ma e ials
A icle
E ec o WEDM P ocess Pa ame e s on Su ace
Mo phology o Ni inol Shape Memo y Alloy
Rakesh Chaudha i 1, Jay J. Vo a 1, Vi ek Pa el 2,3 , L. N. López de Lacalle 4,* and
D. M. Pa ikh 1
1Depa men o Mechanical Enginee ing, School o Technology, Pandi Deendayal Pe oleum Uni e si y,
Raisan, Gandhinaga 382007, India; [email p o ec ed] (R.C.); [email p o ec ed] (J.J.V.);
[email p o ec ed] (D.M.P.)
2
School o Ma e ial Science and Enginee ing, No hwes e n Poly echnical Uni e si y, Shaanxi 710072, China;
[email p o ec ed]
3Depa men o Enginee ing Science, Di ision o Welding Technology, Uni e si y Wes ,
46186 T ollhä an, Sweden
4Depa men o Mechanical Enginee ing, Uni e si y o he Basque Coun y,
Escuela Supe io de Ingenie os Alameda de U quijo s/n., 48013 Bilbao, Spain
*Co espondence: [email p o ec ed]
Recei ed: 30 Sep embe 2020; Accep ed: 31 Oc obe 2020; Published: 3 No embe 2020
Abs ac :
Nickel– i anium shape memo y alloys (SMAs) ha e s a ed becoming popula owing o
hei unique abili y o memo ize o egain hei o iginal shape om he plas ically de o med condi ion
by means o hea ing o magne ic o mechanical loading. Nickel– i anium alloys, commonly known
as ni inol, ha e been widely used in ac ua o s, mic oelec omechanical sys em (MEMS) de ices,
and many o he applica ions, including in he biomedical, ae ospace, and au omo i e ields. Howe e ,
ni inol is a di icul - o-cu ma e ial because o i s e sa ile speci ic p ope ies such as he shape
memo y e ec , supe elas ici y, high speci ic s eng h, high wea and co osion esis ance, and se e e
s ain ha dening. The e a e se e al challenges aced when machining ni inol SMA wi h con en ional
machining echniques. Noncon ac ope a ion o he wi e elec ical discha ge machining (WEDM)
p ocess be ween he ool (wi e) and wo kpiece signi ican ly elimina es he p oblems o con en ional
machining p ocesses. The WEDM p ocess consis s o mul iple inpu pa ame e s ha should be
con olled o ob ain g ea su ace quali y. In his s udy, he e ec o WEDM p ocess pa ame e s on
he su ace mo phology o ni inol SMA was s udied using 3D su ace analysis, scanning elec on
mic oscopy (SEM), and ene gy-dispe si e X- ay (EDX) analysis. 3D su ace analysis esul s indica ed
a highe alue o su ace oughness (SR) on he op o he wo k su ace and a lowe SR on he bo om
po ion o he wo k su ace. The su ace mo phology o he machined sample ob ained a op imized
pa ame e s showed a educ ion in mic oc acks, mic opo es, and globules in compa ison wi h he
machined su ace ob ained a a high discha ge ene gy le el. EDX analysis indica ed a machined
su ace ee o molybdenum ( ool elec ode).
Keywo ds:
shape memo y alloy; ni inol; WEDM; shape memo y e ec ; su ace mo phology;
molybdenum ool wi e
1. In oduc ion
In he p esen scena io o manu ac u ing compe i i eness, he adop ion o new echnologies
is essen ial o o e come he challenges o achie ing componen accu acy, high quali y, accep able
su ace inish, inc eased p oduc ion a e, enhanced p oduc li e, and educed en i onmen al impac .
