1

DÉTERMINATION DU TAUX DE CARBONE PAR ULTRASONS ULTRASONIC DETERMINATION OF CARBON CONTENT

A.Badidi Bouda, R. Halimi et W. Djerir

Division de Caractérisation et d’Instrumentation, Centre de Recherche et Technique en Soudage et Contrôle (CSC), Route de Dély Brahim, B. P.64 ChéragaALGERIA, Tel et fax :

213 21 36 18 50 Email : alibadidi@yahoo.fr

Résumé Dans ce papier, nous proposons une étude expérimentale de l’effet du taux de carbone des aciers faiblement alliés sur la vitesse de propagation et le coefficient d’atténuation des ondes ultrasonores à travers ces matériaux. Nous avons observé des relations simples entre les vitesses, les atténuations et le taux de carbone. Les mêmes observations peuvent être faites avec le module d’Young. Ces résultats, en accord avec la théorie, montre la possibilité d’une caractérisation du taux de carbone par une méthode non destructive: les ultrasons. En parallèle, nous avons étudié l’effet de certains traitements thermiques tels que la trempe et le recuit sur les vitesses et les atténuations. Les résultats obtenus montrent une corrélation entre les traitements thermiques et les paramètres ultrasonores. Ceci ouvre la voie à une caractérisation complète et non destructive des aciers par des méthodes ultrasonores.

Abstract In this paper we are proposing an experimental study of the effect of low alloy steels carbon content on the velocity and propagation attenuation coefficient of the ultrasonic waves in these materials. We have observed simple relations between the velocities and the attenuations according to the carbon content. The same observations can be made for the Young modulus. These results, in conformity with the theory, show the possibility of characterizing the carbon content by a nondestructive method: ultrasounds. In parallel we have studied the effect of some heat treatments such as hardening, annealing and quenching on velocities and attenuations. The results obtained show a correlation between heat treatments and ultrasonic parameters. This opens ways to a more complete and nondestructive characterization of steels by ultrasonic methods.

Introduction An ultrasonic wave undergoes modifications during propagation through a material. This modification is directly related to the intrinsic characteristics of the propagation medium [1]. Such as, the nature of the material, its structure, its state of stress as well as the undergone heat treatments. Each one of the material’s characteristics influences one or more of the ultrasonic parameters such as the propagation velocity or the attenuation coefficient of one or several modes of propagation. In this paper, we study the effect of the carbon content of low alloy steels on the ultrasonic parameters.

2

Low Alloy Steels Steel is an alloy of iron and carbon with the maximum carbon content limited to 2 %. It can contain small amounts of other elements incorporated intentionally or not during its development. One can also add larger quantities of alloying elements. It is then regarded as an alloy steel. The carbon percentage of some of these alloy steels can sometimes exceed 2 %. On the contrary to the cast iron, which contains more than 2% carbon, steel is a ductile metal. It can undergo changes in form by hot/ cold compression or extension. Steel can also be hardened under the effect of treatments other than hardening, for example by cold hammering. The components of steel include carbon, silicon, and manganese. Sulphur and phosphorus, as well as oxygen, are harmful impurities, even with a low content. In many special steels, one can meet nickel, molybdenum and vanadium. Principal element of steel, carbon has a large influence on the physical and mechanical properties. With increase in its content, it offers the possibility of steels hardening, therefore by carbide formation, and reinforces with this fact the wear resistance. The aptitude of hardening of the steel increases as the percentage of carbon increases, whereas ductility, the aptitude for forging, welding and machining decreases. The standardized designation of low alloyed steels affects the addition elements of less than 5% of the most important alloy element.

Hardening Hardening is defined as an operation which consists in subjecting metal to an adapted thermal cycle, including successively:

Heating, intended to dissolve the alloying elements in solid solution at high temperature in the stable phase of austenite.

An adapted cooling mode, carried out starting from a temperature known as of hardening or austenitization temperature, until a lower temperature, and which can be different from the ambient temperature.

Samples We chose three samples of five different steel nuances. The samples are cylindrical with 50 mm diameter and of 20mm thickness. In order to quantify the composition in elements of the samples, an X-ray fluorescence spectrometric analysis was carried out. The results are given on table 1.

