Controlling the structure of aluminium alloys before

casting with the thermal analysis equipment

THERMATEST* 5000 NG III

Evaluation of the grain size

Introduction For a given cooling speed, the size of the
grain depends on the amplitude and
In the casting of aluminium, it is important to be able to quickly assess duration of the undercooling, which
the degree of grain refinement and the type of eutectic structure. If this appears at the formation of primary
is done immediately after the addition of grain refining and modifying aluminium crystals.
agents it gives the foundry the opportunity to make any necessary
adjustments to grain refiner or modifier levels before casting, thus The graphs below show the profile of the
avoiding the risk of costly scrap due to shrinkage, leakage, porosity and cooling curve, during the solidification of
hot tearing caused by poor treatment levels. the primary aluminium crystals in case of
a hypoeutectic alloy.
The thermal analysis of treated alloys is a widely recognised technique
for assessing liquid metal quality before casting. The analysis involves
recording the cooling profile of a sample of liquid aluminium during its Temperature
solidification.

THERMATEST 5000 NG III is a well–proven thermal analysis device θ

designed to quickly predict and subsequently control the structure of
aluminium alloys before casting. θ2
θ1
Theory
t1
Specifically for aluminium alloys, thermal analysis enables foundrymen
to predict the grain refinement and type of eutectic structure before
casting aluminium silicon alloys.
Time
Grain refinement is dependant, to a large degree, on the number of
nuclei of titanium boride and aluminium boride. These finely dispersed Case No 1
particles are highly efficient nuclei that promote a fine equiaxed grain
growth during solidification. The spectrometer analysis gives only the
chemical analysis element by element (Ti, B, etc..) and not any Temperature
combination of the various elements. Thermal analysis on the other
hand shows the alloy behaviour linked to the titanium boride nuclei
quantity, which, in turn, directly influences the grain refinement
efficiency. Whilst the levels of Titanium and Boron are extremely
important they don’t necessarily guarantee a good grain refinement.

As far as the modification of aluminium silicon alloys is concerned, θ1

good modification is essential to avoid leakages, shrinkage and to
improve the machining and mechanical properties of the casting.

Specific elements such as sodium, strontium, calcium, antimony and

phosphorus have a significant impact on the silicon form in eutectic
structure. However, when combined together some of these elements Time
can give an opposite effect to the one expected. The thermal analysis
shows the influence on the structure. Thus any combination of Case No 2
incompatible elements will be immediately detected and shown. This is
a major benefit and enhances the knowledge already obtained from the ❑ θ1 is the temperature at which the
chemical analysis. solidification begins
❑ θ2 is the maximum temperature reached
Thermal analysis is therefore able to assess : at the end of the undercooling
❑ grain size in hypo and hypereutectic alloys ❑ Δθis the apparent undercooling equal
❑ eutectic structure in Al-Si alloys to θ2-θ1
❑ silicon contamination of Al-Cu alloys. ❑ t1 is the duration of undercooling.

02
 Controlling the structure of aluminium alloys before casting with the thermal analysis equipment THERMATEST 5000 NG III
When the undercooling is high and duration medium (case N°1), grain Al-Si 12% Case Study
size is coarse. Al-Si12% is different from other
When there is no undercooling (case N°2), grain size is very fine. aluminium-silicon alloys because it is a
When undercooling is low but duration is high, the grain size can be binary alloy, which only contains two
very coarse. elements (Al and Si), (beside impurities).
Most other alloys have at least 3 major
Nature of eutectic structure constitutive elements.

Temperature
The composition range is such that the
alloy can have either a hypoeutectic,
θ eutectic, or hypereutectic structure.

θ2 For these reasons, the eutectic level is

completely horizontal and undercooling
θ1
duration has no influence on the
t1
structure.

Time The eutectic level temperature is

sufficient to predict the structure if the:
❑ eutectic temperature above 573 °C
The nature of the Al-Si eutectic phase for an alloy of given composition indicates that the modification is
depends on: insufficient
❑ eutectic level temperature θ2 ❑ eutectic temperature below 570 °C is
❑ undercooling value Δθ characteristic of an over-modified
❑ undercooling duration t1 alloy
❑ alloy is perfectly eutectic, then liquidus
The eutectic level temperature θ2 and eutectic are merging
The lower the eutectic temperature is, the finer the eutectic phase will be. ❑ alloy is hypereutectic, then primary
undercooling is nil or of a small
The undercooling value Δθ duration
❑ is nil when the structure is acicular or needle-like ❑ alloy is hypoeutectic, the primary
❑ rises when the structure becomes lamellar undercooling is visible but of a long
❑ goes through a maximum when the structure is lamellar fine or duration
under-modified ❑ in case the alloy is eutectic or
❑ is low when the structure is over-modified hypereutectic, thermal analysis cannot
predict if the structure is lamellar or
The undercooling duration t1 acicular.
The longer this duration is, the better the eutectic phase modification
will be. The graph below shows the nature of the eutectic phase relative For the hypoeutectic alloys, the
to the three parameters (eutectic level temperature, undercooling value undercooling depends on the
and undercooling duration). germination of silicon, which acts on the
eutectic structure. Only in this case, we
can see a correlation between the
Temperature of eutectic
plateau Length of undercooling structure and the undercooling.

Al-Si-Cu-Mg alloys Case Study

These alloys can exhibit a secondary
Value of undercooling eutectic phase depending on the Cu and
Mg content. This eutectic phase Al-Cu
solidifies around 507°C.

It is important not to forget about this

low temperature eutectic phase at around
Acicular Coarse lamellar Fine lamellar Modified Overmodified 507 °C when carrying out heat treatment
structure structure structure structure structure
on these castings.

