Running an Electric Arc Furnace (EAF) is a very complicated process. Large production volumes and extreme energy needs will definitely cause some challenges.  Especially when raw material used in the process has a lot of unknown variabilities and the visibility to the process is limited. When is the optimal time to start different phases of the process? How to respond to excessive equipment wear? And so on.

In the future, these challenges will be even more significant. The global availability of scrap can not match the need of all EAFs and new materials will be taken into use. The use of biochar, worse scrap qualities, and new direct reduced raw materials will increase the variance in the process. Unknown process inputs combined with limited visibility are a challenging combination.

This article discusses some of those challenges and examples of what could be done to lessen them. And what kind of solutions there are for these problems now and in near future.

There are always some levels of uncertainty when operating an EAF process. Especially when timing the different phases of the heat. This is mostly caused by the lack of methods to see into the furnace during the process. Delayed process phases cause a lot of lost potential production. Having power on in the furnace longer than necessary will consume large amounts of unnecessary energy.

Measuring the online melting progression would help. However, the use of conventional measurement methods is very restricted. The extreme conditions inside the furnace make it difficult to use measurement technologies inside of the furnace. Standard cameras or other equipment cannot withstand the heat and slag spray.

Experienced operators might be able to know when to start the different phases of the process based on the sound of the furnace or visual signs. For example, the operator might look at how much light the furnace is emitting outwards.

However, these methods are not very reliable and are dependent on the operator’s experience and knowledge. It also requires extraordinary motivation to carefully follow the process continuously 24/7.

Fortunately, this uncertainty can be countered with emerging technologies. These new technologies enable measuring the melting progression of the scrap in real-time. Currently, only Optical Emission Spectroscopy (OES) has been able to produce this data in:

• Real-time
• Reliably
• Directly from the melting phenomena

With OES significant energy efficiency & productivity improvements can be achieved. With OES timing of different process phases is quite straightforward. Different process phases emit different types of spectrum and intensity of light. This information can be then used to time the process accordingly. For example, it can be clearly seen from the light data when the first scrap load has melted and it is time to load an additional scrap basket.

The wear and tear of the equipment cause expenses but also a lot of downtime for the furnace and lost production. The extreme heat and molten steel consume the refractories and but also other furnace equipment.

In addition to the refractory and equipment wear also electrodes are consumed during the process. Electrode’s break, sublimate and are lost to oxidation. Electrode consumption forms a large part of the costs of EAF steelmaking. Also, electrode prices have been very volatile during recent years. Therefore, it is important to realize this and understand how this could be reduced.

The uncertainty during the process also accelerates this electrode consumption.

Electrode consumption could be lowered with a higher arc voltage which decreases the arc current. This keeps the amount of melting energy at the same level without restricting productivity but increases the length of the arc. A long arc is not a problem as long as there is unmolten scrap to protect the sidewalls of the furnace (or foaming slag/high enough slag layer). Arc hitting the sidewalls causes even quicker wear of the equipment. But if we would have information about the scrap melting progression near the side walls this would not be a problem.

There has been a significant reduction in the wear of electrodes and other equipment running with higher voltage. However, this requires a way to reduce the voltage when the scrap near the side walls has been melted. One method could be to use OES to track the melting progression of this scrap near the side walls.

Learn more about reducing the electrode wear from this article!

After a scrap load has been placed to the furnace, stars the electrode bore-in phase. Sometimes when starting the bore-in the lowering electrodes might face some resistance.

For example, a large piece of scrap metal blocking the path down might cause the electrode to be damaged. This might cause even breakage of the electrode or damage the roof of the furnace. In either case, this would of course cause excessive costs and delays in the process.

One way to ease this challenge is to monitor how well the electrode is boring down to the scrap below the electrode. This could be done for example monitoring the light from the furnace during the bore down phase. Electrode hitting a large piece of scrap produces a lot of light which can be detected easily for example with the ArcSpec system.

For the first time that the furnace is used after relining, the volume of the furnace is precisely known. But how about when the furnace has been used for a while and the refractories have worn due to the abrasion. And when there are material build-ups on the walls and the bottom of the furnace.

In addition, there are challenges in precisely measuring the volume of the scrap load in the basket being loaded to the furnace.

The unknown volume of the furnace and the scrap being loaded can cause some inefficiencies. The furnace is loaded with less than maximum capacity or the furnace might be in some cases overloaded which causes its own problems.

This variation in the scrap loaded in the furnace has also of course an effect on the length of the optimal heat. If the load is a bit heavier the melting process takes a bit longer. And vice versa.

Recent developments in 3D camera systems have reached a state where they can be used to estimate the scrap volume in the baskets. The method is not perfect and gives information only about the scrap load and not the furnace itself. However, having data about the scrap load volumes would be useful to optimize scrap loading practices.

3D camera system to estimate the volume of the scrap loads is also something that we at Luxmet are currently looking at. It could be a very valuable tool with other specific information received from our ArcSpec system. Stay tuned for news on this front!

EAFs typically use recycled scrap. This means that there is a lot of variation between scrap loads which again causes a variation in the steel batch being melt in the furnace.

The usual way of measuring the composition is by taking a sample from the batch. This of course is a good way to get information about the quality and content of the batch, but it slows the process and gives only one data point.

The ideal way would be to measure the process continuously. So that chemical changes in the process could be identified as soon they happen. This would enable improvements in the efficiency of the process and in the quality of the steel. Online analysis of the slag could directly affect:

· Efficient resource usage

· Timing of adjustments to optimize steel composition

· Identification of unusual process conditions

From the University of Oulu, Dr. Henri Pauna in his study showed that the content of the slag could be evaluated from the OES spectrum in real-time. The arc in the furnace excites different elements from the slag and these can be seen in the light’s spectrum.

There are no commercial solutions available yet using this technology but online slag analysis is something that we are developing for our ArcSpec platform.

Illustration of an OES spectrum and the data it contains

Automation has been a big part of the process industry for a long time. However, there have been some difficulties when trying to automate the EAF process. Variation in the scrap loads makes static process models and estimations suboptimal. Limited visibility to the process has been limiting dynamic control. At least until now.

Automation could be used for example in optimizing voltage tap changes and timing process phases. However, how do you automate for example the timing of phases? Timing is typically based on operators’ experience to spot the right visual cues and sounds from the furnace.

Or what if you use a predetermined profile based on historical data and the material content differs from the assumptions?

You need to have a way to track the process in real-time.

Optical Emission Spectroscopy has been proven to work as a way to measure the melting progression of the steel scrap in real-time. This will enable an online analysis of the EAF process for the first time. With this information timing of the phases could be easily automated. This is exactly what the ArcSpec system does.

Overall low visibility of what happens inside the furnace causes many of these challenges. Understandably, compromises must have been done earlier. There has simply not been a practical way to measure the melting progression in real-time. The process is always operated only with the information available.

However, technologies have been rapidly developing in recent years. There are already a couple of methods to measure EAF in real-time. From these OES is the only one to reliably provide real-time data from the melting progression.

The ArcSpec system successfully utilizes the OES technology to optimize EAFs.

We at Luxmet specialize in working with EAF’s and we would be happy to tell you how we have helped steel factories to counter some of these problems.