AN-1137: ADE7816 Theory of Operation (Rev. 0) PDF

AN-1137
APPLICATION NOTE
One Technology Way • P.O. Box 9106 • Norwood, MA 02062-9106, U.S.A. • Tel: 781.329.4700 • Fax: 781.461.3113 • www.analog.com
ADE7816 Theory of Operation
By Aileen Ritchie
This application note provides theoretical information about
the ADE7816 energy and rms calculations and should be used
as a supplementary document to the ADE7816 data sheet.
INTRODUCTION
The ADE7816 is a multichannel energy metering IC that can
measure up to six current channels and one voltage channel simultaneously. The ADE7816 provides active and reactive energy,
along with current and voltage rms readings on all channels.
A variety of power quality features, including no load, reverse
power, and angle measurements, are also provided. The ADE7816
is suitable for use in a variety of metering applications, including
smart electricity meters, power distribution units, and in-home
energy monitors.
Figure 1 shows the signal path for one voltage and current channel
pair on the ADE7816. The ADE7816 has six separate energy
signal paths and, therefore, this signal path is duplicated six times.
Details are provided on each block of the data path in sequential
order, focusing on the theory behind the measurement. The application note is divided into two sections: the first section provides
details about the analog conversion and filter, and the second
describes the digital signal processing.
VRMSOS
X2
1.2V
REF
xWATTOS
VGAIN
VP
PGA2
ADC
VN
Ф
VRMS
LPF
xWGAIN
LPF
HPF
PCF_x_COEFF
IxP
ADC
Ф
xVAROS
HPF
IxGAIN
DIGITAL
INTEGRATOR
xVARGAIN
COMPUTATIONAL
BLOCK FOR
TOTAL
REACTIVE POWER
X2
ENERGY
AND RMS
DATA
ALL
CHANNELS
COMMUNICATION
AND
INTERRUPTS
IxRMS
LPF
IxRMSOS
Figure 1. Signal Path
Rev. 0 | Page 1 of 8
10449-001
IxN
PGAx
AN-1137
Application Note
TABLE OF CONTENTS
Introduction ...................................................................................... 1 Digital Processing..............................................................................6 Revision History ............................................................................... 2 Total Active Power Calculation ...................................................6 Analog Conversion and Filtering ................................................... 3 Reactive Power...............................................................................7 Analog-to-Digital Conversion.................................................... 4 RMS.................................................................................................8 Digital Integrator (Current Channels Only)............................. 5 REVISION HISTORY
2/12—Revision 0: Initial Version
Rev. 0 | Page 2 of 8
Application Note
AN-1137
ANALOG CONVERSION AND FILTERING
voltage channel signal paths are identical, with the exception of
the digital integrator that is included only on the current channels.
The integrator allows direct connection to Rogowski coil current
sensors (see the Digital Integrator (Current Channels Only)
section for more details).
This section focuses on the analog portion of the ADE7816
energy meter and the associated filtering, which includes the
analog-to-digital conversion, high-pass filter, and integrator
(current channel only). Figure 2 and Figure 3 show the signal
path of the analog converter and filter portion of the current
and voltage channels, respectively. As shown, the current and
POWER QUALITY
(SEE THE ADE7816 DATA SHEET)
IRMS CALCULATION
REFERENCE
IxGAIN
PGA GAIN
DIGITAL
INTEGRATOR
IxP
VIN
PGAx
ADC
WAVEFORM
SAMPLE REGISTER
ACTIVE AND REACTIVE
POWER CALCULATION
HPF
IxN
CURRENT CHANNE L
DATA RANGE AFTER
INTEGRATION
CURRENT CHANNE L
DATA RANGE
+0.5V/GAIN
0x5A7540 =
+5,928,256
0x5A7540 =
+5,928,256
0V
0V
0V
–0.5V/GAIN
ANALOG INPUT RANGE
0xA58AC0 =
–5,928,256
ANALOG OUTPUT RANGE
0xA58AC0 =
–5,928,256
10449-002
VIN
Figure 2. Current Channel Signal Path
POWER QUALITY
(SEE THE ADE7816 DATA SHEET)
VRMS CALCULATION
REFERENCE
WAVEFORM
SAMPLE REGISTER
VGAIN
PGA GAIN
VP
VIN
PGA2
ADC
HPF
ACTIVE AND REACTIVE
POWER CALCULATION
VN
VIN
+0.5V/GAIN
0x5A7540 =
+5,928,256
0V
0V
–0.5V/GAIN
ANALOG INPUT RANGE
0xA58AC0 =
–5,928,256
ANALOG OUTPUT RANGE
Figure 3. Voltage Channel Signal Path
Rev. 0 | Page 3 of 8
10449-003
VOLTAGE CHANNEL
DATA RANGE
AN-1137
Application Note
ANALOG-TO-DIGITAL CONVERSION
The ADE7816 includes sigma-delta (Σ-Δ) analog-to-digital
converters (ADCs) to convert the incoming analog signal into
a digital bit stream. When the ADE7816 is fully powered, all
ADCs are active.
