Your SlideShare is downloading. ×
  • Like
Earth resistivity logger (john becker)
Upcoming SlideShare
Loading in...5

Thanks for flagging this SlideShare!

Oops! An error has occurred.


Now you can save presentations on your phone or tablet

Available for both IPhone and Android

Text the download link to your phone

Standard text messaging rates apply

Earth resistivity logger (john becker)



Published in Technology
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
  • Hi
    i build this detector but my software is magna radar and when i transfer data the program is error .please do you have the earth resistivity program?
    my email :
    Are you sure you want to
    Your message goes here
No Downloads


Total Views
On SlideShare
From Embeds
Number of Embeds



Embeds 0

No embeds

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

    No notes for slide


  • 1. Constructional ProjectEARTHRESISTIVITYLOGGERJOHN BECKER Part OneHelp your local archaeological society to PROBESlocate and reveal the hidden mysteries of GROUND LEVELour ancestors. SECTION ANUARY and February 1997 saw the connected across them, current will flowJ THROUGH publication in EPE of Robert Beck’s between them, just as it does through an SOIL Earth Resistivity Meter, an electronic ordinary resistor.tool to assist amateur archaeological soci- The amount of current that flowseties “see beneath the soil” in their search depends on how much resistance the soilfor ruins and other hidden features. interposes between the two electrodes. The The design presented here is based upon value depends on several factors, the soil’sthe same concept as used in Robert’s cir- water content and chemical make-up (i.e.cuit, but it has been considerably simpli- the impurities the water contains), and thefied in terms of the components count and presence (or absence) of non-conductivetheir ready-availability. Significantly, it objects. The relationship is complex, and PLAN VIEWhas also been put under the command of a will not be discussed in detail here,PIC microcontroller and provided with although some experiments which shoulddata logging facilities. The principal fea- give an insight into it are suggested in the Fig.1. Current paths set up by probetures of this design are outlined in Table 1. text file supplied with the software. It is array. discussed more fully by Anthony Clark inDOWN TO EARTH his book. main field, as you will see presently from Before going any further, though, the The current flow through soil is also wishes to “put his cards on the sur- complicated by the fact that it is not flow- The overall current flow between thevey grid”. He is not an archaeologist and ing in a straight line, as it does (in effect) probes is thus not just governed by thehas approached this design purely as an through an ordinary resistor. The current resistance of one direct horizontal path, butelectronic problem to be solved – transmit can simultaneously flow through a multi- by the total resistance of innumerablea signal, retrieve it at a distance and store tude of paths, not only horizontally, but paths effectively in parallel within a givenit for later analysis. three-dimensionally, as illustrated in volume of soil, and each experiencing dif- Along the path to this end, he has Fig.1. It also radiates outwards beyond the ferent values of resistance. Despite theresearched a fair bit, chatted with a local complexity, though, as far as the readingarchaeological society and with EPE read- on a current meter is concerned, theers who have knowledge in this field. Most answer is a single value, and from it animportantly, Nick Tile, EPE reader and assessment of the soil’s relativefriend of the author, has spent several density can be made.months successfully using the prototype foractive archaeological survey work. More onthis in Part 2. Further reference to Nick’ssurveying will be made during this article. A list of useful references is quoted atthe end of Part 2, to which readers arereferred for more information on survey-ing techniques. The main reference sourceused by the author has been AnthonyClark’s Seeing Beneath the Soil.BASIC PRINCIPLES For the sake of readers who have not yetbeen enticed into joining their localarchaeological society in search of knowl-edge about our ancestors and how theylived, it is appropriate to outline how elec-tronics can help us see subterranean fea-tures without ever touching a spade or Prototype Earth Resistivitytrowel. Logger, housed in a plastic When two conductors are placed in case with transparent lid.moist soil with a d.c. voltage source288 Everyday Practical Electronics, April 2003
  • 2. What is being looked for in an electron-ic survey is reliably monitored variations in TABLE 1. WHAT IT DOESreadings across a site, the pattern of which The PIC microcontroller performs the following functions:indicates where different sub-soil features *Generates 137Hz square wave ground-penetrating transmission signalexist. *Converts the received and amplified analogue signal to a 10-bit digital value *Stores each converted value to user-specified non-volatile (EEPROM) memoryUNIFORMITY PROBLEM address representing specific site plotting coordinates A problem arises, however, in that not *Continually displays immediate real-time data and coordinates on alphanumericonly does the soil have resistance, but it liquid crystal display (l.c.d.)also has capacitance and additionally *On request, outputs stored data via serial link to Windows 95/98/ME PC for storageexhibits various electrolysis effects as the to disk and subsequent analysisd.c. current continues to flow, and mostsignificantly, a polarization process takes Other features of the logger include:place, resulting in progressively changing *Switchable output resistance to vary transmission currentvalues on the meter. *Switchable amplifier gain, x1, x10, x100 To be able to take meaningful readings it *Pushswitch selection of survey site row and column coordinates allocation in memoryis necessary to counteract the polarization *Memory capacity for 16384 10-bit samples, representing a survey site grid ofeffect. This can be done by passing an 128 x 128 squaresalternating current through the soil instead *Data storage action under complete user controlof a direct one. With each of the a.c. cur- *Data locations may be overwritten with fresh data if requiredrent’s phases, the polarizing effects of the *Sampled data stays in memory indefinitely, even after power switch-offpreceding phase are reversed, thus causing *Recall of last used survey coordinate when next switched on, allowing survey to bea more consistent current flow to occur in spread over several days or weeksboth directions. *Individually stepped push-button recall and display of recorded samples and their Whilst the soil’s electrolysis process will coordinatesnot be reversed, its effect is likely to be so *Total clearance of memory to zero value upon request, with security feature to helpminute in relation to the polarization effect, prevent erroneous usethat it can be ignored during the relatively *Operable from any d.c. supply between about 9V and 15V, consuming about 25mA.brief time during which current flow read- It is equally suited for use with a 9V PP9-size battery (rechargeable types areings are taken. available), or a 12V car battery (see later) The capacitance effects are also largelyovercome by using an alternating current at Software features for the downloaded memory samples include:a suitable frequency. *Program written in Visual Basic 6 (VB6) *Disk storage under unique dated and timed file namePROBING FREQUENCY *Graphical display of data on PC screen as waveform graphs and value-related The question then arises: at what fre- coloured or grey-scale grid squaresquency should the current direction be *Four screen slider controls allow data to be processed for best visual contrastrepeatedly reversed? Too high a frequency to aid analysiswill cause the soil’s capacitance effects to *Facility to invert data values for viewing as “valleys” or “peaks”“mop-up” and attenuate the alternating sig- *Main screen display as 20 x 20 samples block, with vertical and horizontal panningnal’s amplitude. Too low a frequency will across full 128 x 128 gridagain cause variation in the monitored *Secondary screen displays of separate grid or graph data for full 128 x 128 samplesreadings, albeit smaller than would occur blockthrough using a d.c. signal. *Zoom facility for closer examination of separate graph and grid data It appears that the optimum rate at which *Reloading of previous survey files via dedicated file selection screenthe signal phases must be changed has *Downloaded files stored in format suited for analysis and graphical display viabeen established at around 137Hz Microsoft Excel (found on most PCs)(Anthony Clark quotes 137·5Hz but also *Data may be downloaded to PC as often as required without disrupting its existingsays that 67Hz is used in some equipment). on-board storage (allowing on-going visual display of site progress across longThese frequencies assist in not only the periods)elimination of the polarizing effects, but *Suited to survey monitoring using any of the standard probing techniques (Wenner,also in reducing the affect of other alternat- Schlumberger, Twin-Probe, etc).ing electrical fields which might be presentin the site being surveyed, such as a 50Hzmains frequency, for instance. EPE contributor Aubrey Scoon has researched into this latter aspect and has reported the presence of many other fre- quencies in some locations he has exam- ined, some emanating from a nearby “supercomputer” in one instance. The frequencies of 67Hz and 137Hz (the latter is used in this Logger), are not a multi- ple of 50Hz, nor of the 60Hz mains cycle used in some countries, such as the USA. Thus, by performing rectification or sampling that is synchronised with the trans- mission signal, the effects of these extrane- ous fields can be reduced. They are also min- imised by the use of a differential amplifier, which will be discussed presently. It is worth pointing out, however, that in the suburban garden where the author’s tri- als with this Logger were performed in conjunction with an oscilloscope, residual 50Hz mains currents were not evident. MULTIPLE PROBES The discussion so far has been in rela-Typical example of one of the three analysis screens used by the Earth Resistivity tion to the current flowing between twoLogger’s PC software. The other two show full-screen displays of grid or graph data probes in series with a meter. Over thefor a 128 x 128 samples survey site, with zoom facilities. many years that geophysicists have beenEveryday Practical Electronics, April 2003 289
