NSC LMX2541SQ3030E

July 2, 2009
LMX2541
Ultra-Low Noise PLLatinum Frequency Synthesizer with
Integrated VCO
General Description
Features
The LMX2541 is an ultra low noise frequency synthesizer
which integrates a high performance delta-sigma fractional N
PLL, a VCO with fully integrated tank circuit, and an optional
frequency divider. The PLL offers an unprecedented normalized noise floor of -225 dBc/Hz and can be operated with up
to 104 MHz of phase-detector rate (comparison frequency) in
both integer and fractional modes. The PLL can also be configured to work with an external VCO.
The LMX2541 integrates several low-noise, high precision
LDOs and output driver matching network to provide higher
supply noise immunity and more consistent performance,
while reducing the number of external components. When
combined with a high quality reference oscillator, the
LMX2541 generates a very stable, ultra low noise signal.
The LMX2541 is offered in a family of 6 devices with varying
VCO frequency range from 1990 MHz up to 4 GHz. Using a
flexible divider, the LMX2541 can generate frequencies as
low as 31.6 MHz. The LMX2541 is a monolithic integrated
circuit, fabricated in a proprietary BiCMOS process. Device
programming is facilitated using a three-wire MICROWIRE
interface that can operate down to 1.8 volts. Supply voltage
ranges from 3.15 to 3.45 volts. The LMX2541 is available in
a 36 pin 6x6x0.8 mm Lead-Free Leadless Leadframe Package (LLP).
■ Very Low RMS Noise and Spurs
Device
VCO Frequency
LMX2541SQ2060E
1990 - 2240
LMX2541SQ2380E
2200 - 2530
LMX2541SQ2690E
2490 - 2865
LMX2541SQ3030E
2810 - 3230
LMX2541SQ3320E
3130 - 3600
LMX2541SQ3740E
3480 - 4000
■
■
■
■
■
■
■
■
■
■
■
■
■
■
— -225 dBc/Hz Normalized PLL Phase Noise
— Integrated RMS Noise (100 Hz - 20 MHz)
■ 2 mrad (100 Hz - 20 MHz) at 2.1 GHz
■ 3.5 mrad (100 Hz - 20 MHz) at 3.5 GHz
Ultra Low-Noise Integrated VCO
External VCO Option (Internal VCO Bypassed)
VCO Frequency Divider 1 to 63 (all values)
Programmable Output Power
Up to 104 MHz Phase Detector Frequency
Integrated Low-Noise LDOs
Programmable Charge Pump Output
Partially Integrated Loop Filter
Digital Frequency Shift Keying (FSK) Modulation Pin
Integrated Reference Crystal Oscillator Circuit
Hardware and Software Power Down
FastLock Mode with Cycle Slip Reduction
Analog and Digital Lock Detect
1.6 V Logic Compatibility
Target Applications
■
■
■
■
Wireless Infrastructure (UMTS, LTE, WiMax)
Broadband Wireless
Wireless Meter Reading
Test and Measurement
System Diagram
30073322
© 2009 National Semiconductor Corporation
300733
www.national.com
LMX2541 Ultra-Low Noise PLLatinum Frequency Synthesizer with Integrated VCO
PRELIMINARY
LMX2541
LMX2541 Frequency Chart
Divide
Value
Start
Stop
Start
Stop
Start
Stop
Start
Stop
Start
Stop
Start
Stop
1
1990.0
2240.0
2200.0
2530.0
2490.0
2865.0
2810.0
3230.0
3130.0
3600.0
3480.0
4000.0
2
995.0
1120.0
1100.0
1265.0
1245.0
1432.5
1405.0
1615.0
1565.0
1800.0
1740.0
2000.0
3
663.3
746.7
733.3
843.3
830.0
955.0
936.7
1076.7
1043.3
1200.0
1160.0
1333.3
4
497.5
560.0
550.0
632.5
622.5
716.3
702.5
807.5
782.5
900.0
870.0
1000.0
5
398.0
448.0
440.0
506.0
498.0
573.0
562.0
646.0
626.0
720.0
696.0
800.0
6
331.7
373.3
366.7
421.7
415.0
477.5
468.3
538.3
521.7
600.0
580.0
666.7
7
284.3
320.0
314.3
361.4
355.7
409.3
401.4
461.4
447.1
514.3
497.1
571.4
8
248.8
280.0
275.0
316.3
311.3
358.1
351.3
403.8
391.3
450.0
435.0
500.0
2060E
2380E
2690E
3030E
3320E
3740E
…
…
…
…
…
…
…
…
…
…
…
…
…
63
31.6
35.6
34.9
40.2
39.5
45.5
44.6
51.3
49.7
57.1
55.2
63.5
All devices have continuous frequency coverage below a divide value of 8 (7 for most devices) down to their minimum frequency
achievable with divide by 63. The numbers in bold show the upper end of this minimum continuous frequency range. Below 570
MHz, all devices can be used down to their minimum frequency of Min (fVCO) / 63.