Beyond hese con en ional challenges, he machining o newly de eloped sma ma e ials also equi es
Ma e ials 2020,13, 4943; doi:10.3390/ma13214943 www.mdpi.com/jou nal/ma e ials
Ma e ials 2020,13, 4943 2 o 14
inpu s o in elligen machining s a egies. Fo he manu ac u ing p ocesses, he accu acy o he inished
p oduc is mainly dependen on i s inpu p ocess pa ame e s. The e o e, i is necessa y o con ol he
inpu pa ame e s and secu e hei op imum alues. One such newly de eloped gene a ion o alloys
is he shape memo y alloys (SMAs). SMAs display an excep ional cha ac e is ic o egaining hei
shape when hey a e hea ed. Nickel– i anium alloy is one o he shape memo y alloys, which is known
as ni inol in espec o i s inno a ion a he Na al O dnance Labo a o y (NOL) [
1
]. Shape memo y
alloys we e ini ially e ealed by Swedish physicis A ne Olande [
2
]. He ound ha Au–Cd alloys
e u ned o hei ini ial shape and size a e hea ing abo e a ce ain empe a u e e en a e hei plas ic
de o ma ion in he cold s a e. Ou o numbe s o combina ions o SMAs, ni inol became a amous
shape memo y alloy owing o low p oduc ion cos as compa ed o o he SMAs, sa e and easie
handling, and supe io mechanical p ope ies. Ni inol SMAs ha e been applied in a ious a eas like
ai condi ioning en s, elec onic cable connec o s, and al es [
3
]. O e he las decade, he a eas o
applica ion also sp ead o ae ospace, oil indus ies, au omobiles, and obo ics. These sma ma e ials
possess he main cha ac e is ics o supe elas ici y (SE) and shape memo y e ec (SME) [
1
]. Ni inol is
conside ed an ideal ma e ial in he biomedical ield owing o i s p ope ies like biocompa ibili y and
wea and co osion esis ance [
4
,
5
]. Due o he unc ional p ope ies o ni inol SMAs, hei biomedical
applica ion has p o en success ul by inc easing he possibili y as well as he pe o mance quali y o
minimally in asi e su ge ies. The nickel– i anium SMA is highly biocompa ible, which makes i use ul
in o hopedic implan s, su gical ins umen s, ca dio ascula de ices, and o hodon ic de ices [
6
,
7
].
Ni inol SMA is ca ego ized in o wo di e en ypes, namely shape memo y and supe elas ic. I he
empe a u e o shape eco e y is less han he oom empe a u e, i is ecognized as supe elas ic;
i he empe a u e o he shape eco e y is g ea e han oom empe a u e, i is ecognized as shape
memo y [
8
]. The e a e se e al challenges aced when using con en ional machining echniques o
machine ni inol SMA; hese challenges a e due o se e e s ain ha dening, high duc ili y, poo chip
b eaking, supe elas ici y, bu o ma ion, and high wea and co osion esis ance [
9
,
10
]. Weine and
Pe zold concluded ha he machining o NiTi-based alloys is complex using con en ional echniques
like u ning, d illing, and deep hole d illing. Poo chip b eaking, ool wea , and bu o ma ion ha e
been obse ed while machining shape memo y alloys using con en ional machining echniques
Wi e elec ical discha ge machining (WEDM) is a ype o noncon en ional machining me hod
which is mo e p e e able o o e come hese de ec s [
10
]. The WEDM p ocess is applicable o all
conduc i e ma e ials ega dless o ma e ial ha dness [
11
–
14
]. The WEDM p ocess c ea es a se ies o
spa ks ha helps o emo e he ma e ial om he wo k su ace. These spa ks a e gene a ed be ween
he wi e (elec ode) and he wo kpiece in he p esence o a dielec ic luid. The WEDM p ocess consis s
o a high numbe o p ocess pa ame e s ha should be con olled o acqui e g ea su ace quali y and
con ol he esul ing me allu gical and mechanical p ope ies. Achie ing supe io pe o mance o
ni inol using he WEDM p ocess equi es he op imized pa ame e se ings o inpu p ocess pa ame e s
such as discha ge cu en (I), pulse-o ime (T
on
), pulse-on ime (T
o
), and dielec ic luid p essu e
o machining o ni inol SMA [
15
]. Chaudha i e al. [
10
] iden i ied cu en (I), T
on
, and T
o
as key
inpu pa ame e s o machining o supe elas ic SMA. Thei esul s indica ed ha all he pa ame e s a e
signi ican o achie ing simul aneous ou pu pa ame e s. In ano he s udy, Majumde and Mai y [
16
]
used he WEDM echnique o he simul aneous op imiza ion o mul iple inpu pa ame e s o ni inol
SMA. Soni e al. [
17
] iden i ied he app op ia e le el o pulse-on ime and se o ol age o he educ ion
o mic oc acks. P ope selec ion o inpu p ocess pa ame e s is a key ac o in he WEDM p ocess o
a oid wi e up u e p oblems and he o ma ion o la ge c a e s on he wo k su ace [
18
]. In pas
esea ch, elec odes o b ass and coppe wi es ha e been used o su ace analysis o SMAs o samples
machined by WEDM [19,20].