  • 3. electrically probing the soil in theirsearch for minerals and oil deposits IN IC1 OUT +5V 78L05(since 1946 says Robert Beck), it has R6 COMbeen found that there are better probing D1 k 10k * OUTPUT C2 C3techniques than just using two probes. 1N4001 a 100n 100n R1 R5 RESISTANCE 100kSome of these have been adopted by C1 + 1karchaeologists. 22µ R4 S2 Most of the favoured ones all use four ON/OFF 100Ω TO SK2probes – two for transmission (TX), and S1 (C1, WHITE) (FREQUENCY) R3two for reception (RX). The righthand sec- 8 7 10Ω C4tion of Fig.2 shows one way in which the 22µ +VE 2 +second pair of probes can be used. Anthony 2 C+ OSC 7 N.C. IC3 6 6 TL071Clark says that there are also some tech- 4 C IC2 LV N.C. 3 + 1 7660 5niques that use five probes – with push-pull *B1 N.C. N.C. OUT 4TX across two and the fifth becoming a 9V GND TO RA2grounded reference perhaps? 3 C5TWIN PROBES 22µ + SK1 *SEE TEXT R2 (C2, BLACK) There are several ways in which four 100kprobes are used in relation to each other, 0Vand each with its own merits. Their use is 5Voutlined later, but no quality judgement isoffered here on their appropriateness to Fig.3. Power supply and transmission interface circuit for the Earth Resistivity Logger. mA a wire attached will the other TX probe is connected to the 0V CURRENT V MEASURED do. The probes don’t power line. IC3 is configured as a compara- SOURCE POTENTIAL even need to be tor whose inverting input (pin 2) is tied to inserted very far, just the potential divider chain formed by equal- enough to penetrate value resistors R1 and R2. The resistors are the soil to make connected across the +5V and 0V lines and electrical contact the voltage at their junction is thus 2·5V. with its moistness. The non-inverting input (pin 3) of IC3 is It will be obvious, connected to one of the PIC microcon- of course, that dry troller’s output pins (RA2) and is fed with LINES OF EQUAL POTENTIAL soil will be less a 137Hz square wave, generated by the capable of passing a software, and which alternates between CURRENT FLOW LINES current than moist +5V and 0V. As this square wave repeated- soil. Keep in mind ly crosses above and below the 2·5V refer- A) B) that the surface of ence voltage, IC3’s comparator action the soil can dry out takes place and its output (pin 6) alternatesFig.2. How current flowing between two probes is detected by faster than that between the device’s upper and lower volt-a second pair. below it, and so a age limits, i.e. swinging between about reasonable amount +4V and –4V.various survey situations – but it is worth of penetration should be allowed. Robert Note that the op.amp to which the TXnoting that Clark considers the Twin-Probe Beck allows 200mm with his probe struc- probes are connected (IC3) is short-circuittechnique to be the most favoured for tures discussed in Part 2. protected internally and is unlikely to suf-archaeological surveying, although the With some sites it may be necessary to fer if the probes accidentally come intoWenner technique is said to provide more evenly damp the soil with water before contact with each other while the power isdetailed results. Nick in his extensive use adequate probing can begin. switched on. However, do not sustain suchof the prototype adopted the Twin-Probe contact since it could cause regulator IC1technique. POWER SUPPLY to get hot, and it will shorten the battery The Twin-Probe and Wenner techniques The PIC-controlled processing circuit is charge life.were outlined in Robert Beck’s article and almost irrelevant to the main aspects of soilwere used in the author’s garden tests with monitoring! So first let’s look at the power OUTPUT RESISTANCEthis Logger. They will be discussed in Part supply requirements, and the simple trans- Depending on the probing technique2 in a bit more detail. Suffice to say for the mission circuit, both illustrated in Fig.3. used, experienced geophysicists can deter-moment, both involve placing in the soil a As said in Table 1, the power can origi- mine not only the subterranean density, butreference probe that is connected to the cir- nate from any d.c. source (e.g. battery) also its possible composition. This iscuit’s 0V line (common ground). This is ranging between about 9V and 15V. This is apparently achieved by pre-setting the cur-regarded as one half of the TX probes pair. input via diode D1 to the +5V voltage reg- rent which flows between the two TX To the other TX probe is fed the alter- ulator IC1. The diode prevents distress to probes.nating voltage or current, evenly swinging the circuit in the event of the battery being Robert discussed this in the ’97 text,as a square wave above and below the 0V connected with the wrong polarity. referring to the technique as providing areference value. The function of the TX The regulated +5V output from IC1 “constant current”. It would appear,probes is to set up a field of potential gra- powers the main PIC-controlled circuit, though, that his circuit did not provide adient in the soil, which is then sampled by which must not receive a supply signifi- constant current in the literal sense – samethe RX probes. cantly greater than +5V. It also provides current flowing irrespective of resistive The RX probes are positioned at dis- the positive power to the TX and RX cir- conditions – but rather it provided a currenttances away from the TX probes as dictat- cuits. Both of these circuits additionally limit. It is the same limiting approach thated by the probing technique being used. need an equivalent negative supply. This is has been taken in this Logger design.They are connected to the twin inputs of a generated from the +5V line by the voltage The output from IC3 can be switched bydifferential amplifier, whose output signal inverting chip IC2, which outputs a voltage S2 to the active TX probe via one of fiveamplitude is determined by the difference of close to –5V. paths. These comprise a direct unlimitedin the two input levels. It is this signal path, and four limiting paths via resistorswhich is then monitored by the control TRANSMISSION R3 to R6, in order of 109, 1009, 1k9 andcircuit. OUTPUT 10k9. It is not even necessary to use special Op.amp IC3 is the device which feeds the Readers are referred to the publicationsprobes, any metal object that does not cor- 137Hz alternating signal to one TX probe listed in Part 2 for information on resis-rode and can be inserted into the soil with (the “active” TX probe). As previously said, tive path use. The field tests performed by290 Everyday Practical Electronics, April 2003
  • 4. +5V R21 R7 1M 1k 4 k 5 D4 + GAIN 1N4148 TO SK3 IC4a 7 a R20 (P1, YELLOW) TL074 S3 6 100k R9 R12 R13 R22 100k 100k 100k R19 100k 10k C6 R18 R10 2 22µ 10k 100k R11 R14 C7 + 100k 100k IC4c 1 13 470n 3 TL074 14 + IC4d 12 TL074 + VOUT R16 R17 R15 TO RA3 9 10k 10k 100k R23 100k R8 1k IC4b 8 TO RA0 10 TL074 + TO RA1 k k k D5 TO SK4 11 D2 D3 1N4148 (P2, GREEN) a 1N4148 1N4148 a a 0V 0V 5VFig.4. Differential amplifier that receives, amplifies and conditions the RX probes signal prior to sending to the ADC input of thePIC microcontroller.the author and Nick Tile were carried out C6 to the amplifying stage around IC4d. R17 plus diodes D2 and D3. These are notvia the direct TX path (Nick says he has Here the gain can be switched by S3 part of the required analogue processingnot found the switchable resistance facil- between ×1, ×10 and ×100. In the proto- circuit but were included for use duringity to be useful). In this role, the signal type’s garden tests, the ×1 gain was software development. Their function willamplitude across the TX probes is picked satisfactory across the maximum probe be described presently.up by the RX probes simply as an alter- separation distance that the dense gardennating signal whose amplitude varies flower beds would allow (11 metres)! Nick CONTROLLER CIRCUITaccording to the soil density it has to pass says he prefers the ×10 setting. The PIC-controlled processing circuit isthrough. At this stage the signal is swinging shown in Fig.5. At its heart is a PIC16F876 above and below 0V. It has to be shifted so microcontroller, IC5, manufactured byRECEIVING CIRCUIT that it only swings between 0V and +5V at Microchip. It is run at 3·6864MHz, as set The receiving circuit is shown in Fig.4. the maximum extremes, to suit the PIC by crystal X1. The frequency may seemThe twin RX probes and their received d.c. microcontroller’s limits. This is achieved unusual, but crystals tuned to it are stan-coupled signals are connected via buffering by a.c. coupling the signal via capacitor C7 dard products. Its choice provides greaterresistors R7 and R8 to the respective inputs to the level-shifting potential divider accuracy of the baud rate at which theof the differential amplifier, formed initial- formed by resistors R22 and R23. Diodes logged data is output to the around op.amps IC4a and IC4b and hav- D4 and D5 limit the maximum voltage The software-generated 137Hz squareing a gain of three. The outputs from these swing then fed to the PIC, preventing it wave pulse train is output via pin RA2, andop.amps are summed, still as d.c. signals, from swinging above or below the PIC’s fed to the TX op.amp IC3 in op.amp IC4c, which provides unity gain. limits of acceptance. Pin RA3 is the pin to which the level- The resulting signal represents the It will be seen that two additional signal shifted signal output from IC4d is input.difference between the two input signal paths are provided from the output of The pin is configured by the software as anlevels. It is now a.c. coupled via capacitor IC4a/b and consist of resistors R16 and analogue-to-digital converter (ADC). TEST SAVE UP DOWN MODE DOWNLOAD 2 +5V 7 +VE N.C. D0 S9 S8 8 S4 S5 S6 S7 N.C. D1 20 TB1 9 N.C. D2 +VE +VE 10 N.C. D3 2 21 D4 11 TO R16 3 RA0/AN0 INT/RB0 22 D5 12 D4 X2 T0 R17 RA1/AN1 RB1 D5 L.C.D. 4 23 D6 13 MODULE F OUT RA2/AN2/VREF- RB2 D6 5 24 D7 14 TO D4/D5 RA3/AN3/VREF+ PGM/RB3 D7 6 25 RS 4 RA4/TOCK1 RB4 RS 7 26 E 6 RA5/AN4/SS RB5 E 27 0V 5 3 C8 IC5 PGCLK/RB6 28 0V R/W GND CX 10p PIC16F876 RS232 9 PGDA/RB7 OSC1/CLKIN TO IC7 PIN 11 CX 1 a 11 X1 D6 T1OSO/T1CKI/RC0 R31 3.6864MHz 1N4148 12 T1OSI/CCP2/RC1 10k C9 k 13 10p CCP1/RC2 10 14 OSC2/CLKOUT SCK/SCL/RC3 15 SDI/SDA/RC4 R25 16 8 SDO/RC5 CONTRAST 1k 17 TX/CK/RC6 +V 1 7 R26 1 18 R29 R24 N.C. A0 WP MCLR RX/DT/RC7 N.C. 10k GND GND 10k 10k VR1 N.C. 2 A1 IC6 SCL 6 10k 3 24LC256 5 R27 8 19 R28 R30 N.C. A2 SDA 10k 10k 10k GND 0V 4 TB2 *PROGRAMMER 0V VPP DATA CLK Fig.5. PIC-controlled processing, display and data storage circuit.Everyday Practical Electronics, April 2003 291
  • 5. The PIC repeatedly converts the input COMPONENTS Approx. Costsignal to a 10-bit binary value which it out-puts for display on the 2-line × 16-charac- Guidance Only £45ter l.c.d. X2, as a decimal number. As usual excl. batts & casewith the author’s designs, the l.c.d. is con- IC6 24LC256 256 kilobittrolled in 4-bit mode (and its pinouts on the Resistors See serial EEPROM R1, R2, R9printed circuit board are in his standard to R15, R20, SHOP IC7 MAX232 RS-232order). Its screen contrast is adjustable by R22, R23 100k (12 off) interface driverpreset VR1. R3 10W TALK R4 100W page Miscellaneous Pressing switch S8 causes the PIC to S1, S9 s.p.s.t. min. toggle switch R5, R7, R8,store (Save) the ADC’s 10-bit binary out- R25 1k (4 off) (2 off)put value to the 32 kilobyte (32768 bytes) R6, R16 to S2 2-pole 6-way rotaryserial EEPROM chip, IC6, at the address R19, R24, switchset by the user via switches S4 to S6. This R26 to R31 10k (12 off) S3 4-pole 3-way rotary R21 1M switchchip is another Microchip device, and was S4 to S8 min. push-to-make All 0·25W 5% carbon film or betterfirst demonstrated by the author in his switch (5 off)PIC16F87x Data Logger of Aug/Sep ’99. Potentiometer SK1 to SK4 4mm single-socket,Its device number, 24LC256, indicates that VR1 10k min. preset, round 1 each black, white,it has 256K single-bit memory locations. yellow, green (seeThese are accessed as 8-bit bytes. Capacitors text) C1, C4 to SK5 9-pin D-type serial In other applications, the 24LC256 is C6 22m radial elect. 25V (4 off) connector, female,capable of being multiplexed with seven C2, C3 100n ceramic, 5mm chassis mountingothers of its type, using its A0 to A2 inputs pitch (2 off) TB1, TB2 pin-header strips to suit, orto set each device’s multiplexed address. In C7 470n ceramic, 5mm pitch 1mm terminal pins (2 off)this application they are left unconnected, C8, C9 10p ceramic, 5mm pitch X1 3·2768MHz crystal (2 off) X2 2-line, 16-characterleaving them biased internally. Resistor (per line) alpha- C10, C11 1m radial elect. 