Cirrus Logic CS5460A, Manual

Page: 36 / 64

CS5460A
36 DS284PP4
an upside-down parabola. The T
VREFOUT
for any
particular CS5460A sample is derived from that
sample’s individual VREFOUT Voltage vs. Tem-
perature characteristic. Using the data from the
sample’s VREFOUT Voltage vs. Temperature
characteristic, two lines are calculated, each of
which is defined by two significant data points
along this characteristic. The first line is deter-
mined by connecting the VREFOUT data point
taken at T = -40°C to the VREFOUT data point lo-
cated at the peak of the characteristic (i.e., the top
of the parabola). The second line is defined by con-
necting the VREFOUT data point located at the
peak of the characteristic to the VREFOUT data
point taken at T = +80°C. Determine (in units of
ppm/°C) the slopes of these two lines. The higher
of these two slopes is the VREFOUT Temperature
Coefficient.
The VREFOUT Temperature Coefficient (tempco)
for
a given CS5460A sample is typically
25 ppm/°C of drift. The maximum VREFOUT
Temperature Coefficient that any particular
CS5460A sample will exhibit is typically
60 ppm/°C of drift. But this maximum VREFOUT
tempco value is not guaranteed. If higher accura-
cy/stability is required, an external reference can be
used, in which case the VREFOUT pin can be left
unconnected.
In some cases, the designer may prefer to charac-
terize the energy registration drift of the entire de-
vice over temperature, as opposed to the drift of the
voltage reference only. A “Device-Level Energy
Registration Temperature Coefficient” (referred to
as T
Edevice
) can be defined in accordance with the
temperature coefficient requirement specified in
the IEC 1036/687 Standard.
[Specifically, this
T
Edevice
specification is defined as the CS5460A’s
maximum “mean temperature coefficient” (as de-
fined in IEC1036, Section 4.6.3) over the tempera-
ture range specified per the “Limit range of
operation” for Outdoor Meters (see IEC 1036, Sec-
tion 4.3.1)]. Instead of using the VREFOUT Volt-
age vs. Temperature characteristic, T
Edevice
is
based on the device’s Energy Registration Drift vs.
Temperature characteristic (for any particular
CS5460A sample), which indicates the drift of the
device’s energy registration drift over temperature,
using either the registration of the device’s ener-
gy-to-pulse engine or the Energy Register. Note
that typically this characteristic will have a shape
that is roughly similar to the shape to the sample’s
VREFOUT vs. Temperature Characteristic, shown
in Figure 17. T
Edevice
is defined as the greatest slope
of all possible lines that are defined by connecting
any two data points which lie along the Energy
Registration Drift vs. Temperature characteristic
such that 1) the two data points are separated by a
span of 20 degrees (Celsius) and 2) the temperature
value at the mid-point of the 20-degree span must
be within the temperature range between T = -25°C
and T = +60°C.
4.8 Calibration
4.8.1 Overview of Calibration Process
The CS5460A offers digital calibration. The user
determines which calibration sequence will be exe-
cuted by setting/clearing one or more of the 8 bits
in the calibration command word. For both chan-
nels, there are calibration sequences for both AC
and DC purposes. Regardless of whether an AC or
DC calibration sequence is desired, there are two
basic types of calibrations: system offset and sys-
tem gain. When the calibration sequences are be-
ing performed by the CS5460A, the user must
supply the input calibration signals to the “+” and
“-” pins of the voltage-/current-channel inputs.
These
input calibration signals represent
full-scale levels (for gain calibrations) and ground
input levels (for offset calibrations).
The AC and DC calibration sequences are differ-
enct. Depending on the user’s specific metering
application and accuracy requirements, some or all
of the calibration sequences may not be executed
by the user. (This is explained in more detail in the
following paragraphs). In order to help the reader
to better understand the functionality of each cali-