Service Performance of Multiple Anchors in difficult

ground conditions for both Permanent Anchors and

Temporary Removable Anchor usage

Tony Barley, Single Bore Multiple Anchor Ltd, Harrogate, UK

Mark Edwards, Keller Ground Engineering, Wetherby, UK

Abstract
At the site location in Ashford, Kent, the underlying Hythe Beds typically consist of
alternating layers of hard sandy limestone and weakly cemented sandstone interspersed by
beds of silts or clays of variable thickness. The irregularity of the strata made preliminary
anchor design difficult since it was not possible to predict that inclined wall retention anchors
distributed around the site would encounter either rock bands or entirely clay within the
proposed fixed length. The multiple anchor design offers flexibility which is particularly
beneficial in such varied ground conditions.

Both permanent anchors and temporary anchors were required to support sides of the secant
pile wall cofferdam. Furthermore the full length tendons were required to be removed from
the temporary anchor borehole when the adjacent structural works were complete. The
satisfactory performance of each and every individual anchor (typically 4 "units" per anchor)
in compliance with the acceptance criteria for both load/extension behaviour and creep
characteristic is described. Temporary anchors were load checked during the service period
and prior to successful removal of the anchor tendons.

Introduction
Permanent anchors were required to support one side of a secant pile wall and temporary fully
removable ground anchors were required to provide temporary support to a second length of
secant pile wall on the opposite side of the site. On completion of usage of the temporary
anchor some anchors were lift-off tested to confirm their load performance in service, whilst
the contractor confirmed that movement of the retained wall was within acceptable limits.
Finally the entire steel strand tendons were completely removed from the borehole of the
temporary anchors.

Ground Conditions Fixed Length Design and Supplementary

Safeguards
Typically the anchor contractor was provided with the load per metre run of wall, level of the
single row of anchors and a complete package of drawings, plans and site investigation data.

The geological sections showed "made ground", "drift", "Hythe beds", "Artherfield clay" and
"Weald clay" with the lower three beds being appropriate for anchor founding strata. The
"Weald clay" was dismissed owing to its depth below ground level and the uneconomic

Ground anchorages and anchored structures in service 2007, Thomas Telford, London

Downloaded by [ The Institution Of Engineers Malaysia] on [26/07/17]. Copyright © ICE Publishing, all rights reserved.
 BARLEY AND EDWARDS 381

length of anchors. Consideration was given to the "Artherfield clay" consisting of

predominantly very stiff sandy clay which would require either undereaming or post grouting
to provide anchors of adequate capacity.

However, the mixed soil conditions described as the "Hythe beds", occurring at between 5
and 15m below the existing ground level were selected as being the most suitable stratum for
founding the multiple anchor system. These comprised of medium dense to dense clayey sand
interbedded with Limestone bands, or firm to very stiff sandy clays interbedded with either
limestone or some weak sandstone bands (See Figure 1).

Multiple anchors consist of a number of short unit anchors founded at staggered depth in a
single anchor borehole (Barley 1997). Typical unit anchor fixed lengths are varied from 3.0m
to 4.0m depending on the envisaged strength of the ground and the bond capacity anticipated
in that ground at that unit anchor founding depth. The use of a multiple of short fixed lengths
reduces the effect of progressive debonding encountered in long conventional anchors and
allows more efficient load transfer and hence mobilisation of higher anchor loads.

Owing to the variable nature of the Hythe beds and the possibility that a unit anchor may be
located solely in a clay bed and possibly fail, and then more unit anchors were installed than
the basic design required. For example in certain areas anchors with 3 units each with a 4m
unit fixed length were adequate from a tendon performance viewpoint. However, in order to
ensure the complete anchor fully performed in any expected ground condition then 4 unit
anchors were installed and tested. In addition to this safeguard the actual depth and length of
each unit anchor installed could be varied, based on the ground conditions encountered during
the anchor drilling. These precautionary measures proved effective throughout the entire
anchor works involving 29 permanent anchors with 356 kN working load and 27 temporary
removable anchors with working load of 590kN. During cyclic testing to the required test
loads in actual fact no individual unit anchors failed. Despite the extreme ground variation no
redrilling or testing of supplementary anchors was required.

