Geotechnical Instrumentation News · Perth, Western Australia. The New MetroRail Project is a A$1.5...

Geotechnical Instrumentation News

John Dunnicliff

IntroductionThis is the forty-eighth episode of GIN.Two articles this time. Also responses toquestions that I asked in the previousepisode of GIN.

My ‘Umbrella’ QuestionsIn the previous episode of GIN I askedtwo questions, intending to seek outopinions on how manufacturers and us-ers can cooperate to the maximum ex-tent. I’ve received six responses. Fiveare from representatives of NorthAmerican manufacturers (AppliedGeomechanics, Durham Geo Slope In-dicator, Geokon, Roctest and RST), andone from a user (Arthur Penman).

Cable-free SensorsThe article by Chris Rasmussen tells usabout recent work with cable-free sen-sors. He discusses the design parame-ters that are required for a cable-freesystem, reviews two installations andlooks at some of the lessons learnedalong the way.

Interpreting UnexpectedInstrument DataVerne McGuffey comments that manycasual users might have informationthat would be useful to the profession.He has put together a few ideas aboutevaluating data, and has described someunusual situations where the instrumentwas assumed to be at fault, but was not.

He comments, “Perhaps this notewill invite other users with useful expe-

riences to share them also”. YESPLEASE.

Next Instrumentation Course inFloridaThe next instrumentation course inFlorida will be on March 18-20, 2007 atSt . Petersburg Hil ton (www.stpetehilton.com). Details of the courseare on www.doce-conferences.ufl.edu/geotech.

International Symposium onField Measurements inGeomechanics (FMGM)The 7th International Symposium onField Measurements in Geomechanics(FMGM) will be held in Boston, MAduring September 23-27, 2007. FMGMsymposia have been held every fouryears since 1983, in Switzerland, Japan,Norway, Italy, Singapore and Norwayagain. Information about these past sixsymposia, together with related infor-mation about geotechnical instrumenta-tion, can be found on www.fmgm.no.

The 7th FMGM will concentrate ongeotechnical, structural, environmentaland geophysical instrumentation meth-ods and applications, and will focus onthe following themes and topics:

Theme 1. Case StudiesThe role of field measurements in prob-lem-solving, research, safety assess-ment, risk assessment or improving thedesign of civil engineering structuresand works.

• Case histories and monitoring appli-cations

• Instrumentation for innovating de-sign

• Instrumentation and geo-hazards

Theme 2. State-of-the-Art andFuture TrendsThe latest in measurement technology,equipment, communication methods,data management and interpretation,and visions for future development.• Geotechnical, structural, geodetic,

environmental and geophysicalinstrumentation methods and equip-ment

• Real-time monitoring• Remote monitoring, wireless sys-

tems• Early warning systems• Data acquisition systems• Analysis and presentation software• Performance, cost and reliability

data• Capabilities and limitations• Future trends and needs• Emerging new technologies• Fiber optic sensors• Internet applications• Global Positioning Satellite systems

(GPS)• Automated total stations• Problems and pitfalls• Avoiding electromagnetic interfer-

ence (EMI)• Protecting equipment against dam-

age during electrical storms

Geotechnical News, September 2006 29

GEOTECHNICAL INSTRUMENTATION NEWS

Theme 3. The Business Side ofInstrumentationDemonstrating and quantifying thebenefits of field measurements to pro-ject management teams, owners, engi-neers, contractors, regulators and insur-ers.• Benefits of monitoring to owners• Benefits of monitoring to designers• Benefits of monitoring to construc-

tion contractors• Role of monitoring in risk manage-

mentPlease visit www.geoinstitute.org

and www.fmgm.org for more informa-tion, including the requirements forsubmitting abstracts of papers.

Third International Symposiumand Workshop on Time DomainReflectometryThe third International Symposium andWorkshop on Time DomainReflectometry for Innovative Soils Ap-plications (TDR 2006) will take placeon September 17-20, 2006 at PurdueUniversity in West Lafayette, Indiana.

Details are available at: https://engi-neering.purdue.edu/TDR

Registration is limited to 120 per-sons. A grant from NSF allows for anominal registration fee for up to 20 fulltime students and nontenured facultypersons who hold tenure-track facultypositions, with priority given to womenand under-represented minorities.

