IFMASS8

function of stress fluctuation. The simulation of each of the amplitudes should not be

performed so as to finish firstly with one amplitude and then go onto the other. Each

amplitude should be simulated from the beginning in small blocks, which will be repeated

several times up to failure of the tested object. This will have a positive effect on getting

closer to the effect of the sequence of stress fluctuation to the real state. He suggested a

diagram following which it is necessary to make a simulation of stress change (Fig. 17)

which is obtained by transforming the stress spectrum. As the stress spectrum value

= 10

or 10

is too big, fracture may occur before the simulation on spectrum is

completed, i.e. at

or at a little bigger cycle number

than

. In order to avoid

this possibility, the stress change simulation should be performed in blocks

which are

by the cycle number much smaller than the unit spectrum of stress

. The block of cycles

is obtained by proportional decrease of the cycle numbers in the stress spectrum. The

proportion for this transformation is

N N

n n

∆

;

N n

∆ = ∆

;

= ∆

∑

N N

= ∆

∑

The simulations are materialized by starting from some stress level from the spectrum

(block) centre. Upon completion of the simulation of stress level

, for example, with the

number of cycles

∆

, a transition is made onto a higher stress level

and so on up to

, and then from the biggest stress level towards smaller ones as presented in Fig. 17. If

block

is smaller, it will be repeated several times up to fracturing, i.e. the number of

repetitions is

. Better mixing of big and small stress amplitudes suits the greater

number of repetitions. Too big number of repetitions of blocks imposes the need for the

unacceptable big number of stress level change in the course of the testing process. This

problem may be resolved by electronic, i.e. program management of stress level changes.

2 2

8 7

∆

Figure 17. The sequence of cycle blocks of stress in testing of service

fatigue strength (Gassner’s diagram)

For the determination of failure probability for service fatigue strength, test results are

required for three stress levels

, i.e.

. Two levels are required for the determination of

the direction of the lower part on the service fatigue strength curve. The upper part of this

broken straight line is parallel with the line of the basic (Wöhler’s) strength. Testing is

necessary for one stress level for the purpose of determination of the position (distance)

of this part from the line of basic strength. The highest stress level

′

is chosen in such a

manner that all stress levels in the stress spectrum are greater than the infinite life

strength. The lowest stress level in the spectrum must thereby be significantly above the

infinite life strength, so that the greatest stress in the spectrum

′

is in the zone of the

349