Baseplate - Large Moment - Anchors within flanges - AISC Design Guide 1? 5

[StructureMan44]

I'm attempting to make a 11"x11" baseplate work below a W10x49 subjected to 20k-ft moment and 10kip shear in both directions, and a 2 kip tensile load. I'm trying to place the anchors within the column flanges. The column loading creates a "large moment" situation per AISC design guide 1 section 3.4. When placing the anchors between the flanges what values should be used for n and m? This impacts the equations given for determining the minimum plate width. Also is it likely that the baseplate would have no area in bearing and instead two of the four anchors resist the moment and uplift entirely?

 

[WARose]

Anchor bolts that fall within the (square) perimeter formed by the flange edges theoretically transfer no moment. In actuality, there will be a bit from compatibility as the structure deflects.....but your model should not call this (anywhere near) a fully fixed base.

Also is it likely that the baseplate would have no area in bearing and instead two of the four anchors resist the moment and uplift entirely?

There will be a compression zone under the plate (resisted by bending in the plate) and there will be some uplift resisted by 2 bolts. See McCormac's book on steel design for this situation.

 

[dik]

WARose: I didn't know that. Do you have a source? For small moments, I've often used anchorrods within the flange area.

Dik

 

[WARose]

WARose: I didn't know that. Do you have a source? For small moments, I've often used anchorrods within the flange area.

Pretty much every reference I've seen had the bolts on the outside of the flange perimeter for moment resistance. (See the one I mentioned before: 'Structural Steel Design: ASD Method', Jack McCormac, 4th edition, Fig. A-6, p. 713.)

If you have enough compressive axial load, the plate might be in full contact/bearing (whether the bolts are on the inside/outside or not).....but min. dead load and a wind/seismic load case typically winds up controlling for that aspect (i.e. plate thickness) of the base design.

 

[JLNJ]

While bolts inside the flange limits surely can carry some moment, the connection has a lot of inherent flexibility. If this doesn't adversely affect your sway or frame action moment distribution, then go for it. If you need complete fixity to make the structure work, then this arrangement needs closer scrutiny.

I'd center the compression block under the flange for the moment in the strong direction and distribute the tension accordingly in the outermost bolts. Moment in the weak direction would give me some pause. The base plate will need to work pretty hard. Maybe the moment is small enough that the compression is mainly under the flange tips.

I think you are going to get some large bolt forces. 20'-k x 12"/8" = 15 kips per bolt each direction, so 30k per bolt if added directly. That's going to wreak havoc on your base plate and maybe even your welds.

 

[KootK]

Bolt patterbs inside the flanges can transmit moment. To a greater extent than we'd like in many of our "pinned" connections really. I've long been baffled by the binary emphasis placed on bolts being outside the flange for moment connections. That improves your lever arm and reduces your bolt forces which is great, and often advisable. For strong axis moment on a thin plate, a reasonable lever arm assumes compression centered under the leading flange and tension wherever the rear bolts are. Inside, outside, Spokane... It's just a number. In a way, an inside bolt might actually prove stiffer as the base plate condition is less of a cantilever.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[dik]

JLNJ: I would suggest that bolting outside the flange would be far more flexible that bolts in proximity to the flange and web. No question that the moment capacity is restricted due to the shorter moment arm.

Should have added that bolts outside are a bigger 'tripping' hazard. <G>

Dik

 

[WARose]

The thing about it is: I've questioned the full fixity assumption even when the bolts are on the outside of the flange. Unless we are talking a cantilever, that assumption may not be valid. I've done models taking into account the plate flexibility, foundation rotation, etc.....and it's not quite as "fixed" as you would think.

Just another grey area in what we do.

 

[KootK]

I would suggest that bolting outside the flange would be far more flexible that bolts in proximity to the flange and web

Must be a crazy Canuck thing...

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[WARose]

JLNJ: I would suggest that bolting outside the flange would be far more flexible that bolts in proximity to the flange and web.

IIRC, depends on the thickness of the plate. (And the situation.) The thing about the "inside" deal is: it takes a good amount of rotation to "engage" it (in some cases).

