine system.
· Tonners can be kept in tilted position forming 20-30 degree angle to withdraw maximum liquid chlorine from the tonner.
· No chance should be taken with liquid chlorine system. Help of the filler (chlorine supplier) must be taken in such cases. One volume of liquid chlorine expands to 460 times
· Chlorine neutralization system must be effective and checked from time to time.
· Single person should never take this type of emergency job.
· Always stand by person ready with safety equipments should be present during such operations.

January 24, 2012

Process safety - Seeing and managing

A typical day for today's plant manager is like this: Punch in....login......read emails and answer......collect data for the meetings scheduled......firefight today's issues.......go back home late in the evening! I was just comparing the daily routine i used to do many years ago in the same position: Punch in....go around the plant for at least one hour.......read the log book.....write relevant instructions in the instruction book.....attend the daily plant meeting for discussing and resolving issues....discuss and take instructions from my boss....communicate these instructions to the plant......go back home peacefully, on time! Note: There were no ISO9000,ISO14000,OHSAS18001,PSM,TQM,Six sigma etc in those days! Managing process safety needs management by seeing, hearing and understanding. Unfortunately today's plant managers do not have the time to see the plant.....this is a dangerous trend. Also competency for managing process safety is lacking. Somewhere we seem to have lost our way!

January 22, 2012

Awards and Accidents

Further to the refinery accident during hot work which I had mentioned in my previous blog entry, an article in the Hindu lists out other accidents that occurred in the same refinery in 2009. It also mentions that the refinery won safety awards in 2009 and 2010. There is a disconnect here! Read the article in this link.
While awards are a good way to motivate people, the onus lies on management to sustain and improve process safety performance.Long ago, I had audited an organization that had been granted a prestigious award by an international organization for their safety management system. I visited the plant 6 months after this award. Their safety management was in shambles and I had mentioned to them that the sword is now hanging over your heads.
If everyone goes back home safely everyday and this is maintained, then your process safety management system is working well! Period.

Worker killed in explosion during hot work

An accident at a SRU at a refinery has killed a worker. Read the article in this link. Please see my earlier posts under safe work practices. Hot work should be done with proper precautions and it is sad that repeated accidents are occuring. 

January 20, 2012

Process Safety and Risk management in the high speed age

A good article called "Black swans turn grey -The transformation of risk" by Price Waterhouse Coopers highlights the following:
  1. The boards of big organisations do not fully understand the risks that they are running
  2. In the Internet age, speed and prejudice are all
  3. Checks and balances at board level are critical.
  4. Leadership and culture shape an organisation’s attitude to risk.
I think in a chemical industry "operational risk" is of greater importance that financial and strategic risks. As long as there are human beings involved in making decisions, there is a possibility of a process safety incident that has serious implications for the business. That's why I agree totally with the Baker Panel report suggesting that a person competent to understand process safety be on the board of Directors for Chemical Industries.
Read the article by PWC in this link.

January 19, 2012

Oil rig catches fire

An oil rig in Nigeria has caught fire and it is reported that two workers were missing and two are hospitalised. Reason for the fire is not known. Read about the fire in this link.

January 17, 2012

Learning from Buncefield

The Buncefield incident in 2005 was a wake up call for the industry. A lot of assumptions that were in vogue till then in QRA were overturned. Henry Troth has made a good presentation of the incident mentioning the following:
  • "Take a critical look at your Safeguards, your Prevention and Mitigation Layers – they may not be as effective as you need
  • Tanks should have overfill lines from HHH down to ground level to reduce splashing and vaporizing overflowing fuelTank 
  • Overfill Protection should be SIL rated and proven in use
  • Retrofit water curtains on closely spaced tanks
  • Fire Pump House should not be a source of ignition (classified area)
  • Store portable fire fighting equipment and foam in a ‘safe’ place -stationary equipment usually gets knocked out.
  • Remember – you must keep all Safeguards working as well as the SIS layer(s) – otherwise you are exposed
  • Is a spill all you need to worry about (what could possibly go wrong)?
  • Consider consequences carefully – What will you do if the unthinkable happens?"
Download the presentation from this link.

