Torsional Stress Fractures 1

[MachineryWatch]

Long distance runners are plagued with stress fractures in the bones of their feet due to the constant, repetitive "impact tests" their feet are subjected to with each step. Each step produces an impact which excites the natural frequencies of the bones. The constant excitation of these natural frequencies leads to the stress fractures.

Would a common rotating machine such as a forced draft blower, whose shaft was subjected to a "rub" for a few (5-10) degrees of each rotation also be exciting torsional natural frequencies? Could this constant excitation of the torsional natural frequencies eventually lead to a stress fracture of the rotor assembly? Conventional vibration monitoring methods would be ineffective at detecting the torsional excitation or even incipient damage before failure.

Am I over simplifying? Is my analogy with the runner's stress fractures flawed?

Skip Hartman

http://www.machinerywatch.com

 

[vanstoja]

I'm not a medical man but I'm inclined to believe that runner's stress fractures are more likely induced by extensional rather than torsional vibrations.
On fan shaft torsional vibrations excited by impactive "rubs" I think it depends on the normal applied at the point of impact and on the time duration of the contact. "Soft" rubs won't excite much torsional response but "hard" rubs might. If you've got labirinth seals at you fan's impeller periphery, you may get teeth tip melting without much torsional action. Slip-stick vibrations are significant torsional exciters but seem unlikely to exist for relatively high speed fans. It is sometimes possible to detect torsional responses in turbomachinery by adding an extra accelerometer 90 degrees circumferentially spaced from a triaxial accelerometer array by using the two inline accels that are 90 degrees apart. I wouldn't know how to track torsionally-induced damage though there may be something on it in the literature. There is a fair volume of papers on so-called "packing rub" much of it by A. Dimaroganas in recent years (also called the "Newkirk effect" after a GE engineer from the 20's and 30's). However, it mostly addresses the rotordynamic effects of rubbing contacts including thermal bending of shafts at rub sites. For other info on rubbing dynamics look for papers by Anna Muszynska of Bentley-Nevada who did a monumental study for NASA that runs to 500+ pages (some 30-40 published journal papers).

 

[MachineryWatch]

Thanks for the response and the good information on shaft rubs. I will try to get hold of some of this information.

I really didn't mean to imply that the runners stress fractures were from torsional vibration. Just that they were from continuous excitation via impacts. Like the shaft is subjected to, only the shaft is being excited torsionally via the rubs.

Thanks again.

http://www.machinerywatch.com

 

[vanstoja]

Here's a better cut at the references I cited.
Muszynska,A. etal (1990), "Rubbing Between Rotors and Stators", Technical Support Package, NASA Tech Briefs MFS-27266, Vol. 14 #7, July 1990 (798pp). They were giving away free hard copies of TSPs back then and this one arrived in a fair-sized box. Much of it is in journal and conference papers from the early 90's.

Dimarogonas,A.D.(1974), "A Study of the Newkirk Effect in Turbomachinery", Wear, Vol.28, pp. 369-382

Dimaroganas,A.D.& Sandor,G.N. (1969)"Packing Rub Effect in Turbomachinery, State of the Art Review", Wear, Vol.14, pp. 153-170

Dimaroganas,A.D. (1983), Analytical Methods in Rotor Dynamics, Applied Science Publishing - Chapt. 5, Heat-Flow-Induced Vibration of Rotating Shafts - The Newkirk Effect

Beatty,R.F. (1985), Differential Rotor Response Due to Radial Rubbing, Trans. ASME, J. Vibration, Acoustics, Stress & Reliability in Design, Vol.107, April, pp. 151-160.

Muszynska,A. (1984), "Partial Lateral Rotor to Stator Rubs", IME Conference Publ. CP 1984-10, 3rd Vibrations in Rotating Machinery

 

[vanstoja]

The Muszynska 1984 paper is on pages 327-324. She ran a rotor against a brass screw and got 1X and either 1/2X, 1/3X, 1/4X or 1/5X in the vibration spectra with the subharmonics dependent on the radial force applied. For "light" rubs the contact period was 22 to 28% of the rotational period while for "heavy" rubs the contact could be maintained for up to 60% of the rotational period.

Another interesting report is NASA-CR-159424 by Pratt & Whitney, "Compressor Seal Rub Energetics Study - Final Report, 4/8/77-4/8/78 (139pp). It ranked independent variables as follows:
Incursion rate - Very important
Rub velocity - Moderate importance
Blade thickness - Moderate importance
Incursion depth - Low importance
Abradable density - Low importance

It said, "Incursion rate, the most important variable, increased rub energy and blade wear with increasing values of incursion rate". Highest rub energy and wear rate occurred with combined high incursion rate, low rub velocity and thick blades. Low, mid and high tested incursion rates were 0.0025, 0.025 and 0.25 mm/sec.