Removing interior columns in an existing PEMB 6

[pderrick21]

I've got a project where the existing PEMB has frames at 20' on center. Out-to-out of PEMB frame is 120' with interior columns approximately 20' from the exterior column. The owner wants to remove the interior columns. Has anyone done this?

 

[KootK]

That's a pretty tall order. Anything's possible with enough money and understanding but that sounds as though it would be cost prohibitive.

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.

 

[haynewp]

No, but I would get the information on the original building manufacturer and hire them to analyze the building for the needed reinforcing. We have done this several times for other PEMB modifications. The foundations will also likely be impacted, which it was chosen to install helical piers through core-drilled holes in the existing footings and connect them to the existing columns to address the uplift/gravity in the last one I was involved in.

 

[pderrick21]

I've been told the original PEMB manufacturer is out of business. Any analysis will be from square 1 with field dimensions.

 

[hokie66]

No chance. Wash your hands of the job.

 

[WARose]

Most of the time, a PEMB is just barely making it even with those columns. Taking them out would mean some major beef up work not just for the columns and the girders......but probably the foundation as well.

I wouldn't be surprised if the cost wound up being about as much as a new building.

 

[haynewp]

Another PEMB mfr won't want to deal with it, even though they could. That is my experience anyway if it was designed by another PEMB mfr since they will take full responsibility of the original mfr's design and their steel products. There is an engineer in my area that specializes in existing metal building analysis that is not affiliated with a PEMB mfr.

 

[mike20793]

I agree with hokie.. beware of this one. If you are going to analyze this from square 1, then it is likely not going to be economical for you or the building itself.

 

[JedClampett]

Congratulate the owner on all the money they saved on the original PEMB and walk away.

 

[ajh1]

A few columns out and replace with jack beams, maybe, but all of them, forget it. I wouldn't want to try and do this with any building as you are effectively increasing the moments in the main span to somewhere around double what they were before. Some modifications are doable, others don't make any sense regardless of the structural system.

 

[CANPRO]

I'm a bit surprised at everyone saying to walk away from this, its challenging but not impossible. We don't know the building owners situation - maybe they need the space, moving isn't an option, and they have the budget to make this happen.

I have done something similar a few times, but on a smaller scale and with just one frame (sounds like you have to do this for a more than a few). First question is - do they need the full 120' clear on the inside? If they can lose a bit of space each side of the building that means you can put new columns/foundations adjacent to the existing, inside the building. That shortens your span, simplifies the foundation work and reduces the reinforcing.

When I did this, I turned the roof beams into the top chord of a new truss. For one design, I used back-to-back channels for the bottom chord and sandwiched the existing post, which became vertical members of my truss. This way the whole truss can be built in place without shoring, and when the truss is complete the existing columns can be cut below the bottom chord. The back-to-back channels on the bottom chord also make fitting up the new web members easier since they don't have to be cut to exact lengths.

 

[XR250]

I like Canpro's approach and seems it would remove a fair amount of thrust/overturning from the foundation. Depending on the foundation type, this may compensate for some of the additional gravity load.

 

[JAE]

It seems to me that using CANPRO's approach in utilizing the existing beam as a top chord of a new built-up truss would require a significant amount of field work to verify:
1. Beam geometry and web/flange thicknesses
2. Steel testing to verify Fy of both flanges and welds (in PEMB's they sometimes are different)
3. Connection information including verifying bolt types/grades and diameters.
4. Verification of roof purlin sizes and connections to the beam - they brace the top chord so this must be verified.
5. Capacity of the roof purlins to also accept diagonal strut bracing of the bottom chords.
6. Capacity of the existing end columns for the higher loading - need to stiffen or strengthen?
7. Capacity of the base plates of those columns and the anchor rods.
8. Steel grade of the anchor rods.
9. Embedment of the anchor rods to work with the higher loadings.
10. Footing size verification.

I'm sure there's other things as well. All perhaps do-able (CANPRO did it after all).

Check out Eng-Tips Forum's Policies here:
faq731-376

 

[CANPRO]

JAE, yes there is a lot of field verification required. In my experience, any PEMB work requires a lot of field verification.

As a side note, as mentioned above, the span of the frames on my jobs were considerably less (in the range of 70'-80'). And I think the key to making this feasible, is installing new columns/footings adjacent to the existing.

EDIT - I should also note that another important factor here is making sure it is okay to lose the headroom involved with installing the truss. If you can make the truss deep enough, the stress in the existing beam/new top chord (and the demand on the purlins acting as a brace) could be less than the current configuration

 

[SteelPE]

I have done something similar to as well, but again on a smaller scale. We removed a few isolated columns to make the owners equipment layout work. I did this under the review of an engineer I know who used to work for a large manufacturer. The work was only completed a 1-2 years ago so I don't know how well it will perform as it hasn't been that long. My colleague has a belief that when you do work like this you make sure you never get called back.... ie make it stout.

 

[KootK]

If there is an appetite for reduced headroom, another option is external post tensioning, probably in combination with the truss. That can help to dial out stresses and deflections.

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.

 

[jike]

Not knowing the specific use of the building or why the owner wants to eliminate those columns, is there a possibility of locating a new column somewhere else in the interior other than at 20 feet?

 

[Brad805]

Step one is to get an idea how you might do it and then tell the owner what it could potentially cost. I too have went down this rabbit hole, and when they learn it will cost $200k or so per frame, they learn to live with the column.

 

[pderrick21]

Update....
The existing PEMB was built in the early 80's. Its an existing church that had classrooms around the perimeter with the sanctuary in the middle portion. Now the sanctuary is expanding to the out-to-out dimensions. There will be (2) existing frames that will be modified. The truss idea from CANPRO may be more feasible engineering-wise, but economics may win out. The other option will be to buy (2) new PEMB frames, existing foundation modification, a new roof and re-using existing roof purlins and wall girts (possibly).

 

[ajh1]

Be sure that the full 120' span is actually the main frame. In a lot of cases with a layout as you describe you may be looking at an 80' wide main frame with a 20' lean-to (pinned connections) off each of the two sides. In that case the main lateral resistance is with the 2 "interior" columns, not the columns at the outer walls.