Underpinning, new beams pocketed into existing walls, rowhouse

[DCStructures]

Hi all,

Working on a basement renovation of an existing rowhouse. New steel beams will replace intermediate lally columns. Was originally going to support new steel beams on new steel columns at perimeter, but cost of steel/connections/foundations becoming a problem.

Alternatively, if I pocket the beams into the existing walls, how would you decide whether existing wall foundation (likely brick foundation; bldg is probably 100 years old) can support re-direction of floor loads on new beam? Wouldn't new floor loads be small compared to self weight of wall and other loads already bearing on wall footing? Current configuration involves potentially pocketing beams into party(side)wall, as well as back and front walls.

These would be single isolated pocketed beams, essentially one in each wall. Wouldn't new beam load spread out into wall and along wall footing? Is this enough distribution of new load into existing footing, or would you advice underpinning regardless of how small loads are or how much they are distributed, just based on fact that you don't want to put new loads on an old existing foundation.

If underpinning is advised, would one area of underpinning beneath each new beam be sufficient? How are these underpinning pits sized?

Also, any tips regarding detailing of new pockets into existing walls for steel beams? Any concerns or things to look for with these pockets, particularly at the party walls.

I'm using 1500psf soil bearing capacity per code min. This is in DC.

Much thanks in advance.

 

[ztengguy]

Can you argue that he bearing is higher due to long term loading consolidating the soil and perhaps giving a higher bearing? How much higher is the load?

 

[bookowski]

What do you mean by new beams replacing existing columns... how does a beam replace a column?

Have you seen the basement wall? It could be brick but it also common in the northeast for it to be rubble, a mix of brick and rubble, or other types of local stone (such as brownstone). If it's brick or brownstone then it is often in extremely poor condition at the front and rear (where it's exposed to soil) and may require repairs, the mortar is often toast and possibly the stone/brick. At these locations you also want to be aware that any pocketed beams are likely to experience moisture.

The front and rear walls have been self supporting for their life and not carrying floor load. Adding floor load from a pocketed beam might change this and any settlement telegraphs up your walls. It depends on the magnitude - are you adding 1k or 15k etc. The load probably will spread but again I'd take a look at the wall. If it's loose rubble then I wouldn't be too aggressive with that assumption.

Pocketing into party walls is usually cleaner but be aware that your neighbor may do or have done the same, i.e. you think that you have a 12" wall but often once you get in there it is full of chases, flues, neighbors beam ends etc.

 

[DCStructures]

bookowski- to clarify: removing lally columns inside basement, adding steel beams at roughly midspan of floor joists. new beams will be supported by one new interior column at center of basement and by new pockets in existing walls. walls are brick.

 

[bookowski]

Got it. Typically row houses are 40 to 50 ft in length, I wouldn't put 1/2 that span into the front/back walls unless you are sure that you aren't going to cause a settlement and thus cracking at those walls above grade.

 

[DCStructures]

bookowski- agreed: new D+L at back wall is 20 kips. however new D+L at sidewall is only 4 kips.

 

[Archie264]

If I may piggyback and ask a quick (somewhat) related question:

Bookowski,

I've noticed from this and prior posts that you seem to know you seem a lot about these row houses; would you ever consider opening up the back wall on one of them built as you described? That is, considerably enlarging/combining the openings in back wall of one of structures and supporting above the new opening with a lintel? Temporary needle beams or pinch beams would be required for construction, of course, but I'm curious how it would affect the long-term viability of the structure and/or any other potential pitfalls I haven't thought about. Thanks.

 

[bookowski]

DC - 20 kips is a lot to me. Maybe in your condition it's ok, or as you said you can design underpinning to take this discrete load.

Archie,
No problem. Yes, I do that all the time. The tough part is how you justify the lateral - it's hard to get into all the details because I've heard many different arguments, many of which are reasonable. None of these buildings have a designed lateral system. However, the walls do provide something - and usually the overall renovation includes opening up the interior which also removes any contributing partition walls. The other problem is that everyone wants to do this - so any argument about counting on 'the row' for stability isn't valid in my mind.

I typically provide a new steel frame, so two wf columns and a wf beam (or usually two since the opening is usually 2 tall). I design the frame to take 1/2 the building width of wind. There are some problems with this as the front wall is probably nowhere near as strong, and getting a proper diaphragm in is usually optimistic - but that's what I do anyway. Most people have a bit of a fit when I specify this but most of my clients now expect it and have given in.

Last issue is that typically it's not worth trying to needle and save the wall - at least 1/2 the time I see the wall come down and get rebuilt on my frame. The demo guys can take those walls down in 2 days, then you also get a good connection to your frame, a decent wall, and avoid shoring.

I also see lots of people just put in a lintel and argue that the rest of the building and row could never take lateral anyway. I see the argument - but I'd rather be on the other side of it if anything goes wrong.

 

[Archie264]

Wow, what excellent advise, borne of experience! Thanks!

Gotta laugh at the justification for no lateral...just because it was done inadequately in the past is not a reason to do it inadequately now. Do those wf colums and beam(s) constitute a moment frame? and if so, do you use any special anchoring at the base? Thanks again.

 

[bookowski]

Yeah, although I can sort of see that argument - I'm sure that if there ever was an event only my frame would be standing the rest would be rubble. Still - I prefer that option.

Yes, it's a moment frame. The base gets into another issue and it depends on what you have. The gravity load is usually pretty small that you're picking up. If you've got a good basement wall (sometimes heavy stone in good condition or equiv.) then I have them take down the top of wall a bit with the wall removal and pour a concrete cap beam along there - the base plate gets anchored into that. It's a little bit grey - it's hard to show that the basement wall works for soil lateral... but you have to stop somewhere. If the basement wall is bad I've had them cut a pier into it - this is a tough sell though. In several cases I've had the cellar extended under the rear yard anyway so it gives a chance to get in there and do work. Sort of depends on your condition I guess...

 

[bigmig]

References on mortar of that age place the mortar strength just above that of loose sand. I would remove a sample of brick and mortar and get it tested so you can establish its strength. Then I would establish a close monitoring system at jobs end to confirm things with an identified plan of action for schedule of monitoring. The IBC has an entire section on in-situ testing. In the end, you need to prove to a court that you took professional measure to establish the strength of that brick assembly and that you have counted in the topic of statistics v.s. the samples you tested. See IBC Section 1714 (2009).

 

[Archie264]

Bookowski, great info, thanks again.

 

[wannabeSE]

DC
Have you consider ACI 530-08 section 1.9.7 and commentary for how concentrated loads are distributed in walls. For running bond, the load distributes at a slope of 2 vertical:1 horizontal for the top 1/2 of the wall.