www.national.com
2
LMX2541
Functional Block Diagram
30073301
Connection Diagram
36-Pin SQ Package (Top View)
30073302
3
www.national.com
LMX2541
Pin Descriptions
Pin #
Name
Type
0
GND
GND
Description
The DAP pad must be grounded.
1
GND
GND
2
VregRFout
LDO Output
LDO Output for RF output buffer.
VccRFout
Supply
(LDO Input)
Supply for the RF output buffer.
4
L1
NC
Do not connect this pin.
5
Lmid
NC
Do not connect this pin.
6
L2
NC
Do not connect this pin.
7
VccVCO
Supply
(LDO Input)
Supply for the VCO.
8
VregVCO
LDO Output
LDO Output for RF output buffer.
9
VrefVCO
LDO Bypass
LDO Bypass
10
GND
GND
11
CE
CMOS
12
ExtVCOin
RF Input
13
VccPLL1
Supply
Power supply for PLL N counter.
14
VccCP1
Supply
Power supply for PLL charge pump up current.
15
Vtune
High-Z Input
16
CPout
Output
Charge pump output.
17
FLout
Output
Fastlock output.
18
VccCP2
Supply
Power supply for PLL charge pump down current.
19
VccPLL2
Supply
Power supply for PLL R Counter
20
Ftest/LD
Output
Software controllable multiplexed CMOS output.
Can be used to monitor PLL lock condition.
21
OSCin
High-Z Input
Oscillator input signal. If not being used with an external crystal, this input should be AC
coupled.
22
OSCin*
High-Z Input
Complementary oscillator input signal. Can also be used with an external crystal. If not
being used with an external crystal, this input should be AC coupled.
23
VccOSCin
Supply
24
RFoutEN
Input
25
VccFRAC
Supply
(LDO Input)
Power Supply for the fractional circuitry.
26
VregFRAC
LDO Output
Regulated power supply used for the fractional delta-sigma circuitry.
3
Chip Enable.
The device needs to be programmed for this pin to properly power down the device.
Optional input for use with an external VCO.
This pin should be AC coupled if used or left open if not used.
Tuning voltage input to the VCO.
Supply for the OSCin buffer.
Software programmable output enable pin.
27
GND
GND
28
VccDig
Supply
Supply for digital circuitry, such the MICROWIRE.
29
VccBias
Supply
Supply for Bias circuitry that is for the whole chip.
30
Bypass
Bypass
Put a cap to the VccBias pin.
31
VccDiv
Supply
Supply for the output divider
32
DATA
High-Z Input
MICROWIRE serial data input. High impedance CMOS input.
This pin must not exceed 3.45 V.
33
CLK
High-Z Input
MICROWIRE clock input. High impedance CMOS input.
This pin is used for the digital FSK modulation feature.
This pin must not exceed 3.45 V.
34
LE
High-Z Input
MICROWIRE Latch Enable input. High impedance CMOS input.
This pin must not exceed 3.45 V.
35
NC
NC
36
RFout
RF Output
www.national.com
No connect.
RF output. Must be AC coupled if used.
4
If Military/Aerospace specified devices are required, please contact the National Semiconductor Sales Office/ Distributors
for availability and specifications.
Parameter
Power Supply Voltage
Input Voltage to pins other than Vcc Pins
(Note 4)
Storage Temperature
Range
Symbol
Vcc
Ratings
-0.3 to 3.6
Units
V
VIN
-0.3 to (Vcc+0.3)
VIN
TSTG
-65 to 150
°C
Lead Temperature (solder 4 sec.)
