Geoff Warren examines the equivalence of super heavy dynamic probe (DPSH) and SPT blow count data.
Parameter determination from dynamic probe data has long been an uncertain and debatable subject within the geotechnical community. On a recent project a super heavy dynamic probe (DPSH) was used to obtain data in granular material, chalk and clay. The critical aspect is the conversion of DPSH data (blow count n) to standard density and strength data for design purposes, which in our case were standard penetration test (SPT) N values in granular material and chalk and undrained shear strengths of clays.
Initially it was thought this would be a straightforward matter of getting the equivalence from the drilling companies who supplied the equipment. But as the questions were asked it rapidly became clear that the whole subject of exact equivalence of the DPSH data to SPT N values and undrained shear strengths was a poorly understood area. Clear guidelines on equivalence did not exist and a review of available literature produced surprisingly few papers on the subject.
Only two were identified which contained data and conversion factors that could be directly used: one for granular material and chalk, and the other for clay. A widely promoted and used piece of equipment was producing data that could not be readily used by designers.
To verify the data given in the two papers and obtain site specific data for the project, DPSH probes were carried out in granular and clay soils adjacent to boreholes where SPTs were performed. Direct field correlation was then obtained for these soils. A summary of the results, the equivalence in the papers and the equivalence developed for the design on the project is presented, and comments given. A comparable exercise was not carried out in chalk because strength data was obtained from traditional sample testing, but comments are given on dynamic probing values in chalk and the results obtained.
Dynamic probing has been in use for centuries as a quick method of obtaining information about soil, and a variety of probes have been developed. The equipment was standardised by the ISSMFE Report in 1977, which determined four sizes, DPL (light dynamic probe), DPM (medium dynamic probe), DPH (heavy dynamic probe) and DPSH.
The previous best standard for penetrometers had been the German standard DIN4094 Part 1 in 1974 (followed by DIN4094 Part 2 in 1980). This described six sizes with the largest being the SRS15, which is equivalent to the DPH probe. Dynamic probing was subsequently included in BS1377:1990 and the Department of Transport (DTp) Specifi cation. But these only detailed two probe types, DPH and DPSH.
Some papers on dynamic probing have been written in the UK in the last twenty years, but not as many as might be expected. A good, clear explanation of the history, sizes and size equivalence is given in Nixon (1988) at a conference on Penetration Testing in the UK. But a comprehensive review of European papers has not yet been carried out.
The dynamic probe is a small, compact, economic, convenient, mobile and quick piece of equipment. In the postwar era it was initially used more on mainland continental Europe in countries such as Italy and Germany than in the UK, where there has been a certain scepticism about its usefulness.
Nixon states that prior to 1988 'application in the UK has been negligible'. It is certainly very effective at giving absolute data on soil characteristic boundaries, depth locations of dense soils/soft rocks and investigation of hard spots, soft spots and voids. It is the extrapolation of actual usable (correct) parameters that has been the reason for its comparative lack of use within the UK, despite the fact that density and strength can be and have been obtained. There is also a cultural aspect: dynamic probing has never been part of the traditional ground engineering tools for investigation in the UK.
The inclusion in the BS and DTp specifi cations has seen an increase in use of dynamic probing in the UK. Initially the DPH was the preferred probe but this has gradually been replaced by the heavier DPSH, which is currently widely available from drilling companies.
One example is shown in Figure 1, and it is becoming the preferred and only probe offered by some drilling companies.
There are several reasons for this.
It has a heavier hammer with a larger drop and hammer size: drop length and cone are almost the same as the SPT (with a cone). This was thought to show the DPSH could have a direct correlation to the SPT.
Another factor is the growth of the window sampler equipment on track-mounted rigs which double as dynamic probe rigs. Window sampling has now developed to such an extent that a range of sizes of sample tubes is available. Eg. From 50mm to 100mm. i. e. solid 1m window U100 samples can now be produced (within a plastic liner). Test quality U100 samples can now be obtained by the latest generation of window sampling/dynamic probing rigs in Figure 1, although a slightly greater sampling disturbance could occur and should be taken into account.
Also worth considering is the cost effi ciency for obtaining samples and the density data if only limited depths are required.
The DPSH can probe to about 10m and the window sampling to 8m depending on the soil.