The registration form is available ath t t p s : / / e n g i n e e r i n g . p u r d u e .edu/TDR/ and the registration may bedone online. Any questions should beaddressed to: [email protected]

Some Anglo-American TriviaRecently I did something foolish, ad-mitted it and said that I’d “eat crow”.Then I realized that I had no idea whatthat meant, so pulled out my trusty Dic-tionary of Phrase and Fable and foundthe following for “Eating Crow”:

To be forced to do something ex-tremely distasteful. The expressionderives from an incident during anarmistice of the Anglo-AmericanWar of 1812-1814. A New Englanderunwittingly crossed the British lineswhile hunting, and brought down acrow. An unarmed British officer

heard the shot and determined topunish the offender. He gained holdof the American’s gun by praisinghis marksmanship and asking to seehis weapon. The Britisher then toldthe American he was guilty of tres-pass and forced him, at the point ofthe gun, to take a bite out of the crow.When the officer returned the gun theAmerican in his turn covered the sol-dier and compelled him to eat the re-mainder of the crow.

So now we know!

ClosurePlease send contributions to this col-umn, or an article for GIN, to me as ane-mail attachment in MSWord, [email protected], or byfax or mail: Little Leat, Whisselwell,Bovey Tracey, Devon TQ13 9LA, Eng-land. Tel. and fax +44-1626-832919.

Na zdrowie! (the website says “Polishdrinking toast” – will someone pleasetell me what it means? Maybe it’s some-thing that I shouldn’t print!)

Responses to ‘Umbrella’ Questionsabout Manufacturers and UsersWorking Together

John Dunnicliff

Geotechnical News, Vol. 24 No. 2,June 2006, p 33

In the previous episode of GIN I asked:

1. What is the best way for manufac-turers to help users select the mostappropriate instruments for theirapplication? Manufacturers ofgeotechnical instruments are a re-source that should not be over-looked when selecting appropriateinstruments for any particular pro-ject. However, users can’t expectthat manufacturers will put them-selves in the same position asgeotechnical project designers. It

seems to me that if users have anyuncertainties about which instru-ments are appropriate, there needsto be interaction with manufacturersso that the decision can be made mu-tually. Is this practicable?

2. How can manufacturers help us-ers to learn about the most appro-priate installation methods? Forexample, in the case of KarkhehDam, earth pressure cells andpiezometers. I’ve always contendedthat manufacturers can provide ex-

plicit details about such installationsas strain gages on steel, becausethere are no geotechnical variables.However, manufacturers cannotprovide explicit details on how to in-stal l their instruments ingeotechnical surroundings, becausethey can’t possibly know all the de-tails that are required to do so—thisis the job of the user. But of coursemanufacturers can provide generalguidelines. In the experience ofmanufacturers and users, is there aneed to improve interaction to en-


30 Geotechnical News, September 2006

sure the most appropriate installa-tion methods? If yes, how?

I’ve received the following six re-sponses. The response by Louis Marcil

of Roctest also had an additional discus-sion of the article by Ali Mirghasemiabout unexpected vibrating wirepiezometer data at a dam in Iran

(Geotechnical News, Vol. 24 No. 1,March 2006, pp 32-36).

Gary Holzhausen

My answers to your two questions are“Yes” and “Yes”. My answer to the“how” part of the second question is, “Ifunsure, customers should be encour-

aged to ask questions of the manufac-turers”.Gary Holzhausen, President, AppliedGeomechanics Inc., 1336 Brommer St.,

Santa Cruz, CA 95062,Tel. (831) 462-2801,email:[email protected]

Louis Marcil

I thank John Dunnicliff and AliMirghasemi for raising such an interest-ing discussion about the Karkheh Daminstrumentation results. The diversityand number of possible explanationsput forward by the participants reflectthe complexity and importance of thediscussed subjects. I wish to revisit anobservation made by Elmo DiBiagio, aswell as suggest some answers to the two‘umbrella’ questions raised by JohnDunnicliff.

In his discussion, Elmo DiBiagiomade an important observation, namelythat in high embankments, it is neces-sary to use vibrating wire piezometersmounted in robust housings, to ensurethat the sensing element inside thepiezometer will not be affected by theearth pressure. This was done atKarkheh, where model PWFpiezometers were installed. This modelis also referred to as thick-walled PWS.The distinction between thin andthick-walled PWS was not made in MrMirghasemi’s article. Deformation ofthe piezometers housing is thereforeunlikely to have any effect in this case.

