ADF4157 Data Sheet
Rev. D | Page 18 of 24
APPLICATIONS INFORMATION
INITIALIZATION SEQUENCE
After powering up the part, this programming sequence must
be followed:
1. Test register (R4)
2. Function register (R3)
3. R divider register (R2)
4. LSB FRAC register (R1)
5. FRAC/INT register (R0)
RF SYNTHESIZER: A WORKED EXAMPLE
The following equation governs how the synthesizer should be
programmed:
RF
OUT
= [N + (FRAC/2
25
)] × [f
PFD
] (3)
where:
RF
OUT
is the RF frequency output.
N is the integer division factor.
FRAC is the fractionality.
f
PFD
= REF
IN
× [(1 + D)/(R × (1 + T))] (4)
where:
REF
IN
is the reference frequency input.
D is the RF REF
IN
doubler bit.
R is the RF reference division factor.
T is the reference divide-by-2 bit (0 or 1).
For example, in a system where a 5.8002 GHz RF frequency
output (RF
OUT
) is required and a 10 MHz reference frequency
input (REF
IN
) is available, the frequency resolution is
f
RES
= REF
IN
/2
25
f
RES
= 10 MHz/2
25
= 0.298 Hz
From Equation 4,
f
PFD
= [10 MHz × (1 + 0)/1] = 10 MHz
5.8002 GHz = 10 MHz × (N + FRAC/2
25
)
Calculating N and FRAC values,
N = int(RF
OUT
/f
PFD
) = 580
FRAC = F
MSB
× 2
13
+ F
LSB
F
MSB
= int(((RF
OUT
/f
PFD
) N) × 2
12
) = 81
F
LSB
= int(((((RF
OUT
/f
PFD
) N) × 2
12
) F
MSB
) × 2
13
) = 7537
where:
F
MSB
is the 12-bit MSB FRAC value in Register R0.
F
LSB
is the 13-bit LSB FRAC value in Register R1.
int() makes an integer of the argument in brackets.
REFERENCE DOUBLER AND REFERENCE DIVIDER
The on-chip reference doubler allows the input reference signal
to be doubled. This is useful for increasing the PFD comparison
frequency. Making the PFD frequency higher improves the noise
performance of the system. Doubling the PFD frequency
usually improves noise performance by 3 dB. It is important to
note that the PFD cannot be operated above 32 MHz due to
a limitation in the speed of the Σ-Δ circuit of the N divider.
CYCLE SLIP REDUCTION FOR FASTER LOCK TIMES
In fastlocking applications, a wide loop filter bandwidth is
required for fast frequency acquisition, resulting in increased
integrated phase noise and reduced spur attenuation. Using
cycle slip reduction, the loop bandwidth can be kept narrow to
reduce integrated phase noise and attenuate spurs while still
realizing fast lock times.
Cycle Slips
Cycle slips occur in integer-N/fractional-N synthesizers when
the loop bandwidth is narrow compared to the PFD frequency.
The phase error at the PFD inputs accumulates too fast for the PLL
to correct, and the charge pump temporarily pumps in the wrong
direction, slowing down the lock time dramatically. The ADF4157
contains a cycle slip reduction circuit to extend the linear range
of the PFD, allowing faster lock times without loop filter changes.
When the ADF4157 detects that a cycle slip is about to occur, it
turns on an extra charge pump current cell. This outputs a constant
current to the loop filter or removes a constant current from the
loop filter (depending on whether the VCO tuning voltage needs
to increase or decrease to acquire the new frequency). The effect is
that the linear range of the PFD is increased. Stability is main-
tained because the current is constant and is not a pulsed current.
If the phase error increases again to a point where another cycle
slip is likely, the ADF4157 turns on another charge pump cell.
This continues until the ADF4157 detects that the VCO frequency
has exceeded the desired frequency. It then begins to turn off
the extra charge pump cells one by one until they are all turned
off and the frequency is settled.
Up to seven extra charge pump cells can be turned on. In most
applications, it is enough to eliminate cycle slips altogether,
giving much faster lock times.
