New tool for calculation of heat transfer coefficients

Bild 1 2007-07-06

IVF Industrial Research and Development Corporationwww.ivf.se

New tool for calculation of heat transfer coefficients, microstructure and hardness from the cooling curve

Eva Troell, Hans Kristoffersen, Jan Bodin & Sören SegerbergIVF Industrial Research and Development Corporation, Mölndal, Sweden

Imre Felde Bay Zoltán Institute for Materials Science and Technology, Budapest, Hungary

Bild 2 2007-07-06


Contents• Introduction - Testing and evaluation of quenchants and quenching

systems - The ivf SmartQuench™ system

• A new software module for cooling curve analysis and interpretationThe SQi Inverse module - calculation of heat transfer coefficientsThe SQi Property Prediction module - calculation of hardness and microstructure

• Verifying tests

• Application examples

• Summary

Bild 3 2007-07-06


Original functions ivf SmartQuench• Testing of quenching media Cooling curves

• Evaluation of the cooling curve. A large number of characteristics are recorded or calculated:

Cooling times to certain temperaturesCooling rates at these temperaturesMaximum cooling rate

Bild 4 2007-07-06


New software – SQintegrastructure and functions of the ivf SQintegra software

• Calculation of heat transfer coefficients (for the ISO 9950 probe)

• Calculation of hardness and microstructure in a cross section

Bild 5 2007-07-06


Step 1. Measurement of cooling curve

Bild 6 2007-07-06


2. Calculation of heat transfer coefficient

• Inverse calculation

• Direct calculation

• The HTC determined is valid at the surface of the probe at the same height as thethermocouple

Bild 7 2007-07-06


Step 2. Calculation of heat transfer coefficient –first iteration

Measured, centre

Calculated, centre

Calculated, surface

Point for HTC

Bild 8 2007-07-06


Step 3. Calculation of heat transfer coefficient –completed (addition of time steps)

Measured, centre

Calculated, centre

Calculated, surface

Point for HTC

Bild 9 2007-07-06


Step 4. Heat transfer coefficient vs Temperature

Bild 10 2007-07-06


Step 4. Heat transfer coefficient vs Time

Bild 11 2007-07-06


Temperature difference: calculated - measured

Bild 12 2007-07-06


Step 5. Property prediction

Steel grade

HTC

Cylinder radius

Temp of quenchant

Start temp

Bild 13 2007-07-06


Input Steel dataTTT approach (Réti et al. 1981)

Bild 14 2007-07-06


Step 6. Calculation of microstructure and hardness

Diameter 12,5 mm Diameter 20 mm

Steel SAE 1045

Bild 15 2007-07-06


Polymer quenchant - AquatensidCooling curve

Heat transfer coefficient

Bild 16 2007-07-06


Verifying tests: steels SAE1045 and 42CrMo442CrMo4

0

100

200

300

400

500

600

700

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14Radius, mm (0=centre)

Har

dnes

s, H

V

Test, D28Calculated, D28

SAE 1045 (C45)

0

100

200

300

400

500

600

700

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Radius, mm (0=centre)

Har

dnes

s, H

V

Test, D20Calculated, D20Test, D28Calculated, D28Test, D10Calculated, D10

Bild 17 2007-07-06


Cooling curves and Heat transfer coefficients Oil and water; 30 ºC, 70 ºC

Bild 18 2007-07-06


Calculated hardness compared to meassured Steel SAE1045 quenched in water 30 and 70 ºC

Quenched in water 30 and 70 ºC Quenched in oil 30 and 70 ºC

Bild 19 2007-07-06


• Sensitivity analysis - Parameter prediction

Application examples

Bild 20 2007-07-06


Comparison of cooling curves

Calculated hardness

-500

0

500

1000

1500

2000

2500

3000

3500

4000

4500

0100200300400500600700800900

Temperature, C

HTC

, W/m

2C

FeroquenchDurixol ny

HTC

300

350

400

450

500

550

600

650

0 5 10 15 20 25

Distance from center, mm

Har

dnes

s, H

V

F 40 mmF 30 mmF 20 mmF 10 mmF 40 mmD 30 mmD 20 mmD 10 mm

Bild 21 2007-07-06


Quality of the calculations

• The accuracy of the calculated heat transfer coefficient. • The correctness of the used steel dataIt is important to bear in mind that:• The HTC calculations are valid for the probe used for recording the

cooling curve• The cooling curve is valid only under the particular measurement

circumstances. • The normal variation in the chemical composition within different steel

grades will influence the parameters describing the steeltransformation.

• The calculations are one-dimensional.

Bild 22 2007-07-06


The quality of the calculations made by SQintegra in predicting the microstructure and hardness is dependent on:• The accuracy of the calculated heat transfer coefficient. The

difference between the measured and the calculated curve can be analysed in the software. If needed, additional time steps can be added to increase the accuracy.

• The correctness of the used steel data, particularly parameters describing the IT phase transformations and iso-hardness.

• It is important to bear in mind that:The HTC calculations are valid for the probe used for recording the cooling curve. When transferring the HTC to much larger dimensions or other geometries, the effects thereof must be taken into consideration.The cooling curve is valid only under the particular measurementcircumstances. Flow, temperature and concentration may vary in aquench tank and over a period of time.The normal variation in the chemical composition within different steel grades will influence the parameters describing the steel transformation.The calculations are one-dimensional.

Bild 23 2007-07-06


Summary

With the ivf SmartQuench test system and the SQintegra software module it is possible to:

• Continuously perform quality control of the quenching process

• Calculate heat transfer coefficients for simulation purposes

• Predict hardness and microstructure distribution on cylindrical specimens

• Perform sensitivity analysis on quenchants and the quenching system

• Acquire information needed for troubleshooting

Documents

New tool for calculation of heat transfer coefficients