Anchors were installed nominally 1.0 to 2.0m below the wall crest and inclined at alternate
37.5 and 40° angles to allow utilisation of the appropriate level in the Hythe Beds whilst
maintaining a safe distance between the total fixed length of individual anchors. Anchor free
lengths were between 8 and 10m and a total of the unit fixed lengths summated to 13m.

Tendon Systems
All permanent anchor tendons were fully protected complying with the corrosion protection
requirement of BS 8081thropughout the fixed length, free length and the capped anchor head.
Each 2.5m long unit anchor encapsulation contained two concentric pregrouted corrugated
plastic ducts and each free length strand was surrounded by grease, a lmm thick yellow
plastic sheath and a lmm thick black plastic sheath.

Research work on totally removable anchors was started in the mid eighties by the
investigation of the capacity of a looped strand around a small diameter saddle (Figure 2).
Inspection of a loop after testing did indicate the expected damage to the sheathing but the
sheath in the fixed anchor constitutes a barrier to prevent tendon bond to the cement grout and
not a protective medium (Figure 3).

Downloaded by [ The Institution Of Engineers Malaysia] on [26/07/17]. Copyright © ICE Publishing, all rights reserved.
 382 GROUND ANCHORAGES AND ANCHORED STRUCTURES IN SERVICE

sample deplh(m) casing test amf). tnstru depth reduces legend

no& depth lype& foore ment description (ml level
type from to (m) value range (m)

MADE GROUND: SUff green-brown molted orange-brawn :

medium gravel of ctBK and WaL Rare fine and rnodlum
\
^88<
grave! size pockets of charcoal and brick.
m
1U
2X
1 2 0 - 1.65
1.20- 2.70
nil I tao- 49.75
§§§
3D nil S 8
1 Firm yellow-brown locally motlled dart brown sandy CLAY.
*t .—;

1.65- 2.05

2.30 -Z70ircwrw<Kx3sloo3! wry W i bed of brown

•

4U 2.70- 3.15 1 sBgrfly sandy day.

5X 2.70- 4.20
I 2.70 - 3.70m: soft with a Stae subrounded fine and medium
randstone gravel. -z-:
3.15- 3.80 nil S>1 3.15-3.70m: very soft -

3.70- 47^5
Soft yellow-brown motlled Hghi grey and orange-brown

BO 4.20- 4.65 nil S 6

sandy CtAY Wm e WBetosome angulartosubangular the
to coarse gravel and rare cobble of weak to moderately
ihS .
7X 4.20- 5.20 strong sandstone. *
-irr
4.90 - 4.96m: light grey cobble of moderately strong -
BO 5.20- 5.65 ml S 1 6
9X 5.20- 6.20 5V20 - 5.90m: becomes yeBoc^fcrown and sSft

UJ 5.70 - 5.S0rre tWn bed of yeftwrtrovm clayey Urns to 5.90 " 45.05
coarse sand vrifli some angutarto sooanguter line to -
10D 6.20- 6.65 nl S 1 0 coarse oravel.
X 8 2 0 - 7.20 Firm yeKour-orown spedded green sOflhlry sandy CtAY
Mltri occasional medium gravel size lenses of (inn

8.20 -7.30m: sandy.