TL
+ 260
°C
Recommended Operating Conditions
Parameter
Power Supply Voltage
(All Vcc Pins)
Symbol
Min
Typ
Max
Units
Vcc
3.15
3.3
3.45
V
Ambient Temperature
TA
-40
+85
°C
Note 1: “Absolute Maximum Ratings” indicate limits beyond which damage to the device may occur. Operating Ratings indicate conditions for which the device
is intended to be functional, but do not guarantee specific performance limits. For guaranteed specifications and test conditions, see the Electrical Characteristics.
The guaranteed specifications apply only to for the test conditions listed.
Note 2: This Device is a high performance RF integrated circuit with an ESD rating < 2kV and is ESD sensitive. Handling and assembly of this device should only
be done at ESD-free workstations.
Note 3: Stresses in excess of the absolute maximum ratings can cause permanent or latent damage to the device. These are absolute stress ratings only.
Functional operation of the device is only implied at these or any other conditions in excess of those given in the
Note 4: Never to exceed 3.6 V.
5
www.national.com
LMX2541
Absolute Maximum Ratings (Notes 1, 2, 3)
LMX2541
Electrical Characteristics
(3.15 V ≤ VCC ≤ 3.45 V, -40°C ≤ TA ≤ 85 °C; except as specified. Typical values are
at Vcc = 3.3 V, 25 C.)
Symbol
Parameter
Conditions
Min
Typ
Max
Units
VCO_DIV>1
170
204
VCO_DIV=1
130
156
72
85
mA
VCO_DIV >1
Default Power Mode
84
102
mA
CE = 0 V, Device Initialized
100
250
µA
300
µA
Current Consumption
ICC
Entire ChipPower Supply
Current with all blocks enabled
IPLL
Current for External VCO Mode
(Note 6)
IDIV
Current for Divider Only Mode
ICCPD
Power Down Current
Default Power
Mode
(Note 5)
RFoutEN = LOW
mA
Oscillator (Normal Mode Operation with XO=0)
IIHOSC Oscillator Input High Current for
OSCin and OSCin*
in
VIH = 2.75 V
IILOSCin
Oscillator Input Low Current for
OSCin and OSCin* pins
VIL = 0
-100
fOSCin
Frequency Range
Limited to ½ · fPD when the Oscillator doubler is enabled.
5
dvOSCin
Slew Rate
Single-Ended Mode
150
Single-Ended
0.2
2.0
Differential
0.4
3.1
5
20
MHz
100
Ω
vOSCin
Oscillator Sensitivity
fXTAL
Crystal Frequency Range
VIH = 2.75 V
ESRXTAL
Crystal Equivalent Series
Resistance
This a requirement for the crystal,
not a characteristic of the LMX2541.
PXTAL
Power Dissipation in Crystal
This requirement is for the crystal,
not a characteristic of the LMX2541.
COSCin
Input Capacitance of OSCin
µA
900
MHz
V/µs
Vpp
Oscillator (Crystal Mode with XO=1)
TBD
µW
6
pF
PLL
fPD
ICPout
Phase Detector Frequency
Charge Pump
Output Current Magnitude
104
CPG = 1X
100
CPG = 2X
200
CPG = 3X
300
...
...
CPG=32X
3200
MHz
µA
ICPoutTRI
CP TRI-STATE Current
0.4 V < VCPout < Vcc - 0.4
1
5
nA
ICPoutMM
Charge Pump
Sink vs. Source Mismatch
VCPout = Vcc / 2
TA = 25°C
3
10
%
ICPoutV
Charge Pump
Current vs. CP Voltage
Variation
0.4 V < VCPout < Vcc - 0.4
TA = 25°C
4
%
ICPoutT
CP Current vs. Temperature
Variation
VCPout = Vcc / 2
8
%
Normalized PLL 1/f Noise
LNPLL_flicker(10 kHz)
CPG = 1X
-116
CPG = 32X
-124.5
CPG = 1X
-220.8
CPG = 32X
-225.4
LN(f)
(Note 7)
Normalized PLL Noise Floor
LNPLL_flat(1 Hz)
fExtVCOin
PLL Input Frequency
pExtVCOin
PLL Input Sensitivity
www.national.com
dBc/Hz
dBc/Hz
RFout Buffer Enabled and VCO_DIV > 1
400
4000
RFout Buffer Disabled and VCO_DIV = 1
400
6000
fExtVCOin ≤ 4 GHz
-15
3
fExtVCOin > 4 GHz
-5
3
6
MHz
dBm
Parameter
Conditions
Min
Typ
Max
Units
VCO Specifications
fVCO
ΔTCL
pRFout
ΔPRFout
KVtune
Internal VCO Frequency Range
Mode = Full Chip Mode
This is the frequency before the
VCO divider.
