Machined UB/UC Sections

[Drej]

Has anyone ever come across the application or use of "machined" UB / UC sections for use within nuclear or any other safety-related sector?

By "machined sections" I mean the material is bought in as a billet and machined in-house to produce the UBs / UCs.

The machined sections need to be seismically qualified as part of a support structure, and I'm interested to understand any / all issues associated with this approach.

Thanks for any input.


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com

 

[dhengr]

Why on earth would you want to do that; other than to enrich a favored machine shop and produce a whole bunch of scrap metal in small shavings? What assurance is there that the billet is pure and perfect, so you don’t spend hours machining to run into a defect just short of the finished piece? Actually, rolling is a pretty efficient way to use all the material and turn it into a useable shape which can still be tested for imperfections if you wish. You can fabricate unusual shapes too, and test them, if you need a special shape. Alternatively, for a small special shape, you certainly may machine a part, but usually you make your design fit normal available materials. I believe the aircraft industry, and a few special industries like that do machine special shapes out of larger plates or billets. But not likely something as plane as a UB or UC shape.

 

[Drej]

dhengr - thanks for the reply.

This is currently just an idea kicking around and is being investigated to assess its suitability. Clearly there are issues with it and I'm interested to garner people's thoughts. The issues you've raised have also been raised here, and the problem we have with using standard rolled UB / UC sections is the use of full pen welds (which is required) and also access for certain welds, hence the whole thing to be machined to remove the requirement for welds.


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com

 

[dhengr]

The cross sectional area at the junction of the flange and web, in a rolled section, is solid because of the nature of the manufacturing process, not because it is needed structurally, or from the stress, durability or strength standpoint. And unfortunately, because so many engineers and code and standard writers don’t know how to design quality detailed welds, the default solution in their minds has been that everything must be full pen. welded. After all, they were just trying to replicate a WF shape, for a structural purpose, so not knowing better, the web to flange weld had to be full pen., solid like the rolled beam. Many good arguments can be made that over welding is detrimental and very costly. I have seen this full pen. welds req’rmt. on many mechanical written codes, specs. and stds. and there is just no evidence that it is needed. I would argue against the need for that req’rmt. and show plenty of reports, testing and experience which shows that it is not required. And, I still think I could weld the web/flg. joints on a 6' long built-up WF shape quicker and at less cost than you would machine it out of a solid steel billet. There is nothing difficult about these welds, there are no access problems at this stage, distortion from welding must be managed and controlled, not too tough with experience in these methods. The joinery beyond this point becomes more complicated, but your design should be done to manage and minimize this difficulty. And, machining a complex joint out of a solid is no less difficult.

 

[MikeHalloran]

The cost of machining a part from billet is not what it used to be, mostly because a CNC machine can run mostly unattended, so the skilled part of the labor force can be doing other things.

Rolled sections are nominally cheaper than machined billet, but sometimes getting a short piece of an uncommon section to the place where you need it can be expensive enough to justify hogging from a simpler shape.

This sort of thing can make even more sense if the machining can also produce features that you would otherwise need to add, eliminating the fabrication and assembly cost of those features.



Mike Halloran
Pembroke Pines, FL, USA

 

[Drej]

Cheers. The web-to-flange weld isn't the problem, the spec forces us to provide a full pen weld at the section-footplate interface and also at the section-cross member interface - absolutely no fillets or part pens allowed. This is where we have access problems also. The full pen weld is a blanket requirement in the spec but from a structural aspect a fillet or part pen is more than adequate given the loadcases. Questions for the machined section being raised include mechanical strength properties of this section, specifically that the billet grain structure results in reduced mechanical props because of the unfavourable grain structure compared to a rolled section. Personally I've never had to machine a section for a design and it just seems overkill.


------------
See faq569-1083 for details on how to make best use of Eng-Tips.com