Der Einfluss thermophysikalischer Daten auf die numerische

Transcription

Einfluss thermophysikalischer Daten auf die Simulation
Der Einfluss thermophysikalischer
Daten auf die numerische Simulation
von Gießprozessen
Tagung des Arbeitskreises Thermophysik,
4. – 5.3.2010
Karlsruhe, Deutschland
E. Kaschnitz
Österreichisches Gießerei-Institut
Leoben, Österreich
1
Numerical simulation in metal part production:
¾ Differential equations describing the physical phenomena are
usually highly non-linear and coupled
¾ Analytical solutions are not obtainable
¾ Numerical approximations by means of Finite Methods (e.g.
Finite Elements, Finite Differences, ...)
¾ Need of high computing power (computing times hours, days,
weeks), parallelization of processes
¾ Simulated processes: casting, welding, forging, heat treatment,
rolling, cutting …
¾ Accurate thermo-physical and -mechanical properties are
urgently needed
2
Geometry
Process
at
He
ed
as
re
le
tu
Re
ra te
pe ra
m ng
Te ooli
C
Material
Thermal
Model
Heat transfer
Air gap
Strain
Material
model
Mechanical
Model
ge
n
a
ch n
ed e
e ai
c
s
u g
a tr
nd han
ph s
i
s
es se c
r
t
s ha
p
Shrinkage
Distortion
Residual stress
Cold shots
Microstructure
Segregation
Thermal model: fluid flow, heat transfer, diffusion
Material model: macro-, meso- micro-model
Mechanical model: elasto-, plasto-, visco-model
And Combinations
3
Density and volume expansion of Al-17Si-4Cu:
1.10
2750
Teutectic Tliquidus
Tsolidus
1.08
Density, kg.m
-3
2650
1.06
2600
2550
1.04
2500
Volume Expansion
2700
1.02
2450
2400
1.00
0
100
200
300
400
500
600
700
Temperature, °C
4
Thermal diffusivity of Al-17Si-4Cu:
Thermal diffusivity, mm2 . s-1
70
60
solidus
temperature
50
eutectic
temperature
40
30
20
10
liquidus
temperature
Fit to the measured data
Initial heating
Col 7 vs Col 8
Col 9 vs Col 10
f vs Col 12
0
100
200
300
400
500
600
700
Temperature, °C
5
Influence of volume thermal expansion data on feeding of
castings:
¾ Alloys are (usually) contracting during solidification (up to
several percent)
¾ Shrinkage has to be compensated in order to get a sound
casting
¾ Additional material is provided from an extra reservoir by
hydrostatic or external pressure (feeding)
¾ If feeding is disturbed, shrinkage cavities or porosity will be
found in the final casting – leading to scrap
¾ feeding channels must not been frozen until solidification of
casting (directional solidification towards the feeders)
¾ Precise volume expansion (contraction) data in the vicinity of
the solidification region must be known
6
Example: Liquid and solidification shrinkage due to density
change during cooling – porosity and shrinkage cavity
7
Example: Solidification of a gear housing
8
Volume thermal expansion (contraction) depends also on:
¾ Exact chemical composition – an alloy designation has (wide)
composition ranges for each element
¾ Solidification speed – competition between thermal and mass
diffusion (lever rule, Scheil, back diffusion models)
¾ Global vs. local equilibrium at solidification front – development
of different phases
¾ Melt treatment (inoculation, grain refinement, modification,
degassing
9
Influence of linear thermal expansion data on distortion of
castings:
¾ When solidified, the casting shrinks from its high temperature
dimensions to room temperature
¾ In the mold, the (relatively soft) casting is more or less
constrained and follows the dimensions of the (stiff) mold
(plastically deformed at high temperature)
¾ When the casting is shaked out of mold, it can shrink freely
¾ Different regions of the casting are at different temperatures at
shake-out – shrinkage is locally different
¾ The local shrinkage (elastic part) equals to thermal expansion x
local temperature difference to room temperature
¾ Each region has an individual shrinkage – that can lead to
distortion (plastic contributions ease the stress but increase
distortion)
10
Example: V-shaped high-pressure die-casting – temperature
distribution until shake-out
11
Example: V-shaped high-pressure die-casting – distortion in xdirection during free cooling
12
V-shaped high-pressure die-castings are produced at ÖGI
High-pressure die-casting machine
V-shaped specimens
13
Dimensions of the V-shaped specimens are measured and
compared to simulation
14
Influence of linear thermal expansion data on distortion of
castings:
¾ Different regions of the casting are at different temperatures at
shake-out – shrinkage is locally different (plastic strain can
