Question for you PEMB types... 2

[swearingen]

Please see the enclosed picture.

It is a detail at the top of a number of columns in a large warehouse. The column basically reaches around and holds the main roof member from the side.

What is going on with this detail? Is there a specific reason this is done?


-5^2 = -25 ;-)

http://www.eng-tips.com/supportus.cfm

 

[ajh1]

Assuming these columns are at the endwall of the building, the bracket design you have shown is a cleaner way to transfer lateral forces from endwall wind loads into the purlins than trying to transfer those forces via a flange brace and/or weak axis bending of the roof beam. Loads would be transferred into the column from the endwall via the girts on the wall. This would only normally be done for very high lateral forces. Low level forces can be taken through the flange brace relatively easily. In many cases manufacturers only have brace connections details for single bolts so those may get overwhelmed fairly quickly for tall buildings with higher wind speeds.
There is no particular reason a detail of this nature would be used for an interior column. If the pictured columns are not at the exterior wall, then this detail doesn't make much sense.

 

[swearingen]

It is indeed an exterior wall. The columns with this detail are interspersed with pipe columns that go up and support the roof girders directly at the bottom flange. This is in a hurricane wind zone, so high lateral forces on this wall are to be expected.

Although your explanation makes sense, it seems that beefing up the flange brace would be easier than creating what's going on in the picture.


-5^2 = -25 ;-)

http://www.eng-tips.com/supportus.cfm

 

[KootK]

Although your explanation makes sense, it seems that beefing up the flange brace would be easier than creating what's going on in the picture.

For what it's worth:

1) You're unlikely to find a more competent PEMB pro on this forum that ajh1.

2) The economics of something that is essentially "product" will differ from normal EOR work. Field vs shop, cheap access to all day welding, use of standardized components...

 

[swearingen]

Now that I've thought about it a bit, I guess it makes more sense, even from the cost standpoint.

To push it through using a beefed-up strut from the girder bottom flange into a purlin means that the purlin would also see a greatly increased moment necessitating changes there as well. Keeping the moment at the top of the column, even with a circuitous detail, would keep all other members untouched.


-5^2 = -25 ;-)

http://www.eng-tips.com/supportus.cfm

 

[swearingen]

It dawned on me as I was going back through the pictures of the structure that the explanation by ajh1 doesn't hold water in this case for two reasons:
1. If it were to transfer end-wall loads into the roof truss system, shouldn't the end-wall actually touch the column? It doesn't.
2. To transfer shear loads to the roof truss system, you would need a node at the top of the column. It is clear from the picture that the top-of-column connection occurs at a point with no structure on the other side of the connection. Not a good way to transfer high shear loads.

I guess I'm back to square one. Why in the world would this connection be made?

I've included another picture to illustrate what I said above.


-5^2 = -25 ;-)

http://www.eng-tips.com/supportus.cfm

 

[KootK]

If it were to transfer end-wall loads into the roof truss system, shouldn't the end-wall actually touch the column?

Maybe not touch it but deliver lateral load to it somehow. Girts etc. So yeah, I agree with this. Walls are concrete?

 

[bones206]

It’s hard to see, but the second photos appear to show the strut attaching to the beam near the top and bottom flanges. Perhaps it’s a torsion brace?

 

[MotorCity]

Perhaps it serves to prevent the beam from rolling over. Or, it could serve as a brace point when the top flange goes into compression under uplift loads.

 

[MonsieurR]

This is a typical column for exterior cladding with a parapet wall, it makes a better constructive detail for collecting water and general detailing of the parapet. The roof should be placed at the discontinuity and so indeed it doesn't make sense to use it in the way you are showing. Maybe they where finding the use of an already fabricated column! very cheap.

 

[malikasal]

i don`t think that this column is to serve as wind column, we usually break the roof bracing at the point were the wind column hits the roof girder.
plus, there are no girts connected to it.

the only explanation that i can think of for this type of connection(which we only use if there is a parapet and the extension should go beyond the top of the frame girder) is that they had an already fabricated column and they wanted to use it no matter what, so they added this extension to connect it to the frame girder.)





ôIf you don't build your dream someone will hire you to help build theirs.ö

Tony A. Gaskins Jr.

 

[haynewp]

I agree with it being a parapet support that has had the top cut off and roof extended over it. See attached.

 

[KootK]

Thank you for posting that haynewp. I'd been wondering how the parapet support condition worked now that that seems to be the consensus opinion.

 

[ajh1]

I had missed the view that indicated no girts attached. It obviously can't function as a wind column without the girts. From my view of the pictures it isn't clear that there is a parapet either but then we haven't seen a picture above the roof. I wonder if the building was originally designed with a metal wall instead of the concrete. With a metal wall and girts, the detail would have made sense without a parapet.

 

[swearingen]

To fill in a few holes brought up by the recent replies:

- Walls are CMU to 8' height with framing up to eave covered with corrugated metal exterior and wallboard interior.
- Column connection does connect at top and bottom flanges of main girder.
- There is no parapet wall, nor is there any indication that there ever was one.

I hadn't thought of repurposing a column that was left over from another job. From the evidence I have, that appears to be the most likely scenario.

As a non-PEMB engineer, I have never seen a detail like the one you presented, haynewp. Interesting...


-5^2 = -25 ;-)

http://www.eng-tips.com/supportus.cfm

 

[MonsieurR]

See detail attached

 

[bones206]

Is there a cap plate on top of the stub column? It looks like there is, which would indicate that what we see is likely the original fabricated length of the member.