January 15, 2012

Investigating process incidents

"Aerodynamically, the bumble bee shouldn't be able to fly, but the bumble bee doesn't know it so it goes on flying anyway" - Mary Kay Ash, American Businesswoman
I like this saying as it is very relevant to process incident investigation. In the course of investigating many process incidents, I have come to the conclusion that you need to be like the bumblebee (keep your mind open, and avoid jumping to conclusion!) while investigating incidents. Many chemical process incidents may apparently not reveal the root causes immediately. I have used the event and causal factor analysis/barrier analysis and Man-Technology-Organization analysis to determine the root causes of many chemical incidents. Also, listen to the people who were present during the incident and observe the incident site. Equipment tell silent tales.

January 7, 2012

The hazards of thermal expansion

Many plant personnel do not give importance to thermal expansion safety valves. These are typically small valves and often, the isolation valves are kept closed as they pass. It is human tendency to ignore small things but in process safety it is the small things that cause big disasters!
The CSB had published a good case study on a fatality that occurred due to the bursting of a heat exchanger due to thermal expansion of trapped ammonia. It is worth reading and sharing with all your colleagues. Read it in this link.
The photo at the left is courtesy of the CSB.
Read another incident in this post. 
Read a third incident in this post

January 5, 2012

Cyber security for chemical plants

Dr Trevor Kletz has said "what you don't have cannot leak". This was with reference to hazardous chemicals. It makes sense then and makes sense now. But with the advent of Internet and remote operation of chemical plants, we often think that the best way to keep a plants intranet secure is not to connect it with the Internet. In other words, "when you are not connected , you cannot be hacked". Last year, the cyber attack on a nuclear development facility in Iran, caused centrifuges to speed up but the control room did not display the increased speed! It is a very interesting story and just imagine what could happen if a cyber take over of critical equipment in a chemical plant takes place? Ensure your systems are adequately protected and conduct cyber security drills along with your regular mock drills! Read the very interesting article about the centrifuge incident in this link.

January 4, 2012

Process Safety - Keep it simple

I often think that today we are complicating things too much in process safety, in an already complicated World! This generation of plant operators have been inundated with technology. While some of the technology is excellent, not all of them really help the plant operator. Information overload is the bane of today's PSM programs. When I was a shift in charge at an ammonia plant, we had pneumatic control system (no DCS), but it was so user friendly. I would sit in the center of the control room and at a glance I got to know the plant status. The control room had three operators - one for the front end of the plant, one for the back end and one senior guy looking after both. During emergencies, the senior guy would coordinate the actions very swiftly as he could see the complete plant status just at a glance. The whole ammonia plant had only about 200 alarms that were located on the panel. The critical ones were painted red. By experience we would know which alarm meant what! We never had a serious process safety incident!!I managed to get some pictures to illustrate what I am talking about. 
Our control room looked similar to the one at the left. The table at the center was a flat table. The shift engineer used to sit at this table.The control panel is just below the clock.The flowsheet of the entire plant (called a mimic) was depicted at the panel top

This was a field controller.  See how simple it is! The red arrow is the setpoint.

 This temperature recorder was similar to the one we used to record secondary reformer top temperature and methanator temperature.

The message I am trying to convey is try to keep it as simple as possible. Buy only what you want and not what you get! (This is especially true for DCS and electronic instrumentation)

January 2, 2012

PSM and PDCA cycle

The PDCA cycle is the core of any management system. I have seen a common trend from investigating root causes of many chemical plant process incidents - it appears that many of the root causes are due to following the PDCA cycle in a different way: DCA,no P -DO, CHECK, ACT and No PLAN! Fire fighting efforts take place to immediately attack an issue. Productivity, cost cutting and efficiency improvements often take place without undergoing the management of change process. This is why I feel that while certifications to OHSAS 18001, ISO 14001, Responsible Care etc are good, the sustainability of such certifications become questionable. This is evidenced by fatal accidents in companies that are certified. How do we solve this issue? I believe that the onus lies with the top management. If top management are clear about process safety and its implications, then you do not need any system at all! Unfortunately, there is a huge knowledge deficit about the technical aspects of running a chemical plant and the importance of PSM. Recently I had implemented PSM in a pesticide manufacturing unit where the head of the unit was a hardcore experienced chemical engineer. His understanding of the technical nitty gritties of PSM helped me greatly during the implementation. I will end my ranting by repeating two things that Dr Trevor Kletz has said and is highly relevant to Process safety management: 1.Walk the talk! 2.Keep it simple!

January 1, 2012


To all my subscribers and readers,
Wish you and your family a very happy New Year!