Section 6. cr800 specifications – Campbell Scientific CR800 and CR850 Measurement and Control Systems User Manual
Page 79

79
Section 6. CR800 Specifications
1.1
CR800 specifications are valid from ─25° to 50°C in non‐condensing environments unless otherwise specified. Recalibration is recommended every two years. Critical specifications and system
configurations should be confirmed with a Campbell Scientific applications engineer before purchase.
2.0
PROGRAM EXECUTION RATE
2.1
10 ms to one day at 10 ms increments
3.0
ANALOG INPUTS (SE 1–6, DIFF 1–3)
3.0.1
Three differential (DIFF) or six single‐ended (SE) individually
configured input channels. Channel expansion provided by
optional analog multiplexers.
3.1.0
RANGES and RESOLUTION: With reference to the following table,
basic resolution (Basic Res) is the resolution of a single A/D
(p.
conversion. A DIFF measurement with input reversal has
better (finer) resolution by twice than Basic Res.
3.1.1 ‐‐ 8 10
Range (mV)
1
DIFF
Res (μV)
2
Basic
Res (μV)
±5000
±2500
±250
±25
±7.5
±2.5
667
333
33.3
3.33
1.0
0.33
1333
667
66.7
6.7
2.0
0.67
1
Range overhead of ≈9% on all ranges guarantees full‐scale
voltage will not cause over‐range.
2
Resolution of DIFF measurements with input reversal.
3.2 ‐‐ 8 10
ANALOG INPUT ACCURACY
3
:
±(0.06% of reading + offset
3
), 0° to 40°C
±(0.12% of reading + offset
3
), ‐25° to 50°C
±(0.18% of reading + offset
3
), ‐55° to 85°C (‐XT only)
3.2.1
3
Accuracy does not include sensor and measurement noise. Offset
definitions:
Offset = 1.5 x Basic Res + 1.0 µV (for DIFF measurement w/ input
reversal)
Offset = 3 x Basic Res + 2.0 µV (for DIFF measurement w/o input
reversal)
Offset = 3 x Basic Res + 3.0 µV (for SE measurement)
3.3
ANALOG MEASUREMENT SPEED:
3.3.1 ‐‐ 8 10
‐‐‐Total Time
4
‐‐‐
Inte‐
gration
Type
Code
Inte‐
gration
Time
Settling
Time
SE
with
no
Rev
DIFF
with
Input
Rev
250
_60Hz
5
_50Hz
5
250 µs
16.67 ms
20.00 ms
450 µs
3 ms
3 ms
≈1 ms
≈20 ms
≈25 ms
≈12 ms
≈40 ms
≈50 ms
4
Includes 250 μs for conversion to engineering units.
5
AC line noise filter
3.4
INPUT‐NOISE VOLTAGE: For DIFF measurements with input
reversal on ±2.5 mV input range (digital resolution dominates for
higher ranges):
250 μs Integration: 0.34 μV RMS
50/60 Hz Integration: 0.19 μV RMS
INPUT LIMITS: ±5 Vdc
DC COMMON‐MODE REJECTION: >100 dB
NORMAL‐MODE REJECTION: 70 dB @ 60 Hz when using 60 Hz
rejection
INPUT VOLTAGE RANGE W/O MEASUREMENT CORRUPTION: ±8.6
Vdc max.
SUSTAINED‐INPUT VOLTAGE W/O DAMAGE: ±16 Vdc max.
INPUT CURRENT: ±1 nA typical, ±6 nA max. @ 50°C; ±90 nA @ 85°C
INPUT RESISTANCE: 20 GΩ typical
ACCURACY OF BUILT‐IN REFERENCE JUNCTION THERMISTOR (for
thermocouple measurements):
±0.3°C, ‐25° to 50°C
±0.8°C, ‐55° to 85°C (‐XT only)
4.0
ANALOG OUTPUTS (VX 1–2)
4.0.1 ‐‐ 8
Two switched voltage outputs sequentially active only during
measurement.
4.0.2
RANGES AND RESOLUTION:
4.1 ‐‐ 8 10
Channel
Range
Resolu‐
tion
Current
Source
/ Sink
(VX 1–2)
±2.5 Vdc
0.67 mV
±25 mA
4.2 ‐‐ 8 10
ANALOG OUTPUT ACCURACY (VX):
±(0.06% of setting + 0.8 mV, 0° to 40°C
±(0.12% of setting + 0.8 mV, ‐25° to 50°C
±(0.18% of setting + 0.8 mV, ‐55° to 85°C (‐XT only)
4.4
VX FREQUENCY SWEEP FUNCTION: Switched outputs provide a
programmable swept frequency, 0 to 2500 mV square waves for
exciting vibrating wire transducers.