Recen ly, we conduc ed a Pa e o analysis o WEDM inpu pa ame e s o machining o ni inol
SMA using he hea ans e sea ch (HTS) algo i hm [
21
]. The HTS algo i hm was ound o be e ec i e
o p edic ing and op imizing he inpu alues o all objec i e unc ions. ANOVA es esul s showed
he obus ness o he gene a ed empi ical models o he op imiza ion o mul iple esponses. The key
Ma e ials 2020,13, 4943 3 o 14
ocus o he p esen s udy is on he in luence o inpu pa ame e s on he su ace in eg i y o he
machined su ace. Based on e iewed li e a u e and machining capabili ies, cu en (I), pulse-o
ime (T
o
), and pulse-on ime (T
on
) we e selec ed as h ee key inpu pa ame e s su ace oughness
(SR), ma e ial emo al a e (MRR), and mic oha dness (MH) we e selec ed as ou pu cha ac e is ics.
Fo su ace mo phology, se s o pa ame e s wi h high and low discha ge ene gy le els and a se
o op imized p ocess pa ame e s we e conside ed. In pas s udies, mo e ocus was placed on he
pa ame ic op imiza ion o ni inol alloys. Howe e , no much has been epo ed on su ace in eg i y
a e conduc ing pa ame ic op imiza ion. In he p esen s udy, scanning elec on mic oscope (SEM)
and ene gy-dispe si e X- ay analysis (EDX) we e used o s udy elemen al composi ion, su ace analysis,
and phase analysis a e he machining o ni inol SMA. The aim o he p esen s udy was o p o ide
signi ican inpu o end-use s o he selec ion o WEDM p ocess pa ame e s o ni inol SMA.
2. Ma e ials and Me hods
In he cu en s udy, a Conco d WEDM machine DK7732 wi h EDM oil (Conco d Limi ed,
Bangalo e, India) as dielec ic luid was used o machine he samples o ni inol SMA (P ocu ed om
SMA Wi es, Ahmedabad, India). The chemical composi ion o ni inol SMA (Ni
55.8
Ti) used in he
p esen s udy is shown in Table 1. A ni inol ba wi h a diame e o 6 mm was u ilized o machine
es s o su ace in es iga ion. In a pas epo , he impac o inpu pa ame e s was inspec ed by
cu ing sample sizes o 1.5 mm. A simila size o 1.5 mm was a o ed in he cu en examina ion.
A eusable molybdenum wi e (Conco d Limi ed, Bangalo e, India) o 0.18 mm diame e was used as a
ool elec ode.
Table 1. Chemical composi ion o ni inol shape memo y alloy (SMA).
Elemen Ti Ni Co Cu C Fe Nb C H O N
W (%)
Balance
55.78 0.005 0.005 0.005 0.012 0.005 0.039 0.001 0.034 0.001
The hea ans e sea ch (HTS) echnique has been used o mul iobjec i e op imiza ion o mul iple
a iables like MRR, SR, and MH [
10
,
21
]. The HTS algo i hm unc ions on basis o he ans e o hea
owing o he in e ace be ween he sys em pa icles and he su oundings o achie e he mal s abili y.
The mal s abili y in he ans e o hea be ween sys em and su oundings can be ob ained om a
he modynamically imbalanced sys em. To each he mal s abili y, h ee hea ans e phenomena,
namely conduc ion, con ec ion, and adia ion, con ibu e o a la ge ex en . As such, he h ee phases
o conduc ion, con ec ion, and adia ion a e conside ed du ing he implemen a ion o he HTS
algo i hm. Du ing he implemen a ion o he HTS algo i hm, each hea ans e phenomenon has he
same oppo uni y o hea ans e , and each gene a ion decides any one o hese h ee phenomena
andomly. A bi a ily c ea ed popula ion commences he HTS algo i hm, in which he sys em consis s
o ‘n’ numbe o pa icles (popula ion size) and empe a u e le el (inpu a iables). Subsequen ly,
a bi a ily chosen hea ans e phenomena upda e he popula ion size o each gene a ion. In he nex
s age, an upda ed solu ion ha ing a good unc ional a e ge s accep ed and he wo s solu ion ge s
eplaced by eli e solu ions [10]. The low cha o he HTS algo i hm is shown in Figu e 1.