Setting Bit DB28 in the R Divider register (R2) to 1 enables cycle
slip reduction. Note that a 45% to 55% duty cycle is needed on
the signal at the PFD for CSR to operate correctly. The reference
divide-by-2 flip-flop can help to provide a 50% duty cycle at the
PFD. For example, if a 100 MHz reference frequency is available,
and the user wants to run the PFD at 10 MHz, setting the R divide
factor to 10 results in a 10 MHz PFD signal that is not 50% duty
cycle. By setting the R divide factor to 5 and enabling the reference
divide-by-2 bit, a 50% duty cycle 10 MHz signal can be achieved.
Note that the cycle slip reduction feature can only be operated
when the phase detector polarity setting is positive (DB6 in
Register 3). It cannot be used if the phase detector polarity is
set to negative.
Data Sheet ADF4157
Rev. D | Page 19 of 24
FASTLOCK TIMER AND REGISTER SEQUENCES
If the fastlock mode is used, a timer value needs to be loaded into
the PLL to determine the time spent in wide bandwidth mode.
When Bits DB[20:19] in Register 4 (R4) are set to 01 (switched
R fastlock enable), the timer value is loaded via the 12-bit clock
divider value. To use fastlock, the PLL must be written to in the
following sequence:
1. Use the initialization sequence (see the Initialization
Sequence section) only once after powering up the part.
2. Load Register 4 (R4) with Bits DB[20:19] set to 01 and the
chosen fastlock timer value (DB18 to DB7). Note that the
duration that the PLL remains in wide bandwidth is equal
to the fastlock timer/f
PFD
.
FASTLOCK: AN EXAMPLE
If a PLL has f
PFD
= 13 MHz and a required lock time of 50 µs,
the PLL is set to wide bandwidth for 40 µs.
If the time period set for the wide bandwidth is 40 µs, then
Fastlock Timer Value = Time in Wide Bandwidth × f
PFD
Fastlock Timer Value = 40 µs × 13 MHz = 520
Therefore, 520 must be loaded into the clock divider value in
Register 4 (R4) in Step 2 of the sequence described in the
Fastlock Timer and Register Sequences section.
FASTLOCK: LOOP FILTER TOPOLOGY
To use fast-lock mode, an extra connection from the PLL to the
loop filter is needed. The damping resistor in the loop filter must
be reduced to ¼ of its value while in wide bandwidth mode. This is
required because the charge pump current is increased by 16
while in wide bandwidth mode, and stability must be ensured.
During fastlock, the MUXOUT pin (after setting MUXOUT to
fastlock switch by setting Bits DB[30:27] in Register 0 to 1100) is
shorted to ground (this is accomplished by settings Bits DB[20:19]
in Register 4 to 01switched R fastlock enable). The following
two topologies can be used:
Divide the damping resistor (R1) into two values (R1 and
R1A) that have a ratio of 1:3 (see Figure 22).
Connect an extra resistor (R1A) directly from MUXOUT,
as shown in Figure 23. The extra resistor must be chosen
such that the parallel combination of an extra resistor and
the damping resistor (R1) is reduced to ¼ of the original
value of R1 (see Figure 23).
ADF4157
CP
MUXOUT
C1
C2
R2
R1
R1A
C3
VCO
05874-022
Figure 22. Fast-Lock Loop Filter TopologyTopology 1
ADF4157
CP
MUXOUT
C1 C2
R2
R1R1A
C3
VCO
05874-023
Figure 23. Fastlock Loop Filter TopologyTopology 2
SPUR MECHANISMS
The fractional interpolator in the ADF4157 is a third-order Σ-Δ
modulator (SDM) with a 25-bit fixed modulus (MOD). The
SDM is clocked at the PFD reference rate (f
PFD
) that allows PLL
output frequencies to be synthesized at a channel step resolution of
f
PFD
/MOD. The various spur mechanisms possible with fractional-
N synthesizers, and how they affect the ADF4157, are discussed in
this section.
Fractional Spurs
In most fractional synthesizers, fractional spurs can appear at
the set channel spacing of the synthesizer. In the ADF4157,
these spurs do not appear. The high value of the fixed modulus
in the ADF4157 makes the Σ-Δ modulator quantization error
spectrum look like broadband noise, effectively spreading the
fractional spurs into noise.