16V (2 off)R31 is essential to the correct reading of C12 to C14 10m radial elect 16V (3 off) numeric l.c.d. modulethe device’s retrieved data output value. The 24LC256 data sheet can be down- Semiconductors Printed circuit board, available from theloaded from Microchip’s web site D1 1N4001 rectifier diode EPE PCB Service, code 388; plastic case D2 to D6 1N4148 signal diode with see-through lid, 190mm x 110mm x( 90mm (see text); 8-pin d.i.l. socket (3 off); (5 off) Data stored in the 24LC256 can be IC1 78L05 +5V 100mA 14-pin d.i.l. socket; 28-pin d.i.l. socket;retrieved and downloaded serially to a PC voltage regulator knobs (2 off); 4mm plugs, colours to matchvia the RS-232 interface device (IC7) and IC2 ICL7660 voltage inverter 4mm sockets (4 off); heavy-duty crocodilesocket SK5, in Fig.6. Transfer is initiated IC3 TL071 f.e.t. op.amp clips, with coloured covers to match 4mmby pressing switch S7. Once started, all IC4 TL074 quad f.e.t. op.amp sockets (4 off); robust cable for probes IC5 PIC16F876 (see text); 9V PP3 battery and clip (see32K bytes are sent to the PC in consecutive microcontroller, text); p.c.b. supports (4 off); nuts and boltsaddress order. preprogrammed (see to suit l.c.d. mounting style (4 off each); text) internal connecting wire; solder, etc.DATA SAMPLING The software controls the output of atrain of square wave pulses at the 137Hzrate. Data sampling takes place on each TEST VALUE DISPLAY Software, including source code files, for the PIC unit and PC interface is avail- Resistors R16 and R17, mentioned pre-phase of the output pulse (high and low). viously, allow the PIC to monitor the volt- able on 3·5-inch disk from the EditorialOn each complete cycle, the minimum age on the outputs of IC4a/IC4b for test office (a small handling charge applies –value received is subtracted from the max- purposes, via its ADC inputs RA0/RA1. see EPE PCB Service page) or it can beimum (to establish the received signal’s Diodes D2 and D3 prevent the PIC from downloaded free from the EPE FTP site.amplitude) and the result stored to a 32- receiving damaging negative voltages. The latter is accessible via the top of thebyte temporary memory block. So that Originally, these outputs were intended home page of the main EPE web site atmaximum peak-to-peak values of the purely for development use. However, their Click onreceived square wave have stabilised, the use has also proved beneficial in the out- “FTP Site (downloads)”, then in turn onsynchronous sampling takes place at the door monitoring environment and has been PUB and PICS, in which page the files areend of each peak. retained. The monitored values are dis- in the folder named EarthRes. About once a second, the pulse train played in decimal on the l.c.d. and provide This month’s ShopTalk page providesstops while the 32 sample values are aver- indication of relative probe signal information about obtaining pre-pro-aged, and the l.c.d. display updated. The strengths, and of the loss of connection to grammed PICs.pulse train then recommences for another one or more probes. The PIC program (ASM) was written insecond. This gives the soil time to respond In relation to this test-motivated option, TASM, although the run-time assembly isto the re-application of the a.c. waveform, a second signal strength display option has supplied as an MPASM HEX file, which hasand for the effects of any d.c. currents to be been included via the software. The second configuration values embedded in it (crystalover-ridden. mode displays the XT, WDT off, POR on, all other values off). upper and lower Regarding the PC interface, if you have peak values of the Visual Basic 6 already installed on your +5V 16 signal applied to the machine, you only need to use files +VE C12 SERIAL PIC’s RA3 input. EarthRes.exe and INPOUT.DLL. Copy 10µ OUTPUT 1 2 + The two modes are them into a new folder named C:EARTH + C1+ V+ + SK5 RES, or any other of your choosing on C10 C14 SERIAL selected by toggle 1µ 3 10µ OUTPUT switch S9. Drive C (the usual hard drive letter). C1- 4 5 9 The ability to install to another drive let- + C2+ C11 1µ 5 IC7 MAX232 V- 6 SOFTWARE ter, e.g. Drive E on a partitioned drive, has not been provided with this program. C2- In common with 11 14 many other PIC de- Although the author has previously offered FROM IC5 PIN 17 T1 IN T1 OUT 10 T2 IN T2 OUT 7 N.C. signs, the facility has the option with other VB6 programs, feed- N.C. 12 R1 OUT R1 IN 13 1 6 been provided to pro- back from readers has indicated that the N.C. 9 R2 OUT R2 IN 8 C13 gram the PIC in situ, option is not always reliable with some GND 10µ + via connector TB2. systems. Consequently, it has been 15 Diode D6 and resis- dropped from this program. Readers who 0V tor R25 prevent dis- know how this option can be reliably tress to the +5V line implemented with VB6 are invited to tell Fig.6. RS-232 interface circuit. during programming. the author at EPE!292 Everyday Practical Electronics, April 2003
  • 6. TO PROGRAMMER (SEE TEXT) TO SK2 S2 5 A DATA CLK 4 MCLR RB7 RB6 0V OUTPUT 1 RESISTANCE 2 3 SERIAL VR4 OUTPUT TB2 R1 + SK5 TB1 R R R R 3 4 5 6 C4 IC2 CX 3 k IC3 R25 +5V 2 + 0V 1 C12 5 D4 C5 k D6 a 0V(R/W) 5 + + + 9 a R2 C10 C14 REAR VIEW OF PINS S3 R22 S9 + E 6 IC7 6 C7 S8 RS 4 C11 OUT k 1 R D7 14 3 a a R21 D5 23 D3 R26 IC5 D6 13 A 2 R20 a D2 R27 IN OUT D5 12 C13 1 R19 k k D4 11 + 0V COM C R18 T.P. T.P. C6 3 +5V R R IC1 GAIN 13 + 15 R28 R R R R R R29 24 31 12 IC4 17 16 + C1 R30 R9 R R R11 C2 IC6 7 8 k C9 D1 X1 R14 a R10 C8 TO SK3 TO SK4 TO SK1 TO RA5 0V TO +9V BATTERY TO RA4 S1 S4 S5 S6 S7 S8 S9 ON/OFF UP DOWN MODE DOWNLOAD SAVE TEST 4.3in (109.2mm) 388 2.8in (71.1mm) Fig.7. Printed circuit board component layout and full-size copper foil master track pattern for the Earth Resistivity Logger. If you do not have VB6, you need three are shown in Fig.7. This board is available Double-check the perfection of yourother files, comdlg32.ocx, Mscomctl.ocx from the EPE PCB Service, code 388. soldering and component positioningand Msvbm60.dll, held on our 3.5-inch Assemble in any preferred order, ensur- before applying power. Do not insert anydisk named Interface Disk 1, and in the ing that all the on-board link wires are of the d.i.l. i.c.s until the correctness of theInterface folder on the FTP site (they are included, and that all polarity-conscious +5V output from regulator IC1 has beenalso included with the Toolkit TK3 soft- components are the correct way round. proved.ware). These files must be copied into the The use of sockets for all the dual-in-line To provide a degree of waterproofness,same folder as the other Earth Resistivity (d.i.l.) i.c.s is recommended; it is essential the prototype was mounted in a robustfiles. to use one for the PIC, IC5. Treat all i.c.s plastic box with a see-through lid. The as static sensitive and discharge static elec- l.c.d. was mounted below the lid on theCONSTRUCTION tricity from yourself before handling them, inside. If a metal box with a see-through Details of the component and track lay- by touching the bare grounded metal of an lid can be found, it would provide evenouts for the printed circuit board (p.c.b.) item of earthed equipment, for example. greater durability.Everyday Practical Electronics, April 2003 293
  • 7. pinouts for the latter are shown in Fig.8. It will probably be necessary to adjust its contrast using VR1 before a display will be seen. With power switched on again, check that +5V and –5V are still present where they should be. Switch off immediately if they are not, and correct the cause of malfunction. On line 1 of the l.c.d., the message “SOIL RESISTIVITY” will be displayed briefly before being replaced by some numerical values, with more on line 2. L.C.D. display following switch-on. The final prototype board prior to installation. It is recommended that a case of at least differing lengths and cores. Obviously the50 per cent larger than used in the proto- thicker it is, the lower the loss over longtype should be employed to allow a large lengths, but 50m (say) of such cable is Example display when carrying out soil9V to 12V battery to be adequately housed. expensive, and heavy to drag about. monitoring with S9 switched on to test Probe sockets were 2mm types on the Details of constructing customised mode.prototype, simply because the author had probes are given in Part 2, but in simplethem in stock. It is recommended that 4mm applications four thin metal rods of the With Test switch S9 switched on, the firsttypes should be used. These provide type used in gardens as flower supports can two values on line 1 show the monitored val-greater robustness of the plugged connec- be used. ues present at the outputs of IC4a/IC4b, astions and allow them to be removed readi- detected by the PIC’s ADC Nick recommends the use of restraints TESTING Respectively, they are suffixed by the lettersnear the sockets to prevent the connections Having established that +5V is present B and A, indicating the op.amp to which theypulling out during a survey. on the output of regulator IC1, plug in the refer (as given in the circuit diagram Fig.4). The probe sockets should be colour voltage inverter chip, IC6, and check that With S9 off, the values are the upper andcoded, as should their respective plugs. around –5V is present on its output. lower peak values resulting from the ADCColour suggestions are shown in the circuit Naturally, always disconnect power before conversion of the output of IC4d. They arediagrams of Fig.3 and Fig.4, but may be making component changes. suffixed by the letters H and L (High andchanged to suit availability. It is important If all is well, the remaining i.c.s can be Low). Any value between 0 and 1023 couldNOT to duplicate the colours – doing so inserted and the l.c.d. connected. Typical appear at this time for all four readings.could result in leads being incorrectlyallocated to probes. The use of crocodile clips with colour-coded plastic covers was found to facilitatethe connection of leads to the probes them-selves. Heavy-duty crocodile clips are rec-ommended for ease of use (especially incooler or wet weather!). When testing the prototype, it did notappear to matter whether the probe leadswere screened or not. Consequently, stan-dard lighting or low current cable could beused. Twin-core mains cable was used bythe author and Nick, but in long term sur-veys it might prove more convenient tohave a mix of cable arrangements, of Interior of the case showing the relative positioning of the components. The p.c.b. is the first prototype which did not include the RS-232 device, IC7. The latter canFig.8. The two “standard” l.c.d. module be seen on its own sub-board to the left of the push-switches. It is recommendedpinout arrangements. that a larger case is used to allow a heavier-duty battery to be inserted.294 Everyday Practical Electronics, April 2003
