|Sharp corners in machines don't mix. An ample radius is required in order to minimize the chances of a crack initiating at the point of highest stress concentration. However, what constitutes an "ample" radius?
A few years ago one of these machines, shown in the picture to the left, cracked and failed. The location of that crack was at a sharp machined corner where a dozen brackets intersected the shell. The location of one of those typical sharp corners is highlighted in red. A close up view of one of the sharp corners is shown in the pictures to the lower left and right. The crack initiated at the sharp corner that is highlighted on both sides by red lines. The picture to the lower right clearly shows the very sharp profile that the rib makes with respect to its transition into the bracket. If the designers had located the radius at the red line there would never have been a failure.
|A finite element analysis was performed on the entire machine. The stresses at the sharp corner were well above the yield point of the material. The fact that the model predicted the location of the failure corroborated the model. Therefore, there was confidence that the model could be modified to validate an idea. The idea was that the problem could be reduced if the sharp corner was eliminated by machining a radius into the bracket. However, it was unknown as to what size radius should be used. Therefore, several radii sizes were configured into the model. Each radius was tested in the model individually, and the optimum size was determined. The radius was then machined into each bracket. The picture at the top of the article shows the radius being cut into a bracket.
|A typical finished radius is shown in the picture to the left. It has been checked for pre-existing cracks using liquid dye-penetrant. No cracks were detected. The finite element model showed that the stresses in the region of the new radius were reduced by a factor of ten. Notice how the radius undercuts the bracket. This is important.
On a machine that was different than the one shown above, a radius already existed on it. However, the finite element model showed that the smaller radius and orientation was insufficient to lower the high yield stresses. A liquid dye-penetrant check of the brackets corroborated the model. There were cracks in 10 out of 13 brackets on this machine. A typical crack is shown in the picture to the left. The radius is poorly oriented where it transitions into the bracket. The bracket surface is not on the tangent line with the radius diameter.
The picture above shows the machine that failed. The crack started at the sharp corner and traveled horizontally outwards, eventually connecting with other bracket cracks. Speculation suggests that the designers were aware of the high stresses that sharp corners create. In the picture above, there is a clear indication that a radius was cast into the original machine. However, notice how close the bearing housing is to the corner that was machined into the radius. How much do you want to bet that there was more than one engineering group involved in this design and neither one talked to the other? It is clear that an ample radius would have interfered with the bearing housing location. Therefore, to accommodate the bearing housing location, the radius was cut away by somebody other than the original designers.
Note: For those of you who have eagle eyes, yes, there are two different style filters mixed in the series of pictures. The last two pictures are from one style, while the first four are the other style. However, in both cases, the lack of appreciation for sharp corners is overwhelmingly evident.
All Pictures and Text Copyright © 2001 - 2016 Contact Mr. Adler Adler Engineering LLC of Wyoming USA
Great care has been taken in the compilation of this article. However, no warranty, expressed or implied, including without limitation, warranties of merchantability or fitness for a particular purpose, are given in connection with this article or any article archived on this website. Although this information is believed to be accurate by the author, the author cannot guarantee favorable results will be obtained from the use of this article alone. This article is intended for use by persons at their sole discretion and risk. Since the conditions of product or material use are outside of the author's control, the author assumes no liability or obligation in connection with any use of this information. The author is not liable for special, indirect or consequential damages resulting from the use of this material.
No part of this article or any article archived in this website, or any part thereof, may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, without the prior written permission of the copyright holder R. H. Adler. Nothing contained in this article or any article archived in this website shall be construed as a grant of any right of manufacture, sale, use, or reproduction, in connection with any method, process, apparatus, product, composition, or system, whether or not covered by letters of patent, copyright, or trademark, and nothing contained in this article or any article archived in this website, shall be construed as a defense against any alleged infringement of letters of patent, copyright, or trademark, or as a defense against any liability for such infringement.