Critical design reviews have important function- even down to the dimmensoning and tollerancing on individual parts and assemblies.
Some years back, I was having a conversation with a machinist friend in his shop. We discussed the various aspects of our professions-mine being mechanical and aeronautical engineering-and we got onto the subject of sensible design practices. He related to me an instance where a mechanical design engineer placed a tolerance locating a screw hole at plus or minus 0.0005 inches.
I commented that normal drill bits had a tolerance of plus or -0.003 in diameter and the challenges behind trying to locate a feature on a part to such tight tolerances. Yes, it could be done, but I made the statement that there needed to be a very good reason to do this, as special tooling would be needed not only in making the hole but also in locating it. It would require not simply a standard drill bit but also the use of a reamer to get the tolerances correct as well as very precise positioning on a end mill. This does not even mention the requirements to position the hole and tap the threads, as it would have to be located from an extremely accurate reference datum in both X and Y planes.
We discussed the challenges of threading a hole and maintaining an accurate position following the threading operation. I continued with stating that we would be better off to use dowel pins to locate the two features together followed by screw threads to hold the two parts together. One dowel pin hole needs to be a press fit, and the other a very tight clearance hole. Likely both requiring reaming to different sizes, or using tapered dowel pins, and so on and so on.
Then I asked the question -“why was this feature needed?” The answer I got explained it all. “I just wanted to see if they could do it.” paraphrasing the design engineer for the assembly under discussion when asked the same question. I was shaking my head at this.
Why do such wasteful things? I can hear the conversation now. The machinist will question the need for the features that the designer will insist that it’s in place for this, that, and another reason; effectively just pulling rank on the machinist that has to perform the task. The machinist at this point might be scratching his head by looking at the design and trying to figure out why this feature is in there and ruminating on the rationalizations of the design engineer.
Insisting on features of this nature for prototype parts is one thing, as it will only be built a few times. But if it carries through to manufacturing, the cost of the assembly has gone up for no reason. Specifying a tolerance like this and building it under a manufacturing environment is one thing, but then it must be inspected. Special gauges may need to be built as ordinary equipment cannot get down to this level of precision.
Because of the close tolerance, the temperatures at which this feature is put in needs to be monitored and controlled, as well as the temperatures during inspection need to be monitored and controlled, to say nothing of the coordinate measuring tool needed to make the tolerance checks. Recalling that parts made from steel expand and contract at rates of 0.000006 inches per inch of thickness per degree Fahrenheit, so a 10-degree difference on a feature located 10 inches from a datum line will force this feature out of tolerance by 0.0001 inches. Can the measuring equipment even get down this low? The inspectors must follow the print and would be forced to reject a part out of tolerance even by this tiny amount.
Norman T. Neher, P.E.
Analytical Engineering Services, Inc.
Elko New Market, MN
www.aesmn.org