CA1170756A - Remote seismic data system - Google Patents

Abstract

REMOTE SEISMIC DATA SYSTEM

ABSTRACT

A fiber optic seismic exploration system is provided.
The system includes a plurality of remote data gathering units, a recorder takeout unit, and a central control unit interconnected with dual fiber cables. The cables are terminated at each end with a connector housing a digital, logic compatible transceiver. The digital transceiver includes an optical detector connected to one of the optical fibers and a fiber optic transmitter connected to the other optical fiber. The seismic system uses Manchester II code for data transmission.

Description

GSCC : 1 1 0 7~G

REMOTE SEISMIC DATA SYSTEM

BACKGROUND

The invention relates generally to ~eismic explora-tion systems and, more particularly, to Eiber optic seismic exploration systems.

In canventional seismic exploration systems, s~veral hundred to a thousand seismic sensor groups, each com-posed of one or more individual sensors or geophones, are utilized to obtain seismic data. Usually a sensor group will contain one to thirty geophones electrically inter-connected to form a single data channel. Conventional systems utilize a multiconductor seismic cable containing many conductor pairs, one pair for each sensor group, to transmit the seismic data from the sensor groups to a multichannel data processing and recording unit. As these conventional cables are extremely heavy and bulky, han-dling of such cables is very time consuming and, there-fore, very expensive.
:
~ Recently, the use of fiber optics as a telemetry ; link for remotely distributed~seismic systems has been suggested, for example, see United States Patent No.
4,117,448 to Siems. Siems discloses a fiber optic cable with a set of triple-redundant optical fibers. The fibers . -- ' ~": ' " ' ' .

~7~

-2-are terminated at each end by a cable connector housing a light emitting diode (LED) and a photodetector. Siems suggests the transmission of digital data over the optical fibers. However, as the LED driver circuit and the photo-detector amplification circuit are housecl within the re-mote seismic data gathering unit to which the connector is electrically connected, the digital data is not logic compatible. In other words, the signal amplitude of the digital data is much too low to interface directly with 10` standard logic families, such as transistor-transistor logic (TTL), complementary metal-oxide semiconductor logic (CMOS), and emitter-coupled logic (ECL). Utilization of the arrangement disclosed in Siems is apparently not practical, as the digital data may become attenuated and distorted due to line capacitance, circuit loading and R~I/EMI pickup. Such effect results in ~ostly inaccurate and unuseful data gathering.

It would be advantageous to provide a fiber optic seismic exploration system which has practical application in a field environment. It would also be advantageous to provide a fiber optic cable connector/transceiver which uses digital signal transmission to avoid irreparable attenuation and distortion of control and data signals.
These use~ul results have been achieved by the present invention.

SuM~ARY
The in~ention in one broad aspect pertains to a seismic exploration system, comprising a plurality of remote seismic data gathering units, a fiber optic cable, including a duplex optical link, for interconnecting adjacent remote units, and a seismic cable connector having a digital, logic compatible optical transceiver connected to each end of the cable for mating the cable to the remote unit.
.

....,.. -~

.

-2a- 1~7~756 The invention further pertains to a method for gathering seismic data in a seismic exploration system, comprising the steps of receiving liqht energy representat-ive of a digital, logic compatible signal in a seismic cable connector, transforming the light energy into the digital, logic compatible signal within the connector, and electrically transmitting the diqital signal to an adjacent remote seisrnic data gathering unit.
More particularly, the present invention provides a seismic exploration system which includes a plurality of remote seismic data gathering units (RUs)~ a recorder takeout unit (RTU), and a central control unit ~CCU). The RUs, RTU, and CCU are interconnected by fiber optic cables which are terminated on each end by a seismic ~. ~

. . . ~

' ' , ~ .

.

~17C`~7~;

-3-cable connector housing a digita}, logic compatible optical transceiver. The digital transceiver includes an optical detector and a fiber optic transmitter.

In a presently preerred embodiment, the Eiber optic cable includes two optical ~ibers extending lengthwise through the cable and terminating in the connector/trans-ceiver. One of the fibers is connected to the optical detector of the transceiver while the other fiber is con-nected to the fiber optic transmitter.

Manchester II code is used to transmit data between the central control unit and the remote units. The recorder takeout unit is provided to transmit data to and from the remote units connected to either side of the ~U.
The seismic system oE the preEerred embodimen~ transmits data in multiples o~ eight-bit groups, with each data word ending in a logic low level to conserve energy.

These and other advantages and features of the pre-sent invention will hereinafter appear, and for purposes of illustration but not of limitation, an exemplary embodiment of tbe present invention is shown in the appended drawings and described in the following detailed description.

BRIEF DESCRIPTION OF THE DRAWINGS

FIGURE 1 is an isometric view of a seismic cable connector/transceiver, with the cover removed, in accor-dance with the present invention.

FIGURE 2A is an exploded view of the connector/trans-~; ~ ceiver of FIGURE 1 (without the cover).

' ~ ' ,' , . , ,:

-4- ~ ~ 7~ 7~

FIGURE 2B is an enlarged cross-sectional view of the fiber optic cable which is used with the connector/trans-ceiver of FIGURE 1.

FIGURE 3 is an electrical schematic of the Eiber optic transmitter circuit which is used in the connector/
transceiver o FIGURE 1.

FIGURE 4 is an electrical schematic of the optical detector circuit which is used in the connector/trans-ceiver of FIG~RE 1.

FIGURE SA, SB, and 5C are illustrative digitaL sig-nals transmitted by the connector/transceiver of ~;tGURE 1.
FIGURE 6 is a mechanical schematic of a connector which i9 used in conjunction with the connector/trans-ceiver of FIGURE 1~ -FIGURES 7A-7E are block diagrams of seismic explora-tion systems employing the connector/transceiver of FIGURE
1, Fig.7A appearing witl1 Figs. SA, 5B, and 5C.

FIGURE 8 is an illustrative view of a remote unit which is used in FIGURE 7.
.
FIGURE 9 is a block diagram of the circuitry of the remote unit of FIGURE 8.

FIGURES 10-24 are more de~ailed schematic diagrams of ~arious components of the block diagram of FIGURE 9, Fig. 12 appearinq with Fig.~lO, Fig.I6 appearing with Figs. 14 and l9A, Fig. l9A appearing with Figs.l4 and 16, and Fig. 20 appearing with Fig. 18 and Fig. 25.

.. . . .
:

: , , _5_ ~ ~7~j't75~
FIGU~ 25 i~ a ~ch~matic ~i~.3r~m of the r~corder take-out ~mit-~en-tral control Ullit in-terface shown in FIGUR~
7A, FicJ 25 appearincJ with Fic~s. 18 and 20.
FIGURE 26 is a block diagram of the central control unit shown in FIGURE 7A.