Ma e ials 2020,13, 4943 4 o 14
Figu e 1. Flow cha o he hea ans e sea ch (HTS) algo i hm.
2.1. Conduc ion Phase
Equa ions (1) and (2) a e used o upda e he solu ions o he conduc ion phase:
X0
j,i =
Xk,i +−R2Xk,i, i Xj> (Xk)
Xj,i +−R2Xj,i, i Xj< (Xk);i g ≤gmax/CDF (1)
X0
j,i =
Xk,i +(− iXk,i), i Xj> (Xk)
Xj,i +− iXj,i, i Xj< (Xk);i g >gmax/CDF (2)
whe e
X0
j,i
is he upda ed solu ion; j =1, 2,
. . .
, n; k is a andomly selec ed solu ion; j
,
k; k
∈
(1, 2,
. . .
, n);
i is a andomly selec ed design a iable; i
∈
(1, 2,
. . .
, m); g
max
is he maximum numbe o gene a ion
speci ied; CDF is he conduc ion ac o ; R is he p obabili y a iable; R
∈
{0, 0.3333}; and
i∈{0, 1}
is a
uni o mly dis ibu ed andom numbe [10].
2.2. Con ec ion Phase
Equa ions (3) and (4) a e used o upda e he solu ions o con ec ion phase:
X0
j,i =Xj,i +R×(Xs−Xms ×TCF)(3)
TCF =(abs(R− i), i g ≤gmax/COF
ound(1+ i), i g >gmax/COF (4)
Ma e ials 2020,13, 4943 5 o 14
whe e
X0
j,i
is he upda ed solu ion; j =1, 2,
. . .
, n; i =1, 2,
. . .
, m; COF is he con ec ion ac o ; R is he
p obabili y a iable; R
∈
{0.6666, 1};
i∈
{0, 1} is a uni o mly dis ibu ed andom numbe ; X
s
is he
empe a u e o he su oundings; X
ms
is he mean empe a u e o he sys em; and TCF is a empe a u e
change ac o [10].
2.3. Radia ion Phase
Equa ions (5) and (6) a e used o upda e he solu ions o adia ion phase:
X0
j,i =
Xj,i +R×Xk,i −Xj,i, i Xj> (Xk)
Xj,i +R×Xj,i −Xk,i, i Xj< (Xk);i g ≤gmax/RDF (5)
X0
j,i =
Xj,i + i×Xk,i −Xj,i, i Xj> (Xk)
Xj,i + i×Xj,i −Xk,i, i Xj< (Xk);i g >gmax/RDF (6)
whe e
X0
j,i
is he upda ed solu ion; j =1, 2,
. . .
, n; i =1, 2,
. . .
, m; j
,
k; k
∈
(1, 2,
. . .
, n), and k is a andomly
selec ed molecule; RDF is he adia ion ac o ; R is he p obabili y a iable;
R∈{0.3333, 0.6666};
and i∈{0, 1} is a uni o mly dis ibu ed andom numbe [10].
I is no possible o co e he en i e machining ange in he design o he expe imen . Based on
he li e a u e, sui able anges o inpu p ocess pa ame e s we e conside ed while conduc ing he
expe imen s as ollows:
•Pulse on ime: 35 µs≤Ton ≤55 µs;
•Pulse o ime: 10 µs≤To ≤20 µs;
•Cu en : 2 A ≤I≤4 A.
Howe e , he e is a possibili y ha he op imized pa ame e se ings may be ou side hese
conside ed anges. Fo his, he machining ange was conside ed be ween he ex eme limi s o inpu
p ocess pa ame e s while implemen ing he HTS algo i hm as ollows:
•Pulse on ime: 1 µs≤Ton ≤110 µs;
•Pulse o ime: 1 µs≤To ≤32 µs;
•Cu en : 1 A ≤I≤6 A.
