Friday, July 17, 2015

Case Studies on Some Industrial Accidents: Lessons to be Learnt by Practising Engineers: Part-1

Case Study-1
A Gas Leak Accident that Killed Seven!

This incident happened in a large integrated steel plant in India. In integrated steel plants (ISP) world over, iron ore is first converted to molten iron called hot metal in a chemical process reaction vessel commonly known as the Blast Furnace (BF). ISPs normally have many BFs for producing hot metal all normally working in parallel so that they achieve their targetted production levels. Hot metal production in an ISP is typically to the tune of a few thousands of tonnes per day or a couple of million tonnes per year.

Iron ore is a natural mineral containing oxides of iron and other impurities such as oxides of silicon, manganese, magnesium, calcium, etc. It is nothing but commonly available red soil having the iron content a bit higher for economical extraction of iron. For separating the iron, the iron oxide needs to be melted and reacted with the reducing gas, carbon monoxide freshly produced by the action of oxygen in air with a carbon rich raw material such as carbonized coal commonly known as coke. All these reactions continuously take place in a vertical semi-fluidized bed reaction vessel called the Blast Furnace (For more about the Blast Furnace process of iron making see this wiki article).

Waste process gases emanate from the top of the Blast Furnace, called the BF Gas, which is nothing but the excess air carrying much of the gaseous reaction products such as Carbon monoxide, Carbon di oxide, Hydrogen, etc together with the dust generated during the vigorous exothermal reaction in the furnace. This gas is very hot and contains useful chemical energy due to the presence of Carbon monoxide and Hydrogen. Conventionally, in the ISPs this gas is further cooled and removed of the dust in gas cleaning plants (GCP) for use as a low calorie fuel gas elsewhere in the steel plant. This BF gas is highly toxic because of the presence of Carbon monoxide.

The dust laden BF gas that comes out of a BF is cooled and cleaned in a GCP by spraying water over the gas in closed scrubber vessels. The water used for cleaning the gas entraps the dust present in the gas and becomes a slurry which is then treated in a wastewater treatment plant (WWTP) system. In the WWTP the waste water (effluent lean slurry) is settled for removal of suspended solids, cooled and the clarified water is recycled back for the scrubbing process in the GCP using pumps installed in a pump house.

Blast Furnaces are continuously operating chemical reactors intented for extractive metallurgy. Once it is built and commissioned, it is fully stopped and taken out of production service only for capital repairs. Such capital repairs are done after several years of continuous production. However, minor hot repairs are also carried out periodically by keeping the furnace hot, but without the full fledged production. All repairs are planned with the least shutdown periods, to keep the production losses to the minimum.

A working BF containes several hundreds of molten metal and semi finished raw materials inside its vertical shaft and its bottom belly. This is kept hot and reactive with the action of huge volumes of hot air blast that is is blown inside through the huge bed of raw material stock contained in the furnace shaft through nozzles called tuyeres. The air blast is the most vital process of the blast furnace that is accomplished through large air blowing machines. 

Besides, large quantities of cooling water is continuously required to cool the BF shell. Any failure to supply the air blast or the cooling water can cause serious damage to the BF. If air blast is stopped for some time there is the danger of the molten iron and molten slag getting solidified in the furnance that could cause irrepairable damages. Similarly, stoppage of cooling water can cause the BF shell made of steel plates to melt down. 

ISP operators and management consider BFs as the most critical production facility and ensure all they can to keep the BFs in operation through out is operating life till it is taken out of service for any planned repair period.

A BF plant is a complex conglomerate of many systems, equipment and components, operated and maintained by managers, metallurgists,engineers and technicians who are teamed up in various functional departments and sections. In their anxiety to keep their own systems under top priority, these teams often forget to understand the role of the other supporting  systems and the teams that operate and maintain them. 

Any maintenance to the BF system needs some time of some shut downs to some of the working equipment and systems which may lead to some production loss. Hence, production oriented managements usually keep building the pressure on the various teams to have their maintenance activities completed by taking the least possible time. 

Such pressures often cause some engineers to postpone some maintenance functions which they deem as non-vital to some later maintenance time, only to keep them postponed further in a similar way. Such management pressures also cause some engineers to neglect or by-pass some systems which they consider as irrelevant. Lack of proper understanding of the plant systems by new engineers and teams enhances such risks. This is because the old ISPs have  production facilities with operating life exceeding several decades.

