Ductile iron trends: reducing costs, improving quality.

Even though today's ductile iron foundries come in many shapes and sizes, similarities exist. When it comes to casting production, every foundry looks for ways to reduce costs while ensuring quality. It is difficult, however, to develop new production ideas and innovations.

Examining industry trends allows a foundry to catch a glimpse of what other plants have done in similar environments. In support of this, the AFS Cast Iron Div.'s Molten Metal Processing Committee (5-L) has tallied a survey of North American ductile iron foundries the last three decades (1978, 1988, 1998, in addition to an initial survey in 1957) requesting information on production practices. By studying the results of this survey, foundries can see where they differ from other ductile iron plants and consider changes that could improve quality and production costs. This article will compare the results of these three surveys and identify current ductile iron production trends.

In 1998, the committee sent out surveys to 508 foundries in the U.S., Canada and Mexico and received 187 responses. The respondents have been split into two groups--foundries that produce more than 200 tons of ductile iron/wk (referred to as large foundries), and foundries that produce less than 200 tons of ductile iron/wk (referred to as small foundries). Seventy-four large foundries and 113 small foundries responded. Some foundries were included in both categories because of multiple types of melting groups, treatment methods and treatment sizes.

Furnace Fluctuation

The 1957 survey revealed that 65% of respondents (large and small) conducted their primary ductile iron melting in a cupola. Since then, cupola melting has fluctuated dramatically for both groups. For large foundries, cupola melting dropped to 39% in 1978, increased to 61% in 1988 and again fell to 30% in 1998 (Table 1).

Coreless induction melting has been on the reverse curve, starting at 45% in 1978, dropping to 38% in 1988 and then rebounding to 62% in 1998.

Small foundry furnace choices have remained stable over the last decade. Cupola melting went from 25% in 1978 to 4% in 1988 and rose slightly to 8% in 1998. Coreless induction melting remained at the 84% level in both 1988 and 1998, increasing from 44% in 1978.

The number of foundries using holding furnaces has reduced for both groups, dropping from 92% to 64% for large foundries and 36% to 13% for small foundries from 1988 to 1998. While the use of channel holders and coreless induction holders has remained constant for large foundries over the last 10 yr, the use of arc furnaces and fore-hearth holders has experienced substantial declines due to their inefficiencies when compared to other holders. For small foundries, the use of group channel holding furnaces has been the most unstable, starting at 7% in 1978, rising to 24% in 1988 and ultimately falling to 7% in 1998.

Base Iron Information

For both groups, ductile iron carbon level has remained stable: 72% of large foundries and 63% of small foundries favor levels of 3.6-3.9% (Table 2). The large foundries did show a dramatic shift downward in sulfur levels, While 54% had a sulfur level greater than 0.036% in 1988, only 26% repeated that in 1998. Correspondingly, sulfur levels of less than 0.0 15% increased from 27% in 1988 to 42% in 1998. This change most likely can be attributed to the 22% drop of large foundry respondents melting in cupolas over the last decade.

While the carbon level at small foundries has remained constant, sulfur levels have dropped and now favor less than 0.015%. Although the types of melting in small foundries did not change, the lower sulfur base iron levels suggest that higher quality melting materials are being used. This also is reflected by the increase in silicon levels to the 1.3-1.6% level for both foundry groups, and the decrease of respondents that use the 1.3% level.

Desulfurization Drops

Eight percent of small foundries use cupolas and 8% show some type of desulfurization. For large foundries, the number that desulfurizes is down from 55% in 1988 to 34% in 1998. This reflects the drop in cupola melt respondents. Batch desulfurization has declined in all groups, with the only real increase in the two larger tonnage groupings (2 1-30 tons/hr and greater than 30 tons/hr) performing continuous desulfurization. This can be explained by the increased use of elemental magnesium treatment, a process that negates the need for desulfurization.

For large foundries, the use of calcium carbide has declined from 54% in 1988 to 18% in 1998. The use of calcium oxide and calcium fluoride increased because it is less hazardous than calcium carbide, jumping from 0% in 1988 to 12% in 1998.

Preconditioning appears to have taken a severe blow in the last decade. Seventy-eight percent of large foundries and 85% of small foundries show that no preconditioning is used, an increase of 40% and 67%, respectively. The use of graphite, ferrosilicon and silicon carbide as preconditioners has dropped significantly over the last decade due to the overall lack of preconditioning.

Variety in Methods, Materials

Of all the treating practices, the use of tundish ladles has increased dramatically because of its greater operating efficiency. Accounting for 3% of large foundries in 1988, it rose to 50% in 1998. Using tundish ladles allows for less slag, less temperature loss and a more efficient use of magnesium. With the increase in tundish ladles, the sandwich practice dropped heavily in large foundries, down 16% from 1988 to 1998.

In small foundries, the open ladle practice has dropped 24%, resulting in increased usage of the sandwich practice (up 7%) and the in-flow-through method (up 5%). Large foundries have shown an 8% drop of the in-flow-through method, while in-mold treatment has steadily fallen from 15% in 1978 to 12% in 1988 and finally 7% in 1998.

Magnesium containing cored wire only is being used in 3% of all foundries, while pure magnesium treatment is running at 7% in large foundries and 2% in small foundries. Due to higher costs, it is mainly used in low production situations.

Magnesium ferrosilicon (MgFeSi) use by both groups has remained unchanged over the last 10 yrs-the majority of users stayed in the 5-6% group. Both groups show a 1% usage of 9% MgFeSi. The 2.5-3.5% MgFeSi group is in 5% of large foundries, but has remained relatively unchanged in small foundries at 11% in 1998. The only category that has gained support in both groups is the 6-7% MgFeSi, which has taken some of the 5-6% users over the last decade. This increase allows plants to use fewer pounds of MgFeSi while increasing reactivity. Twenty-seven percent of large foundries and 39% of small foundries do not use any rare earths in their MgFeSi. For those foundries that do use rare earths, low cerium is used in the majority of both groups.

The use of nickel magnesium alloys has been reduced in small foundries over the last 10 yr, down from 20% in 1988 to 1% in 1998. Magnesium ingot or bar has dropped, with 7% of the larger foundries still using this product, mostly in pure magnesium converters.

The rise in tundish ladles has allowed the addition level of MgFeSi to drop off, as 46% of the larger foundries run in the 1-1.5% range. This change has occurred because of the focus on improving production practices and lower treatment temperatures obtainable when using tundish ladles. Because small foundries run higher treatment temperatures, they use fewer tundish ladles and run in the 1.6-2.0% MgFeSi range.

Expanded Post Inoculation Use

Every foundry surveyed indicated that it uses post-inoculation, a significant jump from 1988 when 42% of large foundries indicated that they did not use the process. One possible factor in this change is the use of the in-mold process. Both foundry groups indicated that more than 50% of the post-inoculation occurs in the pouring ladle, Thirty-one percent of the large foundries use in-stream inoculation, a 15% drop from 1988. In small foundries, in-stream usage has increased from 2% in 1988 to 14% in 1998. Inserts continue to have a steady fraction of the post-inoculation practices, appearing in 18% of the large foundries and 14% of the small foundries.

The most prominent post-inoculant being used is a foundry grade 75% FeSi with greater than 0.8% calcium (Ca) and 1.0-1.25% aluminum (Al). This inoculant is being used in 50% of large foundries and 35% of small foundries, only a slight variation from 1988. As in the past, other inoculants used are 50% FeSi, 75% FeSi with 2.0% barium (Ba), 75% FeSi with 3.0% Al, 65% FeSi with Ca and Ba and 75% FeSi with lower than 0.8% Ca and lower than 1,0% Al. Since the 1978 survey, there has been an increase in the variety of inoculants available to foundries because of the increase in the number of ways that iron is poured.

Tying Up Loose Ends: Additional Results

Final Chemistry--Final carbon and silicon levels have remained nearly unchanged from the 1988 survey (Table 3). Large foundries tend to run 3.65-3.84% carbon (C) and 2.45-2.64% silicon (Si) while small foundries tend to run 3.403.64% C and 2.45-2.64% Si. Both groups also show a strong preference for final magnesium levels of 0.03-0.04%.

Types of ductile iron produced--The 1998 survey requested each foundry report ductile iron grades produced. This was the first time this question was asked on the survey, so there is no data prior to 1998. Results were as follows:

* pearlitic castings are produced by 85% of large foundry respondents and 78% of small foundries;

* ferritic castings are produced by 82% of large foundry respondents and 87% of small foundries;

* austempered ductile iron castings are produced by 9% of large foundry respondents and 6% of small foundries;

* silicon molybdenum iron is produced by 7% of large foundry respondents and 2% small foundries;

* zero large foundries and only 2% of small foundries report making niresist castings.

Pouring temperature--In this section, the numbers for the 1998 survey exceed 100% because the 1988 survey asked for a range of temperatures and the 1998 survey asked for a specific temperature. The ranges that were established previously overlap by 50 degree segments and are therefore higher than 100%.

The two pouring temperature ranges that are preferred by both large and small foundry groups are 2500-2600F (1371-1427C) and 2550-2650F (13991454C) for the first iron poured. These results are similar to the 1988 survey, however, there is a steadily growing trend toward pouring in the 2400-2500F (1316-1371C) and 2550-2650F (1399-1454C) ranges for both groups of foundries, Even though there is a trend toward lower pouring temperatures, the variation of temperature is getting smaller and more consistent,

Quality control--More than 70% of each foundry group used either a separately cast coupon or a flow off as their primary source of nodularity control. Some of the other methods used to check the iron were 100% final chemistry check, 100% ultrasonic testing and 100% sonic testing.

The use of a cooling curve analysis appeared in only 1% of the large foundry group while 4% of small foundries used this method for their primary quality control check. Other methods mentioned included ring testing, ball impression, bend testing and cutting up a casting.

Deciphering the Data

In examining this survey, a foundry can see how it compares with current trends. By adjusting various processes, a foundry can reduce costs and improve consistency and quality. Nearly every change that this survey illustrates in the last 30 yr has been an attempt to retain high quality while reducing production costs. An example includes the improvements in magnesium treatment, which have led to higher consistency and reduced defects and scraps. The reduction in holding iron and the introduction of cooling curve analysis also have improved production.

After reading this survey, return to your process and compare your foundry to the average foundries in this survey. Evaluate differences and see if a change could make a positive impact. It is important to realize that not every change will work for your foundry. This survey provides just a glance into ductile iron production, but it does not take specific situations into account. There may be times when a higher cost production practice is the right decision for a foundry.

Table 1.

Primary Ductile Iron Melting & Holding (by % of foundries)

Melting              Large Foundries                 Small Foundries

Cupola         1978     1988      1998       1978     1988      1998
Acid Slag       26       42        20         20        2        6
Basic Slag      13       15        3           5        0        2
Neutral Slag    NA        4        7          NA        2        0
Coreless        45       38        62         44       84        83
Arc             15        4        1          13        6        0
Channel         15        4        12         22       22        8
Other            0        0        0           0        2        3
Holding
Channel         30       46        43          7       24        7
Coreless         6       15        11          5       10        3
Arc              9        8        0           2        0        0
Forehearth      NA       15        5          NA        0        1
Other            9        8        5           5        2        2
Table 2.

Base Iron Information Breakdown (by % of foundries)

Base Iron Information

                          Large Foundries            Small Foundries
                   1978      1988      1998      1978      1988

% Carbon
<3.60               12        19        11        25        14
3.60-3.90           73        77        72        64        86
>3.90               9          4        7         15         4

% Sulfur
<0.015              24        27        42        5         29
0.016-0.025         30        12        15        36        61
0.026-0.035         18         4        7         25         6
>0.036              15        54        26        33         4

% Silicon
<1.30               45        23        8         44        41
1.30-1.60           58        31        47        40        37
>1.60               21        35        34        18        24

Carbon Equivalent
<4.0                NA         8        12        NA        10
4.0-4.3             NA        58        30        NA        57
>4.3                NA        65        47        NA        27

                     Small
                   Foundries
                   1998

% Carbon
<3.60               8
3.60-3.90           63
>3.90               4

% Sulfur
<0.015              38
0.016-0.025         23
0.026-0.035         1
>0.036              9

% Silicon
<1.30               19
1.30-1.60           27
>1.60               27

Carbon Equivalent
<4.0                4
4.0-4.3             42
>4.3                28
Table 3

Final Chemistry of Ductile Iron Breakdown (by % of Foundries)

Final Chemistry         Large Foundries            Small Foundries

% Carbon          1978     1988      1998       1978     1988
< 3.40              3        8         5          5        2
3.40-3:64           3       23        23          5       45
3.65-3.84          55       65        57         47       53
3.85-4.00           6        0         3          2        0

% Silicon

< 2.20              6        0         0         13        0
2.20-2.44          15       31        26         12       18
2.45-2.64          58       65        49         47       59
2.65-2.84          12       15         7         16       25
2.85-3.00           0        0         4          2        6
> 3.00              0        0         3          2        0

Final chemistry    Small
                 Foundries

% Carbon         1998
< 3.40             2
3.40-3:64         35
3.65-3.84         27
3.85-4.00          4

% Silicon

< 2.20             1
2.20-2.44         15
2.45-2.64         35
2.65-2.84         16
2.85-3.00          4
> 3.00             4

For more in formation

"Survey on Ductile Iron Practice," D. W. Brewster, J.E. Foltz, JW.A. Henning, B. Morgansteren, 1989 AFS Transactions, AFS, Des Plaines, IL.

"Survey on Ductile Iron Practice," AFS Molten Metal Processing Committee t12-H), 1979 APS Transactions, AFS, Des Plaines, IL.

This article was adapted from a presentation (02-165) at the 2002 AFS Casting Congress.

For a free copy of this article circle No. 343 on the Reader Action Card.

About the author

Jim Csonka is the Manager of Ferrosilicon Alloys for Hickman, Williams & Co., Brook Park, Ohio, and is also the manager of its Cleveland sales office.

James Woods retired in 1999 from Hickman, Williams & Co., Chattanooga, Tennessee, where he served as Vice President and was a member of the Management Committee.

Gene Muratore is the Senior Foundry Metallurgist for Rio Tin to Iron and Titanium America, Rosemont, Illinois.