Beam Perforation Program.

[pauljohn]

Hi

I am trying to get a 'feel' for how much a beam (rectangle tube steel) is weakened by drilling holes in it and have been looking for free software on the web that might show me this with no luck. I do not wish to solve complex problems but just want to get some idea if I drill a beam to run pipes, conduit, etc. how it might affect the strength and load carrying abilities and how I can keep the strength to a max. I have no previous analysis software experience. Any help finding such a program will be much appreciated.
Thanks for any help.

 

[EdR]

Pauljohn:

Why don't you do this problem by hand and draw a series of curves based on the size of the hole. It's a whole lot easier to do such a simple problem by hand than to try and do it with a computer code.......

Ed.R.

 

[pauljohn]

Hi Ed
thanks for the reply. I think I might of oversimplified my question.(But if you still think it is easy I will return to my textbooks for help!)

Once I get a feel for the weakening process as I enlarge the holes I want to try and strengthen the beam by welding tube into the hole. I would also like to know and understand how this is going to affect the beam in bending and torsional loads and, if I was really lucky, maybe fatigue from cyclic loading.

To be honest I don't know how to do this on paper. Should I post this problem in another forum?

 

[EdR]

Unless I'm not understanding your problem it is just a simple beam bending, beam shear, and maybe torsional problem. Each of these (possibly except the torsion problem) is a simple Mechanics of Materials problem that you should be able to find in any MM text book (or Roarks Handbook)....If you don't understand the behaviour of these problems (as your comment about doing it on paper indicates) then you really need to get a structural engineer involved that does understand the behaviour...

Even doing the problems using some computer code is ill advised if you don't understand the basics of the system you are analyzing.....

Are you really ready to accept responsibility for the consequences of a failure of something that a structural engineer might easily avoid??? Sorry for the chide but just because (I, you, someone else) has a PhD or PE ,etc. added to their signature doesn't mean they are qualified in all areas of engineering......and if your boss, customer, etc. does not understand this then you are working in the wrong place.....

Ed.R.

P.S. I think this is the correct forum....Wait until the first of the week and see if you get any other suggestions..

 

[GregLocock]

While I think analysis by hand may be a valid approach, if the holes are large then the problem rapidly gets unwieldy, and quite possibly only solvable for certain special cases.

So, free FEA software that will do what you want - calculix. use this to build a plate type model of the beam (ie with say 4 elements to a side as a minimum).

From this you'll get stress concentration factors around the holes and the reduction in bending and torsional stiffness of the beam. Your mesh density around the edge of the holes will need to be quite high, the transition from that detailed mesh tot eh general run of the rest of the beam is where you will spend your time. You can then use the SCFs to figure out the effect on fatigue life- actually modelling fatigue life with confidence is possible but still a bit of a black art in my opinion, and I'm not really convinced that the programs that do it are really doing much more than applying SCFs to known stress fields and material properties.

If you want to persevere by hand then I suspect Bruhn will have a chapter on cutting holes in beams, aircraft boys like to do it.

Some idea of the relative size of the holes and the section would be worthwhile - if the hole is much smaller than say 10% of the beam depth then hand analysis is probably sufficient.






Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[pauljohn]

I shall give a little more info, if only to calm the fears of the ‘responsible’ engineers on this forum and show them that I am not one of those ‘irresponsible’ engineers trying to deceive his boss/customer (both myself btw)! ; )

I am working on a small boat (50’) where the builder thought it necessary to use up his left-over stack of steel tube (2x3, 2x4, etc) for various types of reinforcement. It’s easy to see that the tube in places is more than adequate and could be lightened by some enthusiastic drilling but being a ‘keen engineer’ I am also trying to solve two problems at the same time by using the holes as convenient places to run hoses and wiring (seems like an obvious thing to do). I don’t have a problem with most of the drilling but on occasion it might be necessary, depending on the size of hose, to drill a 3” hole in a 4” beam or maybe a 4.5” hole in a 5” sidewall, and at this point I would like to understand more about how the beam is going to behave after removing this much material.