The ISP where the accident occured have been operating six BFs in parallel with many common facilities such as the pump house, the WWTP, the Air Blower Station, etc. There were independent GCP facilities for each of the six BFs, but the waste water treatment facility and the pump house that circulated the water were common to all. There existed one group of pumps that recycled clarified water from the WWTP to the individual GCPs through a common header pipeline. These pumps were located in a common pump house which was located sufficiently away from the BFs and their GCPs. The pumps were located in the dry-well of the pump house which was a few meters below the general ground level.

There were three pumps in this group and normally two were kept in operation and one was kept as a standby. Each of the pump discharge pipes connected to a common collector pipe kept at the pump floor, through their own isolation valves and non-return valves. A common pipeline connected to the common collector pipe distributed the clarified water to all the six GCP scrubbers. Individual branch pipelines supplied water to the scrubbers where the water used to be sprayed to the hot BF gases for cooling and cleaning through various kinds of nozzle systems. The piping system had several motor operated and manual valves for controlling and isolating the water supplies to the GCP systems as per requirement.

The GCP scrubbers were the process equipment where the BF gas and the water interacted. As the water had to spray over the gas, the water pressure was kept much higher than the gas pressure.

Several departments worked to keep the BF production and maintenance operations to go on uninterrupted. There were the BF department with their own sections responsible for the mechanical maintenance, electrical maintenance, instrumentation maintenance and other supporting departments such a the Energy Management Department responsible for the GCPs and the Water Management Department responsible for the pump house, the WWTP and the main water supply pipelines. The staff and engineers of all departments worked in shifts on a 24x7 basis.

Some operating personnel of the Water Management Department had been aware of a minor leak in the common collector pipe located in the main pump house to which the three GCP supply pumps were connected, for some time. The leak had happened in the lower side of this fairly large pipe with a diameter of about 900 mm. This pipe piece was supported on the floor of the pump house and there was hardly any space between the pipe bottom side and the floor. This restrained the operators and the shift managers to assess the real condition of the pipe. While the pipe was quite okay on the top side, there was some corrosion at the bottom side and there was some pin-holes from which water had been leaking.

In the last several decades, this pipe piece was never replaced during any maintenance and the concerned department knew about its old age and probable deterioration due to ageing and corrosion. The pipe bottom was specially vulnerable to external corrosion as this side was difficult to be painted due to the small working clearance. Besides, this side was exposed to water accumulations in the pump house floor that happened occassionally due to various reasons. This also was one of the reasons for the probable external bottom side corrosion of the pipe.

For replacing this pipe piece, the Water Management Department needed to take a shut down to the GCP water supply main header pipe which invariably affected the production from six BFs. The production managers over the years had developed a type of technical  superiority in all decision making due to which it was difficult for the Water Management Department to convince the top authorities about the necessity for a production shut down for replacing a small length of old water pipe. The engineers of the Water Management Department had developed a submissive attitude over the years which prevented them to talk authoritatively about their technical requirements and convince the top authorities of the steel plant. They had almost developed a withdrawn attitude towards their professional vigilance  to  critically study the various kinds of risks involved in their functions!

The inter departmental meetings, for quite some time, had been stressed towards achieving production targets and better production performances with attention towards removing bottlenecks. Perspective assessments and critical analysis of risk scenarios had undergone a decline.

This caused the top officials of the Water Management Department taking temporary actions for rectifying the leak, instead of reporting the issue to higher management authorities with a proper risk assessment study to convince the authorities for an immediate shut down of the BF-GCP systems for taking up the work of replacing the leaking old water pipe. The temporary leak plugging acts done by the department by welding etc further aggravated the condition of the leaking pipe.

And within another couple of days, in one shift, all of a sudden the pipe ruptured. Water splashed out from the rupture and the GCP water supply pressure dropped. The shift personnel of the water management department informed their seniors and the matter got communicated to all the other concerned engineers and managers of  the Energy Management Department and the BF Department. The heavy leakage of water from the ruptured pipe was now flooding the pump house and the pumps had to be stopped to prevent flooding. Those concerned officers, engineers and operating people of the steel plant rushed to the pump house to physically assess the rupture in the collector pipe.

But without them ever realizing, a major tragedy was about to happen shortly. As the water pressure in the GCP water supply pipe decreased due to the rupture and the subsequent pump stoppage, the water in the pipe line began to drain out through the rupture as the rupture point was the lowest point. Soon the water in the supply main got emptied and the toxic BF gas began to fill the water pipe from the scrubbers and began to leak out through the pipe rupture in the pump house.

Carbon monoxide is an odorless gas and a swift acting toxin. The pump house is a closed enclosure and the toxic gas was getting mixed with the air in the pump house. This posed a two fold danger. First the pump house became a toxic gas chamber. Any one who inhaled now got affected by carbon monoxide with imminent lose of consiousness and possible death. Secondly, the pump house was getting an explosive mixture of fuel gas. Certain low concentrations of monoxide gas with air is highly exlosive and it could explode like a bomb causing major damage with a minor spark any where!