The op imum alues o selec ed esponses we e ound a T
on
o 40
µ
s, T
o
o 12
µ
s, and cu en
o 1 A [
21
]. Fo hese op imal condi ions, he esponses o o he ou pu a iables we e p edic ed,
and a alida ion es was pe o med using hese p edic ed p ocess pa ame e s. The HTS algo i hm
was ound o be capable o success ully p edic ing and op imizing he p ocess pa ame e s as he
di e ence be ween he p edic ed and measu ed alue was negligible. In he p esen s udy, he e ec
o inpu p ocess pa ame e s a hese op imum le els was s udied in de ail on he su ace in eg i y o
he machined su ace. Figu e 2shows he slide- lushing mechanism ( om abo e) used in he cu en
s udy du ing he WEDM p ocess. To unde s and he a ia ion o SR in di e en a eas o he machined
su ace, a h ee-dimensional (3D) digi al mic oscope wi h a noncon ac - ype p obe (Keyence VHX-600,
China) was used. Keyence VHX-600 digi al mic oscope (Keyence VHX-600, China) was used o eco d
he a ia ion o SR alues. The eco ded SR alues in he cu en s udy ha e an a i hme ic a e age
oughness (Ra) in
µ
m. The su ace mo phology o he machined samples was examined using SEM and
EDX equipmen om Tescan (Vega Tescan, India). E chan (82 mL H
2
O+14 mL HNO
3
+4 mL HF)
was used o ni inol SMA.
Ma e ials 2020,13, 4943 6 o 14
Figu e 2. Flushing mechanism o he wi e elec ical discha ge machining (WEDM) p ocess.
3. Resul s and Discussion
3.1. Analysis o SR Using 3D Digi al Mic oscope
The sample machined using he op imized se o pa ame e s (T
on
o 40
µ
s, T
o
o 12
µ
s, and cu en
o 1 A) was subjec ed o 3D analysis. Di e en ial scanning calo ime y (DSC) (Ne zsch, Selb, Ge many)
es ing e ealed ha he WEDM sample machined using he op imized se o pa ame e s had e en ion o
shape memo y e ec when compa ed o ha o he s a ing base ma e ial [
21
]. Figu e 3shows he op
su ace o he machined sample, wi h SR alues indica ed a a ious loca ions. The op su ace o he
machined sample shows he highes alue o SR, which is indica ed by g een colo ; he bo om su ace o
he machined sample shows he lowes alue o SR, as indica ed by blue colo . While ed colo indica es
he highes SR in he image, hese poin s we e no aken in o accoun as hey a e ou side o he machined
su ace. As indica ed by he wi e a el di ec ion shown in Figu e 3, SR was measu ed on he su ace om
le o igh . Figu e 4shows he machined su ace o supe elas ic ni inol SMA, wi h SR alues ep esen ed.
I can be clea ly obse ed ha he a e age alue o SR in he cu en s udy (Figu e 3) is less han he
a e age alue o SR ob ained o supe elas ic ni inol shape memo y alloy (Figu e 4) [22].
Figu e 3. Su ace machined unde op imized p ocess pa ame e s.
Ma e ials 2020,13, 4943 7 o 14
Figu e 4. Machined su ace ep esen ing su ace oughness (SR) alues o supe elas ic ni inol [22].
The dielec ic lushing mechanism om abo e was he main eason o ob aining a highe alue
o SR on he op su ace o he sample; i.e., a g ea e a ea o he op su ace is exposed o he dielec ic
luid. This in u n inc eases he discha ge ene gy, which esul s in a g ea e amoun o deb is emo al.
I also inc eases he ma e ial emo al a e due o he highe alue o discha ge ene gy. Howe e ,
a he bo om
su ace o he sample, he supply o dielec ic luid is lowe , which minimizes he
discha ge ene gy and SR o he bo om egion. An inc ease in dielec ic luid p essu e inc eases he
discha ge ene gy which inc eases he SR and o ms he la ge c a e s on he su ace o he wo kpiece [
23
].
