RF3809
Theory of Operation and Application Information
RF3809 design accommodates use in a variety of applications:
• Linear driver from 450MHz to 2500MHz
• 2nd/3rd stage high linearity LNA, with noise figure in the 3dB to 4dB range from 800MHz to 2200MHz
• High efficiency (>50%) output stage for non-linear applications
Nominal data sheet shows specification for V
CC
=V
BIAS
=V
REF
=8V. RF3809 can easily be configured for 5V operation, with a
simple bias resistor change at V
REF
.. “Bias Table” on page 5 shows resistor values for V
CC
=V
BIAS
=V
REF
=5V. Generally speak-
ing, 5V data will compare to that for 8V as follows:
• 3dB to 3.5dB reduction in OP1dB
• 0.4dB to 0.5dB increase in small signal gain
For operation at other than 5V, bias R can be calculated as follows (V
CC
=V
BIAS
=V
REF
=5V is used here to illustrate, operation at
different voltage is determined with same methodology).
1. Use nominal 8V case as a starting point: V
CC
=V
BIAS
=V
REF
=8V, I
REF
=15mA, I
CQ
=258mA. Target condition will be to
achieve same I
CQ
with V
CC
=V
BIAS
=V
REF
=5V.
2. Using evaluation board with separate lab supplies on (V
CC
/V
BIAS
) and (V
REF
), set V
CC
/V
BIAS
=5V, V
REF
=8V. I
REF
is main-
tained at 15mA, and I
CQ
drops from nominal value of 258mA.
3. V
REF
can then be increased >8V until I
CQ
is restored. I
REF
increase to 23mA is required (as seen in “Bias Table” on page 5).
4. At this point, pin voltage at V
REF
is calculated (or measured with DVM): V
PIN
=V
REF
at eval board input – I
REF
*bias R=10.8–
0.023*300=3.9V.
5. Next, calculate new bias R for V
REF
=5V: Bias R=(5–3.9)/0.023=47.8Ω. See “Bias Table” on page 5, standard resistor
value=47Ω is called out. In this way, bias R can be calculated for any V
CC
=V
BIAS
=V
REF
configuration. The maximum I
REF
limit for RF3809=30mA.
Junction-to-case thermal resistance (R
TH_JC
) is shown versus output power in the graph section of this data sheet. The graph
was generated with nominal V
CC
=V
BIAS
=V
REF
=8V, I
REF
=15mA, where ambient temperature=85°C. Using this curve along
with operating condition, junction temperature can be calculated. Resultant T
J
for this case yields MTTF≥100 years. Standard
RF3809 evaluation boards are matched for high efficiency at O
P1dB
. To ensure reliability for operation at high power, output
match achieving equivalent or better efficiency on system board should be the goal.
Typical s-parameter responses for each evaluation board are shown within the data sheet. These boards were matched with
two specifications in mind:
• Output load impedance set for optimum OIP
3
/ACP (Adjacent Channel Power for commonly used modulation standards).
• Output load impedance set for high efficiency at O
P1dB
, with ruggedness (survival) into output 4:1 VSWR.
In some cases, low power operation being one, it may be desirable to improve output return loss seen on evaluation board.
This can be done with output match adjust. The result will be an increase in small signal gain. Tradeoffs between return loss,
gain, OIP
3
, and compression point can then be considered in obtaining optimum performance for a particular application.
Finally, infrastructure qualification report for RF3809 can be obtained by contacting RFMD.
Rev A4 DS080221
7628 Thorndike Road, Greensboro, NC 27409-9421 · For sales or technical
support, contact RFMD at (+1) 336-678-5570 or sales-support@rfmd.com.
7 of 22
RFMD [ RF MICRO DEVICES ]