For simplicity, the block diagram in Figure 4 shows a first-order
Σ-Δ ADC. The converter is composed of the Σ-Δ modulator and
the digital low-pass filter.
it is possible to shape the quantization noise so that the majority
of the noise lies at the higher frequencies. In the Σ-Δ modulator,
the noise is shaped by the integrator, which has a high-pass type
of response for the quantization noise. This is the second technique
used to achieve high resolution. The result is that most of the noise
is at the higher frequencies, where it can be removed by the
digital low-pass filter. This noise shaping is shown in Figure 5.
ANTIALIAS FILTER
(RC)
CLKIN/16
INTEGRATOR
+
C
+
–
VREF
LATCHED
COMPARATOR
–
.....10100101.....
1-BIT DAC
DIGITAL
LOW-PASS
FILTER
NOISE
24
0
2
A Σ-Δ modulator converts the input signal into a continuous
serial stream of 1s and 0s at a rate that is determined by the
sampling clock. In the ADE7816, the sampling clock is equal
to 1.024 MHz (CLKIN/16). The 1-bit DAC in the feedback loop
is driven by the serial data stream. The DAC output is subtracted
from the input signal. If the loop gain is high enough, the average
value of the DAC output (and, therefore, the bit stream) can
approach that of the input signal level. For any given input value
in a single sampling interval, the data from the 1-bit ADC is
virtually meaningless. A meaningful result can be obtained only
when a large number of samples is averaged. This averaging is
carried out in the second part of the ADC, the digital low-pass
filter. By averaging a large number of bits from the modulator, the
low-pass filter can produce 24-bit data-words that are proportional
to the input signal level.
The Σ-Δ converter uses two techniques to achieve high resolution
from what is essentially a 1-bit conversion technique. The first is
oversampling. Oversampling means that the signal is sampled at
a rate (frequency) that is many times higher than the bandwidth
of interest. For example, the sampling rate in the ADE7816 is
1.024 MHz, and the bandwidth of interest is 40 Hz to 2 kHz.
Oversampling has the effect of spreading the quantization noise
(noise due to sampling) over a wider bandwidth. With the noise
spread more thinly over a wider bandwidth, the quantization noise
in the band of interest is lowered, as shown in Figure 5. However,
oversampling alone is not enough to improve the signal-to-noise
ratio (SNR) in the band of interest. For example, an oversampling
factor of 4 is required just to increase the SNR by a mere 6 dB
(one bit). To keep the oversampling ratio at a reasonable level,
512
FREQUENCY (kHz)
1024
HIGH RESOLUTION
OUTPUT FROM
DIGITAL LPF
SIGNAL
Figure 4. First-Order Σ-∆ ADC
4
NOISE
0
2
4
512
FREQUENCY (kHz)
1024
10449-005
R
SHAPED NOISE
SAMPLING
FREQUENCY
10449-004
ANALOG
LOW-PASS FILTER
DIGITAL FILTER
SIGNAL
Figure 5. Noise Reduction Due to Oversampling and
Noise Shaping in the Analog Modulator
ADC Transfer Function
All ADCs in the ADE7816 are designed to produce a similar
24-bit, signed output code for the same input signal level. The
24-bit, signed output code is available at a fixed rate of 8 kSPS
(thousand samples per second). With a full-scale input signal of
0.5 V and an internal reference of 1.2 V, the ADC output code is
approximately 5,928,256 (0x5A7540), as shown in Figure 2 and
Figure 3. The code from the ADC can vary between 0x800000
(−8,388,608) and 0x7FFFFF (+8,388,607); this is equivalent to
an input signal level of ±0.707 V. However, for specified
performance, do not exceed the nominal range of ±0.5 V; ADC
performance is guaranteed only for input signals that are lower
than ±0.5 V.