Maximum Allowable
Temperature Drift for
Continuous Lock
RF Output Power
Change in Output Power
(Note 6)
Fine Tuning Sensitivity
2060E
1990
2240
2380E
2200
2530
2690E
2490
2865
3030E
2810
3230
3320E
3130
3600
3740E
3480
4000
(Note 6),(Note 8)
Maximum Frequency
Default Power Mode
VCO_DIV=1
125
°C
2060E
3.5
2380E
2.8
2690E
1.6
3030E
1.2
3320E
0.2
3740E
- 0.3
Fixed Temperature with 100 MHz frequency change at
the output
1
Fixed frequency with a change over the entire
temperature range
0.5
The lower number in the range
applies when the VCO is at its
lowest frequency and the higher
number applies when the VCO is
at its highest frequency. A linear
approximation can be used for
frequencies between these two
cases.
MHz
dBm
dB
2060E
13 - 23
2380E
16 - 30
2690E
17 - 32
3030E
20 - 37
3320E
21 - 37
3740E
27 - 42
MHz/V
2060E
2380E
VCO_DIV=1
HSRFout
Second Harmonic
2690E
3030E
-20
-15
3320E
Default Power
Mode
(Note 5)
50 Ω Load
3740E
dBc
2060E
2380E
VCO_DIV=3
2690E
3030E
-20
-15
3320E
3740E
PSHVCO
PULLVC
VCO Frequency Pushing
600
VCO Frequency Pulling
VSWR 1.7 to 1
(6 dB Pad)
VCO_DIV = 1
±800
VCO_DIV > 1
±60
2060E
1.6
2380E
1.8
RMS Phase Error
Integration Bandwidth
= 100 Hz to 20 MHz
Middle VCO Frequency
100 MHz Wenzel Crystal
Reference
Integer Mode
Optimized Loop Bandwidth
2690E
2.1
3030E
2.1
3320E
2.3
3740E
2.6
O
σΦ
CVregVCO = 4.7 µF,Open Loop
7
kHz/V
kHz
mRad
www.national.com
LMX2541
Symbol
LMX2541
Symbol
Parameter
Conditions
Min
Typ
Max
Units
VCO Phase Noise (Note 9)
fRFout =
Min VCO
Frequency
L(f)Fout
Phase Noise
2060E
fRFout =
Max VCO
Frequency
fRFout =
Min VCO
Frequency
L(f)Fout
Phase Noise
2380E
fRFout =
Max VCO
Frequency
fRFout =
Min VCO
Frequency
L(f)Fout
Phase Noise
2690E
fRFout =
Max VCO
Frequency
fRFout =
Min VCO
Frequency
L(f)Fout
Phase Noise
3030E
fRFout =
Max VCO
Frequency
www.national.com
8
10 kHz Offset
-89.7
100 kHz Offset
-113.7
1 MHz Offset
-134.9
10 MHz Offset
-155.4
20 MHz Offset
-160.3
10 kHz Offset
-86.5
100 kHz Offset
-111.4
1 MHz Offset
-132.8
10 MHz Offset
-153.4
20 MHz Offset
-158.5
10 kHz Offset
-87.9
100 kHz Offset
-112.7
1 MHz Offset
-133.8
10 MHz Offset
-154.2
20 MHz Offset
-159.5
10 kHz Offset
--83.4
100 kHz Offset
-109.1
1 MHz Offset
-130.8
10 MHz Offset
-151.8
20 MHz Offset
-157.5
10 kHz Offset
-86.9
100 kHz Offset
-111.8
1 MHz Offset
-133.3
10 MHz Offset
-154.2
20 MHz Offset
-159.4
10 kHz Offset
-82.3
100 kHz Offset
-108.4
1 MHz Offset
-130.3
10 MHz Offset
-151.1
20 MHz Offset
-156.7
10 kHz Offset
-86.1
100 kHz Offset
-110.5
1 MHz Offset
-132.0
10 MHz Offset
-152.2
20 MHz Offset
-157.1
10 kHz Offset
-82.2
100 kHz Offset
-107.7
1 MHz Offset
-129.4
10 MHz Offset
-150.5
20 MHz Offset
-156.1
dBc/Hz
dBc/Hz
dBc/Hz
dBc/Hz
Parameter
Conditions
fRFout =
Min VCO
Frequency
L(f)Fout
Phase Noise
3320E
fRFout =
Max VCO
Frequency
fRFout =
Min VCO
Frequency
L(f)Fout
Phase Noise
3740E
fRFout =
Max VCO
Frequency
9
Min
Typ
10 kHz Offset
-84.1
100 kHz Offset
-109.1
1 MHz Offset
-130.7
10 MHz Offset
-151.6
20 MHz Offset
-156.9
10 kHz Offset
-82.0
100 kHz Offset
-107.0
1 MHz Offset
-128.5
10 MHz Offset
-149.6
20 MHz Offset
-155.2
10 kHz Offset
-83.9
100 kHz Offset
-108.3
1 MHz Offset
-129.9
10 MHz offset
-150.6
20 MHz Offset
-156.5
10 kHz Offset
-81.6
100 kHz Offset
-106.5
1 MHz Offset
-127.7
10 MHz Offset
-148.6
20 MHz Offset
-154.2
Max
Units
