power based on the output voltage of the detector.
鈭?/div>
V
OUT(IDEAL)
)/Slope
(12)
Figure 31. Output Voltage and Error vs. P
IN
with 2-Point Calibration at
鈥?0 dBm and 鈥?0 dBm
Figure 30 includes a plot of the error at 25掳C, the temperature at
which the log amp is calibrated. Note that the error is not zero.
This is because the log amp does not perfectly follow the ideal
V
OUT
versus
P
IN
equation, even within its operating region. The
error at the calibration points (鈭?2 dBm and
鈭?2
dBm in this
case) will, however, be equal to zero by definition.
Figure 30 also includes error plots for the output voltage at
鈭?0掳C and +85 掳C. These error plots are calculated using the
slope and intercept at 25掳C. This is consistent with calibration in
a mass-production environment where calibration at
temperature is not practical.
Calibration points should be chosen to suit the application
at hand. In general, though, the calibration points should
never be chosen in the nonlinear portion of the log amp鈥檚
transfer function (above 鈭? dBm or below 鈭?0 dBm in this
case).
Figure 32 shows how calibration points can be adjusted to
increase dynamic range, but at the expense of linearity. In
this case the calibration points for slope and intercept are set
at 鈭? dBm and 鈭?0 dBm. These points are at the end of the
device鈥檚 linear range. Once again at 25掳C, we see an error of
0 dB at the calibration points. Note also that the range over
which the AD8318 maintains an error of < 卤1 dB is
extended to 60 dB at 25掳C and 58 dB over temperature. The
disadvantage of this approach is that linearity suffers,
especially at the top end of the input range.
2.2
2.0
V
OUT
+25
掳
C
V
OUT
鈥?0
掳
C
V
OUT
+85
掳
C
ERROR +25
掳
C
ERROR 鈥?0
掳
C
ERROR +85
掳
C
2.5
2.0
1.5
1.0
ERROR (dB)
04853-038
SELECTING CALIBRATION POINTS TO IMPROVE
ACCURACY OVER A REDUCED RANGE
In some applications very high accuracy is required at just one
power level or over a reduced input range. For example, in a
wireless transmitter, the accuracy of the high power amplifier
(HPA) will be most critical at or close to full power.
Figure 31 shows the same measured data as Figure 30. Notice
that accuracy is very high from 鈭?0 dBm to 鈭?0 dBm. Below
鈭?0 dBm the error increases to about 鈭? dB. This is because the
calibration points have been changed to 鈭?4 dBm and 鈭?6 dBm.
1.8
1.6
1.4
V
OUT
(V)
0.5
0
鈥?.5
鈥?.0
58dB DYNAMIC RANGE (
卤
1dB ERROR)
鈥?.5
鈥?.0
鈥?.5
0
5
1.2
1.0
0.8
0.6
0.4
0.2
鈥?5 鈥?0 鈥?5 鈥?0 鈥?5 鈥?0 鈥?5 鈥?0 鈥?5 鈥?0 鈥?5 鈥?0 鈥?
P
IN
(dBm)
Figure 32. Dynamic Range Extension by Choosing Calibration Points
that are Close to the End of the Linear Range
Another way of presenting the error function of a log amp
detector is shown in Figure 33. In this case, the dB error at
hot and cold temperatures is calculated with respect to the
Rev. 0 | Page 16 of 24
ERROR (dB)
0.5