  • 8. At the top right of line 1 is another incrementing beyond 127, or rolling over ability to display values as different inten-number, suffixed by a hash symbol (#). to 127 after decrementing below 0. sity grey-scales was found to be too limit-This is the processed value that, when Pressing Mode switch S6 changes the ed to justify the extra expense (at leastSave switch S8 is pressed, is stored to the position of the asterisk, thus allocating the another £30) and so the facility wasserial memory as a grid value for the +/– switches to that aspect of the grid, i.e. dropped.coordinates on line 2. Switching between vertical (column) or horizontal (row). Had the result been acceptable, again settings using S3, the value will PIC16F877 would have been used with thechange. (During a survey always keep S3 DATA TRANSFER screen, in a manner similar to the author’sat the same setting.) SWITCH Using Graphics L.C.D.s with PICs article Note that if too strong an input signal is Pressing Download switch S7 causes the of Jan ’01.amplified, the op.amp’s output may satu- PIC to send the contents of the serial mem-rate (reach its maximum obtainable level). ory to the PC at a rate of 9600 baud. As EEPROM RESETTINGIn practice, keep the value at the right of previously said, the values for each of the The contents of the serial EEPROMline 1 well below about 500. A value of 16384 possible grid coordinates are stored can be reset to zero when required. As a1023 is the maximum that can result from as two bytes – the MSB and LSB of the 10- security measure (to avoid resetting inap-an ADC conversion, indicating that the bit ADC values. propriately!), the reset routine can onlyADC has received an input voltage equal to No attempt has been made to be selec- be called at the moment that the power isthe power line voltage of +5V. This is an tive about which set of values is sent to the being switched on. With the power off,improbable event as the op.amp output is PC. All 32768 values are sent on each press and hold down Save switch S8,unlikely to swing that high. occasion that S7 is pressed. The transfer then switch on the power. When the mes- takes about 30 seconds. sage CLEARING EEPROM is seen,L.C.D. LINE 2 During transfer, the top l.c.d. line shows release S8. At the left of line 2 are shown the col- the message “SENDING TO PC”, withumn and row values which represent the line 2 blank. Upon completion of the trans-survey grid coordinates, and thus the loca- fer, line 2 shows “SENDING FINISHED”,tion in the serial memory at which the and line 1 briefly displays the “SOILprocessed IC4 value is stored. They are RESISTIVITY” message again, beforesuffixed C and R respectively. An asterisk clearing to once more show the valuessymbol (*) will be seen to the right of one being sampled.or the other of these coordinate values Line 2 remains with its last message Example display during serial memory(more on setting coordinates in a moment). shown until the asterisk (Mode) switch S6 resetting. At the right of line 2 is shown the value is again pressed, to once more show thethat is currently stored at the specified coordinate values.memory address. During the survey it will On line 2 will be a progress count dis-normally show 0 as each new coordinate is play as the software writes zeros to allselected. When the Save switch S8 is 32768 EEPROM data locations. It is apressed the display will change to repeat somewhat lengthy process, taking aboutthe number that has just been saved to the three and half minutes. This is due tomemory as a 2-byte value. At any time dur- numerous essential delays that are builting the survey, the coordinate switches into the writing procedure.may be used to recall the values that are Example display when downloading The software for the EEPROM writingstored for each grid location. stored data to a PC-compatible and reading was originally downloaded There are three switches for coordinate computer has been completed. from Microchip’s CD-ROM for use in thesetting. Two of them, S4 and S5, respec- PIC16F877 Data Logger referred to earli-tively increment or decrement the value Check that all the switches perform as er. It is recommended that you do notbeside which is shown the asterisk. The intended. It is not necessary to have probes attempt to modify Microchip’s coding torange is 0 to 127, rolling over to 0 after connected at this time, and it does not mat- speed the resetting process! ter that the serial download will not be des- On completion of the resetting, which tined anywhere – the PC’s data reception also resets the column and row values, the side of things will be covered in Part 2. screen briefly shows the SOIL RESISTIV- ITY message and proceeds in the normal PROGRAMMED ASIDE way as described earlier. Incidentally, experiments were made using a graphics l.c.d. instead of an NEXT MONTHExample of display when Save switch alphanumeric one, to see if survey data In the final part next month, the PC-S8 is pressed. In this case saving 28 to could be illustrated by the unit as an in- compatible Windows software is describedEEPROM location 41. built 20 × 20 grid display. However, the and probing methods discussed. NEWSAGENTS ORDER FORM Please reserve/deliver a copy of Everyday Practical Electronics for me each month Signed ....................................................................................................................................................................................... Name and Address ................................................................................................................................................................... ................................................................................................................................................................................................... ................................................................................................................................ Post Code ................................................ Everyday Practical Electronics is published on the second Thursday of each month and distributed S.O.R. by COMAG Make sure of your copy of EPE each month – cut out or photostat this form, fill it in and hand it to your newsagent.Everyday Practical Electronics, April 2003 295
  • 9. Constructional ProjectEARTHRESISTIVITYLOGGERJOHN BECKER Part TwoHelp your local archaeological society to error interception repertoire as pro-locate and reveal the hidden mysteries of grammed by the author – the PC itself will report any such unlikely events.our ancestors. MAIN PC SCREEN Part One last month we discussed the There are six screens associated with the The main screen offers several optionsI N principles of earth resistivity monitor- Logger’s VB6 program: to enable you to analyse the data received ing and described the construction of a from the Logger. It must be said, though,circuit through which this could readily be * Main screen as shown below, through that the facilities offered through theaccomplished and the data stored for com- which sectional analysis of the survey Windows Excel software supported byputer analysis. This month we detail the data is performed most PCs probably exceed what thissoftware that can help in this analysis, and * Full graph screen on which all 128 × screen can offer – more on Excel later.then examine some of the soil probing 128 download amplitude values are There are two main areas on this screen,techniques. The latest updates to the soft- displayed graphically, in oscilloscope as seen in its screen-dump photograph.ware are then discussed, followed by fashion (bottom photo on next page) To the right is a 20 × 20 grid block ofbriefly considering the ethics of surveying * Full grid screen on which all 128 × 128 squares, arranged so that the vertical axisand some practical advice and a list of fur- download values are displayed as grid represents the survey site columns, and thether reading. squares having amplitude-related hues horizontal axis the site rows. The site data or greys (top photo on next page) values determine the colour or grey-scalePC SOFTWARE * Download screen through which data appearance of each of these squares. Two The Earth Resistivity Logger’s PC soft- retrieval from the Logger is initiated scroll bars are provided which allow theware is written in Visual Basic 6 (VB6). It * Directory screen through which previ- grid data coordinates to be panned verti-has been proved under Windows 95, 98 ously recorded survey logging files can cally and horizontally so that all 16384and ME. It has not been tested with be loaded for on-screen analysis values of a 128 × 128 survey grid can beWindows NT, XP or 2000 as the author * Error Message screen – which hope- viewed in 400-sample blocks, seamlesslydoes not have these systems. fully you will never see! This comes joined. Readers who wish to try running the into action if the VB6 software detects The range of coordinates from the gridsoftware under the last three systems may various types of error (such as trying to matrix displayed is stated below it. Tofind benefit from reading Mark Jones’ arti- load a named file which does not know the precise coordinate of any square,cle Running TK3 under Windows XP and exist). It does not intercept errors add the values (numbered 0 to 19) indicat-2000, published in Oct ’02. which occur outside VB6’s specified ed alongside the edges of the matrix. GRID COLOURING There is a choice of four options regard- ing the colour shade range, as shown at the left of the screen. The lefthand bargraph display shows the grey-scale range avail- able, from white to black, 36 shades in all, representing values from 0 (white) to full black (35). The second bargraph shows a 36-value range of “rainbow” colour shades arranged in the order that VB6 offers them in the system’s own (peculiar!) numerical order. They are allocated by the program to rep- resent 0 (top) to 35 (bottom). Bargraph 3 is a monochrome scale of colours essentially in the green range but with varying intensities of red added. The 36 shades are again numbered from top to bottom as 0 to 35. The 8-colour bargraph shows the “pri- mary” colours offered by VB6, numbered 0 to 7, top to bottom. The quality and definition of the scale shades may vary depending on the qual- ity of your VDU. The scales are selected by clicking on Example of the prototype’s revised main screen displays and facilities. the “radio” buttons above them. In360 Everyday Practical Electronics, May 2003
  • 10. Note that if an original sample value of 0 is found, a dash line (–) is shown in place of a numerical value. This allows recogni- tion of any survey site squares for which a sample has not been taken. WAVEFORM DISPLAY The large vertical display area towards the left of the screen shows the sample val- ues plotted as graph waveforms. There are 20 lines (each numbered) representing the numbered grid rows to their right. Horizontally, the lefthand end of each line corresponds with the lefthand side of the grid row. The two sliders to the left of the graph area allow the plot values to be varied in the same way as with the grid, with the same multiply/divide and add/minus options. Thus the display can again be manipulated to show the survey site fea- tures to best degree, in this case as differ- ing amplitude waveforms.Example of the full screen grid display, using a zoom value of x9. With zoom at x1 The range of sample graph values isall 16,384 grid squares are shown. The contrast will show more clearly on screen changed at the same time as the grid’sthan it may on this printed page. coordinate range is set via its scroll and pan sliders. Below the graph area is another tick box,practice, the greyscale and monochrome These two controls allow the optimum Fill Graph Blanks. When the box isbargraph provide the clearest indication of shades or colour to be shown that best unticked, any zero values in the originalsample value relationships. illustrate the sample value relationships. (unmodified) samples are not plotted on The values which are actually obtained Even seemingly similar readings can have screen, indicating any survey site squaresfrom the survey site could, as said previ- their values manipulated to increase the that have not been sampled. With the boxously, fall into the range 0 to 1023. Two contrast. ticked, the zero values are plotted so that aslider controls are provided so that the val- Above the two sliders is a Show Values continuous graph line is shown. Clickingues logged can be suitably displayed as tick box. When unticked, just the colour the box alternates the modes.comparative values within the grid squares. shades are shown. When ticked, the equiv- To the left of the graph display are 20They are to the left of the grid squares, alent numerical value of the scale shade, numbered tick boxes. These allow selectedjointly captioned Graph, with sub-captions from 0 to 35, or 0 to 7 as appropriate, is graph lines to be hidden (no tick) to makeof / (forward slash symbol) and minus. displayed within the squares as well. the viewing of the data in the other linesClicking