Regarding the first ‘umbrella’ ques-tion, my opinion is that good ways formanufacturers to help users to selectappropriate instruments include:• At very the least, advising users

when an inappropriate range, model,or type of instrument has been speci-fied

• When possible, suggesting instru-ments that are most likely to yieldgood results for the user’s specificapplication

• Ultimately, refusing to quote an or-der to avoid being associated with apoorly designed instrumentationproject.Regarding the second ‘umbrella’

question, I feel that manufacturers canplay a valuable role in helping users tolearn about appropriate installationmethods. The instruction manuals pro-vided by manufacturers constitute ahelpful tool. They must be preparedwith great attention so that they prop-erly describe general guidelines relativeto installation methods. The use of ex-plicit images (drawings, photos) shouldbecome widespread considering that, inmany cases, the installers may not speakEnglish. For important and complexprojects, manufacturers may even beasked to send employees on site to assistduring installation or to train directlythe people who will conduct theinstallation.

Manufacturers can then play a posi-tive role for both selecting the instru-ments and helping users to learn aboutinstallation methods. This comes,among others, from their capacityclosely to relate the design of the instru-ments to their performance on site.Manufacturers who realize the impor-tance of such a role will devote maxi-

mum resources in order to improve theirknowledge of applications, to selecttechnically skilled distributors, to main-tain close relationships with specialistsin instrumentation, etc.

Users who are aware that manufac-turers can help will then seek their assis-tance. However the users must keep inmind that they are ultimately responsi-ble for the decision-making on whichinstruments will be used and how theywill be installed. The manufacturers arenot substitutes for the geotechnical pro-ject designers; they can only give gen-eral guidelines. Every engineeringwork is unique and requires specificcare as to the selection and installationof instruments.

In conclusion, I feel that constant ef-forts should be made to develop goodinteractions between manufacturersand users. This will help users select in-struments and learn about installationmethods, and in return it will give valu-able feedback to the manufacturersabout the performance of their instru-ments. This interaction is in thelong-term interests of both parties.

Louis Marcil, Product Line Managerand Technical Services, Roctest Ltd,,665 Pine Street, Saint-Lambert, Qc,Canada, J4P 2P4, Tel. (450) 465-1113,email: [email protected]



Rick Monroe

It is in any manufacturer’s best interestto work closely with the geotechnicaldesigners. Their success is our success.Given the chance to participate in theselection of instruments, the manufac-turer can manage expectations of instru-ment performance and suggest alterna-tives when the instrument is marginalfor the application.

Most manufacturers provide instal-

lation guidelines in their manuals,but users need access to those manualsbefore they purchase the instru-ments. Websites provide an excellentway to distribute both manuals andother information to users. When usershave read the manual and looked at sup-plementary technical notes on the man-ufacturer’s website, they can ask moreinformed questions and bring up issues

that the manual does not address. Thishelps both the manufacturer and theuser.

Rick Monroe, Applications Specialist,Durham Geo Slope Indicator, 12123Harbour Reach Dr., Mukilteo, WA98275, Tel. (425) 493-6200, email:[email protected]

Arthur D. M. Penman

In relation to the two umbrella ques-tions about instrumentation, I wouldsay that the only persons who can de-cide on any required instrumentationare those who are designing the damand know the weaknesses they need tocover. When they know what they want,they then look at manufacturers’ cata-logues to find an instrument that will dowhat they want. The manufacturer cananswer questions about exactly whattheir instrument can do but they cannotbe expected to give instructions on in-stallation because those details dependon the conditions on site.

For dams being built of layers of fill,placed and compacted on site, it is gen-erally better to use instruments withhorizontal connections. Vertical unitscause a nuisance to the placing machin-ery and may be damaged duringnight-shift and also leave a poorly com-pacted zone around them. Vertical incli-nometers are fine to observemovements of natural slopes, but not fora dam during construction.

In general it is not practicable for aninstrument manufacturer to supplyAND install their instruments because,even if they have enough experience ofinstallation to be able to do that, theycan never be at the site at just that mo-ment when construction had reached

the stage for installation. In practice it ismuch better for engineers working onthe site, and with a high enough senior-ity to be able to make demands on thecontractor, to make the installation, orany rate, supervise installation. Themanufacturer must have supplied therequired instruments so that they are onsite when needed, and will work.

Often the instruments are supplied,particularly to dams overseas, with nospare parts, so that they cannot be re-paired on site. A case that I know aboutis with an instrument to measure settle-ment under a dam. A simple water-bal-ance unit was pulled through a tubebuilt under the dam during construc-tion, and readings being taken at spe-cific positions. The unit contained athin-walled rubber bag and no sparebags were included. The bag was foundto be perished and we had to scour theregion to find a shop selling balloons.We needed the cylindrical type balloonand after miles of travel we came acrossa shop with balloons hanging outside.They very kindly went through theirstock, picking out this type and gave usseveral as spares.

