I have a unit that is bolted down into a vehicle. The unit has a mass of 45 kilograms and is around 500mm wide x 300mm high x 450mm deep. C G at centre. The unit is attached to the vehicle by means of 450mm long aluminium angle brackets running along each side. These brackets are approximately 50 x 60 x 4 mm thick. Each angle bracket is attached to the vehicle using 3 off vertical M6 steel bolts. The vehicle is assumed to be very stiff.
The unit is subjected to g loads of 7.4g forward, 2.25g sideward and 4.4g upward. A quick calculation, making simple assumptions, and with just the 7.4g forward load gives a shear stress of 7.4 x 9.81 x 45 N / 120 mm^2 = 27.2 N/mm^2.
When a shell model is constructed and run in Mechanica, with spot welds representing the bolts, the results give a shear stress in the rear bolts of 600 N/mm^2, or almost 20 times the simple calculated value.
On the basis of the analysis results the owner of the vehicle was informed that his 6 bolts were inadequate, but he wouldn't buy it.
Increasing the number of bolts to 7 per side and ramping their material spec up would have meant that they would just about survive, according to Mechanica.
There seems to be such a large difference between simple calculation and Mechanica result. It seems logical that the Mechanica derived stress should be reasonably close to the simple calculated value. Are we not always taught to back up analysis results with a simple calculation?
Unfortunately there is no way that I can check the analysis because it was carried out by a third party, who would want to be paid for any further work. Being an ex Mechanica user I would dearly like to figure out what is happening to give such high stresses. The model is easy to construct and quick to run.
Any of you Simmies had similar strange experiences with unusually excessive shear stresses appearing in bolted joints, and managed to figure out why?