Data Sheet

ADL5511

APPLICATIONS INFORMATION

analog.com

Rev. E | 21 of 26

For applications that are not response time critical, a relatively

large capacitor can be placed on the FLT4. There is no maximum

capacitance limit for CFLT4.

Figure 50 shows how output noise, rise time and fall time vary

vs. CFLT4 when the ADL5511 is driven by an 1.9 GHz LTE carrier

with a bandwidth of 10 MHz (LTE Test Model E‑TM1_1_10MHz,

peak-to-average ratio = 11.99 dB).

Figure 50. Output Noise, Rise and Fall Times vs. CFLT4 Capacitance, 10 MHz

BW LTE Carrier (LTE Test Model E-TM1_1_10MHz) at 1.9 GHz with PIN = 0

dBm

ENVELOPE TRACKING ACCURACY

The envelope tracking accuracy of the ADL5511 is measured in

terms of the higher order distortion of the envelope output when

the RF input signal is AM modulated using a low-harmonic sinusoid

at a given frequency. Such an input sinusoidal envelope has been

generated using the ADL5390 multiplier modulator. This generates

a double sideband AM modulated signal of a known modulation

index. In this measurement, the ADL5511 acts as free-running AM

demodulator without requiring a local oscillator to demodulate the

signal.

Figure 51. THD on VENV vs. RF Input Level; 1900 MHz RF Input, AM

Modulated by a 20 MHz Sine Wave (Modulation Index = 0.25), VENV Output

AC-Coupled into a 50 Ω Spectrum Analyzer Load

Figure 51 shows such a plot total harmonic distortion (THD) of the

VENV output vs. RF input power for the modulation index of 0.25.

As the input power level increases, the THD improves until it

sharply degrades at an input power level of approximately 13 dBm.

This sharp decrease is caused by the clipping of the AM signal’s

peak envelope. Figure 51 also shows carrier leakage at VENV in

dBc with respect to the input carrier amplitude.

This measurement, when conducted over the full input power range

of the ADL5511, suffers from measurement inaccuracies of the

input modulated signal due to the spectrum analyzer’s noise floor

and therefore does not accurately reveal the ADL5511’s limitations

at the lower end of the measurement range. In addition to this, the

process of generating an AM signal for this test (using the ADL5390

multiplier) is not perfect and resulted in a source signal whose

envelope was not harmonically pure.

TIME DOMAIN ENVELOPE TRACKING

ACCURACY

The envelope tracking accuracy of the ADL5511 can also be as-

sessed in the time domain by looking at the input peak power levels

that cause clipping.

The usable rms input power range of the ADL5511 varies depend-

ing on the desired accuracy level and the peak-to-average ratio of

the input signal. Figure 4 shows the linear operating range of the

VENV output when the RF input is driven by unmodulated sine

waves at various frequencies. This shows operation up to rms input

levels of approximately 19 dBm. If the signal has a peak-to-average

ratio that is greater than the square root of two, the usable input

range on RFIN will decrease. In general, the maximum input power

for linear operation should be determined by the peak envelope

power (PEP) of the input signal. Figure 52 shows the time-domain

response of the VENV output to a 900 MHz LTE carrier with a

bandwidth of 20 MHz (Test Model E-TM1_2_20MHz). The signal

level of the carrier (7 dBm rms, 19 dBm PEP) was deliberately

increased until clipping was observed at the VENV output.

Note that the peak envelope power of a signal is derived based on

the rms level of the signal during a peak cycle, that is V p-p/√2. For

example, a signal that achieves a peak voltage of 10 V (or 20 V

p-p) has a PEP of 30 dBm. According to this definition, the PEP of a

sine wave is equal to its rms power level because it has a constant

envelope.