It would be fair to say that DPSH has now grown in use as a direct result of the improvement of winSPT N values, soil description and groundwater regime (from 2mbgl to 4mbgl).
Two differences were identified, geological strata and clay content.
The strata in Aldershot South comprised Windlesham Formation and Bagshot Formation, and in Aldershot North Camberley Sand Formation. The clay content in Aldershot South is 30% compared with 20% in Aldershot North.
Clay layers were only encountered in Aldershot South, and the Huntley equivalence in Table 1 appeared to agree with the site description of the clay.
DPSH probes were carried out in chalk at the same time as good quality U100 window samples were taken. Using the equivalence of n = 0.31N in C&M, equivalent SPT N values of 6 - 30 were obtained.
CIRIA C574 (2002) suggests that N values less than 25 are considered unreliable for density determination. As the results showed that most equivalent N values were less than 25 this data was of no real value for density determination. However the cores produced good samples for density determination and associated chalk strength information.
These strengths were in agreement with those obtained from core cutter samples in trial pits. The cores were a good demonstration of current window sampling techniques, which have improved significantly in recent years. The CIRIA report considers dynamic probing a crude technique that should be used in conjunction with coring and obtaining decent samples.
In sands from the only relevant published paper C&M and Aspire site data, no agreement was found on equivalence of DPSH blow counts to SPT N values. However similar trends of a decrease in N/n with a general increase in density (SPT N) were noted in the site data and paper with a near agreement at the upper density values of SPT N = 40. Site specifi c data is clearly of use, but it is only appropriate for large sites where boreholes can be used to correlate the values.
For clay, the published paper Huntley and Aspire site assessment appeared broadly in agreement.
Dynamic probing can therefore provide an early indication of undrained shear strength. Traditional sampling of clays is still required to give more reliable strength values.
For chalk the probe results from the C&M paper are reasonable.
SPT N data is not considered useful at levels below N = 25, and in any case density testing of samples is the preferred method of assessing chalk strength.
As with any dynamic probe data, all locations identifi d from low results should be given special consideration in the foundation design recommendations.
There are still very few publications on the DPSH probes, so DPSH probing is unable to provide general, clear and usable design data. More research into equivalence (N/n) should be carried out and suggested areas of further investigation for granular material are clay content of sands, gravel content and geological strata correlation.
DPSH plots are currently being produced and presented to geotechnical designers without site specific equivalences. These have very little meaning apart from the broad, crude interpretation of hard, soft or void, and the relative uniformity/non uniformity of ground conditions. Designers should be aware of the limitations of the data, particularly in the light of the increased use of the current window sampler/dynamic probe rigs.
Geoff Warren of consultant KBR is Lead Geotechnical Engineer on Project Allenby / Connaught for Aspire Defence Capital Works.
1 BS1377 (1990). Methods of testing soils for civil engineering purposes, British Standards Institution, London.
2 DIN 4094 Parts 1 and 2 (1974 and 1980).
Dynamic and Static Penetrometers: Application and Evaluation, Deutsche Norm.
3 Cearns P.J. and McKenzie (1988).
Application of dynamic cone penetrometer testing in East Anglia, from Penetration Testing in the UK, Thomas Telford, London (For granular material and chalk).
4 Huntley S.L. (1990). Use of a dynamic penetrometer as a ground investigation and design tool in Hertfordshire, from Field Testing in Engineering Geology published by Geological Society Engineering Geology Special Publication No.6. (For clay).
5 Nixon I.K. (1988). Introduction to Papers 1013, from Penetration Testing in the UK conference, Thomas Telford, London.
6. Lord J A, Clayton C R I & Mortimore R N. Engineering in Chalk, CIRIA Report C574, Department of Trade and Industry, London 2002.
dow sampling technology and the introduction of combined dynamic probe/window sampling rigs.
Theoretical equivalence of DPSH and SPT values and the problems The whole subject of parameter determination from dynamic probing results is considered with much scepticism by many people, including Clayton in Site investigation second edition, 1995. Clayton adds, 'The interpretation of probing results in terms of soil parameters is, apparently, carried out on the basis of locally derived correlations, none of which appear to have become widely or internationally accepted.'
Interestingly, this view is repeated on some current engineering discussion websites. This viewpoint and the limited papers on this aspect illustrate that parameter determination from dynamic probes has hardly advanced to a useful level at all.