3.5.0
PERIOD AVERAGE
3.5.0a
Any of the 6 SE analog inputs can be used for period averaging.
Accuracy is ±(0.01% of reading + resolution), where resolution
is 136 ns divided by the specified number of cycles to be
measured.
INPUT AMPLITUDE AND FREQUENCY:
3.5.1 ‐‐ 8 10
Volt‐
Input
Signal
Peak‐Peak
Min
Pulse
Max
age
Gain
Range
Code
Min
mV
6
Max
V
7
Width
µs
Freq
kHz
8
1
10
33
100
mV250
mV25
mV7_5
mV2_5
500
10
5
2
10
2
2
2
2.5
10
62
100
200
50
8
5
6
Signal to be centered around Threshold (see PeriodAvg()
instruction).
7
Signal to be centered around ground.
8
The maximum frequency = 1/(twice minimum pulse width)
for 50% of duty cycle signals.
5.0
RATIOMETRIC MEASUREMENTS
5.1 ‐‐ 8 10
MEASUREMENT TYPES: The CR800 provides ratiometric
resistance measurements using voltage excitation. Three
switched voltage excitation outputs are available for
measurement of four‐ and six‐wire full bridges, and two‐,
three‐, and four‐wire half bridges. Optional excitation polarity
reversal minimizes dc errors.
5.2 ‐‐ 8 10
RATIOMETRIC MEASUREMENT ACCURACY
9,11
Note Important assumptions outlined in footnote 9:
±(0.04% of Voltage Measurement + Offset
12
)
5.2.1
9
Accuracy specification assumes excitation reversal for
excitation voltages < 1000 mV. Assumption does not include
bridge resistor errors and sensor and measurement noise.
11
Estimated accuracy, ∆X (where X is value returned from
measurement with Multiplier =1, Offset = 0):
BRHalf() Instruction: ∆X = ∆V1 / VX.
BRFull() Instruction: ∆X = 1000 x ∆V1/VX, expressed as mV•V
‐1
.
Note ∆V1 is calculated from the ratiometric measurement
accuracy. See manual section Resistance Measurements
(p.
for more information.
Offset definitions:
Offset = 1.5 x Basic Res + 1.0 µV (for DIFF measurement w/
input reversal)
Offset = 3 x Basic Res + 2.0 µV (for DIFF measurement w/o
input reversal)
Offset = 3 x Basic Res + 3.0 µV (for SE measurement)
Note Excitation reversal reduces offsets by a factor of two.
6.0
PULSE COUNTERS (P 1–2)
6.0.1
Two inputs individually selectable for switch closure, high
frequency pulse, or low‐level ac. Independent 24‐bit counters
for each input.
6.1
MAXIMUM COUNTS PER SCAN: 16.7 x 10
6
6.2
SWITCH‐CLOSURE MODE:
Minimum Switch Closed Time: 5 ms
Minimum Switch Open Time: 6 ms
Max. Bounce Time: 1 ms open without being counted
6.3
HIGH‐FREQUENCY PULSE MODE:
Maximum‐Input Frequency: 250 kHz
Maximum‐Input Voltage: ±20 V
Voltage Thresholds: Count upon transition from below 0.9 V to
above 2.2 V after input filter with 1.2 μs time constant.
6.4
LOW‐LEVEL AC MODE: Internal ac coupling removes dc offsets
up to ±0.5 Vdc.
Input Hysteresis: 12 mV RMS @ 1 Hz
Maximum ac‐Input Voltage: ±20 V
Minimum ac‐Input Voltage:
6.4.1
Sine wave (mV RMS)
Range (Hz)
20
200
2000
5000
1.0 to 20
0.5 to 200
0.3 to 10,000
0.3 to 20,000
7.0
DIGITAL I/O PORTS (C 1‐4)
7.0.1
Four ports software selectable as binary inputs or control
outputs. Provide on/off, pulse width modulation, edge timing,
subroutine interrupts / wake up, switch‐closure pulse counting,
high‐frequency pulse counting, asynchronous communications
(UARTs), and SDI‐12 communications. SDM communications
are also supported.