the sub-captions with the mouse Clicking the box alternates the two modes. clearer. As with all tick boxes, clickingcursor toggles them to show X (multiply) If a particular shade is too dark to read them again alternates between on and off.and add, and back again on the next click. the value, move the mouse cursor over it With the lefthand control, moving theslider causes the basic sample values to be and a “Tool Tip Text” box will appear, stat- ing the value. Tool Tip Text box messages INVERSION Below the Show Values tick box iseither multiplied or divided by the slider’s appear for various functions on screen if another tick box, Invert Values. When sur-value, according to its sub-caption mode. you move the cursor over them. veying, less-dense soil produces higherSimilarly with the second slider, adding To the bottom of the screen below the values than dense soil. High values pro-or subtracting the slider’s value. grid are two text lines. The first shows the duce darker shades on the grid squares andMultiplication/division take place first in the actual range of the sample values, the sec- lower troughs on the graph routine, followed by add/minus. ond shows the range after correction. The Invert Values tick box allows the value relationships to be swapped, high becoming low, low becoming high. This allows denser soil conditions to be dis- played more darkly on the grid than for less-dense soil, and the graph lines to show peaks rather than troughs (valleys). Clicking the box alternates between the two modes. The default is for inversion (tick on). REFRESH BUTTON The VB6 program allows the main screen to be minimised and shifted in the usual Windows-type fashion. Because VB6 does not regard the graph lines as being “permanent”, these can be fully or partial- ly erased by the act of minimising or shift- ing. To restore the graph lines on-screen, click the Refresh button. It is also necessary to use the Refresh on the Full Grid and Full Graph screens to action various value selection changes. FULL DISPLAY SCREENSExample of the full screen graph display. The zoom is at x9 to emphasise the con- There are two buttons, marked Full Gridtour lilnes. The samples cover a maximum area of 16 rows x 26 columns, each and Full Graph. They respectively causesample representing one square metre. The data is the same as in the full grid dis- the selected full screen mode to be dis-play and is in eight colours on the screen. played. On both, value manipulation andEveryday Practical Electronics, May 2003 361
  • 11. inversion are available as on the main elapsing time before a time-out error offering the option to try again or cancelscreen. So too is colour mode selection. occurs. the download. Above and to the left of the grid and If the time-out occurs before data is Occasionally, the PC software thinksgraph areas are two sliders. When received, you are offered the options to that data is arriving immediately followingclicked, these display the survey site grid cancel the download, for the PC to try the click of OK in the Download messagecoordinates to which their arrows point. downloading via its other COM port box, even before the Logger’s DownloadTheir position is also used by the zoom address (there are two allowed for, COM1 switch S7 has been pressed. The reason hasslider facility at the top left. and COM2, at addresses 2F8h and 3F8h), not be found. It is a rare situation, though, There are 10 values of zoom selectable or to retry downloading from the same and in this event the PC software almostaccording to the ratio Zoom / 2 + 0·5, with COM port address. immediately experiences a time-out as dataa range of ×1 to ×5·5. The slider arrow If you choose that the other COM port does not continue to arrive, and then offerspositions determine the origin point on address should be tried, this address is you the option to try again.which the enlargement is made. Intercepts stored to the EarthResSettings.txt file, It is worth waiting a couple of secondsare included in the program to keep the dis- which resides in the same folder as the rest after OK has been clicked before pressingplay within the bounds of its frame. of the Earth Resistivity software. It is then switch S7, in case this situation is about to recalled next time you run the program. occur. Once the Logger has started to sendDOWNLOADING DATA It is permissible to change the COM port data the process must run its full course You require a standard serial cable, of address within this file if you wish (via and cannot be halted. The same applies tothe type used with normal modems (D- Windows Notepad for instance) – it is the the PC routine, it too cannot be interrupted,range 9-way male to female, straight first entry in the file. Take care not to upset and will continue until a time-out has beenthrough), for data transfer from the Logger. the positions of the other lines in the file. experienced.It should have a connector suited to your These lines set various parameters for the It had been hoped to provide a bargraphPC at one end, and a 9-pin male plug at the program each time it is loaded and run. to graphically show the progress of theother. Adaptors (25-pin to 9-pin) are avail- When the Logger starts to send data download. Regrettably, it was found thatable if an existing modem lead has a 25-pin before the time-out ends, and the PC on slower PCs the software is incapable ofmale plug. begins to receive it, the countdown simultaneously updating the bargraph (or other visual forms of timing) and reliably inputting the serial data. Consequently, this option has not been provided. It takes about 30 seconds to download the full 32768 bytes. A starting time is displayed on screen below the primary time-out bargraph. When the download has been success- fully finished, a routine combines all the double-byte values into single 16-bit bina- ry values. These are converted to decimal Screen displayed if synchronisation is and combined into lines of text data, each not correct. value separated by a comma. Each line Download option screen. holds the data for one survey site row (128 bargraph halts and a confirmation that values). There are 128 lines, representing To download data serially from the data is being received is displayed. The the number of survey columns.Logger, first click on the Download Data full 32K block of Logger data (16384 This data is then output to disk, to a filebutton at the bottom of the main screen. samples) is downloaded at 9600 baud, and whose unique name is in the form of theThis causes a message screen to be dis- initially stored into temporary memory following example:played, asking if you want to continue with locations.the download, or cancel the call and re- During this process, another time-out EarthRes 12JAN03 10-27-35.TXTshow the full main screen. countdown of about one second per data If the OK button is clicked the small byte is monitored. If this period is exceed- in which the date and time (hh-mm-ss) isDownload screen is displayed. As advised ed the program assumes that the download that applying at the moment that the file ison the previous message you now have is complete (the PICabout 30 seconds in which to press the has stopped sendingLogger’s Download switch S7. During this data), or that the serial30 seconds or so, a bargraph shows the link has been broken. Because the down- load is asynchronous (i.e. no handshaking), an error checking rou- tine has been includ- ed. When the PIC starts transmitting, it first sends several zeros followed by the message RESISTY.Countdown bargraph while waiting for When the PC pro-data to start coming from PIC. gram finds that the one-second serial time- out has occurred, it checks through the first 20 downloaded bytes to see if these values contain the ASCII coded RESISTY mes- sage. It also assesses whether the download quantity is correct. If neither fact is correct, the screen Example of the folder directory screen through which filesScreen displayed in the event of data displays a message from any folder path can be selected according to a filterednot received due to COM port failure. box stating so, prefix option.362 Everyday Practical Electronics, May 2003
  • 12. created. (The Logger itself has not beenprovided with date or location recordingoptions – it is up to you to record this infor-mation in some other way.) The file is heldin the same folder that holds the rest of theEarth Resistivity software. Having saved the file, the software splitsthe recombined values into a matrix of reg-isters whose coordinates correspond withthose used during the site sampling. It is these values that are used for displayvia the main screen’s graph and grid areas.They are plotted there immediately theDownload screen closes. Simultaneously,the grid matrix location coordinate slidersare reset to zero. The value correction slid-ers are left as previously set, allowing var-ious sets of file data to be recalled fromdisk for viewing under the same correctiveconditions. On return to the main screen after theDownload, the name of the current fileloaded (in this instance that just saved) is Example of using Windows Excel to display data graphically.displayed in bold towards the screen’s bot-tom right. The formatting simply entails using Move this cursor anywhere over the commas to separate the sample values, darkened area and left click to reveal theLOADING SAVED FILES which are expressed as normal text charac- Chart Wizard – Step 1 of 5 window. The To load the program with data from a ters (e.g. 1234). darkened area reverts to normal black onpreviously saved file, click on the Inevitably, there are many versions of white, surrounded by a dotted box, possi-Directory button. This displays a multi- Excel and specific use details that apply to bly “shimmering”.function screen through which files in any all of them cannot be given. The chances Ignore the options offered and just clickfolder on any installed disk can be selected. are, though, that the use will be similar to Next, to show the Chart Wizard – Step 2 ofIt is a modified version of the Directory that on the author’s main PCs. The follow- 5 window. Select (left click) one of thescreen originally designed for use with the ing is the procedure he uses for Excel 97: graph type options offered, the “3-Dauthor’s PIC Toolkit TK3 software (Nov Load Excel, using Windows’ Find button Column” option is suggested.’01), and since used in modified form with to locate it if necessary – on the author’s Click Next to show the Chart Wizard –other VB6 programs as well. PCs it is at Step 3 of 5 window. Select one of the chart It will not be discussed in detail here as type variants on offer, the one numbered 6,the screen has a NOTES button which calls C:MSOfficeExcelEXCEL.EXE perhaps.up a Windows Notepad text window Click Next to show the Chart Wizard –through which you can read the detail of Now follow the path File (in top tool- Step 4 of 5 window, and an illustration ofthe Directory screen’s use. bar), Open, Select folder, set File Type to the Sample Chart selected will be seen. In brief, you can change drives and Text Files, then double click on the Ignore the right hand option boxes andfolder paths, set a “filter” option to only required EarthRes file name to load it. A click Next, to show the Chart Wizard –show files having a specified prefix in Text Import Wizard – Step 1 of 3 window Step 5 of 5 window. Now just click ontheir name, and recall previously select- is now shown, with the first several import- Finish.ed paths through a History box. To select ed values on display. Select the Delimited The graph type selected will now be dis-a file, double click on its name in the option as the active “radio” button. played on the main Excel screen, with therighthand display area. This causes it to Click Next to show the Text Import