Effect of Vibration in Dense sand

I wouldn't expect any significant effects; I've done the same thing in weaker sands and shallow water tables with great success.

Are you worrying about an existing structure nearby?

I am sharing my actual experience here:
The 1.5m dia drilled shaft is around 29.0m long. While inserting the temoprary inner liner by vibrohammer, which was inevitable to prevent side collapse, there was no problem. Then afterwards bore is cleaned and reinforcement placed. Concrete is poured upto around 26.0m and withdrawal of liner by vibrohammer was commenced since ordinary vertical pulling was not successful. After extracting of about 10.0m of liner it was observed that cage has sunk by about 2.0 inside the bore and so for concrete. Again the sinking happened whenever withdrawal of liner was taken up in stages. More importantly, it was occuring for other piles. Also it is noticed that segregation is taking place in the high slump self compacting concrete due to vibration.

Waiting for suggestions/ advises.
Thanks

You have no effect on the dense sand layers : it is too dense to be compacted further with a top vibrator.
Your piling construction method cannot guarantee the level of the top of the reinforcement cages nor the cut-off level.
You should modify your concrete formula in order to avoid segragation. If your high slump is obtained with water, reduce the amount of water and use a plasticizer.
You should not vibrate close to fresh pile, try to implement a working sequence which doesn't submit a fresly cast pile to vibrations before 48 hours.

BigHarvey is correct. The problem lies with the concrete, not the soils. Using water to increase slump is a bad idea for high slump concrete; using plasticizers will result in a much better product. Lymon Reese and Mike O'Neill recommend a slump of 6 to 8 inches (15 to 20 cm) for drilled piers; I prefer 8 to 10 inches (20 to 25 cm) for drilled piers deeper than about 30 to 35 feet (~10 meters) or when the concrete will be placed under or through (a "little&quot water.

I'm not sure that segregation is as much of a problem for piers as it is for other structural elements. But that's just a hunch, not the results of any study of piers constructed and later exposed that experienced aggregate separation. In some ways it goes against conventional wisdom. But so does concrete with a 10 inch slump...

The rebar cage can only sink if the material at the bottom of the shaft is somehow losing it's bearing capacity. If the rebar only sinks when the vibro hammer is in use then one would certainly suspect the vibro. What is the pore pressure at depth? How sure are you that the SPT of 40 is uniform everywhere? The vibro will cause sand susceptible to liquifaction to locally lose strength, resulting in the rebar settling along with the concrete. With a web search you might find case histories. In this situation it has been necessary to hang the rebar cage from a third crane line while pulling casing and placing concrete at the same time. When the casing is up the cage is supported at ground level until the concrete sets.

Thanks guys for your advises.

I am also feeling in line with krd. There might be a temporary localised liquefaction due to vibration which is eliminating the shear strength of foundation sand. If the base soil was weak, as Contractor is also claiming, the reinforcement alongwith 26.0 m concrete would have sunk by its own weight prior to vibration. The "N" value in that sand layer in all the boreholes is 40 and more. Actually I am not sure about the mechanism of liquefaction in dense sand. Contractor is using plasticizer in the concrete mix. To tie the casing with crane is not possible continuously since tremie pipe could not be placed to pour the concrete.

I am searching websites, but of no much use.

I think we are mixing two problems. One is the level of concrete and reinforcement and the other one is segragation of concrete.
For the first one, I can't see anything wrong apart from the installation method chosen. When you withdraw your temporary casing after concreting, the concrete level will be lower because you have at least to replace the volume of the casing you are removing !
For a 1500 mm diam pile I would use a wall thickness of 15 mm at least. The unit weight of such a casing is about 580 kg per m. If you remove 10 meters it is 5800 kg of steel or 0.75 m3. Over a surface of 1.77 m2, this gives a height of about 40 cm and we are talking about 2 inches !
Same for the cage : it is usual to have the bottom of the cage about 20 cm above pile toe, which leaves room for the cage to follow the concrete.
As I said in my first thread it is one of the problems of this installation method : you cannot guarantee cage and concrete level with accuracy ( In some cases the fresh concrete will displace the weak soils and you can experience overbreaks which are far more than just the steel volume replacement.
For segregation, if there is a plasticizer, the quantity might not be sufficient. You should try to see how the slump changes with time : if you experience major changes after a short period ( say 1/2 h) then you should increase the plasticizer quantity, you should also check your fines content. We generally put a minimum of 400 kg of cement in France for deep foundations concrete.

