Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Suitable code for machinery fatigue phenomenon on structures 1

Status
Not open for further replies.

Fran67

Industrial
Jun 26, 2019
27
0
0
FR
Hello everybody,

I work on fatigue phenomenon on steel structures supporting high rotating machinery (10 000 - 30 000 RPM).
I used to work with Eurocodes for steel structure design, but I'm not sure that this code is suitable for structures with a large number of cycles, due to the high rotating speed of machinery.
Do you know if it exists an other suitable code (or a specific chapter of EN 1993), to evaluate durability of structural members, welds and bolts, in response to machinery vibrations?
I'm searching (if it is possible) for a simplified method, for example a method which let to avoid fatigue calculation if specific conditions are fulfilled (a max stress < x*Fy for example)

For my application, vibration frequence is high, but vibrations rates are very low, and then stress on bolts and structural members too, and I presume and hope that there is a possibility to avoid a complicated fatigue calculation on steel structue, it is the Reason why I hope to find a simplified method.

Thank you for your responses

F
 
Replies continue below

Recommended for you

For structural steel, AISC has fatigue criteria in every edition of the steel manual. 20,000 cycles is considered the threshold for required checks in structural steel. (For whatever lifetime the structure is expected to go.)
 
I think you're looking for the Constant Amplitude Fatigue Limit (CAFL). Look at EN1993-1-9. You'll match your specific details to the ones shown in Table 8.1, and find the associated design curve (Figure 7.1). You'll see it goes flat at some point - the stress range associated with that is your CAFL.

As for the speed, I don't think you need to be too worried. The lab tests for the CAFL are usually run pretty fast (200+ Hz). Otherwise the tests last for months and grad students don't get their degrees... At 30,000 RPM you're still pretty close to that speed. I'd start thinking about that once you're in the ultrasonic fatigue range(10+ kHz), and that's mostly just because at that point you can start heating things up and accelerating fatigue.

Bolts will probably be your governing factor.
 
Status
Not open for further replies.
Back
Top