High-Pass Filters (HPFs)
The ADC outputs can contain a dc offset. This offset can create
errors in power and rms calculations. High-pass filters (HPFs)
are placed in the signal path of the phase and neutral currents
and of the phase voltages. The HPFs eliminate any dc offset
on the current or voltage channel and have a cut-off frequency
of 0.2 Hz.
Rev. 0 | Page 4 of 8
Application Note
AN-1137
0
–50
Due to the di/dt sensor, the current signal must be filtered before
it can be used for power measurement. On each current channel,
there is a built-in digital integrator to recover the current signal
from the di/dt sensor. The digital integrator is disabled by default,
but it can be enabled by setting Bit 0 (INTEN) of the CONFIG
register (see the ADE7816 data sheet). Figure 7 and Figure 8 show
the magnitude and phase response of the digital integrator.
When the digital integrator is switched off, the ADE7816 can be
used directly with a conventional current sensor, such as a
current transformer (CT).
PHASE (Degrees)
The flux density of a magnetic field that is induced by a current
is directly proportional to the magnitude of the current. The
changes in the magnetic flux density passing through a conductor
loop generate an electromotive force (EMF) between the two
ends of the loop. The EMF is a voltage signal that is proportional
to the di/dt of the current. The voltage output from the di/dt
current sensor is determined by the mutual inductance between
the current-carrying conductor and the di/dt sensor.
0
100
1000
500
1000
1500
2000
2500
FREQUENCY (Hz)
3000
3500
4000
Figure 7. Combined Gain and Phase Response of the Digital Integrator
–15
–20
–25
–30
30
35
40
35
40
45
50
55
FREQUENCY (Hz)
60
65
70
45
60
65
70
–89.96
–89.97
–89.98
–89.99
30
Note that the integrator has a −20 dB/dec attenuation and
approximately −90° of phase shift. When combined with a di/dt
sensor, the resulting magnitude and phase response should be
1
10
FREQUENCY (Hz)
–50
–100
MAGNITUDE (dB)
Figure 6. Principle of a di/dt Current Sensor
0.1
0
PHASE (Degrees)
+ EMF (ELECTROMOTIVE FORCE)
– INDUCED BY CHANGES IN
MAGNETIC FLUX DENSITY (di/dt)
10449-006
0.01
10449-007
MAGNETIC FIELD CREATED BY CURRENT
(DIRECTLY PROPORTIONAL TO CURRENT)
50
50
55
FREQUENCY (Hz)
10449-008
The internal digital integrator is designed for use with a Rogowski
coil or other di/dt sensor. A di/dt sensor detects changes in the
magnetic field caused by the ac current. Figure 6 shows the
principle of a di/dt current sensor.
a flat gain over the frequency band of interest. However, the di/dt
sensor has a 20 dB/dec gain associated with it and generates significant high frequency noise. An antialiasing filter of at least the
second order is needed to avoid noise aliasing back in the band of
interest when the ADC is sampling (see the ADE7816 data sheet).
MAGNITUDE (dB)
DIGITAL INTEGRATOR (CURRENT CHANNELS
ONLY)
Figure 8. Combined Gain and Phase Response of the Digital Integrator
(30 Hz to 70 Hz)
Rev. 0 | Page 5 of 8
AN-1137
Application Note
DIGITAL PROCESSING
The ADE7816 contains a fixed function digital signal processor
(DSP) that computes all power and rms values. It contains
program memory ROM and data memory RAM.
The program that is used for the power and rms computations is
stored in the program memory ROM, and the processor executes
it every 8 kHz.
TOTAL ACTIVE POWER CALCULATION
Electrical power is defined as the rate of energy flow from source
to load, and it is given by the product of the voltage and current
waveforms. The resulting waveform is called the instantaneous
power signal, and it is equal to the rate of energy flow at every
instant of time. The unit of power is the watt or joules/sec. If an
ac system is supplied by a voltage, v(t), and consumes the current,
i(t), and each of them contains harmonics, then
∞
v (t ) = ∑Vk 2 sin (kωt + φk)
(1)
k=1
The ADE7816 computes the total active power on each channel
by first multiplying the current and voltage signals on each.
Next, the dc component of the instantaneous power signal is
extracted using a low-pass filter (LPF), as shown in Figure 11.
If the voltage and currents contain only the fundamental
component, are in phase (that is, φ1 = γ1 = 0), and correspond
to full-scale ADC inputs, then multiplying them results in an
instantaneous power signal that has a dc component, V1 × I1,
and a sinusoidal component, V1 × I1 cos(2ωt). Figure 9 shows
the corresponding waveforms.
