Pit Undperinning - Jacking or Shims (or none)

[bookowski]

I have two similar projects using two different geotech/SoE engineers and comparing the two sets of underpinning details.

Both projects involve underpinning of existing brick masonry walls/footings on soil (not rock). They are both for complete underpinning, i.e. all walls. Both projects are using pit underpinning with concrete piers.

One set of drawings specifies hydraulic jacking with very detailed instructions for monitoring movement and when to stop jacking. The other set (different engineer) specifies driving steel wedges at the top of underpinning, there is not very specific information about how much wedging/movement should occur to consider the shimming complete.

I am trying to figure out which is better and when jacking would be necessary.

To someone not familiar with the design of underpinning it seems like jacking should be necessary. It doesn't seem likely that pounding steel wedges would sufficiently load the underpinning piers.... but then again this seems to be done often so I guess it does work?

As a related question - does anyone have any good resources for the design of underpinning? I tried looking around to answer the above questions but this does not seem to be covered in many foundation books.

 

[PEinc]

I have design and constructed many, many underpinning projects. For concrete pit underpinning, I have never shimmed or jacked, and then grouted, a building. I have shimmed or jacked when needling a building in order to remove deflections from the steel beams. If you properly dry pack an underpinning pier, there will be sufficient force provided to transfer the building load to the pier. Grouting is not the same as dry packing. Grout is fluid; dry pack is a stiff, damp mixture of sand and regular Portland cement.
The attached file is an excerpt form the "Bible" of underpinning.

www.PeirceEngineering.com

 

[bookowski]

Both methods specify dry pack, not grouting.

What is the method of force transfer to the soil below the underpinning if you don't shim (or jack)? It seems like the same concept as jacking beams for a transfer.

Lets say that in service there is 2ksf below the footing. You dig a pit, fill it to 3", dry pack. There is not going to be 2ksf now below that new pier. Eventually there will have to be some settlement to load the soil. It seems like the shimming is one attempt at this, the jacking is a more substantial version.

I was googling and found a presentation by struct. eng. of ny (seaony) and on one slide titled 'steps to improve underpinning results' they show and note 'steel shims for load transfer'.

I was wondering how shimming was even sufficient but it sounds like you have ok results with nothing. I don't see how it works - but I can believe that it does.

 

[aeoliantexan]

Edward White wrote a chapter on underpinning for Leonard's book "Foundation Engineering", McGraw-Hill, 1962. It has a lot of detail on the method you described. Perhaps you can find a copy.

More recent technologies such as micropiles, ACIP piles, compaction grouting, and jet grouting offer choices not available when these references were written, but the pier method probably has not changed much.

Dry packing probably depends on founding the piers on good material that won't allow much settlement. Compressible soils may need jacking.

 

[PEinc]

aeoliantexan, I already attached an excerpt from Prentis & White's 1950 book, Underpinning. Most other books that talk about underpinning just include information from this book.

bookowski, details for shimming, wedging, and jacking of underpinning piers seem to be most often specified for projects in the NY area, as you can see from the SEAONY PP presentation. However, from my experience and from many other references, shims, wedges, and jacking are infrequently used for concrete pier underpinning. Significant load transfer to the pier can be achieved by dry packing. Ramming the dry pack into place can significantly and sufficiently load the pier. Earth Retention Systems Handbook by Macnab and Temporary Structures in Construction, 3rd Ed., 2012, by Ratay both describe transferring building load by dry packing. In Chapter 11 of Ratay's book, chapter authors Arland and van Leeuwen, (both Metro-NY'ers) state, "The building loads are transferred onto the piers by means of "drypack," a mixture of sand and Portland cement with very little water that is rammed into the space between the existing wall or column footing and the hardened concrete underpinning pier, placed from within the approach pit. This method is more completely described subsequently." The authors then include a detailed 10-step procedure in which Step 9 describes dry packing and step 10 describes an alternative to dry packing. None of the steps includes shimming, wedging, or jacking.

In Winterkorn & Fang's 1975 Foundation Engineering Handbook, Chapter 22 - Underpinning, page 630, the author states, After the (underpinning pier) concrete sets overnight the pit is dry packed with a mixture of one part sand and one part cement with just enough water to make it hold its shape when a handful is squeezed." They then state, "Experience has shown that special nonshrink grout mixtures do not make enough difference to warrant their considerable additional expense. Wedging and jacking of pits are not necessary because of the large mass of the pit and side friction of the concrete against the ground makes this operation difficult."

www.PeirceEngineering.com

 

[bookowski]

I'm in the NY area, so I guess that's why I'm seeing the shims or jack details. I have always seen at least one of those methods used. I've done several projects similar to the one in your photos, I've never seen one without some type of pre-loading. Although I have seen several that have had significant movement. In fact I had one where the tiebacks pulled the underpinning down causing a lot of movement.