1967pradyot said in his original post:
After extracting of about 10.0m of liner it was observed that cage has sunk by about 2.0 inside the bore and so for concrete.

Do you mean the rebar cage sank 2 meters, or 2 cm? If the answer is 2 cm, or even 20 cm, I don't think you have a problem. But if the answer is 2 meters...I think you have a problem with your field investigation. This could be a construction issue also, but I doubt that. I have a hard time seeing rebar penetrate 2 meters of moderately dense sand under its' own weight - even with casing extraction with a vibratory "hammer." I've witnessed lots of pier installations to similar depths and in similar sands with no such effects.

How much did the concrete level drop?

I had understood 2 inches. A two meter drop corresponds to 0.4 m for the volume of the steel casing and 1.6 m for overbreak which is 2.8 m3 of concrete for a theoretical volume of 17.7 m3 of concrete which is an overbreak of 16% which is a little bit high but not exceptional. It is difficult to measure fresh concrete level with an accuracy better than 0.25 m !
If the cage is going down two meters it means that it must have collapsed. have seen people tryng to prevent cage to lift by putting weights on the cage : this ends up with the right level on top but the cage is crushed and the toe is lifted. It must work the same in the other direction. A simple way to check it would be to install sonic coring tubes in the cage and check them on their full length after concreting.

Cage collapse is also a strong possibility. Perhaps the most likely culprit -

Since all of 1967pradyot's dimensions were metric, I wasn't including inches as one of the potential units of measurement for the "2.0" value. Clearly, we need to hear from 1967pradyot at this point.

1.)Dense sands (and n=40 is dense) will become loose when vibrated excessively. Trying to vibrate a 26 m. cassion in N=40 material must be difficult at best. The driving may be causing softnig of the toe and should be investigated, as well as softning from driving adjacent piles.
2.) Topping off of piles is not uncommon as the material settles into the pile.
3.) The critical period for the pile is not while the pile is plastic, but just after it sets. If too much vibration occurs within the first 24 to 48 hours, the concrete will be severely comprimised.
3.) Cages get heavy quickly, especially if you are dropping concrete. If the bars are spliced, tie wire may not be sufficent. It may be nessacry to use rated bar splices. Note that only enough bars need to be spliced so as to develop the weight of the lower cage. It is probably not necessary to splice all the bars.
4)Assuming the cages are self supprting as described in 3, two pairs of holes can be blown in the side of the pile at the top of the pile. A bar can be slid through each pair and the cage secured to the bars.

There's a difference between theoretical and in practice. I realize that theoretically the sands can go 'quick', but remember that the overburden isn't zero - it's equal to the weight of the 26 meter column of fluid concrete above. And a lot of the vibro's energy is dissipated before it reaches the pile tip.

I've been around this type of installation for more than two decades, and I have never seen the bottom go 'quick' and the rebar sink two meters into the underlying sand. Not in material with N=40, anyway. I'll have to see hard evidence to believe that one. If the sands are going quick, then the sands aren't really dense - or the material isn't just sand. Cage collapse seems much more likely.

In any case, a lot more investigation is warranted. At least one of the offending piles needs to be excavated. This seems drastic until you consider the problem from a performance standpoint. After all, the very integrity of the sands and piers - and their ability to support the structure - is in question. A little coring and pile tapping just won't do -