1st Stone Wall - Dynamic Test (04/10/05)

Download TEST VIDEO

This specimen was built with irregularly shaped stones to a height of ... The quality of the construction of the specimenwith was low due to the fact that only a limited number of stones were available for its construction. The infill space was irregular because the stone spacing varied from 0 to 10cm.

The test was initially run with a constant vertical acceleration of 0.2g and the specimen remained stable. Then the acceleration was increased to 0.7g and one of the leaves of the specimen fully collapsed and the second leave collapsed only partially. An acceleration of 1.8g was used to fully fail the specimen.

In this case the infill didn't seem to have a major roll, except from the fact of destabilizing the stones that were partially supported on it. An additional destabilizing effect of the infill could have been to facilitate the sliding of the stones once they started moving with respect to each other.

In the video it can be clearly seen how high frequency vibrations can cause an irreversible displacement of the stones, that ultimately can lead to failure of the system.

1st Wedged Bricks - Dynamic Test (04/03/05): Wedge-shaped bricks

Download TEST VIDEO (Interesting failure!)

This specimen was built to a height of 9 rows (50cm) with the peculiarity that the second and third rows from the bottom contained wedge (angle:10 degrees) shaped bricks. The test was run at a constant acceleration of 1.7g (55Hz) and no infill was added during the test.

The failure observed, as expected, was the outwards sliding of both leaves at the level of the wedged bricks. This test can be conpared to the previous one, which was built with regular bricks and was exited at 4g and still failed by overturning.

It is interesting to observe (pic 6) how the wedged bricks moved with respect to each other, while the bricks above them don't do so (pic 2, 5).

7th Dynamic Test (04/03/05) : Initial maximum acceleration = 4g? / 55Hz

Download TEST VIDEO

This specimen was built to a height of 9 rows (50cm). The initial vertical acceleration of 4g was kept constant throughout the test with the intention of inducing a sliding failure. Some sliding was observed, but the overall failure was again by overturning of one of the leaves.

Observe that in spite of the high maximum acceleration (4g) only two bricks moved with respect to the rest...

6th Dynamic Test (03/21/05)

Download TEST VIDEO

This specimen was built to a height of 9 rows (50cm). Initially, the vertical acceleration was kept low at ....g and then when the infill had settled by about 8cm the acceleration was increased to ....g and kept constant to failure. During this test no infill was added.

The interesting fact to observe in this test is that in spite of the infill having settled by over 12cm its lateral pressure was still able to cause overturning the right leaf. This is because tilting of the leaf reduces the lateral pressure needed to cause overturning.

5th Dynamic Test (03/20/05)

Download TEST VIDEO

This specimen was only 8 rows high (44cm). It was initially vibrated with a maximum vertical acceleration of about 0.9g. This was enough to initiate the failure of both leaves simultaneously. But after the infill started compacting its height was reduced by 10cm and the lateral pressure on the masonry was reduced proportionally. Without interrupting the test some infill (about 4% of the total) was added, and the vertical vibration increased to a maximum of 1.3g (the top bricks started to "walk" around as seen on the post-failure picture), which finally caused failure by overturning of the left leaf.

Note: The specimen intended for this test failed statically while infilling it at a height of 44cm... because one of the leafs was not a-plumb. This is an excellent reminder that the irregularities or deficiencies found in the real world constructions can detrimentally control the failure of a structure. The results obtained from idealistic tests and models should be evaluated with caution.

4th Dynamic Test (03/20/05)

Download TEST VIDEO

Starting from this test a different signal converter was used, which reduced significantly the noise in the data obtained. Also a second accelerometer was added to measure the horizontal acceleration, which is usually one fourth of the vertical acceleration

This specimen was built to a height of 50cm and it was statically stable. A very small vibration (less than 0.4g) was enough to cause the compaction of the inner core and the slow failure of the specimen. No infill was added during the test. In spite of a significant reduction of the infill height (11cm), the overturning motion of the left leaf continued while the vibration was on. This overturning occurred about the second row (11cm from the base).

Improved data collection system (03/19/05)

The noise in the data was reduced to ...% of the readings thanks to the new signal converter (see pictures of reading).
In addition an additional acceleromenter was included to measure the horizontal acceleration. This turns out to be a constant fraction of the vertical acceleration (?)

Vertical acceleration signal during 4th dynamic test (x10)

6th Static Test (03/19/05)

Download TEST VIDEO

As the previous specimen, in this specimen the infill was not contained in a bag and the bricks layers where staggered to avoid a continuous vertical joint. The main difference was that the fine sand was replaced by gravel size... and the lateral boundary conditions where replaced to improve the visibility of the specimen during failure.

The failure of the specimen initiated by slightly displacing the left leave (see video) and them by the overturning of the right leave at an infill height superior to that of the previously tested specimen with sand infill. This difference was expected because the friction angle of gravel is... which is higher than the one of the sand. This increase in friction angle of the material reduced proportionally the infill lateral pressureThe specimen started failing when the 10th row (55cm) was fully filled in, but then the overturning stopped because of the reduction of infill height (3cm reduction). The core was replenished and then the specimen failed. I should have waited a little longer to see if the specimen would have failed without replenishing the core.