_TAILED DESCRIPTION
Throughout the following description, similar refer-ence numerals refer to similar elements in all figures o~
the drawings.

Referring to FIGURES 1 and 2, shown is a seismic cable connector/transceiver 10 in accordance with the present invention. The connector 10 has a shell or cover 12 which attaches to a mounting member 14. A rubber gasket 16 is provided to form an envlronmental seal between the member 14 and the cover 12.

A hollow rectangular member 18 i9 secured to the mounting member 14 for r~eiving one end o~ a fiber optic cable 20~ The member~ 14 and 13 may be a single integral piece The cable 20 extends through a gland 22 which is environmentally connected to one end of the rectangular member 18 by a nut 24.

The iber optic cable 2Q has a jacket 26, preferably made of polyurethane, which surrounds two optical fibers 28. A buffer tube 3~ surrounds each optical ~iber 28 to aid in preventing the optical fibers from being damaged when they are looped or otherwise bent within the cable 20 or connector 10. A strength member 32 extends through the 3~ center of the cable 20 and is looped~around an annular extension 34 secured to the mounting member t4. The strength member 32 is rigidly secured to the member 14 by a screw 36. The strength member may be a Kevlar fiber, an Aramid fiber, a high tensile strength plastic fiber, or the like and is provided to relieve ten~ion from the fibers when external forces are applied to the cable.

Each end of the cable 20 also includes four twisted pair wires 38 for carrying the analog output signals of four sensor geophone groups, as more fully discussed below in connection with FIGURE 7. The four twisted pair wires 38 are soldered or otherwise connected to elght connec-tions oE a sixty-one pin electrical connector or plug 40 which is secured to the mounting member 14 at the end opposite the rectangular member 18. The electrical connector 40 may be of any conventional type such as that manufactured by Bendix. A filler material 42, such as Kevlar filler, is provided within the fiber optic cable 20. In the presently preferred embodiment, the twisted pair wires 38 and the buffer tubes 30 ~containing the optical fibers 28) are helically wound around the ~3trength member 32 to provide additional flexibility o~ the cable and to aid in relieving tension and compression force~ on the fibers.

As shown, one of the optical fibers 28 is terminated in a fiber optic transmitter module 4~ through a fiber optic connector 46. The fiber optic connector may be any conventional type, such as an Radiall Model No.
F706.015.000, while the optical transmitter module 44 may be any conventional type such as Model No. SPX 4140 manufactured by Spectronics.

The second of the optical fibers 28 is terminated through a fiber optic connector 46 in an optical detector 48. Both the photodetector 48 and the optical transmitter 3Q 44 are mounted on a printed circuit board assembly (PCB) 50, which is secured to the mounting member t4. The PCB
50 contains transistor-transistor logic (TTL3 circuits which activate the transceiver as more fully discussed t below in connection with FIGURES 3 and 4. The PCB 50 is connected to the plug 40 by a plurality of wires (not shown), as discussed below.

- .

.
.

-~I'7~`'75~

When assembled, the cable connector/transceiver 10 provides an environmentally sealed optical cable connector which protects the transceiver and the ends of the optical fibers rom being affected or damaged during field use.
S The connector is readily connectable to a remote seismic data gathering unit or box (FIGURE 7) via the plug 40. A
lever or wrench 52 enables the connector 10 to be locked into place when plugged into a box.

Ref~rring now to FIGURE 3, an electrical schematic diagram of the optical transmitter module is shown. The fiber optic transmitter 44 includes AND gate ~2, acting as a bufer, connected to the base oE an N-P-N transistor 64 over line 66. The emitter of transistor 64 is ~rounded through a resistor i58. The collector of tran~istor 64 is connected over ~ine 70 to a light emitting diode ~LED) 72. A pullup resistor 78 is provided to compensate for the parasitic capacitance created by the electrical connection to the remote unit ( FIGrJ~E 7 ) .

In operation, the output of buffer 62 goes high, i.e., the output becomes a logic 1, when a digital signal appears on llne 82, turning on transistor 64. The LED 7 is energized and transmits light energy over one of the optical fibers.

Referring now to FIGURE 4, an electrical schematic diagram of the photodetector amplification circuit is depicted. A photodiode 88 is connected to a fiber optic receiver integrated circuit (IC) or preamplifier 90. The preamplifier 90 may be a Spectronics Model No. SPX 3620 or the equivalent. As light energy strikes photodiode 88, a ~ias current feeding into pin 1 of preamplifier 90 is modulated. The bias current is converted into a voltage signal within the preamplifier 90, and the voltage signal . ~ :

'75~

is compared to a threshold voltage level. When the threshold voltage is exceeded, a logic 1 appears at pin 12 of the preamplifier. The preamplifier IC contains an automatic gain control ~AGC) circuit which gain ranges the voltage signal representative of the light energy striking the photodiode 88.

A voltage holding capacitor 96 is provided to estab-lish the attack rate of the AGC voltage as required. An 10 AGC override circuit 98 is provided to limit the AGC gain to levels which wi}l not result in noise amplificat:ion sufficiently high enough to produce false logic out:puts~
A transistor 104 is switched on by a low level signal AGC
Con~rol at its base. This enables the AGC override 98 to a~ect the receiver 90 gain. The override is only enabled at times when no data in light form is present at the photodiode 88. Immediately upon receipt of valid data, the transistor 104 is switched off to allow the receiver 90 to gain range to the optimum level for proper data reception.
The output of the fiber optic receiver IC 90 (pin 12) is connected to a non-inverting input of a voltage compara-tor 120 through a high frequency filtering resistor 122. A
symmetry restoration circuit 124 is connected to the invert-ing input of comparator 120. The symmetry restoration cir-cuit is provided to restore the digital output of IC 90 which may be distorted by various system rise time versusfull time differences and/or by pulse broadening effects of the optical fibers.

, FIGURE 5A is an example of a digital signal as it should appear when transmitted from the connector/trans-ceiver 10 to a remote unit ( FIGURE 7) when unaffected by the aforementioned distortion effects. FIGURE 5B shows the digital signal as it would appear if affected by pulse . :'. ' ' : ' ' ~:~'7~'7~i~

broadening~ The symmetry restoration circuit 124 restores the distorted digital signal to its original condition, as shown in FIGURE SC. This is accompliched by a feedback loop which detects unsymmetrical data within the remote s unit, as more fully discussed below, and generates a correction voltage to be applied at TT6 (FIGURE 4).

Referring to FIGURE 6, shown is a mechanical schema-tic diagram of the plug 40. Lines 126 indicate the con-nections for the twisted pair wires 38 (FIGURE 2). Lines128 correspond to the electrical connections between the plug 40 and various points in the circuits o~ FIGURES 3 and 4, as indicated by the alphanumerals.