also happen in the die – some stress is already present at
shake-out)
¾ The local shrinkage leads to distortion
¾ After machining, the stress distribution changes – also the
distortion
¾ Similar problems arise at heat treatment of metal parts
15
Influence of linear thermal expansion and thermal conductivity
data on die life expectance (high-pressure die-casting):
¾ After filling of the cavity the hot melt is in direct contact with the
cooler tool steel
¾ The steel surface is heated rapidly, the material expands in the
vicinity of the surface
¾ The surface layer gets under heavy compressive stress – can
be beyond the yield stress
¾ This leads to plastic deformation of the surface layer
¾ After removing of the casting, a lubricant-water mixture is
sprayed at the surface – rapid cooling
¾ The cooled surface layer gets under tension stress – the
material is now “too short” – cracking of the surface under
tension stress
16
Temperatures at a tool-steel die-surface in high-pressure diecasting:
800
600
Casting in die
Air
Spray Air
cooling cooling cooling
600
500
400
200
0
300
-200
200
-400
Cut of a fraction of die
and melt
100
Die closed
Die open
-600
Die open Die closed
0
-800
0
10
20
30
40
50
60
Time, s
17
Stress, MPa
Temperature, °C
400
Tool-steel die in high-pressure die-casting during solidification:
After 1 second
Temperature
Normal stress in vertical direction
18
After 5 seconds
Temperature
19
After 20 seconds
Temperature
20
Tool-steel die in high-pressure die-casting at open die:
After 30 seconds
Temperature
21
Tool-steel die in high-pressure die-casting during spraying:
After 40 seconds
Temperature
22
Tool-steel die in high-pressure die-casting at the end of cycle:
After 55 seconds
Temperature
23
Examples of die damage:
Thermal fatigue cracks
Fracture of the entire die
2
1
10mm
24
Influence of linear thermal expansion and thermal conductivity
data on die life expectance simulation (high-pressure die-casting):
¾ Plastic deformation at heating leads to cracking during rapid
cooling
¾ Smaller linear thermal expansion lowers thermal stress
¾ Higher thermal conductivity removes heat faster – smaller
temperature gradients – lower thermal stress
¾ just a few percent change in thermal conductivity of the tool
steel changes peak temperature at the surface for several
degrees – that can have a tremendous influence in die life
expectance (up to a factor of 10)
¾ Search for a tools steel with higher thermal conductivity is
ongoing
25
Warmarbeitsstahl von Rovalma HTCS-130:
Wärmeleitfähigkeit, W/Km
60
40
20
Rovalma HTCS-130
Datenblatt STM Stahl
1.2343 (W300 ISODISC)
W360 ISOBLOC
0
0
100
200
300
400
500
600
700
Temperatur, °C
26
Influence of thermal conductivity on microstructure simulation of
castings:
¾ Mechanical properties (aluminum, magnesium) depend on
¾Melt cleaning, degassing, grain refinement, modification
¾Correct fluid flow, feeding technique
¾ Solidification speed (thermal conductivity of the mold)
¾ Rapid solidification leads to fine grain and small secondary
dendrite arms in the microstructure – superior mechanical
properties
¾ Local cooling of crucial regions of a casting
27
Simulation of the cooling time:
Casting with grey iron core
Casting with quartz sand core
28
Microstructure (position 7 mm from surface):
Casting with grey iron core
Casting with quartz sand core
DAS 20 µm
200µm
DAS 61 µm
200µm
29
Conclusions: How in(accurate) thermal conductivity and thermal
expansion data influences the quality of casting simulations
¾ Volume thermal expansion of the solidifying melt can lead to
shrinkage porosity – with inaccurate data cavities can be
predicted when casting is sound and vice versa
¾ Linear thermal expansion is responsible for thermal strain –
with inaccurate data the final distortion of the casting and
residual stresses are not correctly predicted
¾ Thermal conductivity of the mold determines the heat removal
from the solidifying casting – with inaccurate data the
microstructure will develop different in simulation and reality
¾ Thermal conductivity and linear thermal expansion of a die are
responsible for the amount of (cycling) compression and
tension stress – inaccurate data lead to a completely wrong
prediction of cracking (lifetime)
30
31

Der Einfluss thermophysikalischer Daten auf die numerische

Transcription

Similar documents

WARMARBEITSSTAHL HOT WORK TOOL STEEL

Programm - Aluminium Engineering Center Aachen