Optimization of Weld Bead Parameters of Nickel Based Overlay Deposited by Plasma Transferred Arc Surfacing with Adequacy Test

Bhaskarananda Dasgupta Int. Journal of Engineering Research and Applications www.ijera.com
ISSN : 2248-9622, Vol. 4, Issue 7( Version 1), July 2014, pp.201-206
www.ijera.com 201 | P a g e
Optimization of Weld Bead Parameters of Nickel Based Overlay Deposited by Plasma Transferred Arc Surfacing with Adequacy Test Bhaskarananda Dasgupta*, Pinaky Bhadury ** *(Department of Engineering Sciences & Humanities, Siliguri Institute of Technology, Siliguri, W.B) ** (Department of Engineering Sciences & Humanities, Siliguri Institute of Technology, Siliguri, W.B) ABSTRACT Plasma Transferred Arc surfacing is a kind of Plasma Transferred Arc Welding process. Plasma Transferred Arc surfacing (PTA) is increasingly used in applications where enhancement of wear, corrosion and heat resistance of materials surface is required. The shape of weld bead geometry affected by the PTA Welding process parameters is an indication of the quality of the weld. In this paper the analysis and optimization of weld bead parameters, during deposition of a Nickel based alloy Colmonoy on stainless steel plate by plasma transferred arc surfacing, are made and values of process parameters to produce optimal weld bead geometry are estimated. The experiments are conducted based on a five input process parameters and mathematical models are developed using multiple regression technique. The direct effects of input process parameters on weld bead geometry are discussed using graphs. Finally, optimization of the weld bead parameters, that is minimization of penetration and maximization of reinforcement and weld bead width, are made with a view to economize the input process parameters to achieve the desirable welding joint.
Keywords - Adequacy test, Colmonoy; Optimization; Penetration; PTA welding; Reinforcement; Stainless steel 316L;
I. INTRODUCTION
Weld deposition of hard facing alloys is commonly employed to enhance the tribological life of Engineering components subjected to hostile environments. The clamation of worn-out metal parts is demanded worldwide and for this demand PTA hard facing of hard, wear resistant thin surface layer of metals and alloys on suitable substrates is one of the proven surfacing techniques(7).In the recent years, PTA Surfacing is extensively used in applications such as valve industries, Hydraulic machineries, Mining industries, Earthmoving equipment, Chemical, Nuclear and Thermal power plants etc.PTA process can be considered as an advanced Gas Tungsten Arc welding process more widely used for overlay applications. The metal and alloy powder is carried from the powder feeder to the central electrode holder in the arc-gas stream. From the electrode holder the powder is directed to the constricted arc zone, where it is melted and fusion bonded to the base metal. Thus, smooth, thin deposits of overlays can be made through this way of precise control of feed stock by PTA welding process. Colmonoy5, a Nickel based alloy (Ni-Cr-B-Si-C) provides excellent resistance to abrasive and adhesive wear with resistance to corrosion and high temperature oxidation. It is widely applied in Thermal power plants, Chemical industries, Nuclear reactors,
food processing industries etc. In nuclear reactors, use of Co based satellite alloys lead to induced activity which will harm the personnel involved in the maintenance of reactor components. Colmonoy5 is a good alternative to the Co based alloys used (9, 15) in nuclear applications, fabricated wearers instant bushes are made of hard facing alloys for high temperature applications. They report that there is a growing tendency to replace Co based alloys by Nickel based alloys, due to the high cost and scarcity of cobalt based alloys. The problem of cracking susceptibility of nickel based Colmonoy alloy hard faced deposits can be controlled by using preheating and slow cooling during deposition. Compared to other welding processes PTA surfacing requires less quantity of material to be deposited with improved mechanical and metallurgical properties.
The shape of the weld bead geometry is affected by the values of PTA process parameters used in deposition. These process parameters should be well established and categorized to enable automation and robotisation of PTA surfacing. Therefore it is imperative to develop mathematical models to predict the bead geometry and study the direct effects of various PTA process parameters on the weld bead shapes. The selection of welding procedure must be more specific to ensure that adequate bead quality is obtained(8).Further a complete control over the process parameters is essential to produce quality
RESEARCH ARTICLE OPEN ACCESS

