Trillium Seismometer: User Guide
Trillium Seismometer: User Guide
Trillium Seismometer: User Guide
Seismometer
User Guide
Nanometrics Inc.
Kanata, Ontario
Canada
© 2002–2004 Nanometrics Inc. All Rights Reserved.
Trillium Seismometer User Guide
The information in this document has been carefully reviewed and is believed to be reliable. Nanometrics,
Inc. reserves the right to make changes at any time without notice to improve the reliability and function of
the product.
No part of this publication may be reproduced, stored in a retrieval system or transmitted, in any form or by
any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior written permis-
sion of Nanometrics Inc.
Nanometrics, Inc.
250 Herzberg Road
Kanata, Ontario, Canada K2K 2A1
Tel (613)592-6776
Fax (613)592-5929
Email [email protected]
i
Contents
ii
Figures
3-1 Good sensor alignment example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3-2 Mass position by expected operating temperature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3-3 Mass position adjustment access screws. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
4-1 Sensor axis orientations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
4-2 Nominal frequency response . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4-3 Self noise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
iii
Figures
iv
Tables
4-1 Axis orientation and polarity of XYZ outputs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
4-2 Poles and zeroes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
A-1 Performance specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
A-2 Operation specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
A-3 Physical specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
A-4 Power specifications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
A-5 Environmental specifications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
B-1 Connector pinout . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
C-1 Generic sensor cable wiring for CBL13942R2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
C-2 Generic sensor cable wiring for XYZ/UVW ± outputs of CBL13942R1 . . . . . . . . . . . . . . . . . 26
v
Tables
vi
Chapter 1 Introduction
This section provides preinstallation guidelines for Trillium. These guidelines are
intended to help achieve the best possible performance, but are not necessary for all
types of study (e.g. short-period) or site.
24 hours. However the pier may still generate spurious signals as the concrete cures
which can take two to four weeks.
ing, and a continuous outer shield provides good high-frequency RF shielding. The
outer shield should be earthed at the digitiser for safety.
The digital ground (DGND) must be used for the return currents of the control signals
(CAL_SIG, U_CALEN, V_CALEN, W_CALEN, UVW and ACC). Note that the ana-
log ground (AGND) is connected to chassis ground (CHGND) inside Trillium, so if
these signals are already connected at the digitiser, AGND should not be connected
through the cable or else a ground loop will be created.
Once the vault has been prepared, use the following procedures to install the Trillium.
Provided at the end is a checklist for quality control at a typical installation.
3.1 Unpacking
Trillium is shipped in a very sturdy triple-wall coated cardboard box with custom-cut
cushioning foam.
Open the box and set aside the user manual, maintenance kit and the top four layers of
foam to completely expose the sensor. To minimize the possibility of damaging the
sensor, do not remove it from the box until it is ready to be placed directly on the pier.
Be sure to save the foam in case the sensor needs to be shipped again.
After aligning the sensor to north-south, you may find that the sensor may need to be
relevelled and then realigned.
Note that for the most precise alignment possible two 3/8" diameter holes aligned to
east-west are provided in the sensor base. However the north-south alignment marks
will be precise enough for most installations.
The clip level below the corner frequency will be reduced at the temperature extremes.
For example, a mass position of 2V results in a reduction of clip level below the lower
corner frequency which is 6dB lower than normal.
For these reasons, after temperature has stabilized in a new installation, it is good prac-
tice to physically recentre the boom. This maximizes the dynamic range of the sensor
below the lower corner frequency.
This procedure is simple, requiring only a 1.5mm hex screwdriver with a long shaft
(provided in the Trillium maintenance kit) and any 2.5mm hex wrench or screwdriver:
1. Wait at least 4 hours for the temperature of the sensor case to come to equilibrium
with the temperature of the vault before performing this procedure.
2. Measure the voltage on the mass position outputs of the three sensor channels.
This can be easily accomplished when the digitiser is a Trident, by using a laptop
running the Nanometrics user interface. If the mass positions are all within the
range ±0.4V, then there is no need for recentring; otherwise continue to step 3.
3. Set the sensor in short-period response mode by shorting the ACC control signal to
ground. This is also easily done using the Trident user interface.
4. For each channel with a mass position greater than ±0.4V, do the following:
a) Locate the access seal screw corresponding to the channel of interest using Fig-
ure 3-3 as a guide.
Caution Do not remove more than one of the access seal screws at any one
time.
b) Using the 2.5 mm hex wrench, remove the appropriate seal screw. Inspect the
O-ring for damage (for example, cuts or pitting), and get a replacement seal
screw from the maintenance kit if the original seal screw appears to be dam-
aged.
c) Place the seal screw where it will remain clean while the mass is being centred.
d) Insert the 1.5 mm hex screwdriver straight down the hole to where it engages
with the screw head.
e) Adjust the screw by small increments while monitoring the mass position out-
puts until it is less than ±0.2 V. Note that the adjustment sensitivity is approxi-
mately 1turn = 1volt.
f) Using the 2.5 mm hex wrench, replace the seal screw. Be sure to tighten the
screw until there is metal-to-metal contact to ensure a good airtight seal.