5th Static Test (03/18/05)

Download TEST VIDEO

In this specimen the infill was not contained in a bag and the bricks layers where staggered to avoid a continuous vertical joint. This staggered distribution prevented the outwards buckling of the leave. The specimen failed statically by the overturning of one of the leaves at an infill height slightly superior to that of the previously tested specimen without the plastic bag: 49cm vs. 47cm. This difference could be attributed to the higher stability of the specimen due to the staggering of the bricks or/and due to differences in the leave alignment. Due to the fact that the bricks are not perfectly symmetrical in shape, there are small differences in the wall geometries.

The outwards buckling of the wall should be taken into consideration in real structures, because the length to height ratio is much higher than the experimental specimens.

3rd Dynamic Test (03/14/05)

Download TEST VIDEO (8Mb) Interesting failure!

In this specimen the infill was not contained in a bag like in the previous tests. The sand was contained by U-shaped cardboard with paper lips on the sides. The specimen was built up a height of 8 rows (44cm), where it was statically stable.

Testing

The specimen was vibrated at 50Hz and with a maximum acceleration of... The right leaf of the specimen started tilting, but then the lateral pressure diminished due to the fact that the infill compacted (settlement of 4cm). At this point the core space was replenished and the specimen was vibrated again. The failure observed was not only a simple overturning of the leaves, but also horizontal sliding of the bricks as well as outwards buckling of the leave along the middle joint. Previous specimens in which the sand was contained in a plastic bag didn't produce such a failure. The continuous central vertical joint facilitated the outwards buckling, which is not representative of most masonry works. In the next specimen this will be avoided by staggering the bricks.

It is my believe that at a given frequency, the actual vibration magnitude/acceleration won't have a major effect on the compaction of the sand, other than accelerating the process. However, an increase in vertical acceleration will cause a proportional reduction of inter-brick frictional forces preventing sliding.The lateral sliding of the bricks was facilitated by the vertical vibration, which reduced the inter-brick frictional force. This relative displacement between the brick is the kind of phenomena that will be investigated in the upcoming tests of pyramidal or spheroid stones.

Testing Equipment

For this test an accelerometer was installed at the base of the vibrating table and the vertical acceleration was collected during the test. The noise in the data due to the sygnal conditioner about 25% of the maximum registered.

4th Static Test (03/14/05)

In this specimen the infill was not contained in a bag. The sand was contained by U-shaped cardboard with paper lips on the sides. The vertical joints between the bricks were covered with a narrow paper slip to prevent excessive leakage. The fourth static specimen had a height of 9 rows and it failed before completing the infill of the ninth row. Compared to the previous static tests (with the plastic bag) this specimen failed with a height of 47cm (55.5cm the previous). This can be explained by the fact that the bag containing the sand took part of the lateral load by going into tension.

A problem that rose due to the fact that there was no plastic bag to contain the fine sand filling in the core was that the sand leaked out through the brick joints and the lateral boundaries. This leaking decreased to a minimum after 5 minutes of finishing to add the infill. This problem was solved in the dynamic test by slightly gluing the ends of the paper lips to the bricks.

2nd Dynamic Test (03/08/05)

Download TEST VIDEO (9Mb)

This specimen was carefully built, making sure that all the walls were at plumb. The total hight was of 9 rows (50cm).

The failure mode was oveturning of the right leave, but took longer to happen than the first dynamically tested specimen and there was no full collapse. This is due to the fact that when the wall starts overturning the sand hight is reduced (the core capacity is increased) and consequently the lateralpressure is reduced too.

Pre-failure state

2nd and 3rd Static Tests (03/07/05)

The second static specimen had a height of 10 rows (55.5cm) and as soon as it was infilled one leave started overturning and the other leave started buckling outwards along the middle joint.

The third specimen was built first to a height of eleven rows, then infilled to a height of then rows, and finally the eleventh row of bricks was removed. As soon as this row was removed one of the leaves failed in overturning.

1st Dynamic Test (03/04/05)

The specimen was built by adding rows of bricks on their wider side and simultaneously introducing the infill. The slenderness ration of the individual leaves was ....
When the construction of the specimen was finished, the right leave shoved some outwards tilting but was stable.
The specimen was 9 rows or 50cm high.

The failure mode was outwards overturning of the right leave about the second brick. The interesting fact is that a minimum vibration magnitude was enough to cause compaction of the core infill and a slow overturning of the right leave. I babtized these vibrations as the "ZZZ-Vibrations". I assume that the magnitude of the vibration needed to induce compaction is proportional to the grain size.

Notes
-Need to make sure that both the leaves are perfectly a-plumb
-Sliding was hindered because the infill was put into a plastic bag and might have prevented the relative horizontal displacement between blocks

1st Static Test

The specimen was built by adding rows of bricks layed on their narrow side and simultaneously introducing the infill. As a result the slenderness ratio of the leaves was about 10.
The dimensions of the specimen were:
1 leave: Height: 57cm; Width: 39cm; Thickness: 5.5cm
The infill was 9.5cm wide.

The failure mode was the outward overturning and bulging of the left leave.