welds with required bead geometry based on which the integrity of the weldment is known. It has been reported by researchers that in PTA surfacing, process quality can be represented by bead shape (11). Thus the weld pool geometry plays an important role in deter mining the mechanical and corrosion properties of the weld. Therefore, process parameters for obtaining optimal weld-pool geometry. It is reported that five level factorial techniques can be employed for developing models to predict weld bead geometry(1,5,6,12,13).In this work, automatic PTA hard facing is carried out for depositing Colmonoy5 over Stainless steel 316L plates of size 150mm x 90mm x 30mm.The experiments are based on the central composite rotatable design matrix. Regression analysis was used to develop the model. Also, the main objective of this study is to optimize the weld bead parameters that is achieving minimum penetration, maximum reinforcement and maximum bead width which gives best weld joint and determining the corresponding process parameters. Penetration means the depth into substrate through which the filler metal penetrates. Reinforcement means height of filler metal on substrate. Weld width means the width of filler metal on substrate.
II. THE WORK TO BE CARRIED OUT WAS PLANNED IN THE FOLLOWING ORDER
1. Identification of important process parameters 2. Finding the upper and lower limits with different levels of the identified process parameters 3. Development of Design matrix. 4. Conducting experiments as per the design matrix 5. Recording the responses 6. Calculation of regression coefficients and development of mathematical model 7. Development of final mathematical model 8. Calculation of adequacy Test. 9. Presenting the relation of process variables on bead geometry in graphical form. 10. Results and discussions. 11. Optimization of weld bead parameter.
2.1. IDENTIFICATION OF PROCESS PARAMETERS-
The independently controllable process parameters identified based on their significant effect on weld bead geometry to carry out the experimental work are welding current(A),Oscillation width(O),Travel speed(S),Preheat temperature(T) and Powder feed rate(F) gas flow rate and Torch standoff distance. Welding current produce heat for fusion, Pre heat temperature and oscillation width (17) which may affect crack formation during hard facing, have to be properly controlled. Minimum torch travel speed gives better deposition of filler material on welding joints. In deposition of material the most effective process parameters are Welding current, Oscillation width, Travel speed, Preheat temperature and Powder feed rate, thus these process parameters are selected in the experiment and the Gas flow rate and Torch standoff distance are kept at constant levels.
2.2. LIMITS AND LEVELS OF PROCESS PARAMETERS-
It has been observed from different experiments that best deposition of nickel on steel plate occurs when the process parameters vary between the levels given in table 1.The working ranges of all selected parameters are fixed by conducting trial runs. This are carried out by varying one parameter while keeping the rest of them at constant values. The working range of each process parameter is decided upon by inspecting the bead for a smooth appearance without any visible defects. The upper limit of a factor is coded as +2, in between levels is -1, 0,1and the lower limit is coded as-2. The chosen levels of the process parameters and notations are given in Table 1. Table 1: Levels of Process Variable
2.3. DEVELOPMENT OF DESIGN MATRIX-
The design matrix is developed according to the combinations normally taken in Genetic Algorithm process for five process parameters. The design matrix chosen to conduct the experiment is a central composite rotatable design. It consists of 32sets of coded conditions and comprising a half replication of 24=16 factorial design with 6centre points and 10 star points .All the welding parameters at the middle level (0) constitute centre points whereas the combination of each welding parameter at its lower value (-2) or higher value (2) with the other four parameters at the middle levels constitute the star points (2, 10, 14). Thus the 32 experimental runs allowed the estimation of linear, quadratic and two-way interactive effects of the process parameters on the weld bead geometry.
2.4. EXPERIMENTAL SETUP AND PROCEDURE-