value boxes still visible behind it. It can bebe loaded and split for grid matrix allo- Wizard – Step 2 of 3 window. Click the moved around the screen and sized in thecation in the same way that the down- “Delimited Comma” box to reveal a tick. usual Windows style. A small (mobile)loaded file just discussed was split and Click other ticks to become unticked (if Chart selection window will also be dis-displayed. necessary). Ignore the “Text Qualifier” played, allowing different options of dis- One of the author’s files is included with box. play to be selected and manipulated.the software (but with fewer that 400 sam- Click Next to show the Text Import Save the file and its graphical displaysples), plus a much longer one produced by Wizard – Step 3 of 3 window. Ignore the (more than one can be generated on theNick during his survey work. Experiment options offered, but click Finish. screen at any time and placed at differentwith them and the screen’s manipulative The main Excel screen will now be positions) as an Excel-type file with anycontrols. shown, with the survey values allocated to name of your choosing. Alternatively, sim- column and row boxes. ply Exit unsaved if you prefer.USING WINDOWS Left click on one of these boxes, say the It is now up to you to experiment withEXCEL first one at top left, to select the starting Excel’s numerous options, calling up its When the downloaded survey data is coordinate of the matrix area you wish to Help files for more information. It is anoutput to disk, it is formatted to suit its show graphically. With the left mouse amazing package with many uses, andanalysis and display via Windows Excel, a button still held down, move the mouse seemingly ideal for the sort of analysis thatfacility that should be on any PC running downwards and to the right, causing the archaeological survey data calls for. In aWindows 95 and later (search your selected boxes to show white text on a black word – experiment!Windows CD-ROM for it if it is not background as the area is increased. Thealready on your system). first box, though, remains as black on white. PROBING METHODS As well as offering graphing facilities, The construction of the probe assem-Excel provides for mathematical expres- EXCEL GRAPHING blies will be discussed once some of thesions to be computed, making it capable of Release the mouse button when you’ve probing techniques have been examined.being set-up to calculate true earth resis- selected the required area. Now click on There are several probing methods avail-tivity in relation to known resistance and the Chart Wizard icon on the top tool bar (it able through which to obtain grid datacurrent factors. Study Excel’s Help facili- looks like several vertical rectangles, with about a survey site. The author makes noty, and read Anthony Clark’s book. (As an elongated diagonal shape above them – attempt to recommend any one in particu-said last month, though, Nick’s surveys a chimney falling onto a factory?). The lar. You must do your own research intowere done in relation to signal amplitude mouse cursor becomes a similar (but not that, through the references given later, andvalues and not the actual resistance, but identical) symbol plus a cross, representing by chatting to those in archaeological soci-see later.) that graphing mode has been selected. eties who know about such things.Everyday Practical Electronics, May 2003 363
  • 13. finished with, putting it in the next square’s COLUMN opposite top corner, and swap the probe ROW 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 leads to retain the correct order. 0 1 The author surveyed his garden several 2 BASIC GRID LAYOUT, ALSO times in different ways during design 3 4 SHOWING TWIN-PROBE EXAMPLE POSITIONS development, and on each occasion 5 became faster at doing so. On the final sur- 6 7 vey, on an 11 × 7 grid (77 samples) it took PROBES C1 AND P1 REMAIN IN FIXED POSITION. 8 PROBES C2 AND P2 TRAVERSE THE GRID. about an hour and half. 9 10 Of course, during the process of doing 11 12 the test surveys, several methods for speed- 13 ing it were imagined. For a solitary survey- 14 15 C1 P1 P2 C2 or, perhaps the most efficient in terms of 16 speed would be to insert separate probes at 17 18 WENNER ARRAY each corner of the matrix prior to taking 19 PROBES EQUIDISTANT readings. It would then only be required to repeatedly change the lead connections – a seemingly much faster “conveyor belt” C2 P2 AT COORDINATES SEE TEXT system. No doubt, though, having an assis- FOR DISTANCE R19/C3 tant would probably make the moving of THESE PROBES ARE MOVED BETWEEN COORDINATE NODES C1 C2 just two probes a speedy alternative. A perhaps less practical method was C1 P1 (bizarrely?) thought up too – using a THESE PROBES REMAIN IN FIXED POSITION motorised vehicle like a golf buggy with A GRID MATRIX OF SQUARES IS COMMON TO ALL NORMAL SURVEY METHODS. P1 P2 probes attached to the wheels in Boadicea THE ABOVE ALSO SHOWS EXAMPLE POSITIONS FOR PROBES IN THE TWIN-PROBE TECHNIQUE. fashion. This would then be driven back SQUARE ARRAY and forth across the grid, the probes auto- PLACING PROBES AT THE GRID INTERSECTIONS ASSISTS REGIMENTATION OF THE SURVEY TECHNIQUE. matically inserting themselves, and trigger- SQUARES TYPICALLY HAVE SIDES MEASURING 1 METRE. ing the storing of each reading at the correct coordinates! (Well – a chap can dream, can’t he?!)Fig.8. A 20 x 20 grid layout with Twin-Probe example positions, and the positioningof the probes in Wenner and Square arrays. WENNER PROBING Another seemingly useful technique is There are numerous archaeological web are then connected to the probes, using known as the Wenner configuration. In thissites with bulletin boards and “chat-zones” heavy duty crocodile clips seems the method the four probes are arranged in aon-line if you search through the excellent easiest method. straight line, equally spaced apart, say, or other quality internet The Logger’s storage coordinates are set metre between them. Fig.8b shows thesearch engines. It is worth noting, though, to suit the square number, i.e. to R0/C0 in order of arrangement.that Anthony Clark considers the Twin- this case, and a reading saved to the This method is apparently better suitedProbe technique to be that most suitable to Logger’s serial EEPROM by pressing to doing a more detailed survey of thearchaeological work, and is the one used switch S8. matrixed grid site. The principle is that theby Nick with his surveys. The C2/P2 probes are then moved to the TX probes are the outer two. The RX With all techniques, the area to be sur- top corners of the next square, to the right probes are in line between them. The cur-veyed is first marked out as a grid with for example, to be monitored and its coor- rent flows across the TX pair and is pickedtapes or similar, to form squares having dinates set into the Logger, in this case up across the RX pair, the received signalsides of, say, one metre in length (this is a R0/C1. Again a reading is stored to the value varying with the resistance in seriescommonly quoted distance in this context), EEPROM. with the probes in a more direct fashionand probably forming a 20 × 20 matrixed The process continues fully across hori- than with the twin-probe technique.area, see Fig.8. zontally for the width of the marked survey A variant of this technique, the Anthony Clark comments that plastic area, e.g. R0/C19 (the final column of this Schlumberger, in which the probes are notcovered clothes line is also useful for set- row in a 20 × 20 grid). The probes are then equally spaced, is discussed in Anthonyting out a grid matrix. He cautions, though, moved down by one row, and the process Clark’s book. But he regards it as not ide-that it can be difficult to untangle and on repeated, to the left this time, back to ally suited to archaeological site he knows of, it had to be “guarded R1/C0. And then down by another row, andin the presence of sheep, by whom it was so on for all 400 squares. SQUARE ARRAYregarded as a rare delicacy”! Another method is known as the Square WELL ORDERED Array in Anthony Clark’s book. With thisTWIN-PROBE Note that the relative order of all probe method, the TX and RX pairs are placed at The Twin-Probe technique is apparently connections must be maintained during the the corners of the one metre squares, asmore suited to initial surveying of a site, survey. Differences in reading can result if shown in Fig.8c. The four probes areestablishing whether or not it is worth car- the order is changed, hence the earlier rec- moved as a set from square to square.rying out a more detailed survey. ommendation that the plugs and crocodile The transmission signal flows between With this method, the two probes C1 and clips should be colour-coded. the TX pair as before. This time the RXP1 are inserted into the ground, sufficient In practice, it does not matter in which pair pick up the radiated signal at the sameto make electrical contact with it (see earli- direction you move the probes, or whether distance from the TX probes. If the soiler), centrally to and somewhat outside the you start the survey from the top of the grid resistance between the TX and RX pairs isarea to be surveyed. Anthony Clark dis- or the bottom. Note that the PC screen uniform, so too will be the amplitude of thecusses the best distance in his book. regards location 0,0 as being at top left of signal received by both RX probes. Probe C1 is the transmitting probe con- the screen. Tests showed that because the probes arenected to the comparator IC3, via switch “Be methodical and consistent” seems to connected to a differential amplifier, if theS2. Probe P1 is one of the receiving probes, be the key phrase, though – this helps you two input amplitudes are the same, theyconnected to the input of op.amp IC4a. to establish a routine that becomes second will cancel each other out at the final com- Probes C2 and P2 are inserted into the nature, which the author soon found when bining stage (IC4c).ground, to a similar depth, at the far cor- starting his own mini surveys! If, on the other hand, the amplitudes areners of the first square to be monitored, say It was also soon found that it is not nec- not the same, the difference between themtop left, coordinate R0/C0 (row 0, column essary to move both probes on each occa- is that which will be finally output from0). Probe C2 is the 0V reference probe, and sion. Since one is already at the corner of IC4c. In this case, what would be lookedP2 is connected to the input of op.amp the next square, it is only necessary to for is any difference values, indicating theIC4b. The respective leads from the Logger move the probe from the corner now edge of a subterranean feature.364 Everyday Practical Electronics, May 2003