F~GURE 7A i~ a block diagram of a Eiber optic seismic exploration system 130 oE the present inVentiQn. ~s shown, remote data acquisition units ~RU) 132 and recorder takeout unit (RTU) 134 are interconnected through the dual fiber optical cables 20 by the connectors 10. The cables may have eight takeouts 136 for connecting an array of geophone groups 138. Utili2ation of the takeouts in the cables 20 is practical, as unwanted cross-over signals are not a problem in fiber optic cables. The recorder takeout unit 134 may be placed anywhere within the array of inter-connected remote units. A central control unit (CCU) 140is connected to the recorder takeout unit 134. The RTU
serves as an input/output device for the CCU. It is a modified RU without analog circuitry but with additional transceiver support circuitry. The RTU reroutes the optical data from both directions to the CCU, allowing independent line operation in either direction.

.

., .

~ ~ 7~!~S ~

FIGURES 7B-7E are block diagrams of alternate config-urations of the seismic exploration system 130. In FIGURE
7B, an au~iliary array of geophone groups 142 is connected to some of the takeouts 136 in place of a portion of the standard array 138. Such configuration has application where more than one array type i5 needed to gather dif-ferent seismic signal types. For example, the standard array 138 may record low frequency or p-wave data, and the auxiliary array 142 may be used for high frequency or shear-wave data.

FIGURE 7C is a configuration without cable takeouts on the data link cables 20. The geophone groups 1 3a are connected directly to remote units 132, Such a system may be desired when crossfeed between geophone inputs is o~
primary importance, and when other than in-line spreads are utilized.

F~GURE 7D represents a configuration with inputs from standard MIN-MAX* array 139, such as disclosed in U.S~
Patent Nos. 4,024,492 and 4,151,504, which are incorporated herein by reference. MIN-MAX amplifiers 143 are provided between pairs of geophone groups and cable takeouts 136.
FIGURE 7E typifies a MIN-MAX configuration in which the amplifiers are moved to the remote unit preamp inputs. It is to be understood that any desired configuration of geophone arrays may be utilized and still remain within the contemplation of the present invention.

.
* MIN-MAX is a trademark of Geosource Inc., the assignee of the present invention.

--- , . -- , : ,, , ' . .
.. :

s~

Shown in FIGURE 8 is a remote data acquisition unit or box 132. The remote unit or box has a detachable D-C
power supply or battery pack 142 for powering the elec-tronics within the box 132 and within the connectors 10 which are connected to that box. Four sixty-ane pin male connectors 144, two each on opposite sides of the box, are provided. Two of the four connectors, one on each side of the box, are mated with plug 40 of the connector/trans-ceiver 10~ The other two connectors 144 are provided as an alternate connection for geophone groups 138. Rather than connecting the geophone groups to the takeouts 136 in the cables 20, the geophone groups may be connected directly to the boxes 132r as previously discussed in connection with FIGURE 7C. The box 132 is constructed o structural foam, ~hereby considerably reducing the weight of the box.

Each remote unit contains the necessary circuitry to preamplify, filter, gain range, and digitize the analog inputs from the geophone groups. Referring to FIGURE 9, shown is a block diagram of the circuitry of remote unit 132. As shown, eight analog inputs, four from each of two standard or alternate connectors 144, are fed over lines 150 to multiplexer 152, which selects either standard or alternate inputs. The analog inputs are filtered for high frequency and amplified in quad preamps 154 prior to being input into track-and-hold, instantaneous floating point (IEP), and analog-to-digital (A~D) module 156. All analog input signals are sampled simultaneously by the track and hold network, and are gain-ranged from 1 to 32,768 times to near A/D midscale by the IFP, as disclosed in U.S. Patent Nos. 4,104,596 and 4,158,819, which are incorporated herein by reference. The resulting digitized mantissa and gain words for each original input or channel are fed to output .: ` ' , 7~;

formatter 158, which loads the parallel data into a serial output buffer for transmission via optical link 20 to CCU
~40.

Timing and control unit 160 functiona as a controller for the entire remote unit. It receives and decodes con-trol data from the CCU through receivers 162 to initiate the sampling and digitization process. All channels are sampled simultaneously, gain-ranged, and digitized accord-ing to the control logic sequence. At the appropriate time, the timing and control logic provides digital data to transmitters 164 for transmission to CCV 140. Control pulses received from the CCU contain an opera~ion code, a group o~ ive bits which the timing and control unit decodes into RU setup parameters, such as relay selection, K gain~ ~ilter selection, function generator control, etc.

Power supply board 166 utilizes bat~ery pack 142 to develop regulated voltage supplies via DC-to-DC converters.
Power on/off interlock 168 is provided to enable remote unit power-up and operation when one or more cable con-nectors 10 are engaged, as more fully discussed below.
Buzzer 167 is provided primarily for theft protection;
however, it also provides indication o~ faulty operating conditions. Shown in FIGURE 10 is the buzzer alarm logic.
Signal CONNECTORS ON is generated when either of the two connectorttransceivers is attached to the RU. The signal CONNECTORS ON clocks one-shot 300 which momentarily activates the buzzer, indicating that battery pack 142 is not dead.

Unauthorized disconnect register 302 and unauthorized power down register 304 are enabled whenever the RU is powered up. The reset of these registers is controlled by the CCU. When both connectors are removed, signal CONNEC-TORS OFF goes high. If unauthorized disconnect register 302 .~.

- - , - -- :. . . ..................... .
.

75~;

-t3-is set when signal CONNECTORS OFF goes high, AND-gate 306 is triggered, thereby energizing the buz2er. This theft protection is operable with the RU powered up or powered down. When communication to the CCU is interrupted, signal POWER DOWN goes high. If unauthorized power down register 304 is set when signal POWER DOWN goes high, flip-flop 308 and AND-gate 310 are activated, thereby energizing the buzzer.

The buzzer may also be enabled by an external voltage check circuit (not shown).

Referring now to FIGURES 1lA and 1lB, shown is a block diagram of a single channel of quad preamp/filter 154. When test relay 170 i~ enabled to the external inputs position ~as shown), geophone analog input signals are fed into preamp or ~ Gain stage 172. K Gain stage 172 may be remotely programmed to gains o 4, 16, 64, or 256 by register 17~ Low-cut pre~ er 176 receives the ~o output from preamp 172 and serves to low-pass filter the signal prior to its input into switched low-cut filter 178. Low--cut filter 178 may be configured as a O (OUT), 12, 24, or 36 d8 per octave high-pass filter by proper stage selection with switches 180 and 182, which are set by register 184. Low-cut corner frequency is determined by the duty cycle of low-cut frequency clocks 186, as more fully discussed below in connection with FIGURE 12.

