beautypg.com

4 calibration procedure with three measurements, Calibration procedure with three measurements – Maxim Integrated 71M6541 Demo Board User Manual

Page 41

background image

71M6541 Demo Board REV 3.0 User’s Manual

41

Rev 4.0

The calibration procedures described below should be followed after interfacing the voltage and current sensors
to the 71M6541F chip. When properly interfaced, the V3P3 power supply is connected to the meter neutral and
is the DC reference for each input. Each voltage and current waveform, as seen by the 71M6541F, is scaled to
be less than 250mV (peak).

2.3.4 CALIBRATION PROCEDURE WITH THREE MEASUREMENTS

Each phase is calibrated individually. The calibration procedure is as follows:

1) The calibration factors for all phases are reset to their default values, i.e.,

CAL_In

=

CAL_Vn

= 16384,

and

PHADJ_n

= 0.

2) An RMS voltage V

ideal

consistent with the meter’s nominal voltage is applied, and the RMS reading

V

actual

of the meter is recorded. The voltage reading error Axv is determined as


Axv = (V

actual -

V

ideal

) / V

ideal

3) Apply the nominal load current at phase angles 0° and 60°, measure the Wh energy and record the er-

rors E

0

AND E

60

.

4) Calculate the new calibration factors

CAL_In

,

CAL_Vn,

and

PHADJ_n

, using the formulae presented

in section 2.2.1 or using the spreadsheet presented in section 2.3.6.

5) Apply the new calibration factors

CAL_In

,

CAL_Vn

, and

PHADJ_n

to the meter. The memory loca-

tions for these factors are given in section 1.9.1.

6) Test the meter at nominal current and, if desired, at lower and higher currents and various phase an-

gles to confirm the desired accuracy.

7) Store the new calibration factors

CAL_In

,

CAL_Vn

, and

PHADJ_n

in the EEPROM or FLASH memory

of the meter. If the calibration is performed on a Maxim Integrated Demo Board, the methods involving
the command line interface, as shown in sections 1.9.3 and 1.9.4, can be used.

8) Repeat the steps 1 through 7 for each phase.

9) For added temperature compensation, read the value

TEMP_RAW

(CE RAM) and write it to

TEMP_NOM

(CE RAM). If Demo Code 4.6n or later is used, this will automatically calculate the cor-

rection coefficients PPMC and PPMC2 from the nominal temperature and from the characterization da-
ta contained in the on-chip fuses.

Tip: Step 2 and the energy measurement at 0° of step 3 can be combined into one step.