It was reported that an emergency stop was initiated at a printing plant. At the same instant the e-stop action was taken, a loud pop was heard on the drive side of the machine, followed immediately by smoke and fire.
Upon investigation, it was determined that the deck 3 Anilox roll shaft had failed. This failure reportedly freed the gear drive (approximately 80 lbs.) which took out numerous electrical and hydraulic components as the gear drive fell to the floor.
An overall view of the testing debris received is shown in Photograph A.
Photograph A Failed shaft and testing debris.
The Anilox Roll Insert shaft fracture, show in Photograph B, was cleaned in an ultrasonic bath with alcohol as the cleaning medium for SEM examination.
Photograph B Fracture surface of failed anilox shaft.
The fracture mode was clearly identified as that of metal fatigue. An SEM map, showing the location of each SEM examination, is shown in Photograph C.
Photograph C Map showing location where SEM examinations were conducted.
It should be noted that the fracture image in Photograph C has been rotated 90 degrees from that shown in Photograph B. The shaft keyway has been rotated now to the 3 o’clock position in Photograph C.
Of particular importance was the fact that the fatigue fracture initiation sites appeared to be at the shaft periphery, i.e., around the outside surface. These multiple fatigue origins correspond to the fillet where the subject roll insert shaft changes diameter. The mating fracture surface is shown in Photograph D.
Photograph D Mating fracture surface on failed Anilox shaft.
The fillet in which the fatigue failure origins are located is denoted with white arrows in Photograph D.
The hardness of the previously tested shaft slice was performed and hardness test results confirmed the results obtained by a previous investigator.
The chemistry of the sample was determined by optical emission spectroscopy (OES). The carbon and sulfur contents were determined by LECO analysis. The results obtained confirmed that the steel conformed to AISI 1010.
The failure of the subject Anilox Roll Insert replacement shaft is clearly one resulting from metal fatigue. The subject shaft was clearly produced from low strength (Tensile strength = 52 Ksi), low carbon steel.
However, the fracture surface evaluation, visually and SEM, revealed that the origin was on the smaller shaft OD, i.e., in the fillet transition from the larger to the smaller shaft.
No written documentation of the shaft order was received or reviewed. Materials received indicated that a previous shaft failure had occurred and that the subject shaft was the replacement for a shaft that failed previously. There was no indication in the materials reviewed that the repair company was given any engineering drawings and/or material and/or hardness and/or strength specifications to guide them in material/strength consideration. The shaft in question was not immediately installed following the previous shaft failure. Another shaft was in use and no mention was made as to the dispensation of that shaft. Did it fail, exhibit unusual wear, etc.? Therefore, the conditions under which the subject shaft was installed were unknown.
The subject failure was unusual in that the shaft fatigue cracking began at the fillet transition between two different diameter shafts. This indicated that the shaft in question was being subjected to bending stress. When the subject printing press was in operation, these bending stresses manifested themselves in rotational bending. That is, the subject fatigue crack initiated in the fillet transition at numerous locations around the smaller shaft. It was reported that the repairing machine ship fabricated the shaft and thus some other party would be responsible for the installation and/or bending stresses induced into the shaft during installation.
Based on the above study, certain opinions were developed. Those preliminary opinions were:
- The failed Anilox Roll shaft failed by metal fatigue.
- The fatigue failure of the subject Anilox Roll shaft initiated in the fillet transition between two different shaft diameters.
- The subject failed Anilox Roll shaft was made from AISI 1010 steel and/or equivalent.
- No specific instructions were given to the fabricating machine shop with regard to the strength of material required and/or stress field or stress state that would be experienced by the subject replacement Anilox Roll shaft.
- No engineering drawings were given to the fabricating machine shop specifying the size or sharpness of the fillet in the subject Anilox Roller shaft.
- The subject failure was the result of rotational bending fatigue induced by bending stresses introduced during the installation of the subject replacement Anilox Roll shaft. Installation of the subject Anilox Roll shaft was not provided by the fabricating machine shop.