Find the Leak - detect where air is entering the smelter offgas train

Lead Mentor: Jen Parker, Process Engineer

Challenge Description

The Nickel West Smelter (NKS) has a gas cleaning section which removes dust and SO2 from the offgas stream of the furnace before releasing it to the atmosphere. This gas is generated through the flash smelting of sulphide containing concentrates. The faster the feed rate on the furnace, the larger the offgas volume that is generated. The gas cleaning system also intermittently handles the water quench (WQ) gas which is an additional ~45, 000 Nm3/h.

The offgas system is kept under suction to prevent the release of rogue emissions. This means that any hole in the offgas system will suck in air which adds to the total volume of gas which needs to be processed by the offgas system.

Air ingress is an issue for a number of reasons:

• The maximum gas that can be processed by the offgas train is 190, 000 Nm3/h. If this limit is reached the gas volume is reduced by either reducing feed rate to the furnace or stopping water quench of the converters.

• Air ingress drops the temperature of the gas below the acid dew point (250°C) which promotes corrosion of the ductwork and equipment.

• Higher power consumption to process the additional volumes of gas.

User Story

Historically air ingress has been hard to detect because a significant portion of the ductwork in inaccessible without additional equipment or without being isolated. There is also a significant amount of insulation and cladding on parts of the equipment which makes it hard to find points of air ingress as well as to narrow down exactly where it is coming from.

The best way found so far to detect which part of the system has air ingress has been through oxygen surveys performed at varying points along the offgas train. An increase in oxygen concentration generally indicates a point of air ingress. Online oxygen monitoring has been tried however the ports fill up quickly with dust (in less than two weeks) which renders them useless.

Other methods which have been tried have been:

• General walk around to see/hear points of ingress

• Smoke machines/flares to provide a visual indication of air ingress

• Data analysis of temperature and pressure sensors

• High frequency sound detectors.

These methods have worked with varying success but often the air ingress is found through having a person in the right place or the hole has deteriorated to the point that it is very clear where it is. All of the solutions found to date are extremely time consuming to complete.

Impact and Value

Air ingress is estimated to cost $3.67M per year. This is $2.7M from reduced throughput and $970k from increased energy costs. This does not include the increased maintenance costs. So the faster any point of air ingress can be found, the less of an impact it will have on performance.

Additional Questions

• Could the pressure/temperature indication be used to narrow down the areas where air ingress could be located?

• Is there an opportunity to install additional/other sensors to help identify air ingress?