PPRuNe Forums - View Single Post - Eight B787 pulled from service over structural issues
Old 6th Oct 2020, 21:32
  #62 (permalink)  
unmanned_droid
 
Join Date: Oct 2013
Location: UK
Age: 42
Posts: 654
Received 9 Likes on 8 Posts
Volume

Shims above a certain thickness cannot be used as there is a knock-down in joint static strength and fatigue strength. Above a certain thickness, capability falls off a cliff edge. You can also use wet shim up to a certain thickness. This appears to be a limitation of the manufacturing process (process variation). I expect that it was not a pop-up issue. Somebody would have authorised a design manual(s) for the product, and in that manual are a list of all the allowed tolerances for parts in design. The digital model is created as 'nominal' with appropriate spacing built in for positional tolerancing. Above and beyond that, tolerance studies are carried out for large scale assemblies to assess the assembly process. Some of those tolerances are well understood (i.e. machined aluminium), some of them come from new process data. All are as a result of some level of testing and measurement process. Then you get in to the production side of things. There are quality auditing processes on purpose, and some are more fit for purpose than others. That possibly a quality process was abandoned with no replacement or improvement, primarily in order to make rate is cause for concern over the whole assembly process within that particular function/building/company unit. I find it hard to believe that there wasn't a manufacturing engineering task to look in to this one way or another. This could have been because the process wasn't fit for purpose, or because the process would show parts to be out of spec, or because the part condition wouldn't allow the user to carry out the inspection. The worst case scenario is that these measurements were not taken and things that needed a concession carried out on them have not been found and checked. Usually these kinds of things show themselves as a fatigue issue rather than a static strength issue.

This is the problem with commercial aircraft sized composites - part tolerances are more like cheap car assembly or worse than what we're used to (and require for our strength data and analyses) with metallic aircraft, for joints anyway. To compound the issue, B and A want to build airliners like Toyota and Honda churn out cars. Of course they do - there are thousands of single aisle in the order list at A, for instance. Something has to give. You get to choose two of: On Time, On Cost & On Quality.
unmanned_droid is offline