Flooding to the Correction

In the last post, I assured that I will write on the measures which we took to protect the bridges and embankments owing to damages caused by scour. After a devastating rainy season which happened during the project implementation, we were convinced on the additional requirements of the protection measures which were not included in the original designs.

The following photo shows the scale of the damages owing to the flood event. Even though all bridge structures were not submerged by the rising flood level, the approaches were badly damaged.

In an another location, the temporary baily bridge, at the bypass road,  which was constructed to cater the traffic during the construction was totally submerged while the new  bridge structure stands firmly owing to correct choice of HFL during the design.

Note the  red colour  bailey bridge.

The temporary bailey bridge is submerged while new bridge unaffected.

The following bridge is located at a lagoon. Similar to other locations, the bridge approaches have been inundated.

The next photo depicts the scale of the damages in the downstream of the approach roads near to bridge structures.

After witnessing a such a devastating flood , we took concrete  remedial measures  in order to withstand the future flood events.

I keep those for next post as  I feel that this post is lengthy enough for a day :)

The bridge on a river where no single drop of water to be found.

Even in a dry season , very few water paths are  get dry as following one  where not single drop of water can be found.

This bridge is located in  Mankulam - Mulathive road, where I worked as the Consulting Bridge Engineer for the project in 2012/13

A Bridge on a river where you can not find a single drop of water.

While structure was under construction,   few visitors to the site raised their eyebrows on the requirement of a bridge at particular  location. However,  the Hydraulic studies of the area gave us a very different figures which are drastically differ from the depiction of the picture.

I was fortunate to witness the theory meets reality.  During the time of project execution, bridge met its highest flood which it was designed and stood against it.

Same bridge but with many drops of water

The bridge was designed with a 600mm of minimum  freebord height for the high flood level and it is adequately designed as seen in the picture.

However,  at the time of  flood , the bridge was  not in its completed form. The approach roads and the embankment protection measures were not completed. As a result, the embankment was damaged and later it was rectified.

Looking forward to write more in my next post on the subject of  bridge protection  measures  that we took after the rainy season.

The importance of Hydraulic Jump

I caught this video  from FB post..........

This video shows the importance of the  correctly design HYDRAULIC JUMP at the  spill.

I can remember the effort of my Hydraulic Professor who took great effort to teach us the phenomena during my undergraduate studies.

The hydraulic jump is designed to dissipate the energy of the water flow, so that it flows in safe speeds inthe downstream.

Cracks in Prestress concrete Beams

I came across this problem last year. Although the issue has been solved, still I think we can share our views on this.

The problem occurred on 19m long pre tensioned PSC beam.  The cracks, in the range of 0.2 mm – 0.5 mm,  were noticed on the top flanges of the 19 m beams and  near the two lifting positions of the each side ends of the beams as shown in the following figure.

cracks positions of the beam

closed view of the cracks

The crack positions, sizes and the propagation are almost identical in every beams at the site and I had my  uncertainties on the structural integrity of the defected beams.

The cracks have appeared on the places where maximum hogging moment occurs during the lifting of beams. Hence these are considered as flexural cracks ( Not shrinkage or early thermal cracks) and possibly owing to lifting of the beams before the concrete achieve its characteristic strength in Transfer Stage.  

I would highly welcome your opinion on what next to do?

Monolithic Buried Concrete Box Culvert Design: Load Cases and Applicability of Hydrostaic

Three cell box culvert which I designed  for  NRCP

Last few days, I spent most of my work hours on reviewing the designs of concrete box culverts.

As local practice is based on British codes, the most relevant  document is identified as part 2 & part 4 of Bs 5400. The additional guidelines are  provided by BD 31/01.

Appendix A of BD 31/01 shows many diagrams on  load cases which are to be considered. It includes following combinations.

1.) Max. Horizontal loads with Max. vertical loads ( No Traction)
2.) Minimum vertical load with max. horizontal Loads (No Traction)
3.)Maximum vertical load with minimum  horizontal loads ( No traction)
4.)Traction with max. vertical Loads
5.) Traction with min. vertical loads

The checks on stability  is conducted  as follows;

6.)  Sliding : min. vertical loads with buoyancy on foundation slab. ( primary live loads are included with min. factors as it is necessary include traction forces)

7.) Bearing : max vertical loads with buoyancy on  foundation. All loads are nominal.


In all above conditions , it is necessary to concern whether to  include live load surcharge on to  both abutment walls or not.

In the first five of loading cases, BD 31 is silent on applicability of  Hydro static Pressure on walls.

However elsewhere , in cl 3.1.4 it says, /// when appropriate , the effects of hydro-static pressure and buoyancy shall be taken in to account. The increase in pressure on the back of the walls  due to hydro static  pressure at a depth  Z meters below water level shall be taken as 10Z(1-K)   kN/m2///

Please see the following figure which i took form my note book and captured from my mobile camera. it  explains how this 10Z (1-k) is derived.

However BD 31 does not well explain   that how hydrostatic forces do   applicable in the first five load cases.


Contrary , the american LRFD documents which are basically based on Marston's studies on the subject , provide more details on this.

The following picture that referred from a design guidelines of Kansas Department of Highways is self explanatory.