dBc/Hz
dBc/Hz
www.national.com
LMX2541
Symbol
LMX2541
Symbol
Parameter
Conditions
VIH
High-Level Input Voltage
Min
Typ
Max
Units
Vcc
V
Digital Interface (DATA, CLK, LE, CE, Ftest/LD, FLout,RFoutEN)
1.6
VIL
Low-Level Input Voltage
0.4
V
IIH
High-Level Input Current
VIH = 1.75, XO = 0
-3.0
3.0
µA
IIL
Low-Level Input Current
VIL = 0 V , XO = 0
-3.0
3.0
µA
VOH
High-Level Output Voltage
IOH = 500 µA
2.0
Low-Level Output Voltage
IOL = -500 µA
VOL
ILeak
Leakage Current
Ftest/LD and FLout Pins Only
V
0.4
V
Attached to Vcc
-10
0.0
10
nA
Attached to GND
-10
10
nA
MICROWIRE Timing
tCE
Clock to Enable Low Time
See Data Input Timing
25
ns
tCS
Data to Clock Set Up Time
See Data Input Timing
25
ns
tCH
Data to Clock Hold Time
See Data Input Timing
20
ns
tCWH
Clock Pulse Width High
See Data Input Timing
25
ns
tCWL
Clock Pulse Width Low
See Data Input Timing
25
ns
tCES
Enable to Clock Set Up Time
See Data Input Timing
25
ns
tEWH
Enable Pulse Width High
See Data Input Timing
25
ns
Note 5: The LMX2541 RFout power level is programmable with the program words of VCOGAIN, OUTTERM, and DIVGAIN. Changing these words can change
the output power of the VCO as well as the current consumption of the output buffer. For the purpose of consistency in electrical specifications, "Default Power
Mode" is defined to be the settings of VCOGAIN = OUTTERM = DIVGAIN = 12.
Note 6: Not tested in production. Guaranteed by characterization.
Note 7: Consult the applications section for more details on these parameters.
Note 8: Maximum Allowable Temperature Drift for Continuous Lock is how far the temperature can drift in either direction from the value it was at the time that
the R0 register was last programmed, and still have the device stay in lock. The action of programming the R0 register, even to the same value, activates a
frequency calibration routine. This implies that the device will work over the entire frequency range, but if the temperature drifts more than the maximum allowable
drift for continuous lock, then it will be necessary to reload the R0 register to ensure that it stays in lock. Regardless of what temperature the device was initially
programmed at, the temperature can never drift outside the frequency range of -40°C ≤TA≤ 85°C without violating specifications.
Note 9: The VCO phase noise is measured assuming that the loop bandwidth is sufficiently narrow that the VCO noise dominates. The phase noise is measured
with AC_TEMP_COMP = 5 and the device is reloaded at each test frequency. The typical performance characteristics section shows how the VCO phase noise
varies over temperature and frequency.
Note 10: See Typical Performance Characteristics for more information.
www.national.com
10
LMX2541
Serial Data Timing Diagram
30073303
The DATA is clocked into a shift register on each rising edge of the CLK signal. On the rising edge of the LE signal, the data is
sent from the shift registers to an actual counter. A slew rate of at least 30 V/μs is recommended for these signals. After the
programming is complete, the CLK, DATA, and LE signals should be returned to a low state. If the CLK and DATA lines are toggled
while the in VCO is in lock , as is sometimes the case when these lines are shared with other parts, the phase noise may be
degraded during the time of this programming.
