Typical construction errors and how to avoid them: Error #4
This is a continuation of my blog about typical construction errors.
Obstruction thermomechanical stresses
The manufacturers of engines, motors, exhaust systems and burners are clear on the fact that thermal expansion in the high temperature range may only be conditionally hampered, as cracks may occur.
Yet the effort is common from school physics, where a massive bolt breaks when a rod is heated and then contraction is hampered by the bolt.
Almost every student in the physics classroom has seen the enclosed test.
Moderate increases in temperature
But how it looks when the temperature increases are relatively low?
In the case of construction groups consisting of one material, there are usually only a few problems.
It however looks different, when for example, components such as fan impellers or turbo machinery are made of aluminum and are connected to steel shafts and flanges.
Here, even small temperature increases by about 60 °C are enough to generate high stresses when the thermal expansions are hindered.
The classic I have encountered several times in my professional career, consists of an impeller connection, in which bolts are used alongside the screws.
Steel wheels are replaced during a redesign with increased requirements due to the weight of the aluminium / silumin. The experiment shows that the shaft / impeller connection slides or imbalances occur after operation, although the system was previously perfectly balanced. Bolts are now put in place to prevent the displacements and cracks are observed in the holes.
The construction error here is that attempts are made to prevent thermal strains.
The coefficient of expansion of aluminum is about twice as high as that of steel.
For example, solutions can be making a medium between steel and aluminum, which is able to compensate for the strains without sliding movements. A reduction in the diameter of the bolt circle also reduces the thermal stresses.
Temperature shocks often also lead to cracks when heating up or cooling down a machine. This for instance is the case when a hot component e.g. when cleaning is done with cold water or a cold part comes in contact with a hot fluid.
There’s also the classic school experiment in physics, in which a heated stone shatters when poured over with cold water. In this case, the thermal stresses due to the strong temperature gradients within the stone are so high that the breaking stress is exceeded.
A misunderstanding I observed more frequently observed in construction engineering is that a body which is heated evenly (i.e. very slow), shows tensions or unevenly extends. This is only true if several materials with different coefficients of thermal expansion are connected. Otherwise, a body will be simply evenly (i.e. geometrically similar) larger at higher temperatures. Tensions this only occurs if the expansion is hindered e.g. by the connection environment.
Components with different wall thicknesses
When components with different wall thicknesses heat or cooled, the thinner parts arrive at the final temperature faster than the thicker ones, which can also result in high stresses.
Many cases of damage investigated by us could essentially be attributed to inadmissibly high thermomechanical stresses in particular during the start-up or shutdown of plants.
The effects described above can all be estimated and evaluated transient (temporary) FE simulations.
I’m looking forward to your feedback,