Integer Boundary Spurs
Interactions between the RF VCO frequency and the PFD fre-
quency can lead to spurs known as integer boundary spurs. When
these frequencies are not integer related (which is the purpose
of the fractional-N synthesizer), spur sidebands appear on the
VCO output spectrum at an offset frequency that corresponds
to the beat note or difference frequency between an integer mul-
tiple of the PFD and the VCO frequency.
These spurs are named integer boundary spurs because they are
more noticeable on channels close to integer multiples of the PFD
where the difference frequency can be inside the loop bandwidth.
These spurs are attenuated by the loop filter.
Figure 7 shows an integer boundary spur. The RF frequency is
5800.25 MHz, and the PFD frequency is 25 MHz. The integer
boundary spur is 250 kHz from the carrier at an integer times
the PFD frequency (232 × 25 MHz = 5800 MHz). The spur also
appears on the upper sideband.
Reference Spurs
Reference spurs are generally not a problem in fractional-N
synthesizers because the reference offset is far outside the loop
bandwidth. However, any reference feedthrough mechanism
that bypasses the loop can cause a problem. One such mechanism
is the feedthrough of low levels of on-chip reference switching
noise out through the RF
IN
x pin back to the VCO, resulting in
reference spur levels as high as 90 dBc. Care should be taken in
the PCB layout to ensure that the VCO is well separated from
the input reference to avoid a possible feedthrough path on
the board.
ADF4157 Data Sheet
Rev. D | Page 20 of 24
LOW FREQUENCY APPLICATIONS
The specification on the RF input is 0.5 GHz minimum; however,
RF frequencies lower than this can be used, providing the mini-
mum slew rate specification of 400 V/µs is met. An appropriate
LVDS driver can be used to square up the RF signal before it is
fed back to the ADF4157 RF input. The FIN1001 from Fairchild
Semiconductor is one such LVDS driver.
FILTER DESIGNADIsimPLL
A filter design and analysis program is available to help the user
implement PLL design. Visit www.analog.com/pll for a free
download of the ADIsimPLLsoftware. The software designs,
simulates, and analyzes the entire PLL frequency domain and
time domain response. Various passive and active filter architec-
tures are allowed.
OPERATING WITH WIDE LOOP FILTER
BANDWIDTHS
If a wide loop filter bandwidth is used (>60 kHz), fluctuations
in the phase noise profile may be noticed on channels that are
close to integer multiples of the PFD frequency. This is due to
operation of the charge pump close to the dead zone. To improve
the phase noise, a bleed current can be enabled to bias the charge
pump away from the dead zone. To enable this, set Bit DB[24:23]
in Register 4. Using this mode has the added advantage of
improving the integer boundary spurs by 4 dB to 5 dB. Note
that it is also safe to use this mode if the loop filter bandwidth
is <60 kHz.
PCB DESIGN GUIDELINES FOR THE CHIP SCALE
PACKAGE
The lands on the chip scale package (CP-20) are rectangular.
The printed circuit board pad for these should be 0.1 mm
longer than the package land length and 0.05 mm wider than
the package land width. The land should be centered on the pad.
This ensures that the solder joint size is maximized.
The bottom of the chip scale package has a central thermal pad.
The thermal pad on the printed circuit board (PCB) should be
at least as large as the exposed pad. On the printed circuit
board, there should be a clearance of at least 0.25 mm between
the thermal pad and the inner edges of the pad pattern. This
ensures that shorting is avoided.
Thermal vias can be used on the PCB thermal pad to improve
thermal performance of the package. If vias are used, they should
be incorporated into the thermal pad at 1.2 mm pitch grid. The
via diameter should be between 0.3 mm and 0.33 mm, and the
via barrel should be plated with 1 ounce of copper to plug the
via. The user should connect the PCB thermal pad to AGND.

EV-ADF4157SD1Z

Mfr. #:
Manufacturer:
Analog Devices Inc.
Description:
ADF4157 Clock Generator and Synthesizer Evaluation Board
Lifecycle:
New from this manufacturer.
Delivery:
DHL FedEx Ups TNT EMS
Payment:
T/T Paypal Visa MoneyGram Western Union