The results in granular material are considered more useful and reliable, but data in clays has been obtained. Since the standardisation of dynamic probes in 1977, there have been surprisingly few attempts at proposing equivalence of probe data, i. e. blow count correlation to SPT N values, c u values and q c values, which are key parameters in geotechnical design.
There is guidance on equivalence for the DPH equipment in DIN 4094 Part 1. Further examination of this equivalence has been carried out by the authors of six papers.
But there appear to be only two papers covering the equivalence of the DPSH. These are Cearns and McKenzie (1988), which covers granular soils and chalk with some comments on clay, and Huntley(1990), which covers clay giving undrained shear strength values.
Curiously, BS1377 covers the description of the equipment and method of probing, but provides no comment on the meaning or interpretation of the data obtained. No further equivalence of the DPSH has been explored in print since 1990.
For clari cation, DPSH n is the blow count for 100mm penetration and SPT N is the blow count for 300mm penetration. The equivalence is therefore of n 100 and N 300 .Data from anonymous locations in East Anglia was analysed by Cearns and McKenzie (C&M).
No description of the soils was given, so unfortunately the sand and gravel content and type of sand grading is not available. Granular equivalence for sands and gravels is shown in Figure 2, reproduced from the paper.
In chalk, the data obtained produced a best t straight line, described as n = 0.31N. In clay a DPSH n - c u plot is presented but because of the scatter it was not considered meaningful. There is a discussion about the use of n values in relation to cone penetration resistance q c but this was not conclusive (q c is generally used for granular material). Torque is discussed but is not considered critical in soft or rm cohesive soils. No clear conclusions or recommendations for equivalence of n with any parameter in clay are given.
Huntley considers a DPSH probe and produced a DPSH n - clay description (soft, m, stiff, etc).
The details of the clay used were not given. With a torque of 100Nm the n value is reduced by 1.
The Huntley table for equivalent undrained shear strengths for clay with SPT N values added is reproduced in Table 1. The SPT N values shown are the generally recognised equivalences to clay strengths, which were originally proposed by Terzaghi and Peck 1948, and have been added to Huntley's table for completeness. It is noted they are not currently included in BS 5930:1999.
Three drilling companies that carried out window sampling/DPSH drilling, were asked their view on this subject. Two had no view and one quoted a linear 4n = N in granular material.
As one key characteristic required on the current project was density data in sands, and borehole use was limited for economic considerations, it was thought that super heavy dynamic probes could be used to provide this data. But as the investigation progressed it became clear the DPSH n - SPT N equivalence was not 'cast in stone'. To provide usable site speci c data, and to compare with C&M, DPSH probes were carried out adjacent to boreholes in which SPTs were taken.
Fieldwork and Data The work was carried out for the Project Allenby/Connaught. This is an £8bn PFI contract to refurbish six army bases, ve at Salisbury Plain and one at Aldershot Garrison. The work is being carried out by a new joint venture company Aspire Defence Capital Works, created by contractor Carillion and consultant KBR. The site work for the ground investigation was carried out by a site investigation contractor from November 2006 to January 2007.
The data was primarily required for the sands, gravels and clays at Aldershot, but data was also obtained for the chalk on Salisbury Plain.
At Aldershot South Camp in April 2006 the rst control set of investigation points was drilled through ne clayey sands and some clay layers (no gravel). These comprised three pairs of adjacent boreholes and DPSH probes. The dynamic probe rig used was a Dando Terrier 2002 drive sampling rig on Figure 1.
The results were compared to the C&M graph for sands and gravels, and a composite usable equivalence produced from this site specic data, which did not vary too signicantly from C&M. This comprised a new linear equation n = 0.2N for n = 0 - 5, and the C&M curve from n = 13. For the range between, i. e. n = 5 - 13, a straight line adjustment is used. This was used for the Aldershot South area, in particular for the loose or low medium dense soils.
In October 2006, at Aldershot North Camp, a second set of investigation points comprising six further pairs of boreholes and DPSH probes were drilled through similar clayey ne sand with no grave and a further equivalence developed. The results are shown in the graph in Figure 3. These results gave lower equivalence values from n = 0 - 22 and conformed more to C&M above n = 22.
The reasons for this are not clear.
Many factors were the same: the drilling company, rig, operator,