  • 14. It is evident, however, that balanced assemblies described 100mm(zero) readings, when the two input values by Robert Beck 170mm SCREW TERMINALare equal, would only indicate the unifor- should be considered. WITH 4mm SOCKETmity of the terrain in that grid. It would not Schematics of the 20mmindicate whether that uniformity was due original figures illus-to a highly resistive feature or a highly trating these probes 330mm TUFNOLconductive one. Nonetheless, the detection have been redrawn 250mm COLLARof only outlines might in itself be a desir- and are repeated situation. Other than the fol- B) A variant of this technique would be to lowing details, noplace the two TX probes at one end of a additional informa-column, and the RX pair at the other, tak- tion can be a reading and moving both pairs to the Robert’s rigid 930mmnext column, still at the top and bottom frame assembly for BAND OFpoints. This could perhaps yield initial two probes is shown PAINT OR 330mminformation about whether or not a site is in Fig.9. Details of TUFNOL COLLAR TAPE 250mmworth examining more closely. Not being his single probe arean archaeologist, though, the author can- given in Fig.10. 15mm B)not comment on the validity of this. His original text Anthony Clark discusses the above states that the Twin- HOLE FOR PROBEnamed probing techniques in more detail, Probe assembly was 250mm TO PASS THROUGHand describes others. specially developed A) TUFNOL A) COLLAR and that its top mem-PROBE ber is a wooden bat-CONSTRUCTION ten, 30mm × 50mm × During garden tests with the prototype 1050mm, the ends ofLogger, individual metal rods measuring which are bound withabout one metre long by 5mm diameter, self-amalgamating D) 120and with a right-angled bend at the top tape to form hand 0BA TAPPED HOLESwere used as the probes. These were pur- grips. TO RECEIVE SOCKETchased inexpensively from a garden cen- An aluminium HEAD GRUB SCREWStre, their intended use being to support platform is attachedplants. to the centre of this Fig.10. Construction details of Robert Beck’s probes. A recently observed, but not tried, batten to carry thepossibility was in the form of long inex- case that holds thepensive barbecue skewers – seen in a local electronics, secured by rubber bands. The either dry and coat the batten with varnish,supermarket. bottom member is a similar wooden bat- or devise insulating collars of Tuffnol or If you wish to construct purpose-built ten, but this piece must have good insulat- similar material, and fit them where theprobe structures of more durability, and ing properties (to prevent current leakage probes go through the batten.perhaps greater ease of use, the probe between the probes). He suggests that you The top and bottom battens are held together by metal conduit pipes, threaded at each end and secured by lock-nuts. In describing the construction of the 1050mm other probes, he says that none of their L-SHAPED ALUMINIUM dimensions are critical and may be dictat- BRACKET ed by what is to hand. In Fig.10a is shown 30mm a substantial probe made out of stainless THIS AREA THIS AREA steel tubing with a brazed on T-handle and TAPED TO TAPED TO tip which assist soil penetration. This FORM FORM HANDLE HANDLE probe is designed to be used by the opera- tor in the standing position. 4mm PLUGS TO CONNECT A smaller version of Fig.10a is shown in TO RESISTIVITY METER Fig.10b. This has a 4mm screw terminal added, an alternative method of wire NOTE: THE UPPER AND LOWER HORIZONTAL RAILS ARE OF WOOD. tHE LOWER RAIL SHOULD BE DRIED AND connection. VARNISHED. tHE L-SHAPED ALUMINIUM BRACKET IS TO SUPPORT THE RESISTIVITY METER. Probes may be constructed of material other than stainless steel, which is expen- 1000mm sive and a little difficult to obtain, he says (provided it is corrosion resistant of METAL CONDUIT PIPE THREADED AT BOTH ENDS course). An extremely simple probe is shown in Fig.10c and which may be constructed from 6mm diameter metal rod, i.e. brazing or uncoated welding rod, mild steel, silver SOLDER TAGS steel, etc. A depth guide consisting of a band of paint or insulating tape is added and connections are made to the top using a crocodile clip. 30mm The depth stop in Fig.10d is adjustable by means of an Allen key. The material need not be insulating, and could be of 200mm PROBE C2 PROBE P2 metal if desired. SERIAL OCX 500mm Since finalising the Earth Resistivity Logger Part One for publication last 800mm month, reader Joe Farr has provided EPE with a specially written SerialIO.OCX program that allows legal access to VisualFig.9. Support frame for the Twin Probe configuration used by Robert Beck. Basic’s own serial control I/O facilities.Everyday Practical Electronics, May 2003 365
  • 15. This option has previously only been avail- is reported on screen, allowing you to re- A. The first value shown (followed by B)able to readers who have a registered ver- download if you prefer, although minor is that monitored from IC4b pin 8, assion of MSCOMM (as Robert Penfold has “first aid” is provided by the program to described in Part One. The output fromdiscussed many times through his Interface regain sync after that point. It is rare, IC4a is no longer monitored via theseries). though, for more than one loss of sync to l.c.d. Joe’s serial OCX facility will be pub- occur. Such loss should not occur with If current values greater than 62 arelished in full at a later date – probably the Joe’s OCX program. encountered, they are limited to 63, and theSeptember issue. However, a section of It should be noted that readers who wish word MAX is displayed on the l.c.d.Joe’s program has been built into this Earth to make their own changes to the ER Switch S2 may be used to select one of theResistivity (ER) program and is available source code cannot make use of Joe’s OCX other resistors (R3 to R6) in the event thatto readers who are using the input option. For that to be used, the instal- the site being surveyed has greater or less-EarthResist.EXE standalone version. lation of Joe’s full OCX facility is required. er resistance than appropriate to a 1k fixed To use Joe’s option, though, several For copyright reasons this will not become resistor value. Do not change the resistorchanges need to be made to ER’s p.c.b., available to readers until its publication. value during a survey.without which the facility cannot function. Attempting to examine the ER source code The PC program stores the full 2-byteThey are: will generate an error condition because of survey value to disk. On re-input the cur- 1. Cut the track (0V) connecting to IC7 the presence of Joe’s program. Until Joe’s rent value is extracted from the MSB, and(MAX232) pin 13. full serial program becomes available, the the MSB is then limited to one active bit ER program can only be recompiled if (bit 0). The range of survey values is then 2. Connect IC7 (MAX232) pin 13 to 9- Joe’s sub-program (EarthResOCX) and all from 0 to 511. During surveying, the gainpin serial socket SK3 pin 3. references to it in the main program are setting via switch S3 should be chosen to 3. Connect IC7 (MAX232) pin 12 to IC5 removed. keep the values below 511, favouring a(PIC) pin 18 (RC7). Also be aware that this version of ER middle range centred on 256. If a value This action allows the PIC to receive with Joe’s OCX has not yet been proved on greater than 511 is encountered by the PIC,handshake data from the PC. a wide variety machines. If it will not work it is limited to 511 and the word MAX is To set the PIC program to respond to the on your PC, revert to using the normal ser- shown on the l.c.d. ial download option on PIC and PC. Pleasecorrect serial data transmission routine, ini-tiate the following procedure: advise us at HQ if this is necessary, telling CURRENT ON/OFF 1. Before switching on power, press and us the PC and its operating system type. OPTION All three display screens of the PC pro-hold down the Mode switch, S6. SURVEY CURRENT gram now have an extra tick box marked 2. The screen will go into serial path MONITORING Current. When it is ticked, each surveychange mode, alternating at about one-sec- Another feature added to this version is value is multiplied by its associated currentond intervals between a display saying the ability to monitor the current flowing value divided by 10. The theory is that“SERIAL PORT NORM” (original ver- between the transmission (TX) probes. It slight differences in the transmission cur-sion) and “SERIAL PORT OCX “ (Joe’s too requires a small change to the PCB: rent value at each survey grid square affectOCX). the actual value of the received voltage sig- 1. Cut the track between resistor R16 3. Release switch S6 when the mode you nal from the receiving probes. By relating and pin 7 of IC4.require is shown. This mode becomes the these voltages to their prevailing transmis-active path mode and is also stored into the 2. Connect the now-open end of R16 to sion current, compensation is made forPIC’s data EEPROM, to be recalled next the pole of switch S2. variations in the latter. The current valuestime the program is run. are not actual milliamp values, but simply With switch S2 in the R5 (1k resistor) numbers representing the relative current 4. On release of switch S6, normal run- position, current flows from the switch flowing.ning of the PIC program resumes. The ser- pole through the 1k resistor and to 0V via It is suspected, however, that in practiceial path mode may be changed whenever the resistance of the soil. These two resis- the variations make little difference to theyou choose. tances form a potential divider. The square interpretation of the displayed results. To wave voltage at their junction is bufferedPC OCX SETTING by R16 and half-wave rectified by diode repeat the statement made in Part One, the aim of this Logger is to show relative dif- Ensure that the PC is also set for the D2. The resulting peak positive voltage is ferences in signal amplitude across a sitechosen mode, as follows: monitored via PIC pin RA0 operating in being surveyed. It is the differences that Click the on-screen button labelled analogue (ADC mode). The peak voltage then indicate different sub-soil features.Please Read. Accept the option that then depends on the resistance of the soil, and If there are significant differences theyfollows to read the text. Having read it, exit from this voltage value the equivalent rela- are worth physical investigation. If therethe text reading screen to reveal another tive current through the resistance path can are no significant differences, then the siteoptions screen. This allows you to choose be calculated. is probably not worth examining further,between the new OCX option and the orig- To establish an initial reference value unless such techniques as magnetometry orinal (normal) serial mode. Click YES for prior to any survey, switch on the unit. ground-penetrating radar reveal differently.Joe’s OCX, NO if you want to use the nor- Then set switch S2 to the setting that A magnetometer design is currently beingmal serial download as originally written directly connects the pole to IC3 output worked on and will be published in EPE atinto the ER program, or CANCEL to exit pin 6. Do not connect transmission probe some time in the future, but not yetwithout making a change to the serial path C1 to socket SK2 at this time. Press scheduled.used. switch S6 (Mode) and hold it pressed, We shall be interested to learn if you Your choice is recorded to disk and then press switch S8 (Save) and hold it find that the current-monitoring featurerecalled next time the program is run. You pressed until the message REF SAVED enhances the results of your survey. Let usmay change your mind at another time if appears, preceded by a value. Release S8, know via EPE wish, re-entering via the Special Note then release S6. The value shown is nowbutton to do so. stored to the PIC’s EEPROM for present DOT MATRIX Note that the same mode must be select- and future use. Then switch S2 to the 1k A further option added to the PC pro-ed for the PIC and the PC. resistor (R5) path. gram since Part One is the dot-matrix dis- The advantage of Joe’s program is that it During active surveying, the voltage play facility, operative when the Matrixallows a bargraph to display the progress of at the pole of S2 is subtracted from the tick-box on the Full Grid screen is ticked.the data input procedure. It is also likely to reference value and stored as a 6-bit This draws small squares on the displaybe better at detecting input data problems number into bits 1 to 6 of the MSB of the whose dimensions are relative to the signalas it uses a handshake procedure to com- survey value recorded to the external amplitude.municate with the PIC, inputting the 32768 serial EEPROM IC6. The range of cur- The principle is a bit like the dots thatbytes of data in blocks of 256 bytes. rent values acceptable is from 0 to 63, make up a B&W photograph in a newspa- Whilst the original program inputs data and the actual value is displayed as the per (known as half-tone). It will be morethat is usually 100% accurate, there is the second value on l.c.d. line 1 when in Test useful with a large amount of survey dataoccasional loss of synchronisation, which Mode (S9 on). It is followed by the letter on screen than with a small quantity.366 Everyday Practical Electronics, May 2003