Figu es 5and 6
suppo his conclusion o di e ences in SR be ween di e en loca ions o he wo k
su ace.
Figu es 5and 6
ep esen he de ailed SR analysis o he machined sample a he ou e mos
and inne mos pe iphe ies, espec i ely. The numbe o da a poin s o each pe iphe y is di e en
due o di e en ci cum e en ial leng hs o bo h pe iphe ies. A dis ance o 0.43 mm was aken in o
accoun be ween he wo consecu i e SR measu emen s. The ou e mos pe iphe y has a ci cum e en ial
leng h o 20,920
µ
m wi h 48,651 da a poin s, whe eas he inne mos pe iphe y has a ci cum e en ial
leng h o 8345
µ
m wi h 19,406 da a poin s. The lowes alue o SR was ob ained a he bo om su ace.
while he maximum SR was obse ed o be a he op su ace. Howe e , i is ecommended o ha e
iden ical lushing p essu e coming om bo h di ec ions (abo e and below) o he wo kpiece o a oid
any a ia ion in SR alue on he machined su ace. This migh esul in gi ing a highe a e age SR
alue o he machined su ace.
Wi e up u e occu s wi h an inc ease in c a e size. To a oid his p oblem, discha ge ene gy mus
be a an op imum le el. A lowe SR alue can be achie ed a a lowe discha ge ene gy le el. Lowe ing
he alues o pulse-on ime and cu en esul s in a lowe discha ge ene gy le el [
24
]. Howe e , a lowe
alue o pulse-o ime gi es a highe discha ge ene gy le el as i minimizes he ime be ween wo
consecu i e spa ks. Pu suan o he same, lowe discha ge ene gy can be ob ained a lowe alues o
pulse-on ime and cu en and a highe alue o pulse-o ime. The machined sample conside ed in
he cu en s udy was ob ained a op imized pa ame e se ings (T
on
o 40
µ
s, T
o
o 12
µ
s, and cu en
o 1 A). These alues show ha discha ge ene gy was no a a lowe o highe le el as hese alues a e
de i ed o mul iple esponse a iables such as MRR, SR, and MH. Table 2shows he highe discha ge
ene gy le el o an op imized se o pa ame e s in ou p e iously epo ed s udy o he su ace analysis
o supe elas ic SMA in compa ison wi h he cu en s udy [
22
]. Lowe discha ge ene gy a op imized
pa ame e se ings in he cu en s udy gi es lowe SR alues as compa ed o he p e iously epo ed
s udy, wi hou comp omising he op imum alues o o he esponse a iables such as MRR and MH.
The a e age alue o SR in he cu en s udy is less han he a e age alue ob ained o supe elas ic
Ma e ials 2020,13, 4943 8 o 14
ni inol shape memo y alloy [
22
]. This is because o he lowe discha ge ene gy o pa ame ic se ings
o ni inol SMA as compa ed o he discha ge ene gy o supe elas ic ni inol shape memo y alloy.
Figu e 5. Su ace analysis o machined sample a ou e mos pe iphe y.
Figu e 6. Su ace analysis o machined sample a inne mos pe iphe y.
Ma e ials 2020,13, 4943 9 o 14
Table 2. 3D su ace analysis compa ison o p esen s udy wi h pas s udy [22].
Resea che Wo k Ma e ial
Inpu
Pa ame e s a
High
Discha ge
Ene gy Le el
Inpu
Pa ame e s a
Low Discha ge
Ene gy Le el
Inpu
Pa ame e s a
Op imized
Pa ame e s
Se ings
Rema ks
Chaudha i e al.
[22]
Supe elas ic
Ni inol SMA
Ton =110 µs,
To =32 µs,
Cu en =6 A
No epo ed
Ton =65 µs,
To =32 µs,
Cu en =6 A
3D su ace analysis shows highe
alue o SR (1 o 148 µm) on
machined su ace a op imized
pa ame e se ings (Figu e 4).
P esen s udy Ni inol SMA
Ton =110 µs,
To =1µs,
Cu en =6 A
Ton =1µs,
To =32 µs,
Cu en =1 A
Ton =40 µs,
To =12 µs,
Cu en =1 A
Figu e 3o 3D su ace analysis shows
lowe alue o SR (1 o 75
µ
m app ox.)
on machined su ace a op imized
pa ame e se ings.
