Load Test for Segmental Block Retaining Wall 2

[emceditor]

A segmental block retaining wall was constructed with several "oddities" incorporated. At some courses the geogrid has not been engaged with a shear pin per the manufacturers recommendations and using asphalt shingles as shims between courses to correct alignments. About a month after construction the wall has developed a series of vertical cracks: Approximately 1/8" wide at the bottom and tapering to a hairline at the top. The maximum height of the wall is 26'-0". Before having the contractor tear down and rebuild the wall, it has been proposed to test the wall. The wall in its current state has not received the full loading due to a 8" concrete slab and chiller at the top. DOES ANYONE KNOW OF A PROVEN TEST PROCEDURE?

 

[Focht3]

Nope. It cannot be load tested. And the wall is entirely too tall to leave in place and hope that it does not have any future problems.

In my view, if some of the geogrids weren't engaged and the courses are uneven then the wall must be taken down. The broken blocks must be replaced. And the use of non-standard building materials must be prohibited.

I suspect the contractor did his work with no one watching, so his errors and screw-ups went undetected until problems surfaced. This time the work should be monitored - full time.

Am I correct in assuming that the contractor does not have the financial ability to redo the work, and that a performance bond was not required? I suspect that the owner will get to pay for this work twice.



Please see FAQ731-376 for great suggestions on how to make the best use of Eng-Tips Fora. See faq158-922 for recommendations regarding the question, "How Do You Evaluate Fill Settlement Beneath Structures?"

 

[gandersen]

emceditor:

The type of cracking the you describe almost sounds like settlement cracking. Whether or not the geogrid is engaged with the shear pin is probably not a big issue.

I would recommend that you contact a local geotechnical engineer or get ahold of the designer of the wall for an evaluation.

If the issue is settlement, a load test would not demonstrate the long term performance.

Glen

 

[emceditor]

Thanks Focht3 and Ganderson for your feedback. The construction of the wall was monitored on and off. The first time was at course 4 I believe. The contractor said that he had done the the incorrect shear pin engagedment just for the last two courses as it wasd too hard to do it the right way. He was made to tear out and rebuild those two courses. Settlement came to our mind too. We had a heotechnical engineer check the bearing soils - they are fine. The work is was done by a sub-contractor for the general contractor who is bonded. The owner will not have to pay for anything.

 

[Focht3]

The shingles will deform over time, so the movements within the wall are very likely to continue. While the contractor says he only fudged on the last few courses of the wall, the presence of shingles - and cracks - suggests otherwise. I'd also be worried about the geogrid being too short, too widely spaced, etc. And the backfill would also be a concern. Who knows what other goofs this subcontractor made?

Glen's gut on the foundation is a good one, and partially drove my initial response. But my take is slightly different - I suspect poor quality construction, not a classic "settlement" problem. The only way to evaluate that aspect of the construction is to tear the wall down.

Be sure to document the wall's condition thoroughly before you deliver the bad news to the contractor.

By the way, what does the owner have to say about all of this?



Please see FAQ731-376 for great suggestions on how to make the best use of Eng-Tips Fora. See faq158-922 for recommendations regarding the question, "How Do You Evaluate Fill Settlement Beneath Structures?"

 

[cdh61]

emceditor,

Did you check the blocks to ensure no manufacture defects? This would not take much time or money, and I would check because the subcontractor will most likely be looking for others to share in the cost of reconstruction.

regards