I'm not designing the underpinning, I was just trying to satisfy my curiosity and resolve the different details. I spoke to the eng that was using the wedge details on a current project and he said that jacking would be preferable but there is also more chance of creating a problem so they don't use it.

As aeoliantexan stated I would think you need to have very good soil that is not susceptible to settlement. At the end of the day it seems the same as jacking for needle beams as you stated. If you don't "pre-deflect" something then it will need to deflect to take load, and I don't see how drypacking would create any sufficient pre-loading.

I had previously found the Winkerton & Fang Chapter 22 - but the reasoning that jacking is not necessary because it is difficult didn't seem very useful.

I found a copy of the Prentis & White book but it was dated 1917, I couldn't find any 1950 versions. I will check out the other references. Thanks.

 

[howardoark]

This is kind of off-topic, but I had a question for PEInc from a post he made six years ago indicating that trench boxes weren't shoring because you needed to dig a hole to put the box in.

If you dig a trench box in (push it down with the excavator bucket as you excavate soil) would you expect any ground movement behind the box? To get it out, you'd have to pull it out and backfill as you went.

 

[PEinc]

Of course the underpinning needs to bear on good soil. That's the purpose of underpinning. It should not be founded on compressible soil. If you have compressible or unsuitable soil at the bottom of the pit, you need to go deeper or bell the pier for a lower bearing value.
The photo is from a NYC project (in Flushing). The 1950 Underpinning book is rare and expensive.
Dry packing builds up pressure on the bottom of the footing, little by little, over a large area.It is hard to shim and wedge a significant portion of an existing footing. Doing this on a small area beneath a footing can damage the footing, especially if done at the edge of the footing.
You can't believe everything you see in books. I often see details that call for posting up a footing while the pit is being excavated. How, then, does the man fit into the pit with a post in the way and how does he lengthen the post as the pit is dug deeper? Not possible, yet I see this too often.

http://www.PeirceEngineering.com

 

[howardoark]

If we're sharing pictures ... this one is similar to PEInc's, but more of an "after" version.

 

[PEinc]

howardoark, I'm not surprised that system failed. First, there should have been underpinning instead of sheeting. Sheeting (at least "regular" sheeting) is not supposed to be used to support buildings. I don't see any rock bench at the base of the sheeting wall. I don't see where the soldier beams extend to or below subgrade. I don't see any other drilled-in near vertical member at each soldier beam that can support the vertical component of the tieback anchors. A significant portion of each soldier beam has the lagging placed behind the rear flange. Therefore, the compression flange of the cantilevered, upper, beam length has no lateral support and therefore has a lower allowable bending stress than if the beam were lagged at the front flange. There is a sharp, maybe 90 degree, turn in the wall where you have maked "About 30'." There may have been insufficient earth behind this corner soldier beam to prevent the tiebacks from pushing the beam backwards and then causing the lagging to fall out. Multiple bad details in this photo. I hope it was not your design.

http://www.PeirceEngineering.com

 

[howardoark]

Hi PEInc (love your work by the way)

It wasn't my design - I imagine the designer is still being deposed so he's unlikely to opine on where he went wrong - which doesn't mean I haven't done things equally badly, just that I've gotten away with them. I'll find a link to the report when I get back to my office tomorrow - I think the report indicated the problem was with the tiebacks which led to the soldier beams kicking out at the toe and then the earth came out from behind the wall. They were very very lucky the apartment building didn't come down into the hole.

I had a question for you above which you probably didn't see as I posted it 2 minutes before one of your reposes.I am curious about that issue and was wondering what you thought.

 

[PEinc]

This is off-topic to this thread, howardoark, but I have seen trench boxes used as you just described (dig and sink). It is not a very productive method for installation or removal. If you install and remove a box as you described. it probably will protect adjacent paving, utilities, etc., as long as you keep moving the box down while digging. However, in order to push the box down and also remove it during backfill, you need a minimum of side friction on the box (which could also mean insufficient lateral support) or you need a really big excavator to push down or pull up the box. Also, if you dig and sink a box, BOTH ends of the box need to be closed off (with a street plate?) to avoid an illegal, unsupported, deep, vertical cut at each end of the box. This closure may also interfere with any utility line that is being placed in the boxed excavation.

http://www.PeirceEngineering.com

 

[howardoark]

Here's a link to the full report

http://www.nyc.gov/html/dob/downloads/pdf/808ColumbusAve_ROI.pdf

Everyone will be glad to hear that the geotechnical engineer was absolved of all fault. They blamed the inspector for not noticing the soldier beams were founded on sheared rock.

 

[PEinc]

I read the report. Interesting. Thank you.
Again, my opinion: bad details, should have underpinned the building, no rock bench, bad rock! A perfect storm. Also, the report sounds like it was written by someone who really does not know enough about sheeting, tie backs, and underpinning.

http://www.PeirceEngineering.com