The people who rushed to the pump house now was getting affected by the toxic gas and they were simply falling down unconscious. It took others who followed those went earlier down to the pump house some time to realize what was happening. And when they realized it, a number of persons were already affected. Fortunately, the rest of the engineers realized what was happening and they could take immediate preventive measure that could save the pump house from exploding!

Several people got affected and about seven lost their lives due to gas poisoning. It took the steel plant a few days to restore normal production! What production they obtained earlier by post poning a planned shut down all wiped out now due to this accident.Perhaps the loss was much more including precious lives of people!

A series of enquiries followed. Some by the statutory authorities, some by techincal empowered groups of the company. All enquiry committees and commissions brought out their own findings and suggestions. While the technical cause was pin-pointed, the committees apparently failed to recognize the root cause of the problem- the cumulative effect of  technical management errors that had been happening for a long period!

I will now discuss some of these errors that keep happening in Indian process industries which most of the Indian engineers assigned with the task of factory management routinely seem to ignore or do not consider as some thing important to be addressed properly:

1. Indian engineers are not very much comfortable with the concept of unit wise operation of multiple process units and systems. If there are more than one process equipment or system for achieving higher production, the Indian engineers quite often force the plant suppliers and designers to interconnect those systems for higher flexibility of operation and maintenance. For example, it was much safer in the above mentioned case to operate and maintain each of the BF systems as  self contained individual systems instead of an interconnected system of six BFs. The interconnections enhance system complexity for engineers to understand and comprehend causing new comers be confused about all the functions of the plant systems. This causes neglect of some parts of equipment and systems, especially those that are to be operated only during exigencies! Whenever, a plant accident or breakdown take place, the subsequent actions would be to have more interconnections and more stand by units which enhance the system complexity rather than reducing it in actual practice. 

2. Indians quite often are not very comfortable with record keeping. They seldom make efforts to keep all technical drawings concerned with the plant process and systems. Even if such drawings and documents are available, hardly any effort is made to ensure that all concerned engineers study those drawings and documents carefully and comprehend those fully.

3. Design engineers and plant operation and maintenance engineers work in isolated pockets. They are seldom encouraged to understand the philosophies and technicalities involved in each others' area of work. This often causes erroneous assumptions and subsequent errors. For example, in the case discussed above, the original design engineers had provided certain provisions of water sealing and siphoning arresters in the pipelines that prevented any gas to escape through the water lines when water got emptied in the pipelines under rare situations as had happened. But many such systems got removed or erroneously fixed in later years as the concerned maintenance engineers did not understand their functions or principles of operations.

4. The steel plant management had adopted a system of inter departmental transfers of engineers without ensuring availability of experience and competence of engineers and technicians with regard to their own area of work. This had caused experience vacuums in many areas with the teams handicapped with understanding of technical issues and technical communications. In some cases, the junior engineer felt difficulty to communicate a problem to his senior as the senior had less experience in the field and vice versa. In the case discussed, such a situation had caused the concerned people to foresee the problem in advance. Had there been the required experience and competence at all levels, this problem would have never got escalated to a major mishap like what had happened.

5. Of late, the Indian managements of technical facilities are more and more becoming non technical in attitude and nature with the top managements losing their capacities to understand the complexities involved in technical works. As finance, marketting and other non technical fields are becoming more and more important at the board levels of decision making, the voices of technocrats in companies are getting diluted and they no more get any chance to make the non technical board members to understand the technical problems. This causes the top technocrats to either discard their technical thinking and behave like non technical persons. They in turn discourage technical issues to be discussed at top levels. Real technical and engineering problems now get reduced to lower level issues. This seriously enhances the technical risk levels of Indian process plants. This attitude then percolates downwards and adversely affect the technical thought processes of all engineers working in Indian factories and process plants. Indian engineers working in processing facilities of companies now seldom get any reward for proactive or creative professional thinking. On the other hand, erroneous technical conclusions and erroneous punishments to engineers are on the rise now in India making professional engineering and technical work something not preferred by brilliant engineers.

I hope eminent engineers from India would open up their minds to express their thoughts and comments on the above.

1 comment:

Your comments are welcome. Express your opinions publicly, but responsibly. Inappropriate comments promoting vulgarity or hate or spam will likely get deleted when noticed. As a step to promote responsible comments, commenting is restricted to those readers registered as members of this blog site with effect from 25th October 2018