7.0
DIGITAL I/O PORTS (C 1‐4)
7.0.1
Four ports software selectable as binary inputs or control
outputs. Provide on/off, pulse width modulation, edge timing,
subroutine interrupts / wake up, switch‐closure pulse counting,
high‐frequency pulse counting, asynchronous communications
(UARTs), and SDI‐12 communications. SDM communications are
also supported.
7.1
LOW FREQUENCY MODE MAX: <1 kHz
7.2
HIGH FREQUENCY MODE MAX: 400 kHz
7.3
SWITCH‐CLOSURE FREQUENCY MAX: 150 Hz
7.4
EDGE‐TIMING RESOLUTION:
7.5
OUTPUT VOLTAGES (no load): high 5.0 V ±0.1 V; low < 0.1 V
7.6
OUTPUT RESISTANCE: 330 Ω
7.7
INPUT STATE: high 3.8 to 16 V; low ‐8.0 to 1.2 V
7.8
INPUT HYSTERISIS: 1.4 V
7.9
INPUT RESISTANCE:
100 kΩ with inputs < 6.2 Vdc
220 Ω with inputs ≥ 6.2 Vdc
7.10
SERIAL DEVICE / RS‐232 SUPPORT: 0 to 5 Vdc UART
7.12
SWITCHED 12 Vdc (SW12)
One independent 12 Vdc unregulated terminal switched on and
off under program control. Thermal fuse hold current = 900 mA
at 20°C, 650 mA at 50°C, and 360 mA at 85°C.
8.0
CE COMPLIANCE
8.1
STANDARD(S) TO WHICH CONFORMITY IS DECLARED:
IEC61326:2002
9.0
COMMUNICATION
9.1
RS‐232 PORTS:
DCE nine‐pin: (not electrically isolated) for computer connection
or connection of modems not manufactured by Campbell
Scientific.
COM1 to COM2: two independent Tx/Rx pairs on control ports
(non‐isolated); 0 to 5 Vdc UART
Baud Rate: selectable from 300 bps to 115.2 kbps.
Default Format: eight data bits; one stop bits; no parity.
Optional Formats: seven data bits; two stop bits; odd, even
parity.
9.2
CS I/O PORT: Interface with telecommunications peripherals
manufactured by Campbell Scientific.
9.3
SDI‐12: Digital control ports C1, C3 are individually configurable
and meet SDI‐12 Standard v. 1.3 for datalogger mode. Up to ten
SDI‐12 sensors are supported per port.
9.5
PROTOCOLS SUPPORTED: PakBus, AES‐128 Encrypted PakBus,
Modbus, DNP3, FTP, HTTP, XML, HTML, POP3, SMTP, Telnet,
NTCIP, NTP, Web API, SDI‐12, SDM.
10.0
SYSTEM
10.1
PROCESSOR: Renesas H8S 2322 (16‐bit CPU with 32‐bit internal
core running at 7.3 MHz)
10.2
MEMORY: 2 MB of flash for operating system; 4 MB of battery‐
backed SRAM for CPU usage, program storage, and final data
storage.
10.3
REAL‐TIME CLOCK ACCURACY: ±3 min. per year. Correction via
GPS optional.
10.4
RTC CLOCK RESOLUTION: 10 ms
11.0
SYSTEM POWER REQUIREMENTS
11.1
VOLTAGE: 9.6 to 16 Vdc
11.2 ‐‐ 8 10
INTERNAL BATTERY: 1200 mAhr lithium battery for clock and
SRAM backup. Typically provides three years of back‐up.
11.3
EXTERNAL BATTERIES: Optional 12 Vdc nominal alkaline and
rechargeable available. Power connection is reverse polarity
protected.
11.4
TYPICAL CURRENT DRAIN at 12 Vdc:
Sleep Mode: 0.7 mA typical; 0.9 mA maximum
1 Hz Sample Rate (one fast SE meas.) mA
100 Hz Sample Rate (one fast SE meas.): 16 mA
100 Hz Sample Rate (one fast SE meas. with RS‐232
communications): 28 mA
Active external keyboard display adds 7 mA (100 mA with
backlight on).
12.0
PHYSICAL
12.1
DIMENSIONS: 241 x 104 x 51 mm (9.5 x 4.1 x 2 in.) ; additional
clearance required for cables and leads.
12.2
MASS / WEIGHT: 0.7 kg / 1.5 lbs
13.0
WARRANTY
13.1
Warranty is stated in the published price list and in opening
pages of this and other user manuals.