11
www.national.com
LMX2541
Typical Performance Characteristics (Not Guaranteed)
PLL Phase Noise
30073307
The above plot demonstrates the PLL phase noise of the LMX2541SQ3700E operating at 3700 MHz output frequency, phase detector frequency of 100 MHz,
and charge pump gain of 32X. The loop bandwidth was made as wide as possible to fully expose the PLL phase noise and reference source was a 100 MHz
Wenzel crystal. This measurement was done in integer mode. To better understand the impact of using fractional mode, consult the applications section.
The measured noise is the sum of the PLL 1/f noise and noise floor. At offsets below 1 kHz, the PLL 1/f noise is dominates and changes as 10 dB/decade. The
noise at 1 kHz is dominated by this 1/f noise and has a value of -103 dBc/Hz. In the 100 - 200 kHz offset range, the noise is -113.7 dBc/Hz and is dominated by
the PLL noise floor. It can be shown that if the effects of the loop filter peaking and the 1/f noise are subtracted away from this measurement, it would be about
0.6 dB better.
If the phase detector frequency is changed with the VCO frequency held constant, the PLL noise floor will change, but the 1/f noise will remain the same. If the
VCO frequency is changed, both the 1/f noise and PLL noise floor change as 20 dB/decade.
Divider Noise Floor vs. Divider Value Value
(fVCO = 3700 MHz, Various values for VCO_DIV)
30073326
When the divider is engaged (VCO_DIV >0), then the entire system phase noise is reduced by a factor of 20 × log(VCO_DIV). However, the noise floor of the
divider will also add to this noise as is visible at far offsets. Note that the noise floor for Bypass mode is lower because the VCO divider is not engaged.
www.national.com
12
PLL Normalized 1/f Noise vs. Slew Rate
(KPD = 32X)
30073308
30073309
PLL Normalized Noise Floor vs. Charge Pump Gain
(Slew Rate = 2000 V/μs)
PLL Normalized 1/f Noise vs. Charge Pump Gain
(Slew Rate = 2000 V/μs)
300733010
300733011
13
www.national.com
LMX2541
PLL Normalized Noise Floor vs. Slew Rate
(KPD = 32X)
LMX2541
VCO Phase Noise Degradation vs. Temperature and Offset
(VCO Relocked at Each Temperature
Vcc = 3.3 V, AC_TEMP_COMP = 5)
30073312
The above plot shows how much the VCO phase noise typically change over temperature relative to room temperature. The typical values for represent an
average over all frequencies and part options and therefore there are some small variations over part options and frequencies that are not shown .VCO phase
noise numbers room temperature are reported in the electrical specifications. A negative value indicates a phase noise improvement.
Relative VCO Phase Noise Over Temperature Drift
(AC_TEMP_COMP = 25, Vcc = 3.3 V)
Temperature
Phase Noise Change in Celsius for Various Offsets
Lock
Current
10 kHz
100 kHz
1 MHz
10 MHz
-40
-40
+0.4
-2.0
-1.6
-1.8
-1.6
-40
25
+0.3
+0.5
+0.5
+0.5
+0.4
-40
85
+0.9
+2.0
+2.4
+2.5
+2.3
25
-40
+0.2
-2.2
-1.7
-2.0
-1.8
25
25
This is the default condition to which these other numbers are normalized to.
25
85
+0.6
85
-40
+0.2
-2.2
-1.7
-1.9
-1.8
85
25
+0.2
+0.2
+0.3
+0.2
+0.2
85
85
+0.6
+1.8
+2.2
+2.3
+2.1
+1.5
+2.0
+2.0
20 MHz
+1.9
The above table shows the typical degradation for VCO phase noise when the VCO is locked at one temperature and the temperature is allowed to drift to another temperature. A negative value indicates a phase noise improvement.
www.national.com
14
30073317
30073316
Output Power vs. VCOGAIN and FREQUENCY
(VCO_DIV = 1, TA = 25 °C, Vcc = 3.3 V, OUTTERM = 12 )
Output Power vs. OUTTERM and FREQUENCY
(VCO_DIV = 1, TA = 25 °C, Vcc = 3.3 V, VCOGAIN = 12 )
30073315
30073314
The above plots show the trends in output power as a function of temperature, voltage, and frequency. For states where VCOGAIN
and OUTTERM are not 12, the table below shows how the output power is modified based on these programmable settings.
