Click here to see this article as it appears in Modern Casting.

Cupola furnace operation is a transformation process, meaning numerous inputs are transformed into outputs, primarily iron, slag and the stack gases. This cycle is straightforward and somewhat automatic. Feedback loops made through testing and analysis take more effort, but they ultimately help you improve operations (Fig. 1).
The cupola melter is a shaft furnace in which charge materials are fed into the top and continuously melted and discharged at the bottom  as long as air is blown  into the shaft. Depending on the size of the cupola and volume of air being blown, it can be 30 minutes to an hour before the material charged into the top comes out the bottom. Measurements help give a clear picture of the cupola operations so you can better control the output of the furnace.

While all inputs have an effect on the operations, many cannot be changed by the operator. These include whether the refractory used is acidic or basic, whether the air is cold or hot blast, the type of charging mechanism, charge materials, iron demand, whether it’s lined or unlined, and weather conditions. However, a number of inputs can be changed by the operator day to day and hour to hour and these include the charge makeup, blast volume, blast temperature and use of oxygen (if so equipped) .  

Decisions about how to change the inputs can be made with the help of monitoring what happens during the conversion process, such as the backpressure of the air going into the furnace, the water temperature to keep the cylinder cool, the blast temperature, and other visual and audio clues. Additionally, analyzing the outputs (iron, slag and emissions) gives a picture of how the furnace is melting. For the iron, you measure temperature, chemistry and volume. Are you getting too much or too little iron based on the current melting conditions? Slag is monitored through floor testing and lab analysis, and stack gas analysis provides information on the cupola furnace emissions.

Making and keeping records gives the operator the power to look at what has changed in the process that led to an unexpected and unwanted result and make adjustments accordingly.

Checking Composition
Continuous measuring and recording of data is preferred, although not practical for all. At the least, consistent sampling should be used to keep tabs on how the furnace operates. The fundamental requirement is that the sample is representative of the whole, or the test results become meaningless and misleading.
For iron, several methods are available to obtain compositional  information, and they don’t always seem to agree. The composition of the iron coming out of the cupola can be measured in spectrometer and combustion lab tests and via thermal analysis. In addition, operators can conduct chill tests on the shop floor—it’s a simple test to use in between the other tests and gives a visual indication of the potential of the metal to solidify as white iron rather than gray iron (Fig. 2).

Changes in operational conditions can affect each of the tests differently.  When the test results disagree with each other or what you are expecting to come out of the furnace, it is an indication that something has changed and needs to be investigated.

Tests on the iron give you the composition of your end product. Testing the composition of your slag may give you clues as to why the iron composition or element recoveries might be out of whack.

Cupola operations have to balance the refractories of the furnace with the materials it is going to be in contact with in the shaft, especially the oxides. If the cupola has an acidic lining, which all cupola operations in North America do, a basic slag will neutralize the refractory and eat it away—meaning it will end up in your slag and reduce campaign life and increase costs.

Slag composition is also an indicator of operational health. Certain elements are more prone to be lost during melting. Ninety-five percent of a pound of chromium added to the charge  will likely  end up in the melt. But elements like silicon and manganese typically  lose considerably more. Fluctuating chemistries in your iron could be related to changes in your slag.

Like the iron, slag can be tested through visual observations, such as color, appearance and texture, as well as through lab analysis. Slag analysis is put into terms of basicity. A typical target basicity for North American cupola operations is 0.5-0.6, although some very good operations may be as low as 0.3 or as high as 0.8.  Whatever  number your facility targets, you should be hitting it consistently. This target basicity is the point where you have balanced the composition and melting of your slag with the furnace operations. The cupola needs to be hot enough to reach that melting point—the slag has to be fluid to get it all out of the furnace.

Effective and Efficient
The goal is to be effective and efficient in your cupola operation. Effective cupola operation means delivering the correct quantity of iron at the correct composition and temperature for the pouring demands.  Efficiency in cupolas is measured by how many tons of iron can be melted per ton of coke, which is the fuel used to heat the furnace.
Cupola operators are armed with a few tools for controlling the cupola, but success is limited by the accuracy of the measurements and the ability to control these inputs.  Primary importance should be given to the blast air.  Where is it measured?  Are there leaks in the ductwork? Is the airweight controller maintained and calibrated on a regular schedule? When changes are made to charge makeup, how close can the charging system get to specified targets?  How consistent in composition and cleanliness are the charge materials from load to load or during various times throughout the year?  How well are the materials kept segregated in the charge yard? The measures of iron chemistry, melt rate and temperature are interrelated. Operators can’t change one of those things without affecting the other two.

An efficient furnace will use the lowest coke percentage to meet the required iron temperature. The more air blasted into the furnace, the more iron can be melted in an hour. Many cupolas running today were designed with a cross-sectional air blasting area to melt a designated number of tons in the furnace per hour. This does not change, even if present-day operations might call for a lower volume per hour to meet the iron demand of the molding and pouring operations.

To illustrate this, Fig. 3 (digital edition) is a blast rate chart for a cupola furnace. The x axis is a blast volume expressed as a ratio of the blast volume (CFM) divided by the cross sectional area of the cupola in the melt zone in square inches.   The red line intersecting the peaks of the various coke percentage curves is called the “sweet spot” as it represents the most efficient blast rate to run the cupola (approximately 2.0-2.2.) In this chart, the lowest coke percent is 7.5, but the iron temperature required to meet the pouring requirements of a particular metalcasting facility is 2,600F. (1,427C)  In this case more coke must be used (to about 9.5% ).  Reaching 2,600F can only be accomplished at the peak in the mid-point of the curve which corresponds to a blast volume ratio of approximately 2.0.  Now suppose the metalcasting facility no longer needs the tonnage of iron produced by blowing at the ratio of 2.0, but only needs the tonnage produced by blowing at a ratio of 1.5.  This would produce iron at approximately  2,550F (1,399C), which is not suitable for pouring.  The coke percentage would need to be raised to about 13% to reach the 2,600F required. The cupola is less efficient, but there’s nothing the operator can do about that. Many of the existing furnaces today are too big for what is needed yet it is a significant investment to change it.

What Do You Do?
Keeping records of your cupola operations aids your decision making when something goes wrong with the melt. Flow charts for situations like cold iron, insufficient melt rate or high back pressure are valuable tools to help new operators make better decisions as conditions change.   For example, one of the worst things that can occur in a cupola operation is for the coke bed to get too low, causing both low iron temperature and chemistry issues.   This will take 30 to 60 minutes to correct by putting more fuel at the top.
So what are the indications the coke bed might be too low? The carbon equivalent, carbon or both are decreasing, the chill tests are trending up, spout temperatures are trending down, and in severe conditions unmelted metallics are seen through the tuyere.

One method to check the coke bed is to increase the blast by 10% and wait 10 minutes (Fig. 4). If the spout temperature increases, the coke bed is too high for the current blast volume. If the spout temperature increases and then decreases (typically below the initial temperature), the bed is too low.
Successfully and efficiently running a cupola cannot be covered in one article, or one training session, but the general rules of operation boil down to three things: 1. Turn it on, let it run, drain it out; 2. Run the cupola at its sweet spot. 3. If following rule 2 makes you violate rule 1, efficiencies will suffer.

Operators must keep track of all inputs and outputs in the melting process, using that information to adjust those inputs that can be changed as necessary.