Fixed base on tube steel colum 3

[dakota99]

I am designing a portal frame with tube steel columns that has a fixed base. Right now I have the base connection detailed with a 12”x12”x.75” base plate with 3/8” gussets going up 6” on the tube column and welded to the base plate and four anchor bolts. The numbers show that this should work I am trying to resist a 10ft-k moment. I am just not sure if this connection provides enough rigidity to be considered fixed. I would like to switch to w flange columns but unfortunately this all I have to work with.Comments welcome.

 

[cntw1953]

It is extremely difficult to have steel column with true fixity. Either I will design the frame for both conditions -column fixed & column pinned, which is very conservative and often non-desirable, or try to come up with reasonable column fixity and design the frame with rotational spring at the column bases, quite tedious this way though.

 

[ishvaaag]

Just some documents related to the issue...

http://www.solutionsforstructuralst.../670/31f56275-10b0-4756-9eda-e5e21d1362a7.pdf

http://heron.tudelft.nl/53-12/1.pdf

http://people.fsv.cvut.cz/~wald/Clanky v Adobe (Pdf)/CB_numerical-Liege_COST-C1.pdf

http://www.aisc.org/assets/0/544/550/668/418ad1c2-11e0-4a97-9ef5-062499545aa9.pdf

http://heron.tudelft.nl/53-12/4.pdf

http://conf.ncree.org.tw/Proceedings/i0951012/data/pdf\4ICEE-0069.pdf

http://heron.tudelft.nl/53-12/2.pdf

http://www.springerlink.com/content/0u5u7h8812168046/

http://www.iitk.ac.in/nicee/RP/2004_Monotonic_LateralLoad_13WCEE.pdf

http://www.ncree.gov.tw/iwsccc/PDF/28 - Hitaka.pdf

 

[DaveAtkins]

In my opinion, if you design everything as fixed (bottom of column, weld of column to base plate, base plate, anchor rods, and footing), then it is fixed.

And I think you can come up with a simpler detail--try a thicker base plate without the gusset plates.

DaveAtkins

 

[Stillerz]

AISC Design Guide 1 covers this

 

[cntw1953]

Dave:

I will do the same as you said if it is for concrete structures. However, I reserve my doubt on most of steel connections, it only needs a little slippage to cause huge loss in capacity. Old school thinking, maybe.

 

[JedClampett]

I'd stay away from fixed supports, altogether. It's much easier to fix your steel to steel connections than to fix the base. The main problem is the anchorage. When you get down and dirty with ACI 318, Appendix D, you're likely to have to deal with load factors on your moment (1.6 or 1.7); capacity reductions on your loads (anywhere from .65 to .9) and significant capacity reductions because of spacing and edge distances. And that's not even considering the stiffness impact of the bolt area stiffness vs. the column stiffness.
Anytime I see one of my co-workers designing frames, I recommend assuming the supports pinned. When they don't do it, they invariably come back and say, "I wish I would of listened."

 

[Stillerz]

It sounds like he is dealing with a flagpole situation. Its hard to get away from a fixed base in that case.
I too avoid moment bases on buildings when ever I can. It complicates the steel desing as well as the foundations. On portal frames I try to use pinned bases with moment conn.'s on the beams.

 

[Lion06]

It's true that it's difficult to get true fixity with any steel connection. A beam to column moment connection is limited by the stiffness of the column or beam (depending on which member is under consideration), but you use them when you need to.

I have many times had to fix the bases of moment frames just to get drift under control. That's a lot less expensive than throwing more steel weight at the columns/beams.

Another thing I've done is grab as many frames as you can with wind moment connections. They are cheap moment connections that do help laterally, especially with drift. These are tougher to do with HSS columns.

Your detail (with the stiffeners) should provide plenty of rigidity, but I would make the plate thicker and lose the stiffeners - that's an expensive detail. A baseplate is cheap, stiffeners and all the extra welding isn't.

Jed,
I don't disagree in principle, but I would say that you always have the ability to lap the anchor rods with pier/footing reinforcement to get away from the tension breakout requirements of App. D. This is typically only a problem with a pier. I rarely have a problem with anchorage into a footing with fixed base moment frames.

 

[Tomfh]

Strength issues aside, "fixed" baseplates often have far more give than a true rigid connection, hence you can end up with more deflection than you bargained for.

 

[rowingengineer]

why not design it as a semi-rigid connection and capture the rotation if only approximately.

Arguing with an engineer is like wrestling with a pig in mud. After a while you realize that they like it

 

[Stillerz]

We make many assumptions in design. Most designs are not possible without some assumptions. Base fixity is a basic design assumption. To me, this is just another case of making a science project out of a moment baseplate design which has been laid out in many texts and design guides. In fact, the lighter the columns, the more readily acheiveable base fixity is.

You can knit pick designs as much as you want, but ultimately we are basing much of our design on assummed loading conditions that I cannot believe are any more accurate than, say, assumming a fixed base....or pinned, for that matter.