The Department of Transport of Massachusetts  LRFD guidelines provides following recommendations on the load cases;

1.) Maximum  vertical load in deck slab and Max. outward load on the wall ( DL max+ Earth vertical max+earth horizontal minimum+hydrostatic max)

2.) Minimum vertical load and maximum load on inward wall of the abutment ( DL Min + Earth verticle Min+ earth horizontal max)

3.)  Max vertical load  and max load in inward wall of the abutment ( Dl max+Earth verticel max+Earth horizontal max+ live load  max)

It is clear that there  is not  general agreement  between distinguish  design guidelines on the load cases and more specifically on the  hydrostatice loads and the  uplift /buoyancy.

I am of the view that hydrostatic pressures become critical when a sudden discharge  flows through the barrel without no time allowing Ground Water level to  rise. Such a phenomena could happen in the cases of upstream flood or owing to spill of upstream reservoir.

In the same line of argument , water level can be differ in two sides owing to rise of ground water level but stream flow is at regular levels. I am in doubt whether is this could happen in real?

In other cases hydrostaic forces are in balance or have little impact on the total effects.

construction of the 3 cell

Modeling of Slab Column joint

In structural engineering, the Flat Slab is identified as a slab which is supported generally without beams by columns with or without column heads. However, I have understood that Flat Slab designs are not quite popular in Sri Lanka, may be owing to its complexity in design, detail and construction. But things what I am going to discuss here are equally important in any other column slab design problem such as pile cap etc.

Generally the slab can be analyzed by dividing it in to longitudinal and transverse frames consisting of columns and strips of slab. This is identified as Equivalent Frame method and frames are analyzed by moment distribution. However these manual methods are things for the past. More sophisticated and user friendly Finite Element Methods (FEM) and software are widely used to solve the problem.

Plate mesh and the columns : STADD Pro

3D view of the Model

However it can be seen  in FEM that the support moments in slab column joint are sometimes unacceptable  owing to its unrealistic high values. The anomaly is varying as accordance with span, column sizes and slab thicknesses.

Longitudinal Bending Moment Diagram

Transverse Bending Moment Diagram

However , most FE software offers  'peak smoothing' techniques to addresst this problems.

In General, the number of elements used in a FEM model can greatly affect the accuracy of the solution. As the number of elements, or the fineness of the mesh, is increased, the accuracy of the model increases as well. However in this particular Slab Column problem finer mesh would deteriorate the quality of the results more.

Deflection pattern : scale is adjusted

The solution for this problem  lies on the  realistic modeling of boundary conditions  of the slab.

In this regard, I usually  adopt two methods;

  1.) As the first method, Slab may modeled as a plate/shell element while column may modeled as a three dimensional solid object.  By doing so, we get closer to the real world scenario of the slab column joint. However analysis of this model may be time consuming and required higher degree of computer resources.

2.) In the second method, the flat plate is modeled as a plane of finite elements, and columns are simplified to “pin” or “fixed” supports applied at the elevation of the plane of the slab. Sometimes these supports are modeled as springs with a finite elastic stiffness to improve the behavior of the model.

In this method it is essential to connect nodes rigidly which lies within the area of the column section to the end of the frame member. This technique is otherwise referred to as a master-slave technique, where the end of the frame member column is the master joint, and the joints in the slab within the column area are slaves.

This technique offers an advantage over the use of three dimensional solids with respect to computational efficiency  and  Analyzing time of the model.

I hope this will help you in FEM modeling.

Is excess cover good in insitu RC structures?

  -->

Many times, during my site inspections, I have found fault with the Contractors for providing an inadequate cover for the reinforcements in R.C structures.  Obviously inadequate cover for the R.F has to be rectified immediately as less cover causes the larger crack widths in concrete and thereby it exposes the reinforcement to surrounding environment to get corroded. This is an important factor to consider in which bridge structures as it withstand severe environmental conditions.

 Furthermore, an inadequate cover leads to unsatisfactory fire resistant as well as the under developed bond strengths between concrete and R.F also.   Hence there should not be any argument whether inadequate cover is acceptable or not. The clear and loud answer is NO!!!!!

Then what about the other side of the story? In other words,  is provision of  an  extra cover than the specified limits  acceptable? Oki doki…take a little time and come up with an answer………………..

Many distinguished concrete specifications are very clear on this matter. The tolerance for the concrete cover is between +10 mm to -10 mm. However some inexperienced contractors, especially small scale subcontractors, do posses a misconception that increased concrete cover is acceptable and allowed to go unnoticed. This is wrong…….

What is the underlying theory behind the upper tolerance limit of the concrete cover? The upper limit of the concrete cover governs by its effect on Crack widths. Generally the Crack width of the concrete is governed by three factors namely 1.)  Tensile strain of the point  2.) Distance to the R.F bar from the concerned point &  3.) Depth of the tensile zone

In this case, the second factor is significant and it governs the upper limit of the concrete cover.  The increased concrete cover leads to longer distances of R.F bar from the cracking point.  Hence with an increased concrete cover, the crack width also becomes larger.

So cover for concrete should be in its specific limits whether it is plus or minus…keep checking for that….

Oki guys…that’s it for today…Looking forward to see your comments on this topic…. Cheers!!!!!