The experiments are conducted by using a PTA welding machine, fabricated by National Automation Calcutta. Experiments are conducted by company and corresponding data are taken from their experiments. The PTA system has six modules namely traverse carriage unit, oscillator unit, powder feed unit, water- cooling unit, turntable unit and torch unit. All the units are supporting the effective functioning of the total PTA system. The experiments are conducted according to the design matrix at random to avoid systematic errors creeping into the system. Colmonoy 5 is deposited over stainless steel 316L plates 15cm x 9 cm.x3cm. Torch standoff distance, oscillating frequency, plasma/central gas flow rate, shielding gas flow rate and powder/carrier gas flow rate are kept constant respectively at 10mm, 72cycles per minute, 3.5 lpm, 12 lpm and 1.5 lpm while hard facing.
2.5. EXPERIMENTAL OBSERVATION-
The hard faced plates were cross sectioned at their midpoints to get the test samples. Those samples are prepared by the usual metallographic polishing methods and etched with aquaregia solution for carrying out weld bead geometry measurements. The profiles of the weld beads and the bead dimensions i.e. Penetration (P), reinforcement (R) and weld width (W) are measured. The observed values of P, R and W are given in Table: 2.
Table 2: Design Matrix and Observed Values of Weld Bead Dimensions
III. DEVELOPMENT OF MATHEMATICAL MODELS
The response function representing any of the weld bead dimensions like penetration ,reinforcement etc, can be expressed as Y=f(A,O,S,T,F),Where Y is the response or yield .The second order polynomial (regression equation) use to represent the response surface for k factors are given by: Y=b0+ bixi ki =1+ bii xi2ki =1i≠j+ bijki ,j=1xixj (1) 3.1. THE SELECTED POLYNOMIAL FOR FIVE FACTORS CAN BE EXPRESSED: Y = b0 +b1A + b2O +b3S + b4T + b5F +b11A2+ b22O2 + b33S2 +b44T2 + b55F2 + b12AO+b13AS + b14AT + b15AF + b23OS +b24OT+b25OF + b34ST +b35SF +b45TF (2) where b0 is free term of the regression equation, the coefficients b1, b2, b3, b4, and b5 are linear terms, the coefficients b11, b22, b33, b44, and b55 are quadratic terms, and the coefficients b12, b13, b14, b15, b23, b24, b25, b34, b35 and b45 are interaction terms. 3.2. DEVELOPMENT OF FINAL MATHEMATICAL MODEL: The co-efficient of the polynomials has been found by using MATLAB. Software. The final mathematical models determined by the regression analysis are as follows: P=0.95063+0.252A−0.038O−0.107S− 0.063T−0.134F+0.027A2+0.0368 O2− 0.025S2+0.111T2−0.016F2+0.072AO− 0.028AS+0.159AT−0.027AF+0.110OS− 0.091OT−0.064OF+0.122ST+0.092SF+ 0.023TF (3) R=3.514−0.003A−0.179O−0.074S+ 0.020T+0.048F−0.049A2−0.014O2− 0.045S2−0.074T2−0.012F2+0.089AO− 0.026AS+0.036AT+0.051AF+0.053OS+ 0.043OT−0.059OF (4) W=18.96+0.856A+1.11O−0.346S− 0.220T+0.093F+0.164A2+0.164O2+ 0.189S2+0.033T2+0.064F2+0.253AO− 0.303AS+0.284AT−0.088AF+0.212OS− 0.080OT−0.171OF+0.105ST+0.1714SF− 0.165TF (5) It was found that the reduced models were better than the full model, because the adjusted R square values and the standard error of estimates of reduced models were higher and lower respectively than the full models. The value of adjusted R square and standard error of estimates are given in the Table 3.

