Wide-Flange Brace to Wide-Flange Column Connect. Design... 4

[jochav5280]

I am a new engineer and have found it fairly difficult to find a good connection design resource (unfortunately, they don't seem to teach but the basics in Undergrad level courses). Currently, I need to design a Wide-Flange (W10x49) brace that comes into a Wide-Flange (W10x30) column at a steep 13 degree angle relative to the vertical position. I've been advised by my counterparts to just make the connection "stout" and not to worry about it; but I don't think this is the responsible way to work. Any advise on what resources I might be able to use in this situation and or what limit states as per the AISC spec I need to check? Thank you!!!

 

[Lion06]

Is this a knee brace?

 

[kslee1000]

Do you checked geometry? For such steep angle, it will take a lot of load, and shear (at column face) would be the focus of design.

 

[csd72]

There have been a few posts on this type of thing recently, I would suggest you use the google search option at the top of the page and look at advice from similar topics (start with the words brace connection).

The AISC specification section J covers design of this type of connection.

I would look at, at the very minimum, the following points:

Bolts
Welds
bolt bearing
plate tearout/shear
plate buckling
Transfer of these loads to the column/base plate.

I would definately recommend that you do a full calculation the first time you encounter a connection. You will soon get a feel for what is critical.

 

[jochav5280]

Thanks for the advice; I guess what I am specifically worried about is analyzing the additional load that will be transferred by the incoming bracing to the column. The column is currently stressed to 80% capacity and I would like to evaluate if the additional force introduced by the brace will overstress the connecting column flange or column cross-section in general (the incoming brace is bringing roughly 150 kips of compression into the column). I was contemplating looking at the columns max. allowable stress it can take in compression as well as evaluating what the max. allowable strong axis bending stress it can take using (My/I), and then adding the additional stressed caused by the incoming brace. I'm just not sure exactly how the stress distribution will look. I tried looking up some resources for this type of connection and have been unsucessful thus far, but will keep trying. Last week I spoke with a P.E. from AISC, and he had very little to contribute.

 

[csd72]

If the centre line of the brace and the column intersect at the center line of the supporting member then there will be no additional stress in the column, the force will go directly into the supporting member.

 

[hokie66]

On a connection that steep, I would tend to provide an end plate on the brace and a welded bracket on the column. That way you don't have to design a long connection depending on bolts in shear and long fin plates acting as columns.

 

[spats]

Obviously, the column takes the vertical component of the brace load in compression. The column can also experience bending from the brace depending on connection geometry. Your column sounds a little light for the brace size.

An excellent reference for your situation, and steel connections in general, is Handbook of Structural Steel Connection Design and Details, by Akbar R. Tamboli (Mc Graw-Hill).

With brace connections, it's best (and easiest) to use the uniform force method, which eliminates bending in the connection. The uniform force method is explained in the book.

 

[frv]

jochav5280-

The advice you have been given by your counterparts is not necessarily bad advice. You will find that for something like that, the money you save by detailing an efficient connection is far outweighed by the time it takes you to properly analyze and design it; consequently, your time is not being utilized efficiently.

In an marketplace that's ever more competitive, you need to find areas where you can save a little time. I'm certainly not suggesting doing anything even remotely suspect. Sometimes, however, you may be able to make a few simplifying, conservative assumptions, use a little bit more material and save yourself a lot of time.

I'm sure you've seen something like this already. Columns are a good example. In our area, from a strictly structural perspective, columns could frequently be much more efficient than what we use. For ease of constructibility, however, they are typically oversized. Again, this goes to the bottom line. The quicker you can build something and the less hassle you have connecting other members to it, the quicker the owner can occupy the building and begin receiving revenue (or at least stop losing revenue on whatever building the new one is replacing).

For a new engineer, it may be worthwhile to go through the exercise once so that you can get a feel for how much an efficient connection differs from a larger, simpler, one.

 

[spats]

frv,

Good Lord! You're advising jochav5280, a self-professed "new engineer", to take a shortcut on something he doesn't even understand! It's not a matter of an "efficient" connection vs a "conservative" one, it's a matter of having a connection that actually can take the forces present. 150 kips of vertical brace force is a serious amount of load.

I occasionally design connections for steel fabricators, particularly bracing connections. Sometimes the steel detailer takes a "wack" at the connection first, and ask me to verify their "design". You wouldn't believe the differences between their "design" and mine in a lot of instances!

 

[frv]

spats-

I urge you to fully read my post. Please tell me where I advised jochav5280 to take a shortcut.

My post was intended to address his comment regarding the "responsible" way to work. If you had taken the time to fully read my post, this would become obvious.

I am a fairly inexperienced engineer myself. I have never once taken a shortcut or even made a simplifying assumption without satisfying two things: 1) I have designed whatever it is before properly without any simplifications and 2) I fully understand the effect of my simplification on the design.

My only intent was to point out that the advice he received was not unreasonable, as his post insinuates.

 

[csd72]

frv,

You do make a valid point in that engineering time often costs more than the connection itself.

But I would argue that an engineer needs to learn how to design things properly before they can learn to do things efficiently.

