The present invention relates to a method of determining the fatigue threshold value of a material, whereby a test specimen of the material which has a crack notch is subjected to a time dependent alternating load causing crack propagation which by a gradually decreasing load mean value is retarded and whereby it is ensured that the distance between two points, one on each side of the crack notch, alternates between two values. The invention also relates to a device for carrying out this method.

Concepts relating to fracture mechanics are gradually receiving increasing attention in analysis and design of how constructions should be shaped to preclude the initiation of cracks, or should cracks have originated, to prevent propagation of the same due to fatigue. A quantity of considerable importance in the mathemati¬ cal- formulation of the growth of a crack is the so called stress intensity factor. This quantity is ^" further described in an essay titled "System for Determining the Critical Range of Stress Inten- sity Factor llecessary for Fatigue-Crack Propagation" by K Jerram and E Priddle. The essay was published in Journal of Uechanical Engineering Science, Volume 15> number 4, 1973, pp 271-273. In the essay is also described a method and a system for automatic gradual reduction of a cyclic load on a test specimen until a fatigue crack stops growing, by which the threshold value of the range of the stress intensity factor can be determined.

A disadvantage of the method described in the essay is that the ratio between the miniπum, ⁷ Pmm. , and the maximum, ^~ P_max. ⁷ values of the cyclic load varlesduring the reduction, of the load, cf Fig 1a, which means that the measured threshold value will depend on said alternating ratio.

The object of the present invention is to remove the above men- tioned disadvantage when determining the fatigue threshold value of a material. This is carried out by the method and the device accounted for in the characteristic part of the appended claims.

The invention will be further_explained below with reference to the appended drawing on which the previously mentioned Fig 1a shows the course of a time dependent alternating load according to the initially mentioned essay. Fig 1b shows the output signa from a positioning gauge by which is detected the variation of the distance between two points one on each side of a crack notch (CGD) in a fatigue test specimen. Fig 1c depicts a time dependent alternating load according to the present invention. Fig 2 shows one embodiment of a conventional test specimen used in fatigue testing. Fig 3 shows schematically an arrangement for fatigue testing of a test specimen according to Fig 2.

In conventional fatigue tests a test specimen according to Fig is subjected to a time dependent alternating load P. The test specimen has a crack notch 1 which propagates on condition that the alternating load is of sufficient magnitude. One can define two points 2, 3 on the test specimen positioned one on each sid of the crack notch 1. During the test one tries to let the dis¬ tance (CO ) between the points 2 and 3 alternate between two values under the influence of the load variation, see Fig 1b. A a consequence of this variation and the propagation of the crac the load P will decrease as shown in Fig 1a. The ratio Pmm. /Pma in the following designated as the R-value, alternates during t test as a conseαuence of the constant value of Pma.n. This means that the threshold value will be determined for a final value o R which can not be predicted at the start of the fatigue test.

According to the invention, one makes sure that the R-value is constant during the fatigue test.

This is effected by selecting an R-value and by choosing the ma mum value, P , of the load P in such a way that propagation occurs. It follows that Pmm. is then also determined. As before the distance between tha points 2, 3 is made to alternate betwe two values as shown in Fig 1b.

In Fig 3 _? the numeral 4 is a loading device by which a test

specimen 5- _» of the same kind as the one shown in Fig 2, clamped in the loading device, is subjected to a time dependent alterna¬ ting load P. A control system 6, accounted for below, controls the load causing it to change with time as shown in Fig 1c.

The loading device 4 contains a load frame 7 and two jaws 8, 9 for clamping of the test specimen 5. The upper jaw 8 is via a load transducer 10 connected to a spindle 11 threaded into the load frame. The spindle 11 ^' has a hand wheel 12 by which the dis- tance between the jaws can be adjusted. The lower jaw 9 is con¬ nected to item T3 which is slidably mounted in the load frame 7 and operatively connected 'to a hydraulic cylinder unit 14 _? shown dashed in the figure where as well the piston-rod 15 of the unit is shown. • -

The control system 6 contains two calculation loops, one of which has the task of producing a signal corresponding to the instanta¬ neous Pmm. -value of the load. The other is arranged to produce a control signal for controlling the hydraulic cylinder unit 14 with reg ^α ard to the instantaneous Pmax- and Pmm. -values of the load. The first mentioned calculation loop consists of a calculation unit 16 connected to a manually adjustable set point unit 17 for setting a predetermined R-value, and the previously mentioned load transdu¬ cer 10. The other calculation loop consists of a control unit 18 connected to the hydraulic cylinder unit 14. To the control unit are connected the computation unit 16, a set point unit 19 for manual adjustment of P and a positioning gauge 20 arranged to the test specimen. The positioning gauge 20 may be optical or mechanical and serves to produce a signal corresponding to the change in the distance (CGD) between the points 2, 3 in Fig 2.

When determining the fatigue threshold value applicable to the material of the test specimen 5 the following procedure is used.

The set point units 16 and 19 are manually preset to given values of E and P respectively. The signals from the load transducer max ^*

10 and the set point unit 17 are processed in the calculation unit 16 and a value ? . corresponding to the preset value ∑'_ _a - _c is received. With the aid of the output signals from units 16 and

20 as well as an exterior control signal suggested in Fig 3 as arrow 21, which for instance can be a sinus-, square- or saw—to signal, the control unit 18 produces a signal which causes the hydraulic cylinder unit 14 to subject the test specimen to a lo which changes according to the control signal _* The ratio betwee the minimum and τnaxi.τmw value of the load will thus be kept con stant and. equal to the preset R-value. The maximum and minimum values of the load are made to assume such values that the dis¬ tance (CGD) between the points 2, 3 is kept constant in spite o the crack propagation.

The invention must not be considered to be limited to the above described specific application. It has a number of applications which all fall within the scope of the invention. It should be obvious that such conditions which must normally be satisfied i similar determinations of threshold values, for instance the co dition of constant temperature, must be satisfied as well when determining the fatigue threshold value according to the presen inventio .