Table: 3 Comparison of Square multiple R values and
standard error for full and reduced models:
IV. CALCULATING ADEQUACY OF
THE MODELS DEVELOPED:
The adequacies of the models were calculated using
ANOVA Technique. If the calculated value of F ratio of
the models exceeds the standard tabulated value of the F
ratio for a desired level of confidence say (95%), the
model can be considered adequate within the confidence
limit. The calculated results of ANOVA are given in the
Table 4.
Table: 4 Calculations for Adequacy test:
V. RESULTS AND DISCUSSION
The developed mathematical models can be
used to predict the dimensions of the weld bead.
Based on these models, the effects of process
parameters on bead dimensions are computed and
plotted by varying one parameter and keeping
others at middle value.
Figure 1: Direct effect of process variables on
penetration
5.1. Direct effects of Process variables on
Penetration:
Figure 1 shows that Penetration (P) increases
significantly when welding current increases. This is
attributed to the fact that heat input to the base metal
increases when current is increased. Penetration
decreases as Oscillation width increases. This is due to
spreading of heat resulting from more melting of base
metal. Penetration decreases steadily with the increase
of Travel speed as less amount of powder is deposited
per unit length of bead. Penetration decreases to a lower
value when pre heat temperature increases but
afterwards penetration increases with further increase of
preheat temperature. This may be for the reasons that at
lower preheat temperature the heat received from
plasma arc will not spread in the stainless steel substrate
due to lower thermal conductivity resulting in
cushioning of arc. Penetration decreases with increase of
Powder feed rate.
Figure 2: Direct effect of process variables on
reinforcement
5.2. Direct effect of Process variables on
Reinforcement:
It is evident from Fig 2 that Reinforcement (R)
increases initially and then decreases with increase of
welding current. Reinforcement decreases with increase
in oscillation width. This could possibly due to the fact
that the deposited metal gets distributed along the width
resulting in decreasing in reinforcement. Reinforcement
decreases steadily from highest value with the increase
of travel speed. This is due to reduce amount of powder
deposit per unit length of bead. Reinforcement increases
with increase of preheat temperature to its peak value
then decreases with further increase of preheat
temperature. Reinforcement increases steadily with
increase of powder feed rate as more powder is
deposited per unit length of weld bead.

Figure 3: Direct effect of process variables on weld
width
5.3. Direct effects of Process variables on
Weld width:
Figure 3 shows that Weld width increases with
increase of welding current .This is due to spreading
of heat resulting of melting of more metal. Weld
width increases steadily with increase in oscillation
width due to decrease of Penetration. Weld bead
decreases with increase of torch travel speed .This
may be for reduced heat input per unit length of weld
bead when travel speed is increased. Weld bead width
reduces as the Preheat temperature increases. Weld
width decreases initially but increases with increase
of powder feed rate.
A better weld should have bead parameters as follows
1. Penetration should be minimum.
2. Reinforcement should be maximum.
3. Weld width should be maximum.
The bead parameters are found independently
considering their equations obtain from mathematical
modeling as objective function. The best weld bead
parameters are found by analyzing the graph between
weld bead parameters and process parameters .The best
weld bead parameters and the corresponding process
parameters are as follows:
a) Minimization of depth of penetration-
Optimum value of process parameters:
Predicted response for penetration Pmin= 0.463mm
b) Maximization of height of reinforcement-
Predicted response for reinforcement
Rmaximum=4.29mm
c) Maximization of weld width-
Predicted response for weld width
Wmaximum=26.58mm
Hence the best of bead parameters for different sets
of process variable is as follows:
1. Penetration=0.463mm
2. Reinforcement = 4.29mm
3. Weld width = 26.58mm
VI. OPTIMIZATION
It is very difficult for an operator to perform the
welding process to get best bead parameters using
different sets of process variable. Hence it is necessary
to find a single set of process variable which will
provide the optimal bead parameters and are very close
to the best bead parameters find with different sets of
process variable. Here optimization of the bead
parameters is done by considering the three sets of
process variable obtained for best bead parameters as
mentioned above. All possible combinations of process
variables are carried by using MATLAB software and
the corresponding bead parameters are recorded,
neglecting negative ones. It can be seen that there are 24
sets of valid process variables giving as many sets of
bead parameters. It is found from results that optimum
set of bead parameter is as follows:
The process variables and corresponding optimum
bead parameters are given below:
Table 5: Optimized values of bead parameters and
corresponding process variables

