MCP3918A1-ESLASHML datasheet(63/88 Pages) MICROCHIP

63 / 88 page

 2014 Microchip Technology Inc.

DS20005287A-page 63

MCP3918

8.6

Energy Measurement Error

Considerations

The measurement error is a typical representation of

the non-linearity of the ADC (see Section 4.0

“Terminology and Formulas”

for the definition of

measurement error). The measurement error is

dependent on the THD and on the noise floor of the

ADC. The measurement error specification on the

MCP3918 can be improved by increasing the OSR (to

get a better SINAD and THD performance) and, to

some extent, the BOOST settings (if the bandwidth of

the measurements is too limited by the bandwidth of

the amplifiers in the sigma-delta ADC). In most of the

energy metering AC applications, high-pass filters are

used to cancel the offset on each ADC channel (current

and voltage channels), and therefore a single-point

calibration is necessary to calibrate the system for

active energy measurement. This calibration is a

system gain calibration, and the user can utilize the

EN_GAINCAL bit and the GAINCAL_CH0 register to

perform this digital calibration. After such calibration,

typical measurement error curves like the ones in

Figure 2-7 can be generated by sweeping the current

channel amplitude and measuring the energy at the

outputs (the energy calculations here are being

realized off-chip). The error is measured using a gain of

1x, as it is commonly used in most CT-based

applications.

At low signal amplitude values (typically 1000:1

dynamic range and higher), the crosstalk between

channels, mainly caused by the PCB, becomes a

significant part of the perturbation as the measurement

error increases. The 1-point measurement error curves

in Figure 2-5 have been performed with a full-scale

sine wave on all the inputs that are not measured,

which means that these channels induce a maximum

amount of crosstalk on the measurement error curve.

In order to avoid such behavior, a 2-point calibration

can be put in place in the calculation section.

This 2-point calibration can be a simple linear

interpolation between two calibration points (one at

high amplitudes, one at low amplitudes at each end of

the dynamic range) and helps to significantly lower the

effect of crosstalk between channels. A 2-point

calibration is very effective in maintaining the

measurement error close to zero on the whole dynamic

range, since the non-linearity and distortion of the

MCP3918 is very low. Figure 2-6 shows the

measurement error curves obtained with the same

ADC data taken for Figure 2-5, but where a 2-point

calibration has been applied. The difference is

significant only at the low end of the dynamic range,

where all the perturbing factors are a bigger part of the

ADC output signals. These curves show extremely tight

measurement error across the full dynamic range

(here, typically 10,000:1), which is required in

high-accuracy class meters.

MCP3918A1-E/ML bảng dữ liệu(PDF) 63 Page - Microchip Technology

MCP3918A1-E/ML bảng dữ liệu(HTML) 63 Page - Microchip Technology

Html Pages

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88

bảng dữ liệu tải về

Link URL