Printer Friendly

Microprocessor charging control unit streamlines iron melting operations; a control-advisory system for cupola charging indicates proper inputs of iron, steel and coke.

Microprocessor Charging Control Unit Streamlines Iron Melting Operations

A control-advisory system for cupola charging indicates proper inputs of iron, steel and coke.

Cupola furnaces provide the bulk of molten iron in many ferrous foundries. In the United Kingdom, for example, over 70% of the cast iron industry's output is melted in cupola furnaces.

The prime energy source for the cupola is coke, and it usually represents a considerable portion of a foundry's total energy bill. Inconsistent charging of coke and metallics to the cupola results in loss of coke bed height and subsequent performance deterioration. This reduced performance leads to: * increased post-melting treatment of the iron to counter the higher sulfur levels resulting from loss of temperature; * increased downtime within the foundry; * increased coke consumption as large amounts of coke are added in an ad hoc manner to replace the depleted coke bed.

To ensure standard cupola performance it is essential to regulate and control the inputs of steel, iron and coke to the cupola.

Conventional Cupola Charging

In a typical charging system used in one British foundry, the crane driver and coke operator independently try to load to given targets with no communication or control. This system has several inherent problems. Brought-in steel scrap comes in greatly differing sizes that range from .5 kg to 200 kg. Achieving a nominal target load, (say 450 kg), is difficult and leads to understandable inaccuracy.

The operator usually tries to correct any error in steel loading with a corresponding adjustment of the iron load. For example, if 500 kg of steel had been charged, then 400 kg of iron charge would be attempted instead of the nominal 450 kg target figure. (The iron charge is easier to manipulate than the steel charge because of its constituent pieces being smaller in size.) However, because more energy is required to melt steel, a coke deficiency would then arise. Should steel be deficient, excess coke would be added.

The coke charge should not deviate greatly from its target load due to its regular size. However, since the operator works independently, the coke charge remains unchanged irrespective of the total metal charge or the steel/iron ratio.

Microprocessor-Controlled Charging

A proposed microprocessor-controlled charging system incorporates standard weighing equipment along with thumb wheel ratio switches to preset required steel/iron and coke/metal ratios, a central processing unit (CPU), digital display units and a control box.

The CPU records the steel charged and advises on inputs of iron and coke. The advice is offered on an updating memory principle that is based on the makeup of the previous six charges (six charges equal one cupola load). A typical cycle would operate as follows. 1. Steel is loaded to the weigh hopper. This activates the load cell and sends a signal to the CPU. When the crane operator has loaded approximately 450 kg, a button in the cab is pressed enabling the CPU to record the amount of steel in the hopper. 2. After processing the steel input, the CPU offers advised loads for iron and coke at the two operators' displays. These advised values are computed based on the steel charge and the previous charges to the cupola. 3. Both operators then load the iron and coke to their advised targets until reasonable accuracy has been obtained. Iron is loaded into the weigh hopper, while the coke is directly put into the charge hopper. 4. The coke operator activates a hydraulic mechanism that tips the weigh hopper's contents into the charge hopper. Activating this hydraulic mechanism also signals the CPU to record the weight of contents in the weigh hopper and calculate the actual iron loaded.

Comparing the Two Systems

To examine the performance of the proposed system, a process model program was written for a desktop computer. As previously discussed, errors will occur between required and actual loads. To account for these errors, the model adds or deducts a 0-10% random error to or from the advised iron load. Similar problems arise for the coke load and errors of 0-1% are generated to model this.

The model stores information in groups of six charges and continuously updates advice based on preselected ratios and actual loads previously charged.

The conventional charging system was monitored over a number of days by recording the coke and metallic inputs to the cupola. Figure 1 shows the variation of steel/iron cupola ratios for both conventional and microprocessor advised systems.

Figure 2 is a histogram of coke/metal cupola ratios, also for both the conventional and microprocessor advised systems. These graphs show that an updating memory system is capable of operating within much tighter limits than the conventional charging system. This greater control is shown by the target (or required) ratio being almost coincidental with the actual charge ratio.

The conventional charging procedure displayed in Fig. 2 shows a shortfall of charged coke/metal ratios from the target mean. During monitoring, the performance of the cupola was erratic and approximately 16 kg of extra coke per charge was supplied to correct this. This additional coke ties up with the shortfall of actual and target means of coke/metal ratio.

The consistent charging practice that the microprocessor-based system provides would likely establish a reduced optimum target value that does not need additional coke charges to overcome cupola performance deterioration.

In terms of reduced coke consumption alone, the microprocessor-based advisory system may pay for itself within 18 months in foundries where current charging is well supervised and annual coke consumption is approximately 2000 tonnes. Greater savings could be achieved where more wasteful charging exists.

The microprocessor-based system was compared only with the performance of one particular charging procedure. Some may think this is unrepresentative because there are numerous stockyard layouts and charging methods in the iron casting business. However, these differences represent variations upon a theme rather than any significant change in method. Thus, the proposed microprocessor-based charging system should be applicable throughout the cast iron industry. [Figures 1 to 2 Omitted]

Anthony Griffiths, Stephen Lloyd University of Wales Cardiff, United Kingdom
COPYRIGHT 1989 American Foundry Society, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1989, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Griffiths, Anthony
Publication:Modern Casting
Date:Dec 1, 1989
Previous Article:Understanding insulating properties of permanent mold coatings; although mold coating application involves many complex factors, recent research has...
Next Article:Aluminum casting process finds new applications; developments in the Cosworth Process have spread its application to such fields as the aerospace and...

Related Articles
Cupola modeling project nearing phase II.
New research advances melt control.
Melting and pouring.
Cupola bed key to quality melt.
The state of the cupola.
Bringing the cupola into the computer age.
Considering the cupola.
Improved cupola performance a major issue for operators.
Controlling the humidity of cupola blast air.
The AFS cupola process model: a computer tool for foundries.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters