The Use of Copper Shells by Twin Roll Strip Casters: TMS Light Metals March 2010

Download as pdf or txt
Download as pdf or txt
You are on page 1of 6

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/304990092

The use of copper shells by Twin Roll Strip Casters

Article  in  TMS Light Metals · March 2010

CITATIONS READS
5 684

5 authors, including:

Dionysios Spathis John Tsiros


Helenic Aluminum Industry SA ELVAL
10 PUBLICATIONS   29 CITATIONS    9 PUBLICATIONS   5 CITATIONS   

SEE PROFILE SEE PROFILE

Aristeidis Arvanitis
PPC SA
4 PUBLICATIONS   36 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Development of Mo-Si-Al intermetallic alloys View project

All content following this page was uploaded by Dionysios Spathis on 07 August 2016.

The user has requested enhancement of the downloaded file.


Light Metals 2010 Edited by: John A. Johnson
TMS (The Minerals, Metals & Materials Society), 2010

THE USE OF COPPER SHELLS BY TWIN ROLL STRIP CASTERS


Dionisios Spathis1 John Tsiros1, Aristeidis Arvanitis1 , Hans-Gunter Wobker2, A. Clemente3
1
Helenic Aluminium Industry – Elval – GR 320 11 Inofyta-Viotia - Greece
2
KME Europe AG – Klosterstrasse 29 , Osnabruck – Germany
3
Castcom Ltda 53230-630 Olinda - Brazil

Keywords: Twin roll casting, copper shells, strip casting productivity

Abstract

Twin Roll Casting (TRC) of aluminum alloys is a well established


process, which exists over several decades. The productivity of
the process has been largely limited by the heat transfer capability
of the steel shells used on the caster rolls. In recent years,
developments in the manufacture of shells based on special
copper alloys have opened a window for productivity increases
and quality improvement. One or both rolls of a caster can be
lined with copper shells, to take advantage of the higher thermal
conductivity of Cu relative to Fe, which provides for an increased
cooling capacity of the working rolls.

This paper presents a review of the equipment/operation,


productivity increase and product/quality performance by using
both copper/copper and copper/steel combination shells.

Introduction

The Twin Roll Casting (TRC) is well known as a relatively low


hourly productivity process compare to traditional DC casting/hot
rolling. Although many attempts have been done to improve the
TRC productivity, the only really possibility was the use of high
thermal conductivity materials on the exterior surface of the roll. Figure 1 – Alusuisse Baby Caster
The use of copper and cooper alloys were first considered. Copper
shells were first used on Alusuisse Baby caster for automotive Therefore this technology has demonstrated potential for doubling
bearing alloys AlSn12. (Figure 1) [1] the caster productivity and improving strip quality and final
products properties.
Other casters users, for example, Granges Aluminum, ran trials
replacing the steel shells by copper alloys[2]. Caster manufacture, Copper Sleeves in TRC
Pechiney, has also run many tests using copper and copper alloys.
Although they have established the fact of significant advantages The main issues regarding the use of copper shell in terms of their
in casting speed by adopting copper shells, they also faced many influence on productivity, strip quality, downstream process and a
problems with mechanical behavior of these roll shells on the steel final 8006-alloy foil product performance are described below.
core. Due to the limited hardness, ductility and especially strength
at higher temperatures of the available copper alloys the attempts 1. Roll Caster Productivity
made to apply copper shells in strip casting at that time was not
fully successful.[3] Low TRC productivity is a result of the difficulties to remove heat
from the strip during the cooling process. Steel alloys have at least
In general, the main constraints for copper shells use were: six times less thermal conductivity compared to suitable copper
alloys (35 W/(m.K) versus 200-300 W/(m.K, respectively). The
• Limited copper shell hardness
steel thermal conductivity is the major limiting factor to the caster
• Low strengths at higher temperatures
productivity.
• Limitations on shell/core shrink fitting, which results on
uncontrolled shell slippage on the core
Many of the common alloys 1XXX, 3XXX and 8XXX series
• Very low shell life, compare to conventional steel shells were cast using Cu/Cu and Cu/St combination. The productivity
(increased costs). achieved was in general like shown in the Table I below, but still
there are possibilities for increases.
KME started in 1998 the development of copper alloys shells for
aluminium strip casting.[4,5] Nowadays, most of these
constraints have been overcome and/or were well improved.

747
Table I – Roll Caster Productivity Improvement The total roll surface area had to be covered by release agent so
Roll Shell Strip Gauge Productivity (Ton/h.m) double guns were used. Their traveling speed, distance from the
rolls and flow pattern were adjusted somewhat differently from
St/St 5 to 10 mm 0,7 – 1,5
what was used with slower running steel shells.
Cu/St 5 to 10 mm 1,7 – 2,2
Cu/Cu 5 to 10 mm 2,2 – 3,2 Due to the high metal flow, the tip internal baffle layout was
changed. Metal flow disruptions and large temperature gradients
The cast strip thickness is recommended to be above 6 mm, to can result in strip defects.
better help minimize the separating forces on the Cu shell. Also
the shell lifetime increases when decreasing the thermal cycle.
4. Products

Final products produced with Cu/Cu and St/Cu shell combinations


include:
Alloy 5005 for trailer sheet
Alloy 3105 for building sheet, welding tubes, roofing sheet, etc
Alloy 3003 for panel sheet, deep drawing quality
Alloy 8006 for fin stock, smooth wall container, container stock,
and high strength household foil
Alloys 1050, 1145, 1100, 1200, 1235 and 8011 for fin stock, foil
stock, PP cap, container stock, converter foil, lids, etc

5. Metallographic Analysis and Product Characterization

The strip’s through-thickness microstructure using Cu/Cu shell


rolls was similar to the microstructure found on the copper side
Figure 2. Actual tilting caster working when casting with Cu/St shells combination. The production case
with Cu/St shell combination below highlights the differences between the various shell
combinations.
2. Shell Life
Typical 8006 alloy for household foil was casted in a twin roll
Shell life and its economical impact has been one of the main caster using one copper shell and one steel shell. Full width strip
development activities. KME and partners (copper shell users, samples at the caster exit gauge of 6 mm were collected from
mainly) have introduced new shell manufacturing techniques and various casting campaigns for metallographic examination.
special heat treatments to improve the shells resistance to fatigue.
(Figure 2). The cast strip was cross sectioned parallel to the casting direction.
Micro hardness measurements using a diamond pyramid indenter
Shell life has been increased from 3-5 days/campaign up to 15- and 100 g load were carried out across the strip thickness.
Electrical conductivity measurements were carried out on both
day campaigns. To maximize the fatigue life, it is recommended
cast strip surfaces to monitor the amount of alloying elements in
to use maximum shell thickness possible, considering the caster the solid solution.
frame. The thermal conductivity is not affected by the copper
shell thickness. Also if the outer diameter of the roll has a 5.1. Optical Microscope Examination Results of As Cast
limitation then the steel core diameter can be reduced to increase Microstructure. Figs 3-4 show an approximately 200 micron thick
the Cu shell thickness. outer surface zone of somewhat coarser dendrites (~12 µm mean
cell size) observed on the surface solidified on the copper shell.
There was a sharp transition between the outer coarser dendrite
3. Equipment Modification zone and the finer (~5 µm) inner structure. At the surface area on
the copper shell side the eutectic-rich regions are not sheared.
No major caster and auxiliary equipment modification were
necessary to run the caster with copper shells. The same re- Near the metal surface solidified on the steel shell, the eutectic-
shelling procedure used for steel shells was applied to Cu shells. rich regions are highly sheared due to the rolling action before the
The casting parameters to be monitored were the same as for steel metal strip exits the nip of the caster (Fig 5).
shells, but due to the high casting speed, special care was taken
Optical microscope examination of the as-cast strip showed that
with the anti-sticking spray system. Since the release agent plays the segregation channels caused during final solidification of the
an important hole in the TRC process, with copper shells it casting sump were located at the 2/3 of the strip thickness and
became more important due to the substantially increased casting were closer to the steel shell. The two-steel-shells cast structure
speed. A system, such as peristaltic pumps can guarantee a usually has a segregation channel (often termed CLS) located at
reproducible and controlled release agent application is the centerline of the strip thickness if the tip is positioning at
recommended, example that one using. center of the roll gap.

748
The segregation degree level is lower (shorter and thinner eutectic
bands) compared with the material cast using typical two steel
shells (Fig 6).

St side

Figure 3. Coarser dendrites in a surface zone on the


side solidified on the copper shell

Cu side

(a) (b)
Figure 6. Centerline segregation comparison
for St/St (a) and Cu/St (b) shells

The surface grains were not worked by rolling the side solidified
on the copper shell (Fig 7). At the side solidified on the steel
shell, grains were heavily worked and fine recrystallized grains
are observed (Fig 8).

Figure 4. Typical transition zone between coarse


and fine dendrite areas (copper shell side)

Figure 7. Grain structure near the surface


solidified on the copper shell

5.2. Results of the Micro Hardness Testing Across the As Cast


Figure 5. Worked of the fine eutectic zones Gauge. Micro hardness measurements were carried out across the
on the side solidified on the steel shell as-cast strip thickness (Figure 9). The VICKER hardness test was
with a constant load of 100 gr.

749
The micro hardness values of the as-cast material produced by the
copper shell was significantly higher relative to the material • The EC values measured at the steel side of the strip
produced using the steel shell. produced with either the one steel shell or the two steel shells
The as-cast surface region which solidified on the copper shell were similar.
showed higher micro hardness values relative to the surface
solidified on the steel roll. Table II. EC measurements (MS/m), frequency 250 KHz
Steel Copper Steel Steel
1 25,12 24,5 24,98 25,2
2 25,36 24,4 24,96 24,9
3 25,29 24,48 25,28 25,3
4 25,21 24,53 25,27 25,3
5 25,15 24,54 25,2 25,3
6 25,22 24,34 25,2 25,4
Mean values 25,23 24,47 25,15 25,23

5.4. Mechanical Properties Results and Microstructure At Final


Gauge Of 8006 Household Foil. Table III shows the mechanical
properties results obtains at final gauge of household foil
produced with one copper shell in -O temper condition at 10 and
13 µm final gauge.

Figure 8. Surface layer showing highly worked grains at the Table III. Mechanical properties of final foil product
surface solidified on steel shell
UTS YS Elongation Burst Strength
(N/mm2) (N/mm2) (%) (kpa)
70
107,8-125,2 88,8-95,7 5,4-6,8 110-123

60
Microhardness (HV)

6. Additional Comments

Slightly higher mechanical properties were attained by using


50 Cu/St or Cu/Cu shells compared with St/St shells, unless a high
temperature homogenization treatment was used.

40 When casting with Cu/Cu and St/Cu, the copper side strip surface
Copper/Steel Shell doesn’t have grain growth during subsequent high temperature
Steel/Steel Shell homogenization heat treatment (Figure 10). This is good news
30 when considering manganese-containing alloys like 3003, which
0 1 2 3 4 5 6 7 has the tendency for grain growth during heat treatment.
Strip thickness (mm) Further investigation is needed to understand if grain growth
Figure 9. Microhardness comparison St/St vs Cu/St for 8006 restriction during high temperature heat treatment occurs in other
alloy alloys.

5.3. EC Measurement Results On Both Sides of the Cast Surface. The change in the CLS position when casting with Cu/St
As-cast surface conductivity measurements were carried out to combination had no effect on foil quality.
determine the level of solute elements trapped in the aluminum
matrix during the solidification process. Higher conductivity
values means less manganese, iron and silicon elements dissolve The displacement of the centerline has no negative effect on thin
into the solid solution. gauge foil quality such like pinholes, holes, web breaks, etc.

Table II shows conductivity values of top and bottom surface of At final gauge of 10 µm and 6,5 µm foils of alloys 8006 or 8011,
the as-cast 8006 alloy strip produced with one copper shell. The the constituents particles are very fine and well distributed across
conductivity data of as-cast surface of 8006 alloy produced using
the foil thickness Figure 11 and Figure 12.
two steel shells has been included in the table for comparison.

The EC measurements gave the following results:


• The lowest conductivity was measured at the copper side of
the cast surface.

750
Conclusion
(a)
1. The latest KME development has allowed the use of a copper
shell for aluminum strip casting, due to the higher achieved
thermal crack and fatigue resistance, which reflect in longer shell
life.

2. Typical productivity improvement for the use of combination


Copper/Steel shells is around 40 to 60%, and for Copper/Copper
shells is about 70 -110% (depending on alloy and each plant
operation). There may still potential for future improvement.

3. The cast stock quality improvements are confirmed by


centerline segregation and particle size reduction due to the higher
cooling rate. Consequently, final product quality was improved.
(b)
4. Relatively minor (but important) caster modifications, such as a
different tip design and a spray system for roll-strip release, are
necessary to obtain the maximum casting speeds allowed by the
Cu shells. The capital investment for these modifications is
very low for the benefits obtained from the increased productivity.

5.With the large productivity improvements, there is an overall


economic benefit of copper shell production even considering the
higher cost of the copper shell.

6. The use of copper shell provide potential for design of new


alloys / unique final product properties.
Figure 10. Grain structure after homogenization at high
temperature; alloy 3003. (a) steel shell side (b) Copper shell side
References

1 . Lauener Engineering private communication with author

2. E. Ness, S. Slevolden - “Casting and annealing structures in


strip casting aluminum alloys”. ALUMINIUM 55, May 1979,
319-324

3. J. Chateau, P. Solignac, “Doubling the productivity of a roll


caster – Is it possible? – Is it economical?” LIGHT METALS
1989, 667-672

4. H. Wobker et al. “Copper shells for Twin Roll Casting”.


LIGHT METALS 2006

Figure 11. Alloy 8006 – 10 µm foil from Cu/St shells 5. Kolbeck et al . Cast-rolling Plant; US Patent US7,028,748; Apr
18,2006

Figure 12. Alloy 8011 – 6,5 µm foil from Cu/St shells

751

View publication stats

You might also like