Drilling methods for soil sampling in clays

[nick269]

Hello all

I've just been assigned to a project in Vietnam, where a site investigation in unconsolidated deposits (predominantly clay) is currently in progress. Drilling is with small coring rigs, advancing the hole with single barrel core barrels and bentonite as drilling fluid. This is my first time to see such a method for geotechnical exploration, although I have quite a bit of expereince with this method for drilling water wells. I have also worked with diamond-bit rigs in granite using only water (bentonite was forbidden because we had to hydraulically test the fractures). For water wells, we always utilize downhole logging methods to give us data concerning groundwater. There is no such provision here. How can we get reliable data about water strikes and artesian conditions without geophysical logging? I've given instructions to check fluid levels every morning, evening, and whenever it seems they are losing fluid or the returning fluid flow is increasing, but these drillers are not experienced enough to tell. And how much contamination is taking place with Shelby tube and split spoon samples? Again, the deposits are mostly clay, and the lab tests are the basic indices, direct shear box, and oedometer consolidation.

Nick

 

[BigH]

I take it that the coring is for advancing the hole - then when you reach a sampling depth, you will take a thin-walled tube or split spoon. With the drilling mud, you will get some contamination at the interface with the soil but you would 'clean' it out when the sample is examined (for split spoon - you would toss out the last 10mm or so - for the thin-walled tube, you would similarly clean out the mud and a bit of the top part of the soil. You will get some smear about the side of the sample as well and clean this up - the sample you will test will not be contaminated to any extent so long as you make sure you clean off the muck. I would try to keep the split spoon sample "logs" as such rather than just shoving into a bag.

As for water pressures - you would install standpipes - pvc tubing with slotted ends - suitably wrapped with geotextile cloth. Properly seal the hole with bentonite plugs (balls of bentonite you drop in). Though the hole is full of bentonite, if the standpipe is wrapped, I don't see this being a problem - you will measure the 'pressure' inside the standpipe. See Hvorslev book on Soil Exploration.

Enjoy Vietnam - I haven't worked there but have worked in neighbouring locales and with Vietnamese contractors.

 

[nick269]

Thanks a lot BigH

I didn't think it was too much of a problem for the samples. Except perhaps for very permeable sands, there might be some contamination.

We are taking SPT/split spoon, and driving in Shelby tubes for 'undisturbed' samples.

Standpipes are not in our contract (I guess the person creating the scope of works just assumed they were going to drill with casing), and the foundation engineer hasn't shown much concern about water levels and seepage, although the water table in the area is known to be quite high.

Nick

 

[BigH]

nick269 - even if they wee drilling with casing, you wouldn't, in clay, be able to determine the groundwater level (or varying piezometric pressure levels in various strata) in clays during the time frame of drilling - i.e., observing at night and then 12 hours later at the beginning of the day. The only possible way would be to leave an open borehole and that is affected by precipitation as well as 'averaging' any varying piezometric levels in your investigated column. While the standpipes might not be in your scope - you can jerry-rig up some in the field pretty cheaply (I've even used garden-hose before). remember is that you might have different piezometric levels in various strata and unless you properly seal off zones, you will only get an 'average' of the zone. I think you are right that with soft surficial clays, the groundwater level would be at or near the surface - the only thing is that it might be perched when compared to the regional levels or underlying levels (say a stiffer clay that has drained to a pervious lower layer). Usually, for embankment settlement computations and the like the choice of high 'gwl' will give conservative estimates of settlement (i.e., higher than real) given that other factors are known and consistant.