Pre-loading beams

[mijowe]

We have a project where we are taking out a column in a steel office building and supporting it with two new transfer beams supported by two new columns. We have an engineer in the office who likes to pre-load the beam for dead load in these situations (sit the beam in place, jack up the supports until the beam deflects the calculated dead load deflection, then lock it into place) This, in theory, eliminates any settlement at the transfer and in turn any possible damage to finishes above. The other option would be to size the beams to limit deflection, likely getting beam sizes significantly larger than if they were preloaded.

Do people like either of the two options above better, and what are your brief reasons?

 

[rb1957]

I have no comment on the settlement issue, but how do you account for this preload in your calcs ?

another day in paradise, or is paradise one day closer ?

 

[SlideRuleEra]

If beam depth is not an issue, I prefer deeper beams (higher moment of inertia) rather than relying on a detailed installation specification. When beam depth can be increased, beam weight (cost) will probably not be much higher.

Even if beam depth must remain unchanged and a heavier beam of the same depth will limit deflection, I consider that preferable. Besides, labor and equipment to pre-load are not "free".

http://www.SlideRuleEra.net

http://www.VacuumTubeEra.net

 

[mijowe]

rb1957, I would take the beam with the anticipated building self weight only and with the beam in place jack up the beam ends with a force equal to the reactions. I would try to match the anticipated deflection as a check while jacking. By doing this you have essentially loaded the beam for the self weight, no real change in the beam loading.

Sliderule a tend to agree, Jacking requires skill and experience from a contractor that I am not sure I will get.

 

[slickdeals]

There have been a few good articles recently on preloading in the Structure Magazine.

http://www.structuremag.org/wp-content/uploads/2016/05/SF-FifthAve-Jun161.pdf

http://csengineermag.com/article/preloading-approach-to-column-removal/

 

[MotorCity]

My first preference would be to use a larger beam to minimize the deflection. Or why not just have the beam cambered in the shop? By preloading, I think the only thing that will get jacked up is the cost.

 

[dhengr]

Mijowe:
It seems to me that you have been asked to do the almost impossible, and then people are telling you, ‘but, don’t make it difficult’ or don’t expect to hire a contractor who knows his butt from a hole in the ground, he might have to think a little in the process. If the contractor can’t grasp/imagine this process, I don’t know if I’d want him on the job anyway. You might even specify that you want to observe and help with this aspect of the project.

I think your coworker/engineer is right on the money with his thinking. After all, you have to unload the existing column before you can remove it, and this could be done with shoring all over the place, but that’s not always so easy either, and involves jacking/lifting too. You certainly don’t want to start loading the new beams from where the structure is now. Although, you haven’t given much info. on what the real structural conditions are. You say two new beams and two new columns, and I assume you mean one beam on each side of the existing column, and then a new column out at the ends of these paired beams. You must have a fairly concentrated load at the existing col., based on the framing system, or you wouldn’t need the col. That means you camber the paired beam set over only a 3-4' length near center span (the existing col. location?), with a fairly small radius of curvature. The rest of the beam lengths remain fairly straight, if not cambered, up, slightly (exact concave shape down, to be determined ). This way the new paired beam system will pick up the load above almost exactly like the existing column does now. You bring in the two paired beams, either side of the existing col., bolt them together with a few cross frame stiff. pls. and lift them until contact at the center col. Then, you start jacking 18" inside to new end col. locations, leaving some working room for new cols. These two locations must already be fairly strong (or shored from below) to take the two new cols., and you already know about how much end delta you’ll get to unload the existing col. At this point, you are already carrying the D.L. and some L.L. from above by the two new paired beams. When the new cols. fit, with a .25" shim pl. atop the col. cap pl. (some thickness to be determined), bolt the paired beams to the new col. cap pls. and you are essentially done. Anything else allows the existing structure above to deflect too much before the new beams really start picking up their loads. What could be easier to accomplish the impossible?

 

[PEinc]

I'll choose the other option. I prefer to use a smaller beam that can be more easily installed INSIDE a building. Then wedge or jack the beam(s) until the existing beam starts to move slightly upward. Then place shims to provide sufficient bearing along the new beam(s).

http://www.PeirceEngineering.com

 

[Ingenuity]

Harped external PT tendon/s to steel beam/s, 'balancing' the uplift to the estimated 'to-be-removed-column' downward load?

I have done this a few times for concrete column removal with new concrete beams, but not too common for steel.

 

[rapt]

If there is going to be significant deflection in the new transfer beams, yes I would do something. We used to use flat-jacks or small rams at the location of the column to ensure that the old member remains at its previous level and the beam is essentially preloaded.

 

[Ingenuity]

We used to use flat-jacks

rapt brings up a very useful 'tool' for such applications. Flat jacks (one of the many original patents by Eugene Freyssinet) are a very compact/high-capacity method of pre-loading.

I have used them on several projects - from small 8" dia to giant 36" diameter ones.

We have our own equipment, but the hydraulics are readily available for rental in the US from WB Equipment Link. I think WB also provide site-services too, across the US.

For smaller-diameter permanent flat jacks it is possible to pre-load the structure via flat jack/s with hydraulic fluid, then (without de-loading) pressure pump in 2-part epoxy resin into the flat jack using a simple hydraulic pump and 'rocket' whilst flushing/bleed out the hydraulic fluid. Avoids have to place shim/packing plates etc.

Here are some photos of a project we were involved with for a transfer beam with new micropiles that required pre-loading:

It does require some specialized experience, but not too complex.

 

[WillisV]

Pre-loading the beam is definitely the way to go if at all possible - essentially eliminates all deflection issues.