  • 16. MISCELLANY A few other “tweaks” have also beenadded since Part One. The text and demo circuit for someexperiments referred to on Part One page 1are now accessible via buttons at the bot-tom left of the screen. Two other buttons allow you to examinethe survey data as text files, one showingthe twin-byte values separately, the otheras the full combined value. These valuesinclude the current values as well. All three display screens have also beengiven a “pre-subtract” box, allowing you tosubtract, say, the minimum value receivedfrom all other values, enabling relevantdata to be extracted from any overall biaslevels. Because this PC software will be usedwith the Magnetometer currently underdevelopment, two “radio” buttons allowselection of whether Earth Resistivity orMagnetometer data will be processed.Ensure that the Resistivity button is the oneselected.ETHICS It was said in the opening paragraph inPart One that the original Earth Resistivity Nick’s survey was done not far from where John Constable painted this BoatMeter published in EPE was an electronic Building Near Flatford Mill scene. The contours in the full graph screen showntool to assist amateur archaeological soci- earlier clearly indicate a trench comparable to that in this painting. Illustration courtesy ofeties. So too is this Logger design. Whilst there is nothing to stop anyonefrom carrying out surveys on their ownproperty, there are considerable ethical To find a local society, look in the tele- The primary area covered in the surveyissues regarding the surveying of other phone directory, or ask at the library. The is approximately 16m × 26m at maximumland. author’s local library building even has a dimensions. Most of it was covered in one First, other land is not your land, and so display of the artefacts found by the society day, but then rain “stopped play” forany surveying of it requires the permission in his area. It is a region once heavily pop- several weeks.of those who own it. Remember that all ulated by the Romans, with many artifactsland in the British Isles is owned by that have been found on display, and even PRACTICAL ADVICEsomeone. Find out who it is and gain their the ruins of two Roman villas (but left From his experience with the prototype ofpermission before you proceed. where they were found!). this Logger, and from his general surveying Secondly, do not dig without an Only a few hundred yards from the activity, Nick offers the following advice:archaeologist’s involvement. If you have author’s house a Roman corn drier waslocated through your earth resistivity sur- recently found by his local society. 400 * For extensive survey work the batteryvey something that proves to be a site of yards further on are the ruins of a Roman needs to be bigger than PP3 sizeany importance, your unsupervised digging bath house. It is quite probable that his * The case should be larger than in thewill certainly destroy information that is garden is on a site where Roman’s once prototype and a better shape to carrynecessary to fully interpret the site. trod. about Earth resistivity surveying is essentially Although his survey graphics did not * Do not use small plugs and socketsnon-invasive except for the slight intrusion show anything other than known modern * The sockets need to be solidly mounted,made by the probes just into the surface. features, and probably including builder’s possibly on a metal base of someMany landowners could well be as inter- rubble of recent decades, perhaps he’ll one description, and include strain relief, it’sested as an archaeological society in know- day do a more detailed survey and then call surprising how hard you have to pulling what history might lie beneath their in the archaeologists to uncover an amaz- 50m of cable laying on wet grass!land, especially if they are approached in a ing find – one way to get the garden dug * Lay the survey out accurately, based onpolite manner and it is explained to them for him! a 3, 4, 5 triangle to get the lines perpen-that the resistivity surveying is just a mat- dicular. Bamboo canes make goodter of sticking some shallow probes in the NICK’S SURVEY markers for the 1-metre grid intersec-soil. Nick was fortunate enough to be permit- tions. If using clothes line with metre Remember that some locations are des- ted to survey a site made famous by marks, beware that rope stretches.ignated as Scheduled Ancient Monuments English artist Constable (before he was Survey tapes (typically 30m) need to beand that permission to carry out any form promoted to Sergeant says reader and carefully cleaned after use, or they getof research on them requires official friend John Waller – quoting an old Goon full of dirt and can jamapproval. Experienced local archaeological Show line!). * Keep perfect track of what you have sur-societies will know where these sites are Constable painted several pictures of veyed, it is horribly easy to lose track ofand whether or not surveying permission is sites at and around Flatford Mill in Essex. the grid section that you have justlikely to be granted, and if so, by whom. If One of them is his Boat Building Near recordedsuch information is not already known to a Flatford Mill, which is reproduced here. It * Probe around the site at random beforesociety, enquiries at the local town hall is near that site that Nick surveyed and his you start to make sure that you are setshould provide answers. results are those illustrated earlier in the up to keep the Logger’s values roughly full graph and full screen illustrations. around 250JOIN A SOCIETY They reveal very clearly the sub-surface * Try and get it all done in a day – a show- It really is in your interests to join an features that could have been bays cut into er overnight throws in a step functionarchaeological society if you are not the ground where boats might well have because you are then working in thealready a member. It is also in the society’s been tied up. Much of the site, though, is area that the rain will have penetratedinterests if you join them and they then now overgrown with trees, preventing ade- * Coil everything neatly – controlling 50have the use of your Logger! quate survey. metres of cable across a plot is trickyEveryday Practical Electronics, May 2003 367
  • 17. * Buy high visibility cable in case some- * Use compass bearings, fixed physical Applied Geophysics, Griffiths and King, one tries to trip over it! features, corners of buildings, drain cov- Pergamon Press. 1965. (ISBN unknown).* Colour code the probes – you need to be ers etc, or triangulate from fixed points Seeing Beneath the Soil, Prospecting consistent if the survey is in an open area. Most Methods in Archaeology. Anthony Clark.* Ask permission, most people will be archaeologists work north to south. Routledge. 2000. ISBN 0-415-21440-8. chuffed to bits that you want to do the This is a revised edition of the title refer- survey – but not everybody, and make ACKNOWLEDGEMENTS enced in EPE Feb ’97, and having a differ- sure that you are not somewhere where The author offers very grateful and ent ISBN and publisher. It is the most you should not be appreciative thanks to Nick Tile for carry- informative source used by the author dur-* Make contact with your local archeolo- ing out extensive field tests with the proto- ing the design of this Logger. gy group, they will be very helpful and type, for discussing at length many aspects It additionally covers other earth survey- interested, and may well bite your arm of its use, for lending Seeing Beneath the ing techniques, including magnetometry, off to get you helping them Soil and vetting the script. and provides several further reference The author also thanks those EPE read- sources. It is known to be available for on-* Be prepared to talk to people, you line ordering from and will cause interest if you are ers who provided him with information during the development of this design (in, current price around £25. somewhere public, and they will be alphabetical order!): surprisingly knowledgeable – and probably have all been watching Time Dave Allen for sending an ancient USEFUL WEB SITES issue of ETI containing a rudimentary Various aspects Team of the subject, including further links, access ER circuit using d.c. probing (and yes* Keep your ears open for local stories of Dave, this design could be used for mon- to the magazine Current Archaeology, and to old ruins, you might be the one that re- itoring relative impurity content levels in the Council for Independent Archaeology. discovers something lost to history water). Source of semi- because you happened to take the time Peter Barnes, for vetting the script and mathematical tutorial on earth resistivity to listen to the ramblings of the old guy for several useful email exchanges of and a link to a site called Introduction to in the pub thoughts and circuits, plus comments from Exploration Geophysics.* You can do a survey on your own, but it his archaeologist acquaintance Derek Excellent search is much easier with two of you about using Robert Beck’s design. engine.* Keep a note book that notes the time, Robert Beck, for the original inspiration. place, date, etc of the survey and things Aubrey Scoon, for comments about LOGGING OFF like weather conditions which could stray electrical currents in the soil. The ER software placed on the EPE ftp explain odd results, for example if it ODAS, the Orpington and District site on 17 March ’03, was version V1.2. started to rain half way through the Archaeological Society, and Alan Hart in Look in on the site occasionally to see if any work particular. further updates have been introduced. $* Any survey must have a repeatable base point, or base line so that if you do find FURTHER READING CORRECTION something interesting, you can be sure Applied Geophysics, W.M. Telford, L.P. Crystal X1 should be 3·6864MHz (as in where it was without having to repeat Geldart, K.E. Sheriff, D.A. Keys. Cambridge Fig.5), not 3·2768MHz as in the compo- the survey! University Press. ISBN 0521-20670-7. nents list. EPE BINDERS KEEP YOUR MAGAZINES SAFE – RING US NOW! This ring binder uses a special system to allow the issues to be easily removed and re-inserted without any dam- age. A nylon strip slips over each issue and this passes over the four rings in the binder, thus holding the mag- azine in place. The binders are finished in hard-wearing royal blue p.v.c. with the magazine logo in gold on the spine. They will keep your issues neat and tidy but allow you to remove them for use easily. The price is £6.95 plus £3.50 post and packing. If you order more than one binder add £1 postage for each binder after the initial £3.50 postage charge (overseas readers the postage is £6.00 each to everywhere except Australia and Papua New Guinea which costs £10.50 each). Send your payment in £’s sterling cheque or PO (Overseas read- ers send £ sterling bank draft, or cheque drawn on a UK bank or pay by card), to Everyday Practical Electronics, Wimborne Puublishing Ltd, 408 Wimborne Road East, Ferndown, Dorset BH22 9ND. Tel: 01202 873872. Fax: 01202 874562. E-mail: Web site: Order on-line from: We also accept card payments. Mastercard, Visa, Amex, Diners Club or Switch. Send your card number and card expiry date plus Switch Issue No. with your order.368 Everyday Practical Electronics, May 2003