
+
20 k
380 k
19 k
V = 15 V
V+ = +15 V
4
7
2
3
1
6
5
VOUT
RS = 10
VCM = 265 V
RC = 10
(1)
IMAX = 1 A
SBOS579B – SEPTEMBER 2011 – REVISED JULY 2012
ERROR BUDGET ANALYSIS
The following error budget analysis demonstrates the importance of a high common-mode rejection ratio when
measuring small differential signals in the presence of high common-mode voltages.
Figure 48 shows a typical
current measurement application.
Figure 48. Typical Current Measurement Application
The maximum current through the shunt resistor (RS) is 1 A and generates a full-scale voltage drop of 10 V. All
error sources in this calculation are shown in relation to this full-scale voltage. The common-mode voltage in this
scenario is 265 V and the temperature range is from room temperature (+25°C) to +85°C.
Table 2 shows the
dominant error sources for the INA149 and a competitor device.
Table 2. Error Budget Analysis
ERROR (ppm of FS)
ERROR
SOURCE
INA149
COMPETIOR A
INA149
COMPETITOR A
Accuracy, TA = +25°C
Initial gain error
0.02% FS
0.05% FS
200
500
Offset voltage
1100 V
1000 V
110
100
Common mode
265 V/90 dB = 8380 V
265 V/77 dB = 37432 V
838
3743
Total acuracy error
1148
4343
Temperature drift
Gain
10 ppm/°C × 60°C
600
Offset voltage
10 V/°C × 60°C
20 V/°C × 60°C
60
120
Total drift error
660
720
Total error
1808
5063
If a smaller shunt resistor is used, the full-scale voltage drop is also smaller. A shunt resistor of 1
Ω causes a 1-V
voltage drop with a current of 1 A flowing through it. The error of 1808 ppm for a full-scale voltage of 10 V
becomes 18080 ppm (1.6%) for a full-scale voltage of only 1 V.
This example demonstrates that the dominate source of error, even over temperature, comes from the CMRR
specification of the devices. The common-mode error is 46% of the total error for the INA149 and 74% of the
total error for the competitor device.
Copyright 2011–2012, Texas Instruments Incorporated
17
Product Folder Link(s):
INA149