Low-cut filter 178 is followed by 50 or 60 Hz strap-pable notch filter 188 which is remotely selectable aseither "in~ or Nout" with switch 190. Notch filter 188 consists of two cascaded second order notch filters to provide better than 60 dB attenuation over a 0.2 Hz `~
bandwidth.

. . =

"~ -- - : : . ., ' .

~7~7Sf~

Seventh order switched elliptical anti-aliasing filter 192 follows notch filter 188. The corner fre-quency is remotely selected by the duty cycle of alias frequency clock 194, as discussed more Eully be}ow in connection with FIGURE 12, at one of ten frequencies in two frequency bands of 44 to 500 Hz (standard) or 62.5 to 714 Hz toptional). The standard slope is 96 dB/octave MAX while the optional slope is 76 dB/octave.

Anti-aliasing filter 192 is followed by post-aliasing filter 196 which removes switching transients from the signal. The output oE filter 196 is AC-coupled in~to gain stage 198 which functions as an output bu~fer and gain~
adjust mechanism. The output is then switched to the track-and-hold (FIGURE 13) by control logic 222 using switch 200.

Low-cut filter 178 and anti-aliasing ilter 192 use pulse width controlled signals for controlling filter corner frequencies. Shown in FIGURE 12 is a schematic diagram of the switched filter control circuit 232 which generates the necessary clock pulses and provides the CCU
interface for pulse width control. This circuit also synchronizes tbe filter switching with the control clock, thereby minimizing the transient signal noise received by the track-and-hold.

Clock circuit 312 generates a 640 KHz or 320 KHz clock signal, depending on whether anti-alias filter 192 corner frequency is set above or below 250 Hz (by command word data). The clock signal is divided by 80 in counter 314 to yield an 8 KHz or 4 KHz period for filter control.
The 8 KHz/4 KHz signal is used as a reset signal for counters 316 and 318 and latches 320 and 322. Counters 316 and 318 control the pulce width of the filter control :
~.. ..... .
;
.

.

S~

signals by setting latches 320 and 322, respectively, at a time determined by command word data. The end result is that the filter switching control period is controlled by the antialias corner frequency which selects either an 8 KHz or 4 KHz reset cycle. The pulse widths o~ the filter control signals, which correspond to the respective duty cycles, are determined by the values loaded into registers 324 and 326 (which preset counters 316 and 318, respec-tively) by a CCU command word over line 328.
In the presently preferred embodiment, there lare four preamps/filters per card, a}ong with associated data registers and support circuitry. Two quad preamp carda 154 are provided in each box 132; however, control logic 160 can accommodate a single card.

FIGURE 13 is a block diagram of track-and-hold, IFP, and A/D module 156. Each preamp 154 output is fed into a corresponding track-and-hold (T/H) 202a-202h ~a total of eight) prior to reception of a command word from CCU 140.
When the command is detected, T~H 202a-202h are simultane-ously switched from the tracking mode to the hold mode, thereby providing minimum sampling skew.

According to the established box direction and sam-pling rate, successive channels are multiplexed by MUX
switch 206 to IFP gain-ranging amplifier 208. MUX control is provided by channel address counter 210. The basic purpose of IFP amplifier 208 is to amplify the analog input signals to a value near the full-scale range of A/D
convertor 212, usually between one-half and full scale, and to provide a digital code corresponding to the actual gain applied to the input signal.

~, . . .

The signal is held constant during the gain-ranging process. This process use~ a successive approximation logic sequence to apply appropriate IFP gain stages 208 to amplify the hold sample by the necessary gain 2n, where n = 0, 1, 2,... , 15. Level detector 214 is enabled at the end of each IFP gain-stage time cycle. The amplified signal is then sampled by A/D track-and-hold 216 and con-verted to digital data by A/D converter 212 for transmis-sion along with the gain code generated during the gain-ranging process.

To prevent various stages oE IFP amplifier 208 from introducing significant o~fset errors, a correction volt-age is subtracted from each stage. Periodically, the offset of each IFP stage is detected by oEfset logic 218, and the correction voltage is incremented by a small amount in a direction that will reduce the offset.

MUX averager ~20 is used to correct preamp offset errors prior to gain stage amplification. Each channel's error correction voltage is summed into its held signal to cancel the offset during its MUX time. The correction signals are updated each scan time.

Control logic 222 provides the timing and control for T/H, IFP, and A/D module 156. This logic controls the T/H
MUX timing, IFP gain stage switching, A/D converting, and data register storing.

System size is input from RU timing and control logic 160 (FIGURE 9) to channel address counter 210. Four or eight channels are selected, based on the desired field spread and scan rate. When channel selection other than four or~eight is required, data is still multiplexed to IFP
amplifier 208 as four or eight channels, but unnecessary 11 7~ 7S~

data is stripped before transmission to data link 20 by output formatter 158.

As shown in FIGURE 14, mask control logic 329, which is included in logic 160, controls the digitized analog data to be formatted and transmitted to the CCU. Control is achieved by the CCU loading a mask bit for each channel, which enables the selection of specific channels for data input to the CCU. Two 8-bit registers 330 and 332 are provided so that dynamic sampling may be achieved. As only one register is used at a time, the other register may receive updated mask bits. The transition between registers 330 and 332 is controlled by the CCU.

Each analog channel is dedicat~d to a particular geophone group, and one of eight different MUX address codes 000, 001, 010, ..., 111 is dedicated ~o each channel.
Whenever data is available Erom any one channel, its address code is presented to multiplexer 33~. Multiplexer 334 selects the corresponding mask bit for that channel and presents it to output formatter 158 ~FIGU~E 15). A
nl" mask bi~ allows the formatter to convert channel data for transmission; a "0" causes the formatter to ignore this channel.

Referring to FIGU~E 15, shown is a block diagr~m of the remote unit timing and control logic 160. ~ox "power on" control interlock 168 cycles power to optical receivers 48 (FIGURE 2) in connector/transceiver 10 until transmission is detected, at which time main power supply 166 (FIGURE 9) is ena~led. If no optical reception occurs for a specified period of time, "power on" circuitry 168 senses the inactivity and returns to the cyclic power on/off mode.
;~ 35 ._, ................. . .
' ~ , ~ .
;:, . : ~

. . . . .
: . .

Because connectors 10 are mechanically identical, box direction/box on line control circuit 224 is provided.
Circuit 224 determines the direction of incidence of the first (command) signal and establishes this as the CCU
direction. With this information, the remote unit may be set up for proper preamp multiplexing regardless of con-nector interchange.