INSTANTANEOUS
POWER SIGNAL
p(t)= V rms × I rms – V rms × I rms × cos(2ωt)
0x3FED4D6
67,032,278
INSTANTANEOUS
ACTIVE POWER
SIGNAL: V rms × I rms
V rms × I rms
0x1FF6A6B =
33,516,139
∞
i (t ) = ∑ I k 2 sin(kωt + γ k )
k =1
0x000 0000
where:
Vk, Ik are rms voltage and current, respectively, of each
harmonic.
φk, γk are the phase delays of each harmonic.
Figure 9. Active Power Calculation
The instantaneous power in an ac system is as follows:
∞
p(t) = v(t) × i(t) =
∑Vk I k cos(φk − γk)
(2)
k =1
∞
−
∑V I
k
k =1
k
cos(2kωt + φk + γk) +
10449-009
i(t) = √2 × I rms × sin(ωt)
v(t) = √2 × V rms × sin(ωt)
∞
∑VI {cos[(k − m)ωt
km
k, m = 1
k≠m
+ φk − γm] − cos[(k + m)ωt + φk + γm]}
Because the LPF does not have an ideal brick wall frequency
response (see Figure 10), the active power signal has some
ripple due to the instantaneous power signal. This ripple is
sinusoidal and has a frequency that is equal to twice the line
frequency. Because the ripple is sinusoidal in nature, it is removed
when the active power signal is integrated over time to calculate
the energy.
0
The average power over an integral number of line cycles (n) is
given by the expression that is shown in Equation 3.
∞
0
k =1
∫ p(t )dt = ∑V
k
I k cos(φk − γk)
(3)
where:
T is the line cycle period.
P is referred to as the total active or total real power.
Note that the total active power is equal to the dc component of
the instantaneous power signal p(t) in Equation 2; that is,
∞
∑V I
k k
–10
–15
–20
–25
0.1
cos(φk − γk)
k =1
This is the expression used to calculate the total active power in
the ADE7816 for each channel.
Rev. 0 | Page 6 of 8
1
FREQUENCY (Hz)
3
Figure 10. Frequency Response of the LPF Used
to Filter Instantaneous Power in Each Phase
10
10449-010
nT
MAGNITUDE (dB)
P= 1
nT
–5
Application Note
AN-1137
DIGITAL
INTEGRATOR
IAGAIN
IA
HPF
AWATTOS
AWGAIN
AWATTHR[31:0]
VGAIN
ACCUMULATOR
LPF
32-BIT
REGISTER
V
HPF
10449-011
PCF_A_COEFF
WTHR[47:0]
Figure 11. Active Energy Data Path
Active Energy Calculation
k =1
As previously stated, power is defined as the rate of energy flow.
This relationship can be expressed mathematically as
Power =
dEnergy
dt
(4)
Conversely, energy is given as the integral of power, as follows:
Energy = ∫ p (t )dt
(5)
where iʹ(t) is the current waveform with all harmonic
components phase shifted by 90°.
Next, the instantaneous reactive power, q(t), can be expressed as
q(t) = v(t) × iʹ(t)
∞
+
∞
q(t ) = ∑Vk I k
π
)
2
k
(10)
k
π
2
)}
∞
Equation 7 gives an expression for the instantaneous reactive
power signal in an ac system when the phase of the current
channel is shifted by +90°.
k =1
{cos(φ − γ − π2 )
− cos(2 kωt + φk + γk +
The ADE7816 can compute the total reactive power on each
channel. Reactive power integrates the fundamental and harmonic
components of the voltages and currents. A load that contains
a reactive element (inductor or capacitor) produces a phase difference between the applied ac voltage and the resulting current.
The power associated with the reactive elements is called reactive
power, and its unit is the var. Reactive power is defined as the
product of the voltage and current waveforms when all harmonic
components of one of these signals are phase shifted by 90°.
2 sin(kωt + φk)
× 2sin(kωt + φk) × sin(mωt + γm +
k =1
REACTIVE POWER
k
k m
Note that q(t) can be rewritten as
(6)
The accumulation is performed in two stages (see Figure 11):
first, by an internal threshold and, next, in an external register.
See the ADE7816 data sheet for more information on setting
the internal threshold.
∞
∑V I
k, m=1
k ≠m
where:
n is the discrete time sample number.
T is the sample period.
