Micropile and Soil Nail Wall Underpinning 2

[MTNClimber]

This underpinning project came across my desk for review. I was hoping to get some opinions on the load transfer and resulting deflection of the pile to see if I'm overthinking this. The pile will not be preloaded so the load transfer will cause some amount of settlement of the existing building, which needs to be reviewed and accepted by the building owner's engineer.

Would the anticipated settlement be PL/AE plus lateral deflection of the pile from the batter? Does the theoretical deflection even come close to the actual settlement?

 

[MTNClimber]

Just figured I would give an update on this. The design engineer was able to provide me with details for 4 other similar underpinning projects that were successfully completed without issues. With proof that this method works in real life but not quite on paper, I notified the stake holders of the potential risks and they decided to move forward with this soil nail/battered micro pile underpinning approach.

Construction of the system took about a month, including soils nails for the other 3 sides of the basement excavation. It's been about a month since the pile contractor demobed. Automatic movement monitoring data has shown negligible movement (within the error of the system).

My guess is that the micropiles are barely seeing a load, if any, and the soil under the footing is still doing the heavy lifting. I would like to instrument the piles on the next job just to see how much they are seeing.

Thanks for all the input (especially John).

 

[PEinc]

With proof that this method works in real life but not quite on paper

In my opinion, it HAS to work on paper also! If it does not work and does not work on paper, you are screwed!

http://www.PeirceEngineering.com

 

[EireChch]

MTN - you're talking about it not working on paper, did you try to look at using Plaxis or something more advanced to "make it work"?

 

[MTNClimber]

The issue is there isn’t an accurate way to determine how much load is going into the piles. For the piles to engage, the soil underneath the existing foundation would have to start to compress. But if the soil is stiff enough to support the building load even after they excavate up to the footing, how much load is transferring to the pile? My guess is near 0% on this job.

 

[PEinc]

MTNClimber, the May/June 2023 issue of Foundation Drilling magazine from ADSC has a story by Nicholson Construction where they did this same type of underpinning project for Colgate University's Olin Hall. The story really doesn't give much info such as building loads on the column footings or performance data (vertical or lateral movements). The story does describe the soils as "dense to very dense granular materials with varying percentages of fines and very stiff clays. Groundwater was not encountered in the borings within the excavation depths or encountered during construction." Nicholson also noted that "The initial SOE proposed adjacent to Olin Hall was a secant pile wall, consisting of 24-inch diameter drilled elements.....The university had concerns that the thickness of the secant wall would encroach on the new below-grade space, limiting the useful area of the basement." Therefore, in my opinion, the owner and CM accepted a riskier, lower cost solution that also provided more basement area. There was no mention of how much of the column load the micropiles were designed to support.

http://www.PeirceEngineering.com

 

[MTNClimber]

PEinc,

Thanks for sharing. It looks relatively similar. We had medium-dense outwash sand and gravel, without groundwater. We did not pre-stress any of the nails like they did, which would have been a safer approach. The building we were underpinning did not have its wall torn down like they did for Olin Hall, which made it even more important that the building didn't move.

While I'm here, they installed the foundation wall for the new building and filled the gap between the soil nail wall and the new foundation wall. The automated movement monitoring system has been de-commissioned. Still no complaints from the tenants regarding cracking or other related issues. All-in-all, it seems like it's an effective system to underpinning. I think this approach needs a little more attention from the research community to help rectify the hanging questions regarding actual load transfer to the piles because it still doesn't make sense on paper.

 

[PEinc]

I am not too sure of the effectiveness of pre-stressing soil nails which are installed without any unbonded length. If nails have unbonded lengths, they are tieback anchors, not soil nails. So, if you pre-stress a fully bonded soil nail, you probably are pulling only on the front section of the nail where it probably will fail its way through the soil. Therefore, how effective is that "failed" portion of the nail? For the same reason, I do not believe in testing production nails as FHWA calls for. If the entire test nail length is grouted, you will fail the bond in the upper length of nail. If you use an unbonded length, it isn't a nail. If you grout only the lower portion of test nail so you can test it (and grout the remaining length after the test), you can have serious risk of collapsing the ungrouted portion of the drill hole during the test. Then you have a problem. If, during the test, you try to leave some casing in the ungrouted portion of the nail, you may unintentionally grout some of the casing which will also affect the test. I prefer to install additional, sacrificial test "nails" with an unbonded length which will serve to verify the ultimate grout to soil bond strength. I test these just like I would test tieback anchors.

http://www.PeirceEngineering.com

 

[MTNClimber]

That's a good point regarding pre-stressing. I haven't specified it on any of our soil nail projects but I have seen it specified by others. I'll have to keep that in mind going forward.

For testing, the newer standard (FHWA-NHI-14-007, 2015) talks a bit about issues with load testing. I have to battle a lot of contractors and their engineers on the testing procedure. They want to use a piece of PVC taped to the reinforcing bar to create their unbonded zone, which was the standard when I started. Unfortunately, research (Cadden 2010) has shown this method can create artificially high bond stresses since the grout from the bonded zone can transfer the load to the grout in the unbonded zone. For sacrificial load tests, I make them wash out grout using the tremie to create the unbonded zone. The next day you can check the unbonded length by probing with rebar to find the top of the hardened grout and adjust your loading schedule accordingly. Usually, we are within 0"-2" of our target bonded length.

This washout method is an issue for testing production nails for permanent walls since the hole along the unbonded zone collapses. In these cases, we opt for only testing sacrificial nails for permanent walls.

 

[PEinc]

Unfortunately, research (Cadden 2010) has shown this method can create artificially high bond stresses since the grout from the bonded zone can transfer the load to the grout in the unbonded zone.

This sounds nice but........Why would people worry about this when they don't ever worry about it when testing much higher capacity, more critical, tieback anchors with unbonded lengths?

http://www.PeirceEngineering.com

 

[MTNClimber]

That's usually the response from the other side of the table. PTI has conflicting information regarding what they recommend. Upfront PTI says this:

And later on they say this:

Not sure what they're trying to say.

But regarding soil nails, I'm holding the team to the standards provided in the latest soil nail design manual. If it says not to use the PVC bond breaker as it causes erroneous results, I'm sure as sh*t not going to let it happen, even if it's inconvenient when we talk about tie-backs. I don't make the standards, I'm just here to abide by them.

 

[PEinc]

I would like to hear of others' experiences and problems encountered when testing partially grouted, uncased, soil nails per FHWA.

http://www.PeirceEngineering.com

 

[MTNClimber]

I would too. Feel free to dig through some of the supporting documentation too: [URL unfurl="true"]https://rosap.ntl.bts.gov/view/dot/42934[/url]

It's focused on HBSN but still applies to typical soil nails. They refer to the artificially high bond strengths as the "Doughnut Effect".

 

[MTNClimber]

PEinc,

GTR recently designed an active soil nail underpinning system, without the use of piles, that was featured in the July/August DFI magazine. Looks like they had success with the system. See the article on page 79.

 

[PEinc]

MTNClimber I just lost a long response I had written. So, I will give you my short version.
Soil nailing is not underpinning.
Sometimes, it might work as an alternate to underpinning.
I don't do designs that MIGHT work.
This DFI project had shallow, very competent soil over high bedrock.
With a good contractor and the admittedly conservative design soil nailing worked - this time.
I have designed and built soil nail walls SI ce the 1980's. Never heard of active soil nails.
Soil nails with unbounded lengths are not nails; they are tieback anchors.
I don't believe that Snail can analyze nails with unbonded lengths.

http://www.PeirceEngineering.com