 

[dik]

Must be a crazy Canuck thing...

Just an engineering thing... have had almost no mentoring in my entire career. And in the limit condition (which is the way I've always designed - we were the first group to learn limit states, back in '65). I use a rectangular stress block (like concrete) for anchor rod moment design.

Dik

 

[KootK]

Just an engineering thing..

I was just teasing. We're both Canadian and we both had the same, somewhat controversial stance on this.

In a way, an inside bolt might actually prove stiffer as the base plate condition is less of a cantilever.

I would suggest that bolting outside the flange would be far more flexible that bolts in proximity to the flange and web

have had almost no mentoring in my entire career.

Not everybody needs it. I doubt Stevie Ray Vaughn spent a lot of time getting guitar lessons after school.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[dik]

KootK... yup.

Dik

 

[JLNJ]

Agreed on the flexibility for bolts outside the flanges, especially if located out at the corners.

However, if we say the lever arm from bolt to compression area under the flange goes from 8" (bolts inside the flanges) to 12" (outside) the lever arm increase alone creates great deal more stiffness for a given bolt size (2.25x by the square of the distance). If the plate size stays the same, does this overcome the plate flexibility? Probably.

 

[dik]

JLNJ: For small moments it works well. I was not aware that there was information advocating always putting anchorrods outside the column; that surprised me.


There are added effects of putting the anchorrods outside... the pier may get bigger, and the distance from the anchorrod to the pier surface may be diminished.

Dik

 

[dik]

have had almost no mentoring in my entire career

That's not quite correct... when I started out, I had very good examples of guidance in how 'to look at things' from Arnold Crosier, Paul Krauss, and Ron Lazar, all deceased. But not guidance in how to design.

Dik

 

[canwesteng]

Anchor rods inside the flange have the added benefit of the flange and the web forming a T around the rods to act as a stiffener. If I need to put the rods outside the flanges for additional capacity I often put an additional stiffener, similar to the design of an extended end plate connection.

 

[KootK]

the lever arm increase alone creates great deal more stiffness

I don't disagree with the general sentiment and, presumably, that's part of why all the textbook examples show the bolts outside. It's the great deal part here that I take minor exception to. At best, this great deal is going to be a linear improvement in limiting engagement slack based on the lever arm. So, for normal columns size that's what... 20% - 50% range? And, depending on how you look at it, that improvement in engagement isn't strictly an improvement in stiffness unless we're considering stiffness in the "secant" sense. The difference is simply in how quickly any slack in the system is taken up. And given that I expect my moment base anchor bolts to be nice and tight, I don't see that slack amounting to much. Once the slack is taken up, I don't think that there is a stiffness differential going forward.

I like to debate structural engineering theory -- a lot. If I challenge you on something, know that I'm doing so because I respect your opinion enough to either change it or adopt it.

 

[JoshPlumSE]

There are two concepts at play here and it's useful to separate them.

Concept #1 is the STIFFNESS of the base plate against rotation. This is what I believe WARose is getting at. What is required for the base plate to be considered fully rigid in your analysis? Even when you have anchor bolts outside the flanges, is this really fully rigid? Certainly it is much more rigid against rotation than when the bolts are inside the flanges.

Concept #2 is whether the connection is capable of resisting moment. Absolutely a 4 bolt connection with the anchor bolts inside the flanges is capable of resisting moment. Because of the reduced moment arm, it will be a lower level of moment than if you had the bolts on the outside.

The key question (for me) is whether the analysis assumptions the OP made when determining the 20 k-ft base moment were consistent with the stiffness of the actual connection. If he assumed a fully fixed connection, I'm not sure I agree with that. If, however, he assumed a rotational spring that was somewhere between pinned and fixed then I think that would be more reasonable. Though how to arrive at a reasonable value for this rotational spring would be another discussion entirely.

 

[hokie66]

Fixed and rigid are not the same thing. Thanks, JoshPlum, for that summary, which should be required reading for young players.