Each powered-up RU begins in a repeater mode. As shown in FIGURE 16, after a command word (CW) transmitted by the CCU passes through a RU, signal CW RECEIVED clocks box-on-line flip-flop 336, which switches out receiver lines 338 and connects RU data formatter/encoder 158 and end-of-transmission ~EOT) generator 230 (FIGUR~ 15) to appropriate optical transmitters. The RU has now been taken out of the repeater mode and is awaiting rec~ipt of the EOT signal ~rom tha previous up~tream RV. After detecting this EOT signal ~which is blocked f~om passing through), the RU transmits its encoded data up~tream to the CCU. Thus, the data from this RU i9 inserted behind the data from the previous RU. By controlling when the ; EOT is transmitted from the previous RU, the gap between the two data bursts is controlled.

After the encoded data has been transmitted to the CCU, the RU transmits an EOT signal to the next downstream RU. At this time, signal DATA TRANSMITTED clocks flip-flop 336, which disconnects formatter/encoder 158 and EOT
generator 230 and places the RU back in the repeater mode.
To ensure that the entire command word has been re-ceived by the RU, the RU remains in the repeater mode for a time period of preferably nine microseconds after signal CW RECEIVED is generated. This time period is longer than .~, , . , "
. . .
.

-1 9- ~ '75~

the time required to receive a command word, but shorter than the time lag between the generation of signal CW
RECEIVED and the EOT signal.

Manchester II decoder and serial-to-parallel conver-sion logic 226 (FIGURE 15) functions as an input decoder and data formatter. Manchester II code tMan II code) is utilized because of its "self-clocking~ asynchronous format. All transmissions are burst mode Man II code to conserve power. As shown in FIGURES 17A and 17B, logic 226 decodes the incoming signal and stores it in a paral-lel register for output to command word decoder 223.

As shown, a Manchester II clock pulse ~Man II ck) i9 generated by EXCLUSIVE OR-gate 340 and inverter 342 each time a transition occurs in the Man I~ code. 1'he Man II
ck clocks one-~hot 34~ which has a time constant oE 3/4 bit cell time. The Man II ck also clocks J/X flip-flop 346, with the 3/4 bit cell time tied to the J inpu~. The Q output of flip-flop 346 is Manchester II data (repre-sentative of either seismic data being transmitted up-stream or command data being transmitted downstream). If a Man II ck occurs during the 3/4 bit cell time, the Q
output of flip-flop 340 goes high, indicating that the Man II data is a "1." The 3/4 bit cell time also clocks the Man II data into 33-bit serial-to-parallel shift register 348. If a Man II ck occurs during the 3/4 bit cell time, a n 1 n is loaded into register 348. Otherwise, a non is clocked into register 348.
A preamble of all zeros precedes a command word. As a result of repeated transmission, the leading bits of the preamble may become distorted and decoded as a n 1 . ~ To prevent false 1's from being detected, preamble stripper logic 350 is utilized. When the preamble is received, the . . . ,, ~ - : - -~- ' : - :
, ~7~ 5f~

first transition of Man II code causes flip-flop 352 to go high. This high output is delayed by strip time ~C time constant circuit 354, preferably two microseconds. After the time delay, one-shot 344 is allowed to fire. Flip-flop 352 is reset after all CW data is received.

Command word decoder 228 decodes the parallel command word into the following:

1. Preamble - a 24 to 56 bit all "O's" data train used by optical receiver symmetry restoration circuitry f24 ~FIGURE 4) to maintain input data integrity. RU logic effectively ignore~ this segment of the command word;
tS
2. Sync Code - a two-bit sequence signifyin~
decoder start;

3. RU Address - a nine-bit code unique to each RU.
~ RU logic ignores all codes except all n 1 ~ S~ and its box address. Addresses are assigned during the RU power-up sequence;

4. OP Code - a five-bit code defining the opera-tion to be performed. For example, a box address assignment has Op Code 00000;

5. Data Word - a fifteen-bit code which is loaded into registers for preamp control box set-up (mask), fun~tion generator, etc.;

6. Stop Bit - a one-bit code, a "1" used as a data validity check bit, indicates that the data is good;

.

' `

~L 7'G75~f~

~,

7. Delay - a calculated N bit data delay which allows each RU to dump its data onto the optical link before receiving additional data from the adjacent RU; and

8. EOT - a four-bit code which signifies end-of-transmission of the command word and the start of data output to the CCU.

As stated above, the command word includes a nine-bit code which is used to give each RU an individual identity in order to keep its data separate. As shown in FIGURE
18~ nine-bit register 374 is provided in each RU, afford-ing 5~2 possible dif~erent combinations. Comparator 376 is included to identify any CW box address that matches ~he data in register 374. ~ll 1's detector 378 is pro-vided to detect an all 1's address code~

When the RU powers up, register 374 is cleared so that the box address is n 000000000 . ~ A special command word transmitted by the CCU assigns a binary number between one and 510 to the RU, as determined by the relative position of this RU with respect to the CCU
and the other RUS, which is loaded into register 374.
The numbers one through 510 are used as individual box addresses. An all zeros address code is used only during powering up, while an all ones code is used t~ enable the CCU to communicate with all of the RUs at the same time.

End-of-transmission detector generator 230 ~FIGURE
15) determines when an EOT code is sent to the next RU
so the latter can transmit its data. As shown in FIGURES
?9A and 19B, two individual EOT circuits control the transmission of data from the RUs. One detects the EOT
~ ~ 35 :

" ~ , .
.~.. ~ .................... . . .. .
. ..

~ . ~

5~

transmitted from the previous upstream RU; the second generates an EOT code for transmission to the next down-stream RU.

The EOT detector of FI&URE 19A includes counter 356 which is reset and enabled when a command word is received. Counter 356 is disabled after the EOT code is received. The command word ~CW) precedes the EOT code in the data stream. The time delay between them depends on the quantity of data transmitted by each RU and the position of the particular RU in the line. The more RUs betwean this RU and the CCU, the longer the delay between the CW and the EOT. The EOT code clocks counter 356, thereby triggering formatter/encoder 158 ~FIGURE 15) to begin data transmission.

The EOT generator o~ FIGURE 19~ is partially con-trolled by the CCU. ~ypically, EOT transmission i~ com-pleted approximately one microsecond after data trans-mission is completed. The CCU can change this time differential or gap by modifying the code loaded into register 358. A different code addresses another section o~ programmable read-only memory (PROM) 360, producing a different EOT position with respect to transmitted data (that is, changing the gap).

Counter 362 counts the number of digitized analog channels which are to be transmitted. Counter 362 addresses PROM 360, and the output of PROM 360 presets counter 364. As data is transmitted, counter 362 begins counting, and the output from the counter allows the EOT
code to be transmitted at the correct time.
!

-23~

Preamp control logic 23~, as discussed above in connection with FIGURE 15, determines the frequencies and duty cycles of the low-cut and anti-alias switched filter clocks 186 and 194 (FIGUR~ 11) and decodes and enables all other preamp setup and control functions defined in the preamp block diagram (FIGURE 11 ) ~

Blast command circuit 234 is provided to initiate a contact closure which activates an energy source. As shown in FIGURE 20, the circuit is first armed, and then a "fire" command is issued to cause the contact closure.
The fire command must be issued within one se~Qnd after arming, or the circuit i9 automatically disabled. The "arm" command is a logic low signal on line 3~6 (rleceived from the CCU) that resets timer 368, thereby removing the reset from flip-flop 370. When a logic high signal is received on line 366 ~indicative o~ a ~ire command), flip-flop 370 is clocked to a logic "l," and transistor 372 is turned on, causing the contact closure to the energy source control located outside the RU.

FIGURE 21 is a block diagram of output formatter/
encoder 158. In the presently preferred embodiment, a standard set of output data for each channel lncludes a four-bit gain word and a fifteen-bit data word. Output data from each remote unit is formatted into a sequence of words as follows: (l) a two-bit sync code; (~ a one-bit box fault code; (3) a nine-bit RU address; and (4) one to `~
eight sets of channel data. The total word length is -selected to obtain a "0" logic level at the end of trans-mission to conserve power between transmissions and to utilize eight-bit groups to simplify encoding. The following chart indicates the nine possible word lengths which may be transmitted from a RU, depending upon the number of channels being transmitted: -~: .
.
` ' .

.

C7~

NO. OF BITS
OUTPUT DATA T~ANSMITTED

SYNC ~ FAULT + ADDRESS + O Channels - 12 BITS 16 S SYNC ~ FAULT ~ ADDRESS ~ 1 Channel =~ 31 BITS 32 SYNC ~ FAULT + ADDRESS + 2 Channels =~ 50 ~ITS 56 SY~C ~ FAULT + AnDREss + 3 Channels - 69 BITS 72 SYNC + FAULT + ADDRESS + 4 Channels - 88 BITS 88 SYNC + FAULT + ADDRESS ~ 5 Channels = 107 BITS 112 SYNC + PAULT + ADDRESS + 6 Channels = 126 BITS 128 SYNC ~ FAULT + ADDRESS ~ 7 Channels = 145 BITS 152 SYNC + FAULT + ADDRESS ~ 8 Channels = 164 ~ITS l68 It is to be understood that the data output may have other formats and still remain within the scope o~ the present invention.

As ~hown in FIGURE 21, the formatter circuit includes counter 380 to count the number of channels formatted, PROM 382 to determine how many bits need to be shi~ted (based on the above chart), counter 384 to count how many shifts have occurred, and first-in, first-out IFIFO) memories 3~6 to stora the formatted data until transmis-sion. When the command word is received, the sync bits, fault bits, and box address are loaded into shift register 3880 Simultaneously, the required number of bits to be transmitted are transferred from PROM 382 to counter 38~.
As shift register 388 shifts, counter 384 counts. After ~; eight bits are shifted and presented to memory 386, a clock is generated to load the eight bits into memory.
This process continues until all full eight-bit words are transferred into FIPO memory, at which time counter 380 increments its count. Any left-over bits remaining in the shift register are used to complete the next eight-bit word. Two PIPO memories are provided so that one may be ~ ~ ;

` ~

~L 17~ ri~5~
--:2s--loaded with new data as the other is unloading data to the encoder circuit.

The encoder circuit generates the preamble, accepts and serializes eight-bit words from memories 386, and converts this serial data to Manchester II code. When an EOT code is received from the previous RU, a preamble of specified length (similar to the command word preamble) is generated, after which an eight-bit data word from memories 386 is transferred to shift register 390. The first word transferred is the sync bits, the fault bit, and part of the box address. Register 390 shifts at an 8 MHz rate into Man II coder 392 which generates the bit cell transitions. FIFQ memories 386 unload at a 1 MHz rate which matches the 8 MHz shift rate so that there i9 a continuous data stream out of Man I~ coder 3~2. Encoding stops when a signal from FIFO memory indicates it ls empty. The ~inal bit coded causes the output of Man II
coder 392 to be at a logi~ low level.
Referring to FIGURE 22, Man II coder 392 is enabled by taking the input to AND-gate 394 high. During genera-tion of the preamble, register 390 is cleared, causing all zeros data to be presented to coder 392. OR-gate 396 causes J/K flip-flop 398 to toggle at bit cell times.
Once generation of the preamble is completed, memory data is loaded into register 390. As the data shifts out, flip-flop 398 either changes state at mid-cell time when a n 1 n is presented to coder 392 or remains in its present state because a "0" is presented. After all data is coded, Man II coder is disabled by taking the input to AND-gate 394 low.

.

. ~

S t;

A block diagram of function generator 169 (FIGURE 9) is shown in FIGURE ~3. Test signals are generated sample by sample in digital form by the CCU. This digital signal is then transmitted to the remote units in the command word. In the remote units, the digital ~ignal is con-verted to an analog signal with a near full-scale peak value by digital-to-analog (DJA~ converter ~4a. The resulting analog signal is attenuated to the desired amplitude by digitally controlled attenuators 242 and 244.
The low level analog signal is then fed into preamp oscil-lator inputs 201 for test purposes.

FIGURE 24 is a block diagram of battery pack 142 and power supply 166. The battery pack includes two 12 volt lead-acid battaries connected in series with a common output. When a connector/transceiver is engaged with a rqmote unit connector 144, interlock circuitry 168 is enabled. Optical receivers 48 are cycled on and o~f to conserve power until light transmission is detected.
Signal VREG ENB then goes high to force power on/off control 246 to enable soft-start circuitry 248 and main power relay 250. Power is distributed to two DC-to-DC
converters 252 and 254. Resulting outputs are filtered with high-cut filters 256, 258, and 260 prior to distribu-tion for various box functions. Secondary relay 262 mustbe thrown to enable power to function generator 169.

The recorder takeout unit (RTU) is provided as an auxiliary interface between the CCU and the RUs. The RTU
regenerates Manchester II code being transmitted to and from the CCU. This regeneration is performed with high speed clocks and circuits to reduce bit jitter, allowing - lengthy interconnecting cables to the CCU. The RTU is unnecessary if the distance between the spread and the CCU is relatively short, such as less than 400 feet.

, , .. ,.. . . ~ ..

.

-~ 1 7 ~7 As shown in FIGURE 25, the RTU is connected to the CCU through input/output panel 400 (located in a truck carrying the CCU). Input panel 400 is the link between electronics inside the truck and electronics outsicle the truck. Manchester II code is transmittecl between the RUs and the input panel via fiber optics, but the input panel is connected to the CCU by twisted pair wires. Two fiber optic cables tonly one is shown) are connected between the RTU and the input/output panel. One cable is dedicated to transmitting data between the CCU and the RUs disposed to the left of the RTU while the other cable is dedicated to transmitting data between the CCU and RUS disposed to the right of the RTU. It is to be understood that the spread of remote units may be optically connected to the central control unit via the input/output, such as when the CCU ls within 400 Eeet of the spread, and still remain within the contemplation of the present invention.

Referring now to FI~URE 26, shown i-q a block diagram of CCU 140 and associated equipment. Microprocessor unit (MPU) 270 serves as the main controller for the system.
Input/output truck panel 400 is provided, as discussed above, to interface RTU 134 with Manchester II receivers 274 and Manchester II transmitters 276. Transmitters 276 accept command words from the MPU and convert them to serial data with Manchester II encoding for subsequent transmission to the remote units. Receivers 274 accept asynchronous data input from the remote units and decode the data. Receivers 274 also check for parity errors, i.e., transmission faults~

Formatter 278 formats the seismic data into a stan- ;
dard multiplexed format, writes this data into buffer memory 280, and decodes the data words requested by MPU

~:1'7Q7~6 270 by RU address number. The data formatting is compli-cated by allowances for dynamic sampling which result in some data being stripped out of the data stream, as previously discussed.

Buffer memory 280 is an interleaved set of two lK x 20 random accessory memory units (RAMs) which contains buffers for the seismic data and gain words. The buffer memory compensates for the unsynchronized input/output of the CCU with respect to digital unit 282. This data along with header in~ormation 284 is transferred synchronously to the digital unit. Digital unit 282 includes a recorder~
tape transport 284, and various auxiliary devices.

MPU 270 performs the following tasks:

1. A status table is used to store frequently used line parameters, such as preamp setting, box number, system size, etc. At system set-up time, the MPU controls the output of this data to the remote units.

2. The MPU formats the front panel display, decodes and registers the front panel switches, and converts the binary data into decimal form.

3. The MPU controls the roll-along operation of the remote units to allow automatic move-up, gap size, etc.
4. The MPU utilizes stored programs to set up, run, and analyze various Go/No Go tests for the system as well as system performance tests.

5. The MPU organizes data input/output and timing during the recording process.

. ' .
.

The CCU encodes the command word by the same method as the Manchester II coder, except that the command word is coded and transmitted at a 4M~z rate. The commands to be coded are generated by software. When no communication is required with the RUs, a command word must be sent to keep the RUs powered up so that they may acquire data for transmittal to the CCU. A command word of all zeros ex-cept for a n 1 n in the sync code is usled for this purpose.

The CCU decodes the Manchester II code in a manner similar to that of the RUs, using edge c}ocks, 3/4 bit cel} times, and shit registers ~not shown). While all command words that are decoded by the RU~ are 32 blts long, data transmitted to the CCU or decoding may be 16 to 168 bits long, as previously discussed. To accommo-date these diEferent word lengths, a counter and a PROM
decoder ~not shown) are used to determlne when data i5 available in the shift r~gisters. Two holding regi~ters, one used for the box address and the second used for the seismic data, are used in conjunction wit~ a third shift r~gister to hold incoming data to be transferred to buffer memory 280 while additional data is being loaded into the third register. A channel counter (not shown) is also provided to keep track of the number of channels. The box address, the channel number, and the recording tape format determine the location in the buffer memory in which the data is written. The box address and channel number are also used as an address for map memory 402 which is a random-access-memory (RAM) element. The contents of RAM
402 are used to address buffer memory 280. Buffer memory 280 is divided into two sections, and each section is alternately loaded and unloaded during each scan. RAM 402 addresses (loads) one section of buffer memory 280 while digital unit 282 addresses (unloads) the other section.

,. .. .
- , ' :' - .

.

'7 5 ~

The contents of RAM 402 are generated by MPU 270 from information supplied by the operator and sSored in the status table. Any RU seismic data, therefore, may be placed in any necessary location in buffer memory 280.

Certain verification checks are performed on incoming RU data. First, the number of bits received from each RU
is checked to verify that it was correct. Second, the fault bit is checked to determine if it is set. Third, the box address is checked to make sure it is being received in sequential order. If any of these checks detect faults, the box address of the RU along with the type of error detected is loaded into a memory and a flag is set. ~PU 270 s~ans these flags periodically and notifies the operator when a flag occurs.

There are also checks performed on the seismic data as it passe~ through buffer memory 270. A parity check is made to determine bad memories, and a ull scale check is perPormed on the seismic data to indicata an overload condition. Any detected faults are loaded into a memory, and a flag is set for the MPU.

Auxiliary channels 40A are provided in the CCU for a variety of purposes known to those skilled in the art, such as time break verification. The data is stored in an auxiliary buffer memory (not shown) and then transferred to buffer memory 280 at the appropriate time. A counter (not shown) addresses an auxiliary RAM memory (not shown) as each channel is transferred. Auxiliary channel data may be stored in any location as needed, like the RU
channel data.

'` ' ' ' .~ ~ - ' , ' ' 7~

In operation, a digital command signal may be trans-mitted from the central control unit 140 to the recorder takeout unit 134 (FIGURE 7A). The recorder takeout unit then transmi~s the digital signal to all the remote units by utilizing the digital connector/transceiver 10 of the present invention. The digital signal is electrically transmitted to the transceivers within the connectors 1Od and 1Oe. The digital signal is converted into light energy by the fiber optic transmitter 44 within each connector.
The light energy representing the digital command signal is then transmitted to the adjacent transceivers 10c and 10f, respectively, over one o~ the optical fi~ers 28 in the cables 20. The light energy i~ detected by photodiodes 88 within the connectors 10c and 10f. The light signal is transformed back into a digital signal, which i~3 elec-trically transmitted to the remote units 132b and 132c.The remote units then sense the digital command signal to see if it applies to them and electrically transmit the digital signals to the connector/transceivers 10b and 10g.
The optical fiber transmitters 44 within the connectors 10b and 10g convert the digital command signal into light energy which is transmitted to the next adjacent connector/
transceivers 1Oa and 1Oh, respectively, through the cables 20. The process is then repeated.

Regarding data transmission, the analog output signals of the geophone groups 138 are transmitted to the remote units 132 over the twisted pair wires 38 which are connected to the takeouts 136. In operation, seismic analog data is received by remote unit 132B, for example.
Remote unit 132B transforms the analog data into digital data, which is then ele~trically transmitted to connector/
transceiver 10c. The digital data is transformed into light energy by the optical transmitter 44 in connector 10c. The data is optically transmitted over cable 20b to .
.

connector/transceiver 1Od, where it is received by photo-detector 48. The photodetector trans~orms the optical data back into its digital electrical state and transmits the digital data to the recorder takeout unit 134.

As the remote seismic data system is nonpolarized, it does not matter which end of the cable 20 is connected to a remote unit. Once a cable is interconnected between two remote units, one of the optical fibers will only carry digitized command signals while the other fiber will only carry digitized seismic data signals. The cable 20 may carry command data and seismic data ~in opposite directions) simultaneously, thereby operating as a digital full duplex system. However, the presently preferred embodiment incorporates half duplex mode operation to preserve signal-to-noise performance.

