Issue 14

B. L

obato da Silva

et alii, Frattura

ed Integrità Stru

tturale, 14 (201

0) 17-26; DOI:

10.3221/IGF-ESI

S.14.02

exp cyc

erimental da les. This rela

ta in the sen tion can be e b N pplications a constant am , Eq. 2. The ss is called am

se to correla xpressed as in

te the alterna Eq. 1, wher

te stress and e A and b ar

the numbe e the constan

r of cycles to t and the cur

failure betw ve exponent,

een 10 3 and respectively. (1) evel stresses mum value , Eq. 3. The cycle.

10 6

A S  

a

that and half

Som cha min of

e practical a racterize the imum value the range stre

nd also fatig plitude load mean stress, plitude stre

ue tests in m s. The stres S m , is the ave ss, S a , Eq. 4.

aterials invol s range, S  rage between These are ba

ve maximum , is the diff maximum v sic relations t

and minimu erence betwe alue and mi hat character

m constant l en the maxi nimum value ize one load

(2) (3) (4)

S S  

S 

m

ax

min

m S S ax

S 



min

m

2

S

S 

S



m

ax

min

a

2

tio, R , is defi

An by

d to describe Eq. 5. The re

the mean str lation betwe

ess, a factor en S a , S m e R

acterize the d in the Eq. 6.

egree of sym

metry of the

load, load ra

ned

used to char is expressed

ma S S R  x in m

(5)

. R S R ndition to de express in t ults. m

1 1



(6)

S



a

 e standard co Eq. 1 can be erimental res

Th the exp

termine the he form of E

f Wöhler cur ed Basquin’s

ve is to assum equation. W

e alternating here  ’ f e b a

load, null m re material co

ean stress. T nstants base

hus, d in

parameters o q. 7. It is call

b

 '

(7)

S

N 



ar

f

Me Ini com pro Go fati pro the In hig end sim dat wid mo stre S y .

an stress effect tially, empiri pensate the posed a par odman intro gue data in posed as imp fatigue stren order to ove h mean stre urance limit ilar to SWT, a, Eq. 14. A espread mat del consists i ss, S m , on th According to

predition Mo c models w effect of m abolic repres duced a theo the graphic rovement of gth coefficie rcome the fa sses, Smith, for the load however usi ccording to hematical rel n the substit e limit of fati this model,

dels ere proposed ean stress in entation of retical line t S a versus S m . the previou nt and that th ilure predicti Watson and ratio, R = -1 ng a factor  empiric con ations to des ution of the B gue strength the stress-life

by Gerber the high cyc the Wöhler’s o represent Since 1960, s models. Fat e compressio on’s problem Topper - S , S ar , is expre that makes siderations, cribe the effe asquin’s equ for the rever relation can

(1874), Go le fatigue st limit fatigu the evaluated some mode igue tests ind n normal m under load WT [3] prop ssed in the E possible an a Berkovits an ct of mean s ation’s const se load condi be presented

odman (189 rength, accor e data on th fatigue data ls to determ icate that the ean stress sho conditions w osed a mod q. 13. On th djustment o d Fang [5] tress on the ant, Eq. 7, fo tion, S rt , and by Eq. 8.

9), Haigh (1 ding to Lee e graphic S m , Eq. 11. Ha ine the effec tensile norm uld increase ith relatively el in which is same year f the curve in and more r fatigue beha r a function on the ultim

917) e Sode [1] and Dow ax /S u versus S igh was the t of mean s al mean stre it [1]. low amplitu the equivale , Walker [4] p relation to ecently Kwo vior of endu that will dep ate strength,

rberg (1930) ling [2]. Ge min /S u , Eq. first to plot tress have b ss should red de and relati nt stress to resented crit the experime fie [6] propo rance limit. S end on the m or yield stren

to rber 12. the een uce vely the eria ntal sed uch ean gth,

  

  



m





S

a S  

e



(8)

rt

ar

pressed in fo

rm of power

series, the E

q. 8 can be e

xpressed by E

q. 9:

Ex

  

  

i



1       ! i

  



m

N





 

S

a S  

e





(9)

m

rt

ar

S

i

0



rt

18