This is due o less discha ge ene gy a
op imized pa ame e se ings.
3.2. Analysis o Su ace De ec s
SEM images o he su ace machined using he WEDM p ocess a e shown in Figu es 7–9.
The WEDM p ocess c ea es a se ies o spa ks ha helps o emo e he ma e ial om he wo k su ace
by mel ing [
25
]. The WEDM p ocess consis s o a high numbe o p ocess pa ame e s ha should be
con olled o acqui e g ea su ace quali y ee o mic oc acks, mic opo es, and o he su ace de ec s.
Highe alues o pulse-on ime and cu en esul in a highe discha ge ene gy le el due o an inc ease
in spa k in ensi y. This in u n inc eases he o ma ion chances o mic oc acks, globules, mic opo es,
and o he su ace de ec s [
26
,
27
]. Howe e , a highe alue o pulse-o ime dec eases he discha ge
ene gy le el and he eby also dec eases he o ma ion chances o mic oc acks, globules, mic opo es,
and o he su ace de ec s. The su ace mo phology o he machined su ace a low and high discha ge
ene gies was s udied. WEDM p ocess pa ame e se ings o high discha ge ene gy we e T
on
o 110
µ
s,
T
o
o 1
µ
s, and cu en o 6 A, whe eas pa ame e s o low discha ge ene gy we e T
on
o 1
µ
s, T
o
o
32 µs,
and cu en o 1 A. Figu es 7and 8show he SEM mic og aph o he machined su ace a low and
high discha ge ene gy le els, espec i ely. I is clea ly e iden om Figu es 7and 8 ha mic oc acks,
mic opo es, and deposi ed laye s o med mo e eadily and globule sizes we e inc eased a he high
discha ge ene gy le el in compa ison wi h he su ace mo phology obse ed a he low discha ge
ene gy le el. A he high discha ge ene gy le el, high empe a u e mel s and e apo a es he ma e ial,
and hen i ge s mixed wi h he dielec ic luid. The ma e ial is hen quenched in he dielec ic luid
as i s empe a u e ge s educed a e mixing wi h he dielec ic luid. This quenching phenomenon
esul s in he o ma ion o c acks, po es, and o he de ec s [
28
]. In addi ion o his, high discha ge
ene gy inc eases he ma e ial emo al a e. Due o his excess ma e ial emo al, some o he deb is
pa icles s ick in he wo king zone and ge edeposi ed on he machined su ace. Howe e , quenching
and ma e ial deposi ion phenomena a e ma kedly educed a he WEDM pa ame ic se ing o low
discha ge ene gy when compa ed o hose obse ed a high discha ge ene gy. Due o his eason,
Figu e 7shows li le p esence o mic oc acks, mic opo es, and deposi ed laye s and small globules.
Howe e , he comple e elimina ion o mic oc acks, mic opo es, deposi ed laye s, and globules is e y
di icul as he e will be some amoun o discha ge ene gy a all imes a any pa ame e se ings o
he WEDM p ocess. This shows ha lowe discha ge ene gy is bes sui ed o imp o ing he su ace
in eg i y o he machined su ace. Howe e , WEDM pa ame e s a he lowes discha ge ene gy will
no be able o sa is y all he objec i es. Achie ing highe MRR, T
on
, and cu en equi es using a highe
discha ge ene gy le el while keeping T
o
a a lowe le el. Ob aining lowe SR, T
on
, and cu en equi es
using a lowe ene gy discha ge le el while keeping T
o
a a highe le el. This con lic ing si ua ion can
be e icien ly ackled by op imizing he WEDM p ocess pa ame e s o mul iple esponses. The WEDM
p ocess equi es he op imized pa ame e se ings o discha ge cu en , T
on
, T
o
, and dielec ic luid
o machining o ni inol SMA [
15
]. Ou p e ious s udy o pa ame ic op imiza ion using he hea
ans e sea ch (HTS) algo i hm p o ed o be highly e ec i e in p edic ing and op imizing he WEDM
p ocess pa ame e s [
21
]. WEDM p ocess pa ame e se ings a op imized condi ions a e T
on
o 40
µ
s,