OUTTERM
Change in Output Power in Bypass Mode as a Function of VCOGAIN and
OUTTERM
3
6
VCOGAIN
9
12
15
3
-9.7
-8.4
-7.9
-7.8
-7.9
6
-6.6
-4.5
-3.6
-3.4
-3.6
9
-5.7
-3.1
-1.7
-1.3
-1.3
12
-5.4
-2.5
-0.8
0.0
0.1
15
-5.3
-2.2
-0.3
0.8
1.1
15
www.national.com
LMX2541
Output Power vs. Temperature
(VCO_DIV = 1, VCOGAIN = 12, OUTTERM = 12,
Vcc = 3.3 V )
Output Power vs Voltage
(VCO_DIV = 1, VCOGAIN = 12, OUTTERM = 12,
TA = 25°C )
Output Power vs. Temperature
( VCO_DIV > 1, DIVGAIN = OUTTERM = 12,
Vcc = 3.3 V )
30073321
30073320
Output Power vs. DIVGAIN and FREQUENCY
(VCO_DIV > 1, TA = 25 °C, Vcc = 3.3 V, OUTTERM = 12 )
Output Power vs. OUTTERM and FREQUENCY
(VCO_DIV = 1, TA = 25 °C, Vcc = 3.3 V, VCOGAIN = 12 )
30073319
30073318
The table below shows the RELATIVE output power to the case of VCOGAIN = OUTTERM = 12.
Change in Output Power in Divided Mode as a Function of DIVGAIN and
OUTTERM
OUTTERM
LMX2541
Output Power vs Voltage
(VCO_DIV > 1, VCOGAIN = 12, OUTTERM = 12,
TA = 25°C )
www.national.com
3
6
DIVGAIN
9
12
15
3
-10.2
-9.8
-9.8
-9.9
-9.9
6
-6.1
-4.4
-4.3
-4.3
-4.4
9
-5.7
-2.4
-1.5
-1.4
-1.4
12
-5.5
-2.1
-0.7
0.0
0.3
15
-5.5
-2.0
-0.5
0.2
0.7
16
The LMX2541 is a low power, high performance frequency
synthesizer system which includes the PLL, VCO, and partially integrated loop filter. The following sections give a discussion of the various blocks of this device.
1.5 PARTIALLY INTEGRATED LOOP FILTER
The LMX2541 integrates the third pole (formed by R3_LF and
C3_LF) and fourth pole (formed by R4_LF and C4_LF) of the
loop filter. The values for these integrated components can
be programmed independently through the MICROWIRE interface. The larger the values of these components, the
stronger the attenuation of the internal loop filter. The maximum attenuation can be achieved by setting the internal
resistors and capacitors to their maximum value and the minimum attenuation can be attained by setting all of these to
their minimum setting. This partially integrated loop filter can
only be used in full chip mode.
1.1 REFERENCE OSCILLATOR INPUT PINS
There are three basic ways that the OSCin/OSCin* pins may
be configured as shown in the table below:
Mode
Description
XO Bit
Crystal
Device is used with a crystal
oscillator
1
Single
-Ended
Device is driven with a singleended source, such as a
TCXO.
0
Differential
Use this mode when driving
with a differential signal, such
as an LVDS signal.
0
In addition to the way that the OSCin/OSCin* pins are driven,
there are also bits that effect the frequency that the chip uses.
The OSC_FREQ word needs to be programmed correctly, or
the VCO may have issues locking to the proper frequency,
since the VCO frequency calibration is based on this word.
Word Name
Function
OSC_FREQ
This needs to be set correctly
if the internal VCO is used for
proper calibration.
OSC2X
This allows the oscillator
frequency to be doubled. The
R divider is bypassed in this
case.
30073304
1.6 LOW NOISE, FULLY INTEGRATED VCO
The LMX2541 includes a fully integrated VCO, including the
inductors. The VCO (Voltage Controlled Oscillator) takes the
voltage from the loop filter and converts this into a frequency.
The VCO frequency is related to the other frequencies and
divider values as follows:
Higher slew rates tend to yield the best fractional spurs and
phase noise, so a square wave signal is best for OSCin. