Reliability Engineering Snapshot TM

Illustrated Case Studies in the Industrial World of Failure Analysis, Non Destructive Evaluation, and Predictive Maintenance

Machine Design

February 2001

Eliminating a problem through engineering design requires that the engineering concept be carried out faithfully in the field. As an engineer, one cannot assume that the contractor knows what is trying to be accomplished, and will pick up the oversight of the designer. The designer must first lay out the idea in true scale, and see first hand whether the concept can be executed in the field.

In retrospect, it is hard to decipher what went through the minds of the engineers that designed this solution to a complicated problem eight years ago. There is no written documentation, nor is anyone left in the organization that was on this project. It appears that the idea was to strengthen a failing kiln shell by adding a four foot wide by 3/4" thick reinforcing plate, or "wrapper", all the way around the shell (to the left of the red circles in the picture). It was kind of like an afterthought.

The original problem was a failing shell directly underneath the tire. This section of shell was thicker, measuring in at 1 1/4" thick. The thicker section, or "reinforcing ring" as it is called in the industry, tapered down to the 1/2" thick shell. The reinforcing ring was cracking due to hoop stress overload. The solution to this problem was to add stiffening rings on both sides of the tire (one can be seen in the picture). The engineers should have stopped at that point.

However, they decided to add the wrapper. It was located to the left of the red circles in the picture (the wrapper is removed in this picture). The big mistake was in not drawing out first what they intended to install. If they had done this they would have been able to see the problem that they were about to create. But they didn't, and so the contractor butted the circular wrapper up against the circumferential weld that joined the reinforcing ring to the shell. The bottom toe of the fillet weld that attached the wrapper to the reinforcing ring was not located on the reinforcing ring, nor was it located on the circumferential weld joining the ring to the shell. Instead, it was located on the shell itself. Therefore, looking at a cross section of the varying thicknesses, we have a highly stiffened 1 1/4" thick reinforcing ring followed by a small 1/2" shell section, followed by a build up to a 1 1/4" thick composite section (i.e. shell and wrapper).

In effect what the contractor created was a notch, and this notch acted like a stress riser. In the picture, where the red circles are located, those are the locations of the fillet weld that held the wrapper to the shell. The circles highlight weld repairs of cracks that went all the way through the thinner 1/2" shell. This stress concentration caused the shell to crack in this area continually for the next eight years. At its ugliest point in history, there was a 120⁰ crack running along the circumferential weld.

The wrapper was removed and the surface built up and ground down. The desired transition profile between the reinforcing ring underneath the tire, and the shell, should be gradual and smooth (picture left). This dryer goes through four million stress cycles each year, so it is important that weld toes be ground out and there be no high protruding weld bead profiles. The profile of the old wrapper and its fillet weld is shown in the picture. In effect, the wrapper created a stress concentration at the bottom toe of the fillet weld used to attach it to the shell. If we assume for a moment that the wrapper actually helped to simulate a 1 1/4" thick shell (which it didn't) then we have to look at the additional fact that the region marked in red in the picture is only 1/2" thick. As a designer, one would pick this up immediately when laying the concept out on paper. A 1/2" thick section surrounded on both sides by 1 1/4" thick sections. When looking at the cross section made between the reinforcing ring, the shell, and then the wrapper, the effect was a notch.

The ugly looking gob of weld metal in the lower left is a repair of the stiffening ring. That repair is covered in this month's MATERIAL PROPERTIES.

A portion of the 3/4" thick wrapper still remains in this picture. It will eventually be removed. A portion of the original circumferential weld between the reinforcing ring underneath the tire, and the shell, is highlighted by the red dots. The end of the reinforcing ring is highlighted by the light blue dot. It is easy to see why this region cracked repeatedly over the past eight years. This was only one of the reasons. The rest of the story is in the ASM technical paper I wrote (on the Home page tab bar). This particular weld is highly stressed from the normal loading. So to add insult to injury, a stress concentration was thrown in on top of this circumferential weld when they added the wrapper.