 

[frv]

csd72-

You are absolutely right. That's why I mentioned the fact that I would never simplify anything without first having designed whatever it is without making such simplifications.

Once you understand how something works, you may use your simplifications.

 

[csd72]

Si Comprende

Sorry, I too am guilty of not fully reading your post, my job gets in the way of my eng-tips.

 

[youngstructural]

frv:

I read your post, the WHOLE of your post, and think it is quite good advice. Constructibility is often ignored by engineers who are trying to get the absolute minimum of size... Not always the best course of action. And $1500 of engineering to save $750 worth of material is not good economy. A star for your troubles.

jochav: Save your pencil sharpening for the items that repeat, whether they be columns, beams, trusses, connections, or whatever else.

Cheers,

YS

B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...

 

[spats]

jochav5280 asked a direct question about how to DESIGN the connection, and that's what I tried to address, not to offer philosophies on when one should do a detailed analysis. Besides, I'd be interested to know how one would come up with "a few simplifying, conservative assumptions" for such a connection, other than the use of the Uniform Force Method. You still have to run the numbers for this kind of load. SWAGing doesn't work unless you're a lot smarter than the rest of us.

 

[DWHA]

Well because I too get upset when I post a question and other suggest a way around the issue (and I too am guilty of suggesting ways around the issue). Here is how I would recommend approaching it.

First is the knee brace designed as pinned or fixed? That is the type of connection that you need to detail.

ASSUMEing pinned, I would weld a plate to the w10x30 and then either bolt or field weld it to the w10x49. Remember that if the line of force does not pass thru the centroid of the w10x30 weld you have an eccentric weld. The plate would be designed as a compression member. Size bolts and welds as required.

At the bottom of the W10x30, check the member for compression, the member will not buckle do to added force because unbraced lenght = 0 so just check yelding and possibly local buckling, however a 10x30 does not sound light enough to have local buckling. But check that anyway.

Also check shear in the w10x30, it should not control due to the angle. But you made it sound like you wanted to be complete.

 

[youngstructural]

Okay, let's not let this devolve into banter!

spats: Please do not feel that other people's comments form any kind of personal attack. frv took your criticism in stride and replied calmly. I would also think that addressing any comment made in the original post is not off-topic. Jochav's question might have been about a specific design procedure for this connection, but the motive behind the question is just as germaine. Chicken and the egg, n'est pas?

Designing a connection such as this can easily be done for the moment capacity of the section, peak shear of the overall system, etc, and apply generally through multiple similar situations. Often a design resource is available with fully developped connections for somewhat arbitrary load sets (such as 100% moment, 50% shear or 75% moment, 35% shear, etc). I do not believe frv was advocating neglecting giving the connection consideration, but rather neglecting SPECIFIC design for a particular connection, and applying a general (or previously worked out connection) with a capacity greater than that required. It's a matter of developping engineering judgement, something that will serve us all much better than repeatedly applying detailed design procedures.

Let's keep the hostility to a minimum, and remember that we address our peers. Some courtesy, decorum, and professional respect is in order.

Regards,

YS

B.Eng (Carleton)
Working in New Zealand, thinking of my snow covered home...

 

[frv]

spats-

This is an advice forum. We are free to exchange viewpoints and pointers, even if technically unsolicited. But in the strictest sense possible, you are right- I did not answer his question; rather, I attempted (apparently unsuccessfully) to point out that his assessment of his colleagues was a bit harsh and that there is some validity to their suggestions.

Yes. I do consider myself fairly smart. I believe anyone who graduates with an engineering degree from even a somewhat reputable school has to be smart. I can assure you I am not smarter than everyone on this forum (or -possibly- anyone on this forum, if you ask my wife). Neither are you. That's the pint of it. Together we can be fairly bright.

I ask you to read almost any random thread in this forum and you will find extensive examples of unsolicited advice. In fact, several threads devolve into a discussion on something entirely different than the subject upon which the thread was based. Judging from the amount of people who chime in on these discussions, it seems that, although unsolicited, these tangential subjects are interesting enough.

anyway- feel free to give me the bird from your desk if you so wish. You certainly won't be the last.

Youngstructural- thanks!; this reminds me of the thread on developing plastic hinges from a few weeks ago.

 

[spats]

This is a forum where everyone can certainly express their opinion, and I have no problem with that. It's a good thing!I do however, feel VERY STRONGLY that it is bad advice to a young engineer to tell them how they can dance around an issue, rather than properly addressing it... DWHA seems to agree. If I came across as sounding too critical and intolerant, I apologize, but I have mentored more than a few engineers in my 36 years of designing buildings, and I would never offer them the advice that jochav5280's "counterparts" did. jochav5280 is right... it is not a responsible way to work. I applaud him for his diligence!

Sure there are many cases where conservative is better than designing something to death, and wasting time, especially for a one-of-a-kind detail... this particular instance just doesn't happen to be one of them. I'm sure you would agree, if the truth be told, that jochav5280's "counterparts" probably didn't really know how to answer the question. Maybe they've been doing too much dancing, and not enough designing.