When overseas it is always too late torequest spares from the manufacturerbecause of the time for transport plusport delays. The manufacturers never

manage to provide paperwork that willsatisfy customs. So manufacturerssending their instruments overseas mustpredict what is going to suffer duringtransport and from being left in hot sunplus dust for months after arrival, andprovide secured and protected spares.Of course customs will unpack every-thing for detailed examination, and thenleave it lying loose so it is extremely dif-ficult for manufacturers to get it allright. They rely on site engineers beingaware of the problems and going to theport to receive the goods: this hardlyever seems to happen. Perhaps the man-ufacturers should have staff at the port(airport) to receive the instruments andget them to sites where they are to beinstalled. But this would presumably bemuch too expensive.

So from the point of view of the man-ufacturers, the best they can do is to pro-vide detailed catalogues than answerquestions about expected reliability orother details on the instruments, butthey cannot be expected to give instruc-tions for installation.

Arthur D. M. Penman, Chartered Engi-neer, Sladeleye, Chamberlaines,Harpenden, Herts AL5 3PW, England,Tel . +44-1582-715479, email:[email protected]

J. Barrie Sellers

Manufacturers can be helpful in meet-ing the specific requirements of particu-lar projects. The best and easiest way

for users and geotechnical project de-signers to avail themselves of manufac-turers’ expertise is to contact them di-

rectly by fax, e-mail or telephone, andto request whatever is required. All ma-jor manufacturers have websites which



offer a wealth of information on both in-strument types and installation de-tails. Most manufacturers have techni-cal service groups that can offer adviceand training both in the office and in thefield.

Manufacturing bias can be mini-mized by accessing multiple sources

and, in this way, a consensus of opin-ions can be arrived at and the best ideasincorporated. From the manufacturers’standpoint there is no need to improvemethods of interaction with users be-yond using to their fullest those meth-ods outlined above.

J. Barrie Sellers, President, GeokonInc., 48 Spencer Street, Lebanon, NH03766, Tel. (603) 448-1562, email:[email protected]

Robert Taylor

What is the best way for manufactur-ers to help users select the most ap-propriate instruments for their appli-cation?

By all means call us. Lots of peopledo. Just remember, we are primarilysuppliers, with the limitations whichthat implies.How can manufacturers help users tolearn about the most appropriate in-stallation methods?

This one is less clear. Suppliers typi-cally sit in their factories, know little ofyour soil, access, design objectives, ma-terials, contractors etc, and aren’t usu-ally in a position to research all thosecritical details for you. We can suggestinstallation alternatives, name journalliterature and books, give cautionaryanecdotes, point out theoretical andpractical aspects of the instrumentsthemselves etc—but making the final

installation recommendations is thebusiness of the person(s) who must livewith the result.

Robert Taylor, President, RST Instru-ments Ltd, 200-2050 Hartley Avenue,Coquitlam, BC, Canada, V36 6W5,Tel. (604) 540-1100,email: [email protected]

Experiences Gained from the Installationof Cable-free Sensors for Geotechnicaland Structural Monitoring

Chris Rasmussen

IntroductionFor many years, as instrumentation hasimproved, the Achilles heel of many in-stallations has been the cables from theinstruments to the readout point, whichsometimes includes a datalogger. Ca-bles by their very nature are expensive,time consuming to install and difficultto protect from site conditions. Add intothis the increasing demand for monitor-ing in sensitive areas, often inner cityand historic sites where owners andstakeholders demand minimum disrup-tion and visual intrusion and it becomesincreasingly more and more difficult toinstall the instrument needed for the job.

With the advent of the variouslow-power, license-free radio frequen-cies, approved to national and interna-tional standards in recent years, it hasbeen possible for the first time to con-figure a cable-free interface to the sen-

sors that are used in our profession. Thisgives several advantages, provided thatthe pitfalls of battery life, data security,transmitted range and transmission reli-ability can be addressed adequately.

A number of installations utilizingcable-free sensors have taken place on awide variety of sites in the UK and over-seas during the past three years. This ar-ticle will discuss the design parametersfor a cable-free system, review two in-stallations and look at some of the les-sons learned along the way.