I would also like to know how much of the strength I can regain by welding pipe back into the hole (so I can still run hose through it) on the odd occasion it needs it if the hole is getting close to the size of the sidewall. So my question stems more from a learning point of view rather than an actual problem I am trying to solve as it is probably impossible to estimate any of the forces involved with any degree of accuracy anyway.

Hence my quest of a simple program I could play around with that might give me some insight.

Greg – thanks for the link but I am unable to run that program let alone use it, I fear it is way beyond my limited abilities on software use.

 

[corus]

Castellated beams (ie. with holes in the web) are used quite often when the shear stress is low relative to the bending stresses as most fo the strength comes from the flanges. With such large holes I'd be concerned about localised buckling around the holes and I'd probably treat each half of the 'holy' beam as two seperate beams (say T sections formed from an I beam) with direct loads equivalent to any bending loads on the whole section. As GregLocok says, you'd also have to consider the stress concentration effect of the hole for fatigue assessment, if you have varying loads. Replacing the hole in the web by an equivalent amount of material in a reinforcing pipe (looking at a cross section) maybe a rough guide to size the pipe, but if the region is critical then I'd consider using FE.



corus

 

[GregLocock]

Has anybody got Bruhn to hand? - the idea of cutting HUGE holes in a shear web and then reinforcing with a welded-in ring, or a doubler, seems like the sort of thing you'd do in an aircraft, properly.




Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[111213]

hi,

thi is done easily by modeling a vierendeel frame around
the opening....

statiker

 

[GregLocock]

OK< I looked at one example, a 3 inch hole through the web of a 5 inch box section. Due to lack of nodes and brains I didn't do torsion.

Punching the hole through doubles the general Von mises stress in bending, at the top and bottom of the hole.

It also increases the bending deflection by 25% over an 8 inch section.

My belt and braces solution was to weld a tube through the hole, same wall thickness. This brings the stress down to 120% of nominal, and would also work wonders for torsion I guess. However, the stiffness was still down by 12%, due to ovalling of the tube, so I added a doubler, OD 5, ID 3, over the end. This got the stress and deflection back to where we started.

However, your welder will not be very happy, if there are many of these to do. If I were building for posterity then it would be the way to do it.

These results are very rough, have you found any other approach?






Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Co]

To add to the above valuable posts, from your mentioned size of the holes within the structural members I believe that they may compromise their structural integrity depending on the loads. This can be thought of as an obvious conclusion depending on your loads which I know nothing about.
My experience with thin-walled members (which I assume this to be) leans towards an average nominal stress value that about doubles (or more) as compared with its non holed counterpart. Here, I assumed that the holed 'tube' could be modeled as a thin-walled open cross section. Also note here that in the thin walled structural theory the values can be much higher.
Please note that I'm not sure if this analysis applies to your situation.
Have you thought of reinforcing the members on either side with a brace at the location of the stress concentration?(this was my first thought when I read the problem)

cheers,

 

[Co]

Just a thought. Maybe you are better off using finite elements as mentioned by others. The Comsol Multiphysics version I use actually has an example of a cylinder/tube with a hole in it. They then apply a stress/fatigue analysis, fairly interesting. You may want to look into something similar.
I attached a screen shot.

cheers,

 

[pauljohn]

Thank you for replies. Appreciate your time Greg on running the analysis. I am happy with only losing 10 or even 20% of stiffness. I think I will try and keep the holes down to about 60 or 70% of wall height. Not really following you on what a 'doubler' is and yes I knew from the beginning it was going to be rough estimation at best. But this is a boat and I make lots of assumptions, makes for interesting sailing! Thanks again.

 

[GregLocock]

A doubler is just a ring, welded on to the side of the box. If I were to do many of them I'd do it that way and forget the tube, making the doubler rather larger and thicker.



Cheers

Greg Locock

SIGlease see FAQ731-376 for tips on how to make the best use of Eng-Tips.

 

[Co]

In my opinion rough estimates are not the way to go. Nowadays, we are a little more advanced then that (hint hint).