It is to be understood that the invention will admit of other embodiments, such as a system using a single fiber bi-directional data link. The description of the preferred embodiment is given only to facilitate under-standing of the invention by those skilled in the art and should not be construed as limiting the invention.

.' ~'' ~. - . ' ,

Claims (39)

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. A seismic exploration system, comprising:

a plurality of remote seismic data gathering units;

a fiber optic cable, including a duplex optical link, for interconnecting adjacent remote units; and a seismic cable connector having a digital, logic compatible optical transceiver connected to each end of the cable for mating the cable to the remote unit.

2. The system of claim 1, wherein the optical link extends through a buffer tube within the cable.

3. The system of claim 1, wherein the fiber optic cable further includes a strength member extending length-wise through the cable.

4. The system of claim 3, wherein the fiber optic cable has a jacket with a plurality of seismic sensor takeouts, a twisted pair wire connected to each takeout and extend-ing through one end of the cable.

5. The system of claim 4, wherein the buffer tube con-taining the optical link and the twisted pair wires are helically wound around the strength member for providing additional cable flexibility.

6. The system of claim 1, wherein the optical link in-cludes two optical fibers.

7. The system of claim 6, wherein each of the optical fibers extends through a buffer tube within the cable.

8. The system of claim 6, wherein the digital optical transceiver includes an optical detector termi-nating one of the optical fibers and a fiber optic trans-mitter terminating the other optical fiber.

9. The system of claim 1, wherein the seismic cable connector further includes a multiple pin electrical connector wired to the digital transceiver and a self-locking lever for securing the multiple pin connector to a remote seismic data gathering unit.

10. The system of claim 9, wherein a power on/off circuit is provided in the remote unit for conserving energy, and wherein a plurality of pins of the multiple pin connector are associated with the power on/off circuit.

11. The system of claim 8, wherein the optical detector includes a photodiode and a fiber optic receiver integrated circuit associated with the photodiode and having an automatic gain control circuit.

12. The system of claim 11, wherein the automatic gain control circuit is associated with a holding voltage capacitor and an automatic gain control override circuit for eliminating the effects of noise.

13. The system of claim 11, wherein the optical detector further includes a symmetry restoration circuit associated with the digital output of the receiver integrated circuit.

14. The system of claim 13, wherein the symmetry restora-tion circuit includes a voltage comparator for correcting the duration of the digital output signals of the receiver integrated circuit.

15. The system of claim 8, wherein the fiber optic trans-mitter includes a light emitting diode and circuitry for energizing the light emitting diode.

16. A method for gathering seismic data in a seismic exploration system, comprising the steps of:

receiving light energy representative of a digital, logic compatible signal in a seismic cable connector;

transforming the light energy into the digital, logic compatible signal within the con-nector; and electrically transmitting said digital signal to an adjacent remote seismic data gathering unit.

17. The method of claim 16, further comprising the steps of:

simultaneously receiving a digital, logic com-patible signal in a seismic cable connector;

transforming said digital signal into light energy within the connector; and optically transmitting the light energy through a fiber optic cable.

18. A seismic exploration system, comprising:

a plurality of remote seismic data gathering units;
a central control unit;

a plurality of fiber optic cables for intercon-necting adjacent remote units and the cen-tral control unit; and a seismic cable connector having a digital, logic compatible optical transceiver con-nected to each end of the cable for mating the cable to the remote units and the cen-tral control unit.

19. The system of claim 18, wherein an input/output panel interfaces the central control unit with the remote units, the input/output panel being optically linked to the re-mote units and being electrically connected to the central control unit.

20. The system of claim 18, wherein a recorder takeout unit is disposed between any two adjacent remote units, the recorder takeout unit being a modified remote unit for regenerating digitized data to and from the remote units, the recorder takeout unit being optically linked to the central control unit.

21. The system of claim 20, wherein an input/output panel interfaces the central control unit with the recorder takeout unit, the input/output panel being optically linked to the recorder takeout unit and being electrically connected to the central control unit.

22. The system of claim 19 or 21, wherein the input/out-put panel and the central control unit are disposed in a mobile vehicle.

23. The system of claim 18, wherein each remote unit includes connector-on circuitry for momentarily enabling an audible signal when a cable connector is connected to a remote unit with normal battery voltage.

24. The system of claim 18, wherein each remote unit includes power down circuitry for enabling an audible alarm when communication with the central control unit is unintentionally interrupted.

25. The system of claim 18, wherein each remote unit includes unauthorized disconnect circuitry for enabling an audible alarm when both cable connectors are removed.

26. The system of claim 18, wherein burst mode Manchester II code is used to transmit data between the central con-trol unit and the remote units.

27. The system of claim 26, wherein data is transmitted in multiples of eight-bit groups.

28. The system of claim 26 or 27, wherein each data burst ends in a logic low level to conserve energy in transmis-sion.

29. The system of claim 27, wherein each remote unit is assigned a box address by the central control unit.

30. The system of claim 29, wherein an end-of-transmis-sion code is transmitted by the central control unit and each remote unit after a predetermined time delay follow-ing data transmission.

31. The system of claim 30, wherein each remote unit is in a repeater mode until an end-of-transmission code is received.

32. The system of claim 29, wherein command data transmitted from the central control unit comprises a multi-bit word including a sync code, a box address code, an operation code, a box set-up code, and a stop code.

33. The system of claim 29, wherein each remote unit gathers up to at least four channels of seismic data, the seismic data being transmitted in different word lengths of multiples of eight bits depending upon the number of data channels enabled, each data channel includ-ing a multi-bit code comprising data, each word including a sync code, a fault code, and a box address code.

34. The system of claim 33, wherein each remote unit includes mask control circuitry for selecting which of the available data channels is to be transmitted.

35. The system of claim 32, wherein a preamble of all zero bits precedes command data to provide data integrity and wherein each remote unit includes preamble stripper circuitry to prevent false data from being detected.

36. The system of claim 33, wherein a preamble of all zero bits precedes seismic data to provide data integrity and wherein the central control unit includes preamble stripper circuitry to prevent false data from being detected.

37. The system of claim 18, wherein each remote unit comprises circuitry to preamplify, filter, gain range, and digitize analog seismic data.

38. The system of claim 37, wherein the filtering circui-try comprises switched filters which use a pulse width controlled signal for filter cutoff point control, the pulse width being set by the central control unit, the filter switching being synchronized with a control clock to minimize switching transients.

39. The system of claim 18, wherein the central control unit includes circuitry for performing parity checks, full scale checks, bit count checks, fault bit checks, and box sequence checks.