Radio TransmittersA cable carries an analogue signal fromthe sensor to the readout or datalogger.It was not considered the most secureway forward simply to replace this ca-ble with a radio. This would provide acable-free system by replicating the an-alogue signal with, for example, a fre-

quency modulation (FM) radio, butwould still be prone to interference andother detrimental effects. A more robustway forward was to design a transmitterunit which could be fitted to new sen-sors at the time of manufacture, or retro-fitted to existing sensors. Pictured inFigure 1, this device contains a battery,excitation circuitry for the sensor, sig-nal conditioning, analogue to digitalconversion, and a radio transmitter tosend the digitized reading to a receiver.

At the selected reading frequency(for example hourly), the transmitterawakes from ‘sleep’ mode, energizesthe sensor, waits for the reading fromthe sensor to stabilize, takes a set ofreadings, averages them, digitizes theresult, and sends this along with the ra-dio/sensor ID number (unique for eachsensor) and the transmitter batteryvoltage to a receiver. The transmitter



then returns to ‘sleep’ mode until thenext reading. By using power manage-ment in this way it is possible to have abattery life in excess of ten years from asmall (4” x 1½” square) package. Alter-natively, a more integrated completesensor and radio package in a singlehousing can be used, as shown in Figure2.

Radio ReceiversData from transmitters can be received

in a number of ways, the most basic ofwhich consists of a simple receivingstat ion connected to a PC. Forstand-alone applications there is a re-ceiver/logger which stores transmittedreadings until downloaded either manu-ally or remotely. More sophisticatedstill are loggers with a built in wirelessInternet connection. These loggers usethe third generation cellular data net-work to place received data direct to theInternet. This type has proven to be themost popular and robust, with a veryfast update time where the reading istaken, transmitted to the logger and thenplaced on to the Internet in less than fiveseconds.

As often is the case, users were atfirst reluctant to take up the new tech-nology, due to a misunderstanding ofand a degree of mistrust in the robust-ness and reliability of any radio system -many citing the unreliability of theirmobile phone signal as a reason!

Fortunately for the developers, anumber of projects were faced with theprospect of costly manual monitoringwhere the layout and position of thestructures to be monitored precludedany cabling. Therefore automated oreven manually-read instrumentationwere rejected and the case for a radiosystem became overwhelming. In thefollowing examples a system was pro-

posed and adopted, using the ca-ble-free system, which in both casesresulted in significant cost savings.

Lewes, East Sussex UKFollowing initial development and test-ing, the first sizeable installation wasconducted in Lewes, East Sussex in theUK in late 2003 through to early 2004.Lewes is a predominately Victorian/Ed-wardian town with narrow hilly streets.A new storm water outfall tunnel andassociated underground pumping sta-tions was being constructed, requiringthe monitoring of many separate struc-tures along a zone approximately 1.6Km (1 mile) in length. Figure 3 showsan overview of the project with the tun-nel centerline, predicted settlementzone and the primary structures to bemonitored.

Manually-read sensors were pre-cluded, despite the relatively accessiblenature of the site. The number of sensorsand the amount of data required wouldhave required a small army of monitoringtechnicians to take readings. Sensors ca-bled to dataloggers were not possible be-cause of the large number of buildings,their historic nature and the number ofroads and passageways which wouldhave to be crossed; a large number of dataloggers would have been needed, at asubstantial cost.



Figure 1. Radio transmitter for fitting toexisting sensors.

Figure 2. Integrated senor and trans-mitter to form a complete cable-freetiltmeter.

Figure 3. The Lewes project. Predicted settlement zone in light gray, with tunnelcenterline central with the zone. Primary buildings being monitored in dark gray.

The nature of the site posed somechallenges from a cable-free perspec-tive. The desire was to cover the entiremonitored section with radio re-ceiver/loggers, which would then bedownloaded via built-in cellular mo-dems. As radio waves work best in astraight line without obstructions, acurved site (see Figure 3 above) withmany different buildings made this goaldifficult to achieve. Whilst the transmit-ters have a theoretical line-of-sightrange of up to 250m (820ft), in such acongested environment it was neces-sary to split the monitoring into threesections, each covered by a separate re-ceiver. Despite this, careful positioningof extended-gain receiving antennaswas needed in order to capture all of thesensors for each of the three sections. Anumber of lessons were learned fromthis first full-scale installation of radios.