-z-i
7 . 3 0 - 43.65
120 7.S0- 7.95 7.50 S 1 6 GreerHirown very clayey fine lo coarse SAND.
7 . 6 0 ; 43J5
13X 7.50- 8.50

lit
Stiff becoming very slfffyelktuMirownspecliled green
sandy CtAY. z=z:
X b-Z-
14D 8.60- 8.95 7.50 S 3 0
15X 8.50- 937 8.50 - 9.00m: mottled ligft! grey. wWi occasional gravel size
> pockets of clayey sand.
9.00 41^5
Yellow-brown dayuy finetocoarse SAND.
9.37- 950 8.50 (T250 9.40 41.55
0
C 9.37- 9,90 • Lkjhl grey of moderately weak fine and medium grained
. calcareous SANDSTONE. 9.70 41J5 ::::
YeDowr-brown clayey One toooarse SAND.
16C 9.30-10.80 - 139
10.10" 40.85 -T7.-..
0
Very e«f yeHovMmwm very sandy C W Y locally lending to
\ clayey sand. ruri
i
10.Q0-11.25 I AThei?r\£i.c>
I
17U
10X 10.80-11.80
1 1 1
(i

19D 11.00-1235 I 9.37 S 2 7 11.90 39.05

Figure 1. Illustration of the varied and in some cases the weak nature of the Hythe Beds in
which the 20° inclined fixed anchor lengths were founded

Downloaded by [ The Institution Of Engineers Malaysia] on [26/07/17]. Copyright © ICE Publishing, all rights reserved.
 BARLEY AND EDWARDS 383

• > •

Figure 2 Research work on strand Figure 3 Example of a looped strand

Loading around a steel saddle (Barley 1986) after removal from test arrangement

Figure 4 Tendon removal by Winch Method

After successful trials temporary removable anchor tendons consisted of a number of 180°
looped strands each fully greased and sheathed over its entire length. (Trummer et al 1995).
Within the head of each loop was located a steel saddle ("first size") and a short glass
reinforced plastic compression bar. On completion of usage of the temporary anchor the
lubricated unbonded strands can be easily withdrawn from the grouted boreholes by the use of
a 10T winch (Fig 4) or a crane or where access space is limited by the use of a long stroke
stressing jack

Performance in Service
Permanent anchors were test loaded in accordance with on Site Suitability Test and On
Acceptance Test requirements of BS 8081. This involved application of load cycles up to 1.5
x working load and resulted in compliance with apparent tendon free length requirements and
load hold capacities.

The looped strand temporary anchors were also subjected to the two test methods requiring
load cycles of up to 1.25 x working load and compliance with the same acceptance criteria
(Fig 5).

Downloaded by [ The Institution Of Engineers Malaysia] on [26/07/17]. Copyright © ICE Publishing, all rights reserved.
 384 GROUND ANCHORAGES AND ANCHORED STRUCTURES IN SERVICE

During the works wall movement was monitored by the Contractor and no movement was
detected

On completion of the satisfactory service performance of the temporary anchors 25 out of 100
unit anchor strands were subjected to lift of tests to investigate the potential load loss in this
looped strand system founded on the variable Hythe Beds. A typical example of the
establishment of initial anchor lock-off load (627 kN) followed by a load check at 24 hours
(617 kN),and a lift-off check prior to anchor removal (625 kN) almost a year later is present
in Figure 6.

In other load checked anchors load loss varied from zero to 5% of the original lock-off load.
Hence no unit anchors indicated a loss of load greater than the 10% normally accommodated
in anchorages complying with the acceptable load change requirements of BS 8081.

Strands were inspected after tendon withdrawal and exhibited no indication of corrosion or
corrosive products or indeed any damage to the seven individual wires making up each of the
removed looped strands.

Summary and Recommendations

Subject to the provision of adequate lengths of unit anchor tendons that can be adjusted on the
anchor site, then the anchor lengths can be controlled during anchor drilling to utilise the
stronger grout to ground bond capacity of stratum encountered in such variable materials as
on the Hythe Beds.

Records of movement of the retained wall indicated that no movement took place during the
service life of the temporary fully removable anchors. After 4 and 8 months when the
temporary anchors were surplus to requirements, lift-off checks established that the load loss
ranged from zero to 5% of lock-off load. On completion of lift-off checks the entire length of
the fully lubricated prestressing strand were easily removed from the anchor boreholes. This
completed the special service performance required of the temporary removable anchor
system.