3.5 Checklist
The following checklist can be used as an aid when installing Trillium:
Pier is clear of debris
Sensor is level
4.2 Power
Trillium can be powered from 9V to 36V. Under normal operation the power consump-
tion is approximately 0.6W. On startup the power may briefly rise to 1W.
For very long cables, be sure to account for the resistive voltage drop due to the cable
itself. For example 100m of 24AWG wire has a resistance of approximately 8.4Ω.
Therefore on startup if the input voltage at the Trillium is 9V then the resistive drop
due the peak current of 110mA will be 1.8V so the supply voltage must be at least
10.8V.
The UVW outputs are selected by shorting the UVW input to ground. If this input is
left floating, then the XYZ outputs are selected.
To understand the difference between the UVW and XYZ outputs, refer to Figure 4-1.
The sensor axes have been designed so that they are identical and so that the directions
in which they sense motion are orthogonal. Furthermore the U axis was chosen to be
aligned with the east-west axis when projected into the horizontal plane.
V
Y
U
W
u 2 0 2 x
1
v = ------- ⋅ – 1 3 2 ⋅ y (EQ 1)
6
w –1 – 3 2 z
x 2 –1 –1 u
1
y = ------- ⋅ 0 3 – 3 ⋅ v (EQ 2)
6
z 2 2 2 w
Alternatively, seismic data may be digitised with Trillium in UVW mode and the trans-
formation to horizontal and vertical signals implemented optionally when the data is
processed. This allows for studies and calibrations where both UVW and XYZ data are
required.
∏ ( s + zn )
V⋅s
F(s) = S sensor ⋅ k ⋅ ---------------------------
n - ---------- (EQ 3)
m
∏
( s + pn )
n
0
Zeroes 0 rad/s
51.5
–272 ±218i
Poles 56.5 rad/s
–0.1111 ±0.1111i
Normalization Factor 133310
Normalization Frequency 1 Hz
64
62
60
56
54
52
50
48
46
180
135
90
45
Phase (°)
-45
-90
-135
-180
-2 -1 0 1 2
10 10 10 10 10
Frequency (Hz)
4.5 Self-Noise
Typical Trillium self-noise is plotted in Figure 4-3. Curves indicating Peterson’s new
high- and low-noise models are included for reference.
-90
NHNM
NLNM
-100 Trillium 40
-110
-120
PSD (dB m /s /Hz)
-130
2 4
-140
-150
-160
-170
-180
-190
-3 -2 -1 0 1
10 10 10 10 10
Frequency (Hz)
4.6 Calibration
Calibration inputs are provided to allow for relative calibration of the sensor across fre-
quency and over time.
Since Trillium is a symmetric triaxial sensor, calibration must be performed on the
individual sensor axes (UVW) rather than the horizontal and vertical outputs (XYZ).
Individual axis outputs can be digitized by placing the sensor in UVW mode. This
mode is activated by shorting the UVW input to ground.
Each axis has a separate enable signal (U_CALEN, V_CALEN, W_CALEN) which is
activated by shorting it to ground. All axes use a common input signal (CAL_SIG)
which has a sensitivity of 0.010m/s2/V.
4.7 State-Of-Health
Mass position outputs are provided to monitor the effect of tilt and temperature on the
spring which sets the rest position of the boom. As with the calibration signals, they
represent the state of the individual sensor axes (UVW) rather than the horizontal and
vertical outputs (XYZ). The mass positions are zeroed at the factory at room tempera-
ture with Trillium perfectly level.
These outputs can be used to infer how much power the integrator is consuming to cen-
tre the boom, as well as whether the clip level of the instrument below the lower corner
frequency is being reduced significantly. If the mass positions are all within the range
±0.4V, then there is no need for recentring. Otherwise follow the procedure in section
3.3 “Mass Centring (Optional)” on page 8.
4.8 Troubleshooting
Apply power to the sensor and check that the current stays below 125mA. When the
sensor is level the power should be 0.6W.
If the problem persists contact Nanometrics support (see Chapter 1, “Introduction”).
Parameter Specification
Parameter Specification
Parameter Specification
Parameter Specification
Parameter Specification
The Trillium connector is a 19-pin male military circular type hermetic connector. The
pinout is given in Table B-1.
A generic sensor cable may have been shipped with your sensor. Table C-1 is the wir-
ing key for the standard cable (Nanometrics part number CBL13942R2). This table can
be used as a reference when wiring the generic sensor cable end to a digitiser connector.
Note If you are using the earlier version of the generic sensor cable (Nanomet-
rics part number CBL13942R1), refer to Table C-2 for the wiring of the
XYZ/UVW ± outputs; these are in inverse order to the standard cable. The
rest of the wiring is the same as for CBL13942R2 shown in Table C-1.
Table C-2 Generic sensor cable wiring for XYZ/UVW ± outputs of CBL13942R1