On the technical side refraction,where the radio signals are ‘bent’ by to-pography and structures and reflection,where the signals change direction dueto the physical environment, causedboth unexpected shortening and length-ening of transmitted range. It was dis-covered that by moving the sensor orjust the transmitter by as little as a fewinches, performance could be radicallyaltered as a result of these effects. In ad-dition, different receiving antenna typesneeded careful design and placement tomaximize the distance which could bereliably achieved between the transmit-ters and the receivers, and to minimizethe number of antenna and receivers re-quired. Actual ranges between trans-mitter and receiver varied between100m (330ft) and 750m (2500ft), al-though this range was only achievedwith a specialist high gain, directionalantenna.

In terms of the suitability of the sys-tem to perform the required monitoringtask, this was a considerable success. Inthe early days of the installation, thecontractor proceeded with a great dealof manual survey as a back-up, mostprobably due to a lack of confidence inradio technology as outlined above. Asthe job progressed, it became apparentthat high quality reliable data wereavailable from the cable-free sensorsand therefore the survey work was sub-

stantially scaled down—much to thecontractor’s relief due to the large costsaving.

Aesthetically, the system was a nota-ble success with members of the publicand owners whose buildings were beingmonitored; there was some relief thatholes would not be drilled and unsightlycables run. Instead, a small packagewhich could be painted to match thebuilding was installed. Additionally, thelack of disruption caused by daily sur-veys, particularly where these requiredaccess into buildings (as the majority ofthem did), was welcomed. Such was thepopularity of the sensors with membersof the public that unusually the contrac-tor found himself with many requests toinstall instrumentation rather than re-move it—‘why are you monitoring nextdoor and not me?’!

Perth, Western AustraliaAnother successful project was inPerth, Western Australia. The NewMetroRail Project is a A$1.5 billion(US$1.1 billion) rail line being built toextend services to outer metropolitanareas. It will run from Perth toMandurah, augmenting the existingrailway with more than 80km (50 miles)of new track. It is the largest public in-frastructure project ever undertaken inthe Perth Metropolitan area.

In one of the more challenging as-pects of the project, twin tunnels havebeen bored under the Central BusinessDistrict, opening up opportunities to re-vitalise shopping and business areas in asection of the city, and providing a door-stop service to Perth’s new Conventionand Exhibition Centre.

Leighton Kumagai Joint Venture(LKJV) is the contractor responsible forthe inner city component of works. Thisincludes the construction of:• 770 metres (2500ft) of twin bored

underground rail tunnel• 600 metres (2000ft) of cut and cover

tunnel (3 sections)• New William Street underground

platforms• A below-ground station at The Es-

planadeA comprehensive monitoring pro-

gram was implemented to warn of anypotential damage to buildings and struc-tures along the alignment. In particular,intensive instrumentation and monitor-ing was applied to the buildings underwhich the tunnel boring machinepassed before entering the WilliamStreet station box, which can be seen inFigure 4. Compensation grouting wasused to control any potential settlementof the buildings in the zone of influenceof the tunnel.



Figure 4. Buildings being monitored around the William Street Station Box in Perth,Western Australia.

Instrumentation for the buildings inthis area performed two vital functions:(a) Assess any differential settlement ofthe structures; (b) Control the compen-sation grouting system.

The original monitoring designcalled for manual surveying and a “wa-ter level sensor system” to be used incombination with robotic total stations.On further review, and considering thecomplexity of various structures, it wasrealized that the installation of such asystem was impracticable, both on thebasis of technical logistics for installa-tion in working businesses and cost. Analternative proposal was put forward touse a combination of robotic total sta-tions and electrolevel (EL) beams in“strings”, running from outside the areaof the expected settlement through thelength of each structure. This wouldprovide a “network” of monitoringpoints.

The next two challenges for theLKJV Geotechnical EngineeringGroup were: (a) How to manage theproposed 2.5 kilometers (8200ft) of ca-

ble in such tight constraints without up-setting the property owners andoccupiers?; (b) How to retrieve the datafrom five dataloggers to their networkin near real time?

An innovative design was deter-mined that used cable-free EL Beams.This system had the advantage of re-quiring no cables in the buildings, andallowed the data to be transmitted to asingle receiver in the site office.

Each of the EL beams was fitted witha specially designed radio transmitter,operating on a low transmission fre-quency (434 MHz), as required by Aus-tralian regulations. Data transmissionproved more of a challenge than eventhe Lewes project, with many of thesensor strings being located in base-ments or behind partition walls or ceil-ings. Unlike Lewes, directional andhigh-gain antennas were not the solu-tion; instead it was necessary to install‘repeater’ units. These, as their namesuggests, are tuned to the frequency atwhich the radio transmits, and they thensimply ‘repeat’ any signal they receive

at a higher power than the incoming sig-nal. By using a series of repeaters, lo-cated outside the buildings beingmonitored (in the site compounds), itwas possible to use a single receiverconnected direct to the monitoring PCto receive data from all the sensors, eventhose many hundreds of meters away inbasements. An example of the nature ofthe site can be seen in Figure 5, wherethe basements of the properties shownrequired monitoring, and Figure 6where a line of EL Beams is installed inone of those basements. To receive all ofthese signals from a single antennawithout some form of repeater systemwould have been extremely difficult toimpossible.

