EPOS-N - European Plate Observing System
Seismometers / vault construction
Stations: Upgrade/install notes
Ideas for improvements
Installation / Inventory
Last update: February 17, 2020, at 01:26 PM
In cooperation with Statoil we wish to place (temporarily) a seismometer on one of the three concrete platforms on the Gullfaks field. The seismometer will be positioned at the bottom of a platform leg.
Gullfaks-C concrete platform in 216 m water depth (image source)
Overview Gullfaks oil field - click to enlarge (source: NPD Fact Map)
Gullfaks oil field extension and shape, Gullfaks A/B/C platform locations (source: NPD Fact Map)
Bottom of Gullfaks C platform shaft is classified as Zone 1:
A place in which an explosive atmosphere consisting of a mixture with air of dangerous substances in the form of gas, vapor or mist is likely to occur in normal operation between 10-1000hrs per year. (Source)
The seismometer we use is Güralp model 6TD, a stand-alone unit that combines 3-component sensors, GPS input for time stamping, digitizer and recorder in one small cylindrical package.
The instrument requires 75 mA @ 12 Vdc input (without GPS attached). Mains voltage available is 230 Vac, so a AC/DC power supply will be needed.
Seismometer and AC/DC power supply will be mounted inside an explosion proof Ex-d container.
If mains supply disappears an Uninterruptable Power Supply (UPS) will provide 60 minutes of operation.
Seismograph Güralp Mod. CMG-6TD.
Güralp seismometer mod. 6TD mounted in EEx-d enclosure. Click to see pictures.
Specifications (subset of relevance here; source here):
1) GPS not used. We will use 75 mA in power estimates.
We will probably alternate between two nearly identical instruments, Serial numbers T6303 (512 MByte memory) and T6278 (4 GByte).
Both seismometers will be upgraded to 8½ GByte memory.
Güralp has found that, on average, three 100 samples per second streams consume around 50MB of storage a day.
By this rule, 8½ Gbyte should provide six months of storage.
See also calibration records in section above.
6TD typical frequency response plot (magnitude & phase)
ELC - AL911AE-12V-1A AC/DC power supply.
Power supply selected:
Even though the 6TD uses less then 100 mA (no GPS) we specified a supply capable of 1 Amp output to handle any peak currents at start-up. In addition, it's a linear supply with very low ripple voltage and good load/line regulation figures.
In addition to storing data internally, the 6TD seismometer also transmits serial RS232 data. If we could capture and store these serial data we would achieve an added layer of data security.
This serial data capture device can either share space with the seismometer inside the Exd enclosure, or be placed external to the container - provided there is a signal cable available. The first alternative seems most realistic. In the following some solutions will be outlined.
We are looking for a device which simply captures incoming RS232 serial data stream and dumps them onto e.g. a SD memory card. Requirement summary:
There are several options available:
It is possible to attach what is called "shields" to some common rapid prototyping family of controllers, like Arduiono and Mbed, see next section ...
The MBED is a microcontroller module that has been used successfully in some department projects.
If serial RS232 data signal lines already exists, these can be used to transfer data to laptop placed in nox-ex zone area. In that case, the "Scream" software manufactured by Güralp can be used for data logging.
An external GPS is normally attached to the 6TD seismometer for time stamping purposes.
It is important to notice that the GPS signal consists of two parts: a) An RS232 ASCII data stream, with time, position and other information, and b) a pulse, aptly denoted PPS = "Pulse er second", whose leading edge is synchronized within narrow borders, to start of a new second.
This means that the task of forwarding signals from a remote GPS which is necessary when the seismometer is placed in area where reception of GPS is obstructed - as in our case - gets more complicated. In addition to the serial RS232 data, the PPS must also be transmitted over longer distances.
One connector on the instruments servers several purposes: GPS input (both RS232 serial and PPS (Pulse per second), power input, data output, as described below.
FireWire disk used:
The other FireWire disk (used during readout of data in SFC project) now seems defective.
DIR command before data transfer.
After data extraction is completed, the result of a new DIR command will be:
DIR command after data transfer.
So data seems to have been transferred successfully to the FireWire disk. We now need to transfer them to PC, so we use the disks USB connector.
Windows Explorer reports that the disk is not formatted - we have to use a program from Güralp to extract data - gcfxtract
gcfxtract software to extract data from external hard disk using Güralp file format system.
Clock check: 4 seconds behind UTC.
As two seismometers will be deployed alternately, we wish to be certain that time stamp of data is identical on the two units.
However. there is a time synchronization issue we have to clarify before deployment.
In order to check correct time stamping of data, we first synchronize the two instrumentation systems (each consisting of 6TD seismometer and laptop) to UTC time in this manner:
Both Units Under Test (UUT) placed on concrete pillar in Seismolab, Realfagbygget, adjacent to NNSN-Bergen STS-2 sensor. North-south axis aligned with compass N/S.
6TD clock is 3 seconds behind UTC time (as seen by opening terminal window and issuing CLOCK? command).
Click to enlarge.
6TD clock is 4 seconds behind UTC time (as seen by opening terminal window and issuing CLOCK? command).
List of documentation.