SST-SALEM-3T Analog Devices, Inc., SST-SALEM-3T Datasheet - Page 18

• Manufacturer Part Number: SST-SALEM-3T
• Description: Salem Three-phase Electronic Energy Meter
• Manufacturer: Analog Devices, Inc.

ADSST-EM-3035
DC-Offset
Calibration
Offset
Calibration
(All Three Phases)
The microcontroller now issues 0x45H command on SPI to the
DSP. The DSP sends back Table IV. This table will contain
new dc-offset coefficients. The microcontroller should store
these coefficients.
Procedure
•
•
•
•
Phase Compensation
The ADSST-EM-3035 employs a patent pending algorithm for
phase compensation and non-linearity. This also reduces the
cost of the end product by reducing the cost of the sensing ele-
ments i.e., CT. To compensate for the phase non-linearity in
CTs, the compensation is performed at three current ranges.
The three current ranges for calibration are:
•
•
•
Procedure
•
•
•
•
•
Power up the meter with nominal voltage
Give command for calculation of the coefficient (EFh) to
DSP on SPI.
Receive the coefficient by sending Ox45 on SPI after waiting
at least 1s.
Store the coefficient
20 A > I
7 A > I
1.5 A > I
The ADSSTCOMP.EXE supplied with the chipset is an
executable file for calculation of the phase compensation
coefficients.
Set the voltage equal to 230 V which is the nominal voltage at
all phases.
Inject I
The chipset performs the harmonic analysis by providing
information about the magnitude and phase angle for all odd
harmonics sequenced from fundamental to 21st order. The
DSP sends the phase angle information along with other data
as described in Table IV after sending the command 0x45.
The value of the phase angle for line current A, B, and C is
available at the locations 283, 371, and 459 respectively (say
P
A
, P
B
, P
2
1
1
current at 0.5 inductive (60° lagging) in all phases.
> 1.5 A
3
C
> 7 A
) in the data stream sent by the DSP.
Table X. DC Offset Calibration Data
> 0 A
Command from
Microcontrolled
in Hex
0xEF
Setup Input
Voltage and
Current
V = Nominal
Voltage
I = 0
–18–
•
•
•
•
•
Configuration of Output E-pulses
The ADSST-EM-3035 Chipset provides two pulse outputs
•
•
•
•
•
Inaccuracy of the E-pulse
Higher E-pulse constant is always desirable as it reduces the
testing time. However, increase in pulses/kWh may increases the
error at higher power. The error can be calculated by the
given formula.
General Note About Calibration
•
•
Calculate the normalized lag value (L
as under :
Run ADSSTCOMP.EXE on PC
Feed the normalized lag value during the execution of
ADSSTCOMP.EXE.
The ADSSTCOMP.EXE will provide six coefficients for each
phase and the size of each coefficient is 2 bytes.
The phase compensation should be performed for the three
currents on each phase. These coefficients must be stored in a
suitable location such that DSP can get these coefficients on
power up in the same sequence as shown in Table III.
Configurable for Active energy or Apparent energy
Reactive energy
Table III gives the default conditions and configuration for
first E-pulse.
The E-pulse constant is variable from 1,000 pulses/kWh to
20,000 pulses/kWh.
Example: To set 1,500 pulses/kWh, the new E-pulse con-
stant will be 1,500
It should be noted that ADSST-EM-3035 does not have any
permanent memory and hence all the calibration data are to
be stored by the microcontroller and provided to the DSP at
the time of power up.
Before starting the calibration the meter should be supplied with
the default calibration constants as specified in the Table III.
L
L
L
B
C
A
=
=
=
60
60
60
1 20
1 20
1 20
°
°
°
.
.
.
–
–
–
P
P
P
B
C
A
+
+
+
2
2
2
A
, L
B
, L
C
) for each phase
REV. 0
(1)
(2)
(3)