The lessons learnt at Lewes were ap-plied to the installation in Perth, buthere other unique factors came intoplay. The first of these was a require-ment for 2 minute transmission inter-vals rather than the normal 1 hour. Thisposed problems with both battery lifeand ensuring that the thousands oftransmissions each hour were all cor-



Figure 5. Radio EL beams installed above a false ceiling. Figure 6. Excavation in Perth and buildings being monitored.

rectly received and logged.With careful manipulation of the

power management, battery life was es-timated, with no change of componentsto be two years. To ensure all data arerecoded, the receiver was configured torecord at a higher speed than normal,and as each transmission has a uniquedigital ID, even if two or more transmis-sions arrive at the receiver at the sametime, they were both recorded. Batterylevel is transmitted with each readingand to date, after 18 months is still at75% of capacity for the transmitters.

Secondly, in the summer monthsPerth has very high daytime tempera-tures While this was not a concern forthe transmitters, most of them being in-side buildings, several repeater units in-stalled in full sunlight did shut downafter their internal temperature reached80ºC (176ºF). It was necessary to pro-vide shade for these units to prevent areoccurrence.

As with Lewes, the system found ahigh degree of acceptance from thirdparties, particularly shop owners,where the speed of installation and theability to hide the sensors above false

ceilings and within partitions was wel-comed. There were some problems instock-rooms where shop staff wouldplace boxes against sensors or hangthings from them, leading to some falsealarms, but these were easily resolvedwith a quiet word in the right ear.

ConclusionsSince the introduction of cable-freesensors the technology has found wideacceptance in the UK and many othercountries. With further developmentwork the system will become more so-phisticated and reliable and the cost willinevitably fall. The case for continuingto use cables, currently often favouredon a cost basis, will become evenweaker once this stage is reached.

With care in installation, cable-freesystems do appear to work to a levelwhich both exceeds our design expecta-tions and overcomes our customers’ ini-tial fears in real-world site conditions.There are no serious detrimental effectsdue to poor transmission range, or inter-ference to transmitted signals, evenwhen installed in basements, tunnels orelectrically noisy environments. When

used in sensitive areas, such as railways(where stringent emf standards apply),it is important to have the system certi-fied to the relevant standards. Withoutsuch certification many owners will notallow the systems to be used.

New sensor types are becomingavailable, such as solid state accelerom-eters and pressure sensors. These re-quire very low power and have a farlower cost than traditional force balanceaccelerometers or vibrating wire sen-sors. By combining these new sensorswith radio transmitters, the opportunityexists to advance the use of instrumen-tation beyond that currently experi-enced on many projects . Whencombined with the proliferation ofweb-based data collection and presen-tation, the use of instrumentation willcontinue to grow, particularly in con-gested urban environments.

Chris Rasmussen, Soil Instruments Ltd,Bell Lane, Uckfield, East Sussex,TN22 1QL, UK, Tel. +44 1825 765044,email: [email protected]

Interpreting Unexpected Instrument Data

Verne C. McGuffey

IntroductionThe sharing of detailed information oninstrumentation difficulties in the arti-cle by Mirghasemi (GeotechnicalNews, March 2006, pp 32-36) and thecomments of the seven specialists in thefield (Geotechnical News, June 2006,pp 34-43) led me to believe that many ofus casual users might have similar in-formation that would be useful to theprofession. Perhaps this note will inviteother users with useful experiences toshare them also.

All users sometimes encounter un-expected data from instruments. Thefield staff on many of my projects wereinstantly into the instrument manualtrying to fix the *#* machine. Their ef-forts often resulted in damaged or de-

stroyed instruments and/or recordingequipment, or improper constructiondecisions. My experiences lead me tobelieve that the instruments are seldomreally giving “wrong” data.

I have put together a few ideas that Ifound helpful in understanding strangedata, and to guide methods of interpre-tation and therefore to guide decisionmaking. I have also included a few un-usual situations where the instrumentwas assumed to be at fault, but was not.