Condition for better welding joint: 1. Higher amperage produces higher welding temperature which interns produces better fusion of metal. Better fusion produces better weld 2. More oscillation width produces more weld width. 3. Less travel speed of torch produces better reinforcement. 4. Medium preheat temperature reduces the chance of cracking. 5. More powder feed rate produces more reinforcement. According to the theory the best combination of process variables for optimum bead parameters should be as follows Table 6: Combination of process variables for optimum bead parameters
VII. CONCLUSION
Accordingly the predicted bead parameters for this set are as follows Penetration is 0.640 mm, Reinforcement is 3.1358mm Weld bead is 25.768 mm. This is exactly the same set of bead parameters obtain by the optimization process. Hence it can be said that the optimization process is justifying the theory.
VIII. ACKNOWLEDGMENT
The authors wish to thank Dr. Sudip Mukherjee (HOD, Mechanical Engg Dept. Jalpaiguri Govt. Engineering College). REFERENCES [1] T.T.Allen, R.W.Richardson, D.P.Tagliable, G.P. Maul, Statistical processdesignforrobotic GMAweldingofsheet metal, Welding Journal 81/5(2002)69-172. [2] W.G. Cochran, G.M. Cox, Experimental Designs, John Willy & sons Singapore 1957 [3] C.R. Das, S.K. Albert, A.K., Bhaduri, G. Kempulraj, A novel procedure for fabrication of wearresistantbushes for high temperature application, Journal of Materials Processing Technology141(2003)60-66. [4] H. Eschnauer, Hard material powders and hard alloy powders for plasmas face coating, Thin Solid Films73 1980 pp 1-17. [5] B. Howard, Surfacing for wear resistance, Stephen Helbaet al. (eds.), Modern Welding Technology, Prentice HallInc., New Jersey, 2022pp 721-726
[6] I.S.Kim, J.S.Son, Y.J.Jeung, Control and Optimisation of bead widthformulti-passwelding inroboticarcwelding processes, Australian Welding Journal 46 (2001)43-46 [7] R.Kaul,P.Ganesh,S.K.Albert,A.Jaishwal, N.P.Lalla,A. Gupta C.P.Pal ,A.K.NathLasercladdingof austenitic stainless steel with nickel base hard facing alloy, Surface Engineering19 (2003)pp269-273. [8] K.Marimuthu, N.Murugan Prediction and optimization of weld bead geometry of Plasma transferred arc hard faced valve seat rings, Surface Engineering19/2(2003) 143-149. [9] J.C.Mc.Glone, Weldbeadgeometryprediction – areview, Metal Construction14 (1982)378-384. [10] D.C. Montgomery, Designand analysis of experiments, John Wiley&SonsInc., New York, 2001. [11] P.K. Palani, N. Murugan, Optimizationofweld bead geometry of stainless steel cladding by flux Cored arc welding using Excel Solver, IWS Journal 2 (2005)15-19. [12] A. Scotti, L.Alves, Albuquerque Rosa, Influence of oscillation parameter soncrack for mationin automatic Fe-B hard facing, Journal of Materials Processing Technology 65 (1997)272-280. [13] K.Y.Benyounis and A.G.OlbiOptimization of PTA welding process using Statistical and Numerical approach. (2005) [14] Marimuthu and Murugan Effects of Plasma Transferred Arc Welding parameters on welding bead geometry, in computational materials science and surface engineering vol. 8 pp 51-88 (2006) [15] Keung –Hueng Teseng et.al.Optimization of PTAW parameters. Journal of manufacturing science and production vol.10 pp 155-188 (2008) [16] Siva, K.Murugan and Raghupathy Modelling, analysis and optimization of weld bead parameters of PTAW, computional materials science and surface engineering vol.1 pp 174-182 (2009) [17] R .Logesh, K.Siva, N.Murugan, Optimization of weld bead geometry in PTAW hard faced on austenitic stainless steel plate using genetic algorithm. International journal of advanced manufacturing technology vol.41 pp 24-30

Optimization of Weld Bead Parameters of Nickel Based Overlay Deposited by Plasma Transferred Arc Surfacing with Adequacy Test

More Related Content

What's hot (18)

Viewers also liked (20)

Similar to Optimization of Weld Bead Parameters of Nickel Based Overlay Deposited by Plasma Transferred Arc Surfacing with Adequacy Test (20)

Recently uploaded (20)

Optimization of Weld Bead Parameters of Nickel Based Overlay Deposited by Plasma Transferred Arc Surfacing with Adequacy Test