Samples C Mn Si P S Cu Al Ti Ni Cr

Mo

V Sn

1

0.192

1.170

0.298

0.020

0.026

0.128

0.0179

0.002

0.083

1.071

0.019

0.003

0.009

2

0.268

1.247

0.213

0.029

0.014

0.041

0.0291

0.002

0.054

0.027

0.010

-

0.005

3

0.344

0.418

0.230

0.020

0.012

0.203

0.0497

0.004

3.475

1.798

0.261

0.007

0.007

4

0.385

0.653

0.177

0.015

0.009

0.134

0.0034

0.002

0.066

0.030

0.011

-

0.015

5

0.439

0.421

0.247

0.016

0.012

0.038

0.034

0.003

3.659

1.294

0.236

0.003

0.003

Table.1. Chemical composition of samples

3

The selected nuances are low alloy steels and usually used in industry. Their designations are respectively: 45NCM16, XC38, 35NCD16, 28M6 and 20MC5.

Metallographic analysis The metallographic study of the samples from the rough state with various magnifications has revealed a structure of hypoeutectoïdes steels with 2 micro structural states: → a tempered microstructure for the 45NC16 and the 35NCD16 (fig. 1 and 3). → a ferritic- perlitic microstructure (fig. 2, 4 and 5)

Fig.1. Tempered microstructure of 45NCM16 at 0.439 %C

500 1000

Fig.2. .Microstructure of XC38 (Ferrite + Pearlite) at 0.385 % C

200 500

Fig.3. Tempered microstructure of 35NCD16 at 0.344 %C

500 1000

4

Experimental device The experimental device is composed of an ultrasonic non destructive testing system of materials by immersion: the sample is immersed in a water tank, to analyze the sample under various incidences of sound wave. This manual / automated bench makes it possible to undertake the various stages of characterization of a defect under optimized conditions (detection, discrimination and dimensioning) in the sample (fig. 6).

Fig.4.Microstructure of 28M6 (Ferrite + Pearlite) at 0. 268 % C

1000 500

Fig.5. Microstructure of 20MC5 (Ferrite + Pearlite) at 0.192 %C

200 500

Digital

scope

CONTROL

Transmitter

- receiver

GPIB

PC

echo

tank

water

sample

Z axis

Y axis

X axis axis

Fig.6. Experimental device

5

This system, piloted by a computer, allows acquisition and processing of the ultrasonic signals obtained through the samples.

Ultrasonic parameters measurements VL, L, VT, T : The longitudinal wave velocity (VL) measurement is made by immersion using a 5 MHz frequency probe. The transverse velocity (VT) measurement is carried out in contact mode using a 4 MHz transverse wave transducer. The attenuation coefficients of ultrasonic

longitudinal ( L) and transverse ( T) waves are deduced from three successive basic echoes through the sample [2, 3 and 4]. We give as an example the signals obtained by the longitudinal (fig.7) and transverse (fig.8) waves through the steel sample of 28M6 nuance.

Fig, 8, Transverse wave signal ( sample 28M6)

-0,3

-0,2

-0,1

0

0,1

0,2

0,3

-4,00E-05 -2,00E-05 0,00E+00 2,00E-05 4,00E-05 6,00E-05 8,00E-05 1,00E-04

Time (s)

Am

pli

tud

e (

V)

Fig.8. Transverse wave signal (sample 28 M6) obtained in contact mode

6

Experimental results

The experimental results on the velocities measurements and attenuation coefficients of the longitudinal and transverse waves for the steel samples are given in the tables 2, 3, 4 and 5.

%C Longitudinal velocity Vl (m/s)

0.192 5925 ± 15

0.268 5929 ± 15

0.344 5952 ± 15

0.385 5930 ± 15

0.439 5959 ± 15

1.708 6103 ± 16

%C Coefficient αL (dB/mm)

0,192 0.142 ± 0.010

0,268 0.132 ± 0.009

0,344 0.116 ± 0.008

0,385 0.128 ± 0.009

0,439 0.105 ± 0.007

1,708 0,091 ± 0.006

%C Transverse velocity VT (m/s)

0.192 3206 ± 11

0.268 3212 ± 11

0.344 3231 ± 11

0.385 3219 ± 11

0.439 3232 ± 11

1.708 3309 ± 11

%C Coefficient αT (dB/mm)

0,192 0.174 ± 0.007

0,268 0.169 ± 0.005

0,344 0.166 ± 0.005

0,385 0.168 ± 0.008

0,439 0.163 ± 0.008

1,708 0.141 ± 0.004

Table.4. VT variation with carbon content

Table.2. VL variation with carbon content Table.3. αL variation with carbon content

Table.5. αT variation with carbon content

7

5850

5870

5890

5910

5930

5950

5970

5990

6010

6030

6050

6070

6090

6110

6130

0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6 1,8

lon

git

ud

inal velo

cit

y V

L (m

/s)

%CFig.9. VL variation with carbon content

8

Longitudinal (fig.9) and transverse (fig.11) velocities take an ascending way when the percentage of carbon increases. On the contrary, longitudinal (fig.10) and transverse (fig.12) attenuation coefficients decrease with the increase in the carbon content.

Elastic constants: E, ν From ultrasonic velocities VL, VT and the density, it is easy to know the Young modulus E or rigidity module G as well as the Poisson ratio ν, using the equations (1) and (2). Vℓ =(E /ρ* (1-ν)/ (1+ν)*(1-2ν)) ½ , Vt =(E /ρ * 1/2(1+ ν)) ½ VT = (G/ρ) ½ et VL/VT = [2(1-ν) / (1-2 ν)] ½ The results are given in table 6.

(2)

(1)

9

Concerning the Young modulus, we note a decrease of the Young modulus with the increase in the carbon content.

Vickers hardness For each nuance of steel, we have experimentally measured Vickers [5] hardness. The results are given in table7.

Sample % C ν E (N/m2)

0.192 0.292 215*109

0.268 0.292 208*109

0.344 0.291 212*109

0.385 0.291 207*109

0.439 0.294 194*109

1.708 0.293 198*109

Sample

%C

d1 ( m) d2 ( m) Hv Hardness.

0.192 103.6 103.3 173.6

0.268 103.1 102.3 174.1

0.344 85.8 86.2 250.7

0.385 101.3 101.8 181.7

0.439 79.6 82 284

1.708 87.4 85.9 246.9

Table.6. Elastic Constants of samples

Table.7. Vickers hardness variation with carbon content

10

It is noted that Vickers hardness increases when the percentage of carbon increases (fig. 13). At ambient temperature, steels (Z200C1, XC38, 28M6 and 20MC5) are made principally of mixtures of ferrite and cementite, and the voluminal fraction of the two phases depends on the percentage of carbon. With increase in the carbon content, the volume fraction of carbides is raised leading to increase in the mechanical resistance. The increase in the carbon concentration increases the steels hardness but decreases ductility. We can however conclude; that the presence of carbon in solid solution of insertion seems to be the simplest case of hardening to be envisaged by the techniques of steels characterisation, the rise in the carbon content leads clearly to an evolution of hardness, this evolution of hardness influences the ultrasonic parameters. Hardness makes easy the propagation the speed of the propagating media, while decreasing the attenuation of the energy ultrasonic of both modes.

Conclusion We have studied the interaction between the ultrasonic waves and various carbon contents steel samples. We have measured in experiments propagation velocities and the attenuation coefficients of the longitudinal and transverse waves through these materials. We have showed a correlation between the carbon content in steel and the various ultrasonic parameters, particularly for the propagation velocity of the longitudinal mode which is a parameter whose measurement is easy. It will be interesting to study steels within the range of (0.5-1.6) %C what we couldn’t do because we didn’t have the samples. This study is qualitative. It will be interesting also to complete by a quantitative study with more samples and steel nuances with larger range of carbon content to high light an ultrasonic method to reliably determine the carbon content and to avoid using more expensive classical means. The ultrasounds have proved that they can be an important tool for the experimental determination of the steel carbon content.

Références [1] A. Badidi Bouda A. Benchaala & K. Alem “Ultrasonic characterization of materials hardness”, Ultrasonic, vol.38, 2000, pp224-227 [2] A. Badidi Bouda, S. Lebaili & A. Benchaala, “Grain Size Influence on Ultrasonic Velocities and Attenuation”, NDT&E International, January 2003 [3] C. Gracier et B. Horsten,” Simultaneous measurement of speed, attenuation, thickness and density with reflected ultrasonic waves in plates, IEEE, ultrasonic Symposium, 1994, pp 1219-1222 [4] F. Peters and L. Petit, “a broad band spectroscopy method for ultrasound wave velocity and attenuation measurement in dispersive media”, ultrasonics 41 (2003), 357-363 [5] J Rivenez, A Lambert, Mesure et appréciation non destructives des gradients de dureté. , CETIM Informations n° 97, Octobre 1986