Evaluation ideas• Get good baseline data before you

are at a critical stage of decisionmaking. Take readings hourly, daily,or by the minute, not just at pre-scribed periods. Find out how the in-

strument responds. What are dailyfluctuations due to variations in tem-perature, rain, batteries, differentreaders, etc.? How does it respond toadjacent activities?

• Take very frequent readings at thefirst construction activity that wouldinfluence the instrument. Evaluatethe data for trends, delays, quantifi-cation of first readings, etc. Use thedata to plan the appropriate readingand evaluation schedules. Plan whenthe frequent readings need to be re-peated to reevaluate responses.

• Evaluate the quality of the data atlandmark happenings. For example,start or stop of construction, rain,freezing, atmospheric pressurechanges, hot or dry spells, adjacent



activities.)• Make sure that support data are ade-

quately gathered and included withthe instrument data. For example,time of day; quantification of con-struction in the immediate and adja-cent areas; temperature/atmosphericpressure information; unusual hap-penings like stockpiling, change ofequipment, loss of electric power,etc.

• Look for trends. There are manytrends for different instruments andsituations, e.g. temperature, tidalfluctuations, battery performance,system response time, plus site re-sponse trends. As examples, onepiezometer may be nearer a perme-able layer or zone than another, andconstruction processes or geometrymay encourage arching or stressconcentrations.

Examples of Situations whereInstruments were First Blamedfor Strange Readings• On one project, the electrical instru-

mentation responded to airplanetraffic, apparently the radio contactwith the airport. One was influencedby overhead high tension wires.

• Standpipe piezometers on projectswere not believed to be respondingproperly. I ran falling head and risinghead tests on them to evaluate re-sponse time, leaks, permeability, andget a reasonably accurate value at thetime of readings. Some said a risinghead couldn’t be done, but I havedone them to 100 feet by using airpressure in a smaller diameter tubeto pump the water out of the

standpipe. I put a ¼-inch tube in theexhaust pipe of my old car with somerags, and pumped out 40 feet of wa-ter from a ½-inch poly tubestandpipe by submerging it inches ata time.

• I found that the response time forpiezometer readings to reflect realpore pressures was not consideredon some projects. Sometimes thepore pressure dissipation is similarto the response time curve, causingthe user to believe there has been noresponse from the piezometer. Putt-ing a rod or closed tube into astandpipe provides a quicker re-sponse time to get better informa-tion. You may need special readingequipment if the piezometric level isnot above ground surface. Believingthe instrument to have failed leadsthe user to not consider importantdata when perhaps something majoris happening.

• Natural underground drainage pathsare sometimes not consistent, or asexpected. Data from one project wasfollowing a normal top and bottomdrainage model under first loading.Then it changed to single upwardmodel. Further study showed thegravel layer below the clay was in arock basin and sealed so that after thewhole layer was pressurized, it nolonger would drain downward. An-other project was expected to per-form like a top, bottom and both sidedrainage model. Piezometer datashowed that it would not drain side-ways toward an area that had beenheavily loaded by an old railroadembankment. The soil under the old

embankment was later shown tohave a lower lateral permeability.

• Settlement, as measured with settle-ment platforms, stopped on manyprojects during hard freeze monthsand resumed the original trends inthe spring. It was initially blamed onthe instruments (pipe being frozeninto surface soil), but later platformsfitted with manometers verified thesame trends.

• I visited the site of an embankmentfailure on Sunday. The project staffhad looked at the prior week’spiezometer data and concluded theobserved erratic data were wrong(instrument failure) and that therewas no need to take any action.Therefore they sent me the data inFriday’s mail. The embankmentdropped 8 feet on Saturday and I fi-nally got a call. The previous Tues-day’s data showed unusual changesin pore pressure, indicating that afailure was imminent. The pore pres-sure had gone up faster than the fillwas placed, and as displacements be-came larger (inches to a foot) thepore pressure then dropped down be-low the zero readings and was fol-lowed by an 8 foot embankmentdrop.We often create opportunities for

poor response by using poor installationdetails and techniques. Read and under-stand the red book.

Verne C. McGuffey, NYS DOT retired,22 Lombard St., Schenectady, NY12304, email: [email protected]



do not print key lines

frame is 42 x 13,7

Geotechnical Instrumentation News · Perth, Western Australia. The New MetroRail Project is a A$1.5...

Documents

Transcript of Geotechnical Instrumentation News · Perth, Western Australia. The New MetroRail Project is a A$1.5...