 

[dakota99]

Thanks everybody for your input. After listening to some of the previous comments my design approach is such:

I have designed all of the strength requirements for the frame assuming a pinned base and fixed moment connections at the top. I also checked deflection under this assumption taking into account second order effects. The deflections were high and would require the columns to be mush larger to control deflections with this assumption.

My second analysis assumed a spring constant at the base of the column. I double checked to make sure all strength requirements were met and check deflections taking into account second order effects. The defections were within acceptable range for this application. This is what I used to size the columns for deflection.

I then checked the same frame with a fixed base. I made sure all of my strength requirements were met and checked deflections taking into account second order effects. The deflections were much less from a pinned base and probably over designed with this assumption.

 

[cntw1953]

Stillerz:

You are on both sides of the fence While I align with you more on the previous one, I do not disagree totally on the second, as long as one has looked a problem thoroughly, thus the outcomes would live up/come close to one's expectations, that were based on sound engineering judgement.

I am still wary on assuming full regidity on steel connections, unless a rigorous analysis was done. Avoidance is a good advice.

 

[Stillerz]

cntw1953:
Wary why, can I ask? Do you use your own entirely scientific design approaches? I am not trying to be coy here, but bear with me.
We (I, anyway) do code-based design. We are not designing space ships or even pianos. We make very wide-based assumptions in our designs.
To me, it is very analogous to significant digits inasmuch as you cannot miraculously come up with a perfect design given a starting point of wide-based assumptions. I am not a scientist, I simply practice an applied science...structural engineering. Since there are so many things we don't know and probably can never know, we use judgment.
You can't make chicken salad from chicken...you know the saying.
Does anyone really expect that the EOR do an analysis of base fixity on every structure he might design when the texts and codes he is using to design the structure make assumptions in their theory ? Any engineer doing this is fooling himself.
Finally, i will say, full rigidity is something that can be assumed given due regard to relative stiffness.
For an extreme example consider a very slender column full pen welded to an infinitely large plate, infinitely thick.

 

[rowingengineer]

I’m personally a fan of making assumptions that I can sleep with and not have dog problem. While a fixed connection can be assumed for ultimate limit state situations and comfortably be implemented assuming the components used are able to handle the strain requirements. To ensure this you must back up this with calculations showing such. That means the plate, bolts (both strength and embedment/edge distance), grout, footing ect must be check for strength and strain requirements.

“For an extreme example consider a very slender column full pen welded to an infinitely large plate, infinitely thick”, my query would be what about the footing, soil conditions?

Now let’s get serious, how many time have you been called out to site because an element of a design (by others would hope) has moved beyond the service requirements of the client. Now how many times have you been called out because the element has failed, for me it is about 99% service problems (sometimes by engineers making assumptions that cannot be backup). Thus I put forward, “strength is essential but otherwise doesn’t matter” (one of my favourite quotes).

Given all the calculations you have to do for the ultimate limit state there are only a few more that would be required to design as a semi-rigid connection, Why not take the extra step?

I am not meaning to pick on Stillerz but I would want more information from the OP before I suggested the fixed based assumption is appropriate. Even thou i am sure that Stillerz was NOT in anyway implying this.

I sure if dhengr was to post it would be twice as long as mine reminding people to get all the information before setting people off on course. What if the OP used two bolts located at the centre of the plate with two at the top and bottom, this would surely cause the problem to have to much deflection for our theories to be applicable, if the bolts and plte were design useing linear theroms. While I assume the OP has more sense than this I would like more info myself before I my approval of a fixed plate design assumption.

I would also point out that there are very few articles dealing with shear for fixed or semi-rigid baseplate with CHS columns, with less than 10 bolts.

Arguing with an engineer is like wrestling with a pig in mud. After a while you realize that they like it

 

[cntw1953]

rowing shook some load off my mind.
Happy new year, everyone.

 

[Lion06]

How do you design it as semi-rigid? How do you determine the moment rotation characteristics of the connection?

 

[cntw1953]

Quite tedious for each case/material involved. Usually start with looking into characteristics of deformation under load. It may require to throw a lot of assumptions based on experience and engineering judgement. The effort is trying to catch the worst scenario cloest to the actual, but exact.

Would a steel building fail (total collapse) for no other weakness but the assumption on support? It's not likely, but some localized defects could be the result.

Be prudent while work on frames with fixity, go over the details before make claim. I think there are quite many examples on moment connections could be used as a guide.

 

[rowingengineer]

StructuralEIT
I prefer to use the component method for semi-rigid connections for portal frame base plates, is a bit conservative compared to moment rotation curves comparisons or FEA. However it is effective in getting a number. We had a bit of a discussion in the AS/NZS area and I posted an article I found fairly helpful thread744-260012 you may wish to read (I would post it again but I’m at home). The article doesn’t cover foundations; however there are large amounts of information on stiffness of foundations. As for the OP’s problem, we need a lot more information before we can suggest the spring stiffness.


Arguing with an engineer is like wrestling with a pig in mud. After a while you realize that they like it