Data Sheet

ADL5511

APPLICATIONS INFORMATION

analog.com

Rev. E | 20 of 26

CFLT3= 1

2π×fFLT3×250 Ω

−0.8 pF

(7)

where fLT2 and fLT3 are the desired corner frequencies.

For example, to set the corner frequency to 200 MHz, CFLT2 and

CFLT3 should be set to 1.6 pF and 2.4 pF, respectively. The two

corner frequencies should be set so that they are approximately

equal.

Care should be taken not to set the corner frequency of this

carrier suppression filter too low as it will start to degrade envelope

bandwidth. The ADL5511 has an envelope bandwidth of 130 MHz.

Thus, if the capacitors on FLT2 and FLT3 are so big that the

carrier-suppression corner frequency approaches 130 MHz, the

carrier filtering effort will directly impact the envelope bandwidth.

Thus, the corner frequency should be set low enough so that the

RF carrier is adequately removed from the envelope output while

still maintaining the desired envelope bandwidth. An alternative

option would be to filter the carrier at the VENV output using a

higher order filter.

CHOOSING A VALUE FOR THE RMS

AVERAGING CAPACITOR (CFLT4)

CFLT4 provides the averaging function for the internal rms computa-

tion, the result of which is available at the VRMS output. As already

noted, the on-chip rms filtering corner is internally set by a 400

Ω resistor and a 20 pF capacitor, yielding a corner frequency of

approximately 20 MHz.

For adequate rms filtering, connect an external filter capacitor

between FLT4 (Pin 14) and VPOS (Pin 15). This capacitance acts

on the internal 400 Ω resistor to yield a new corner frequency for

the rms filter given by the following equation:

CFLT4= 1

2π×fFLT4×400 Ω

−20 pF

(8)

For example, a supply-referenced 0.1 µF capacitor on FLT4 re-

duces the corner frequency of the rms averaging circuit to approxi-

mately 4 kHz.

The size of the rms filtering capacitor has a direct impact on the rms

accuracy up to a point. For most accurate detection, the rms filter

corner should be low enough to filter out most of the modulation

content. This corresponds to a corner frequency that is significantly

less than the bandwidth of the signal being measured.

Table 4 shows recommended minimum values of CFLT4 for popular

modulation schemes. Using smaller capacitor values than these

will result in rms measurement errors; using higher values will not

further improve rms accuracy but will reduce the output noise on

VRMS at the expense of increased rise and fall times. In Table 4,

rise and fall times are also shown along with residual output noise.

The recommended minimum values for CFLT4 were experimentally

determined by starting out with a large capacitance value on the

FLT4 pin (for example, 10 µF). The value of VRMS was noted

for a fixed input power level (for example, 0 dBm). The value

of CFLT4 was then progressively reduced (this can be done with

press-down capacitors) until the value of VRMS started to deviate

from its original value (this indicates that the accuracy of the rms

computation is degrading and that CFLT4 is becoming too small).

The recommended minimum value for CFLT4 is roughly inversely

proportional to the bandwidth of the input signal, that is, wider band-

width signals tend to require smaller minimum filter capacitances.

As already noted, the value of CFLT4 sets up an internal low pass

corner frequency, which filters the rms voltage. As carrier bandwidth

increases, a larger proportion of the residual noise (which has

been effectively mixed down to baseband) is filtered away. This re-

sults in smaller capacitances being required as carrier bandwidths

increase.

Table 4. Recommended Minimum CFLT4 Values for Various Modulation Schemes (Pin = 0 dBm)

Modulation/Standard

PEP to RMS Ratio

Signal

Bandwidth

CFLT4 (Min)

Output Noise

Rise/Fall Time (10% to 90%)

W-CDMA, One-Carrier, TM1-64

9.83 dB

3.84 MHz

220 nF

98 mV p-p

82 µs/310 µs

W-CDMA Four-Carrier, TM1-64, TM1-32, TM1-16,

TM1-8

12.08 dB

18.84 MHz

100 nF

140 mV p-p

40 µs/140 µs

LTE Test Model E-TM1_1_4MHz

9.83 dB

4 MHz

220 nF

135 mV p-p

82 µs/310 µs

LTE Test Model E-TM1_1_10MHz

11.99 dB

10 MHz

100 nF

89 mV p-p

40 µs/140 µs

LTE Test Model E-TM1_1_20MHz

11.58 dB

20 MHz

47 nF

90 mV p-p

20 µs/70 µs