The use of removable tendons in temporary work anchorage is becoming more common
owing to the modern advancement and development of ground anchorage technology. In
many cities throughout the world including major conurbations such as Hong Kong and
Singapore it is considered that the abandonment of used prestressing tendons within the
ground is unnecessary and an unacceptable long term contamination of the subsoil.

National and International Codes of Practice for Anchorages should fully acknowledge the
extensive development of such available systems (Herbst 1997) and incorporate an acceptance
criteria for tendon removal from temporary work anchorages.

Downloaded by [ The Institution Of Engineers Malaysia] on [26/07/17]. Copyright © ICE Publishing, all rights reserved.
 BARLEY AND EDWARDS 385

AHCHORHo. I 381

31.7290 - COUNTY S Q U A R E , ASHFORO

W o r k i n g A n c h o r a g e Stressing Results
SBMATEMPORARY REMOVABLE
W o r k i n g L o a d - 57QkN (143kN/Unit)

p-UNITA - » U M I T B

-4-UNITC -KUNITD

10 20 30 40 SO SO 70 80 DO 100 110 120

Corrected Unrt Tendon Extension ( m m )

Figure 5 Proof loading of a temporary removable anchor to Test Load of 720 kN involving
four simultaneously stressed unit anchors.

Downloaded by [ The Institution Of Engineers Malaysia] on [26/07/17]. Copyright © ICE Publishing, all rights reserved.
 386 GROUND ANCHORAGES AND ANCHORED STRUCTURES IN SERVICE

31.7290 CONTRACT - COUNTY SQUARE, ASHFORD

MONO-JACK LOCK OFF - ANCHOR 3I>

ANCHOR 36 WAS LOCKED OFF ON THE 24 AUGUST 2006 & 24 HOUR LOAD CHECKED ON THE 25 AUGUST 2006
'-. V, HOLD ANCHOR UNIT BHUGE UN "RAN' JNIT ANCHO3HEAD CORR. CCUM=NT3
=EROD -SAC LOAD PRESSURE; ME-..; "ENDON MOVEMENT UNIT
STRAND EXTENSION Measured Aaual EiTEF.E OH 4Uni1SBMA»ith2StrandlUnis
|tN) |kN] (psi) (mm) (mm! (mml (mm)
LOCK-OFF
• 10 0 627 76 2351 aeo BB g 86
78 2351 881 84 0 M
78 2351 ae7 74 0 74
78 2391 872 75 0 75
78 23SI 887 S3 D 63
78 2351 871 li 0 61
78 2351 356 49 a 40
T8 2331 849 4a G 4B

IDE 24 Hre S]7 78 2300 1D 1.5 B0

76 2300
78 2350
78 2350
76. 2300
"6 2300
78 2350
78 2350

ANCHOR 36 WAS LOAD CHECKED S. DE

STRESSED ON THE 23 JULY 2007
HOLD ANCHOR UNIT 3ALIC-E
PEROD LOAD LOAD PRESSURE!
STRAND
IB* (kN) ipsi:
108 \IA 625 78 is::
76 2000
78 2050
n 2I0Q
78 2050
76 2000
BO 2100
78 20c1:

Figure 6 Precise load holding capacity of a temporary anchor initially, after 24 hours and
after nearly a year

Acknowledgments
The authors thank Laing O Rourke for their pro-environment approach in specifying the
removal of steel tendons from the temporary work anchors and for their provision of technical
data.

References
British Standard, Code of Practice for Ground Anchorages, BS8081:1989

Trummer F., Stockhammer P., "Der wiederausbaubare Litzinanker, System Keller" Int. Conf
on Anchors in Theory & Practice - Salzburg, October 1995

Barley A.D., "The Single Bore Multiple Anchor System" Int. Conf on Ground Anchorages
and Anchored Structures, ICE London, November 1997

Herbst T., "Categories of Removable Anchors" Int. Conf on Ground Anchorages and
Anchored Structures, ICE London, November 1997

Downloaded by [ The Institution Of Engineers Malaysia] on [26/07/17]. Copyright © ICE Publishing, all rights reserved.