03
 Al-Si11%Cu : this alloy belongs into the hypoeutectic
group like Al-Si10%Mg
Application
Al-Si12%Cu : this alloy behaves like Al-Si12% as it can
have either a hypoeutectic or a hypereutectic behaviour. Grain size evaluation
THERMATEST 5000 NG III measures the following
Silicon detection in Al-Cu5%MgTi liquidus parameters :
The appearance of a second eutectic phase Al-Si indicates
the alloy is polluted with silicon. ❑ temperature θ2 (in °C)
❑ undercooling Δθ (in °C)
❑ duration of undercooling t1 (in seconds).
The THERMATEST 5000 NG III System
For these 3 measured parameters and using a
THERMATEST 5000 NG III is a device designed to predict proprietary model developed in-house by ALCAN,
the grain refinement and the type of eutectic structure THERMATEST 5000 NG III is able to calculate a grain size
when casting aluminium alloys. index, which ranges from 1 to 9.

This equipment is capable, within a few minutes, of Grain size reference pictures (Samples ∅ 6,5 mm)
assessing the melt quality before casting and allows the
foundryman to make the correct additions thus avoiding
costly defects such as scrap due to shrinkage, leakage, TEST OF GRAIN REFINING
porosity, and hot tear. Standards plates with
cotation values

Cotation 0,5 Cotation 1

THERMATEST 5000 NG III device

Compensation cable

Up/down arm

Thermocouple fastening

Tightening screw
Cotation 2 Cotation 3
Locking system

Thermocouple

Protective sheath

1/2

1/2
Crucible

Base

Schematic of Test Apparatus

Cotation 4 Cotation 5

04
 Controlling the structure of aluminium alloys before casting with the thermal analysis equipment THERMATEST 5000 NG III
Micrographies (x200) of Eutectic Structures
Figs NR 1 - NR 7

Alloy AI - Si 7 %

Cotation 6 Cotation 7

NR 1 Acicular structure NR 2 Coarse lamellar

structure

Cotation 8 Cotation 9

Grain refinement is considered optimum when the

undercooling is nil and grain size index is equal to 9.
NR 3 Fine Lamellar structure NR 5 Undermodified structure

However, for certain alloys and thin walled castings in

permanent moulds, we can accept a lower grain size
index (5 to 9), because of the higher cooling rate created
in permanent dies. We then recommend setting a
minimum grain size index for each casting, correlated
with desired elongation or mechanical properties.

For Al-Cu5%MgTi alloys, the absence of undercooling

may not be sufficient to avoid hot tears. A stronger grain
refinement is recommended to improve the alloy’s NR 6 Fine fibrous structure NR 7 Overmodified structure
performance.

Nature of the eutectic structure Silicon detection in Al-Cu5MgTi

Similar to the grain size case, THERMATEST 5000 NG III THERMATEST 5000 NG III is able to detect the existence
measures the following eutectic parameters on the of such a secondary eutectic phase and will indicate that
cooling curve: the alloy is contaminated.

❑ The temperature θ2 (in °C)

❑ The undercooling Δθ (in °C)
❑ The duration of undercooling t1 (in seconds)

For these 3 measured parameters and using a proprietary

model developed in-house by ALCAN, THERMATEST
5000 NG III is able to calculate eutectic structure index,
which ranges from 1 to 7.

05
 The composition of the alloy influences the cooling curve,
System Capability therefore to get the most accurate result the specific Si,
Mg and Cu concentrations in the alloy must be taken into
THERMATEST 5000 NG III is designed to analyse the account. THERMATEST 5000 NG III has a calibration
following alloys and compositions: function that allows the exact chemical composition of
the alloy in any given furnace to be inputted before
❑ Al-Si-Mg alloys: analysis thus giving a more precise result.
Al-Si7Mg, Al-Si10Mg (6%<Si<12%) (0%<Mg<0.7%)
❑ Al-Si-Cu alloys : The other functions are :
Al-Si5Cu3 (4%<Si<10%) (0%<Cu<3.5%) (0%<Mg<0.7%)
❑ Eutectic alloys: ❑ easy printing of cooling curve and results on any
Al-Si12 (11.5%<Si<14%) printer installed under Windows XP
❑ Al-Cu-Mg-Ti alloys: ❑ possibility to transfer data into Microsoft Excel
Al-Cu5MgTi (2%<Cu<6%) allowing easy reporting, batch traceability and
❑ Al-Si-Cu-Ni alloys for pistons: charting to identify trends over time
Al-Si11Cu and Al-Si12Cu (10%<Si<14%) ❑ connection to most company networks using
Ethernet card and RJ 45
❑ automatic saving of all data on the hard drive for
Functions future recovery or analysis.

Cost Aspects

THERMATEST 5000 NG III costs very little to run. Indeed

the only consumable items are the steel protection sheath
for the thermocouple and the thermocouple itself.

Conclusion

To maximise the technical benefits and cost effectiveness

of grain refinement and modification, the foundry needs
to ensure that the treatment agents are used at an
THERMATEST 5000 NG III gives a real time optimum and consistent addition rate.
and simultaneous viewing of the cooling
curve as well as the results as shown The use of the THERMATEST 5000 NG III, which is quick
below. and easy to use by foundry personnel gives this facility,
and with quantitative outputs ensures a consistent level
of casting quality. These values can be used to set the
optimum addition rates for all grain refining and
modification additives and thus to avoid costly scrap due
to shrinkage, leakage, porosity and hot tear.

Screen view during analysis

06