∑V
(9)
∞
π
q(t) = ∑Vk I k × 2sin(kωt + φk) × sin(kωt + γk + 2 )
k =1
The ADE7816 calculates the energy by accumulating the power
over time. This discrete time accumulation or summation is
equivalent to integration in continuous time, following the
description in Equation 6.
v (t ) =
(8)
2 sin (kωt + γ k )
k
∞
π⎞
⎛
i ' (t ) = ∑ I k 2 sin⎜ kωt + γ k + ⎟
2⎠
⎝
k =1
Active energy accumulation is a signed operation. Negative
energy is subtracted, and positive energy is added to total
energy accumulation.
⎧∞
⎫
Energy = ∫ p (t )dt = Lim ⎨ ∑ p (nT ) × T ⎬
T→0 ⎩n=0
⎭
∞
∑I
i (t ) =
+
π
∑V I {cos[(k − m)ωt + φ − γ − 2 ]}
k, m=1
k ≠m
k
The average total reactive power over an integral number of line
cycles (n) is given by the expression in Equation 11.
Q=
1
nT
nT
∞
0
k =1
∞
∫ q(t )dt = ∑Vk I k
Q = ∑Vk I k
cos(φk − γk −
π
2
)
(11)
sin(φk − γk)
k =1
where:
T is the period of the line cycle, and Q is referred to as the total
reactive power. Note that the total reactive power is equal to the
dc component of the instantaneous reactive power signal, q(t), in
Equation 10; that is,
∞
V I sin(φk − γk)
∑
k =1
(7)
k
km
k k
This relationship is used to calculate the total reactive power for
each channel. The instantaneous reactive power signal, q(t), is
generated by multiplying each harmonic of the voltage signals by
the 90° phase-shifted corresponding harmonic of the current in
each phase.
Rev. 0 | Page 7 of 8
AN-1137
Application Note
DIGITAL
INTEGRATOR
IAGAIN
AVAROS
IA
AVARGAIN
HPF
AVARHR[31:0]
TOTAL
REACTIVE
POWER
ALGORITHM
VGAIN
ACCUMULATOR
V
32-BIT
REGISTER
VARTHR[47:0]
HPF
10449-012
PCF_A_COEFF
Figure 12. Reactive Energy Data Path
IxRMSOS[23:0]
CURRENT SIGNAL FROM
HPF OR INTEGRATOR
(IF ENABLED)
X2
IxRMS[23:0]
HPF
10449-013
27
Figure 13. IRMS Signal Path
Reactive Energy Calculation
Reactive energy is defined as the integral of reactive power.
Reactive Energy = ∫q(t)dt
(12)
Reactive energy accumulation is a signed operation. Negative
energy is subtracted from the reactive energy contents.
Similar to active power, the ADE7816 achieves the integration
of the reactive power signal in two stages (see the ADE7816
data sheet).
Equation 14 implies that for signals containing harmonics, the
rms calculation contains the contribution of all harmonics, not
only the fundamental. The ADE7816 uses a low-pass filter to
average the square of the input signal and then takes the square
root of the result. This concept is explained mathematically in the
following formulas:
∞
f (t ) = ∑ Fk 2 sin(kωt + γ k )
Then
RMS
Root mean square (rms) is a measurement of the magnitude of
an ac signal. Its definition can be both practical and mathematical.
Defined practically, the rms value assigned to an ac signal is the
amount of dc that is required to produce an equivalent amount of
power in the load. Mathematically, the rms value of a continuous
signal, f(t), is defined as follows:
Frms =
1 t 2
f (t )dt
t ∫0
Frms =
1
N
N
∑ f [n]
2
(14)
∞
∞
k =1
k =1
f 2 (t ) = ∑ Fk2 − ∑ Fk2 cos(2kωt + γ k ) +
∞
+2
∑ 2 × Fk × Fm sin(kωt + γ k )× sin(mωt + γm )
(16)
k ,m=1
k ≠m
After the LPF and the execution of the square root, the rms
value of f(t) is obtained by
(13)
For time sampling signals, rms calculation involves squaring the
signal, taking the average, and obtaining the square root.
(15)
k =1
F=
∞
∑ Fk2
(17)
k =1
The rms calculation on the current and voltage channel is
performed using the same method. See Figure 13 for a block
diagram of the Irms measurement.
N =1
©2012 Analog Devices, Inc. All rights reserved. Trademarks and
registered trademarks are the property of their respective owners.
AN10449-0-2/12(0)
Rev. 0 | Page 8 of 8