ADL5519ACPZ-R7 datasheet(24/40 Pages) AD

24 / 40 page

ADL5519

Rev. 0 | Page 24 of 40

DEVICE CALIBRATION

The measured transfer function of the ADL5519 at 2.2 GHz is

shown in Figure 60. The figure shows plots of both output voltage

vs. input power and calculated error vs. input power. As the input

power varies from −60 dBm to −5 dBm, the output voltage varies

from 1.7 V to about 0.5 V.

VOUT2

PIN1

PIN2

PIN (dBm)

VOUT1

–60

–50

–40

–30

–20

–10

2.00

1.75

1.50

1.25

1.00

0.75

0.50

0.25

2.0

1.5

1.0

0.5

–0.5

–1.0

–1.5

–2.0

Figure 60. Transfer Function at 2.2 GHz with Calibration Points

Because slope and intercept vary from device to device, board-

level calibration must be performed to achieve the highest

accuracy. The equation for output voltage can be written as

VOUT = Slope × (PIN − Intercept)

(6)

where:

Slope is the change in output voltage divided by the change in

input power, PIN, expressed in decibels (dB).

Intercept is the calculated power at which the output voltage

would be 0 V. Note that an output voltage of 0 V can never be

achieved.

In general, calibration is performed by applying two known

signal levels to the ADL5519 input and measuring the corre-

sponding output voltages. The calibration points are generally

chosen to be within the linear-in-dB operating range of the

device (see the Specifications section for more details).

Calculation of the slope and intercept is accomplished using the

following equations:

Slope = (VOUT1 − VOUT2)/(PIN1 − PIN2)

(7)

Intercept = PIN1 − (VOUT1/Slope)

(8)

Once slope and intercept are calculated, an equation can be

written that calculates the input power based on the output

voltage of the detector.

PIN (Unknown) = (VOUT1(MEASURED)/Slope) + Intercept

(9)

The log conformance error of the calculated power is given by

Error (dB) = (VOUT(MEASURED) − VOUT(IDEAL))/Slope

(10)

Figure 60 includes a plot of the error at 25°C, the temperature

at which the log amp is calibrated. Note that the error is not 0 dB

over the full dynamic range. This is because the log amp does

not perfectly follow the ideal VOUT vs. PIN equation, even within

its operating region. The error at the calibration points of −35

dBm and −11 dBm is equal to 0 dB, by definition.

Figure 60 also shows error plots for the output voltage at

−40°C and +85°C. These error plots are calculated using the

slope and intercept at 25°C. This is consistent with calibration

in a mass-production environment, where calibration over

temperature is not practical.

ADJUSTING ACCURACY THROUGH CHOICE OF

CALIBRATION POINTS

In some applications, very high accuracy is required at one

power level or over a reduced input range. For example, in a

wireless transmitter, the accuracy of the high power amplifier

(HPA) is most critical at or close to full power.

In applications like AGC control loops, good linearity and

temperature performance are necessary over a large input power

range. The temperature crossover point (the power level at which

there is no drift in performance from −40°C to −80°C) can be

shifted from high power levels to midpower levels using the

method shown in the Temperature Compensation Adjustment

section. This shift equalizes the temperature performance over

the complete power range. The linearity of the transfer function

can be equalized by changing the calibration points.

Figure 61 demonstrates this equalization by changing the cali-

bration points used in Figure 60 to −46 dBm and −22 dBm. This

adjustment of the calibration points changes the linearity to greater

than ±0.25 dB over a 50 dB dynamic range at the expense of a

slight decrease in linearity at power levels between −40 dBm

and −25 dBm.

Calibration points should be chosen to suit the application at hand.

In general, however, do not choose calibration points in the

nonlinear portion of the log amp transfer function (greater than

−10 dBm or less than −40 dBm, in this example).