
90E36
POLY-PHASE HIGH-PERFORMANCE WIDE-SPAN ENERGY METERING IC
Calibration Method
32
August 24, 2011
5
CALIBRATION METHOD
5.1
NORMAL MODE OPERATION CALIBRATION
Calibration is done per phase and there is no need to calibrate for the
all-phase-sum (total) parameters. The calibration method is as follows:
Step-1: Register configuration for calibration
-
Start to configure the System configuration Registers by writing
-
The 90E36 automatically reset the configuration registers to their
default value.
-
Program all the system configuration registers.
-
Calculate and write the checksum to the
CS0 register.
-
Write 8765H to the
ConfigStart register (enable checksum check-
ing).
-
System may check the WarnOut pin to see if there is a checksum
error.
The start register and checksum handling scheme is the same
throughout the calibration process, so the following section does not
describe the start and checksum operation.
Step-2: Measurement calibration (per-phase)
-
First calibrate offset at I = 0, U = 0 for current or/and voltage;
Configure calculated channel Gain (The user needs to program
the PGA gain and DPGA gain properly in order to get the calcu-
lated gain within 0 to 2 in step-1).
Read Irms/ Urms value.
Calculate the compensation value.
Write the calculated value to the offset register.
-
Then calibrate gain at I = In (Ib), U = Un for current and voltage;
Read Irms/ Urms value.
Calculate the compensation value.
Write the calculated value to the Gain register.
Step-3: Metering calibration (per phase)
-
First calibrate the Power/ Energy offset.
U = Un, I = 0.
Read full 32 bits (or lower 16 bits) Active and Reactive Power
Calculate the compensation values
Write the calculated values to the offset registers respectively.
-
Then calibrate Energy gain at unity power factor:
PF=1.0, U = Un, I = In (Ib).
Connect CF1 to the calibration bench;
User/ PC calculate the energy gain according to the data got
from calibration bench
Write the calculated value to the Energy Gain register.
-
Then calibrate the phase angle compensation at 0.5 inductive
power factor.
PF=0.5L, U = Un, I = In (Ib), Rated frequency = 50Hz, or 60Hz
according to the application;
CF1 connected to the calibration bench;
User/ PC calculate the phase angle according to the data got
from calibration bench;
Write the calculated value to the Phase angle register.
5.2
PARTIAL MEASUREMENT MODE CALIBRA-
TION
The calibration method is as follows:
Step-1: Set the input current to zero and measure the current mean
value (set MeasureType = 1, write 1 to the ReMeasure bit (b14,
PMCon-fig) to trigger the measurement. Refer to the
PMIrmsA register). Negate
(16-bit) and then write the result to the offset register.
Step-2: The output of Partial Measurement result =
ADC_input_voltage *PGA_gain*DPGA_gain*65536 / 1.2. For instance,
a 150 mVrms signal (from CT) with PGA = 1 gets 8192 in the RMS result
register.
Step-3: The user needs to do its own conversion to get meaningful
result. The scaling factor in user's software could be calibrated device
per device.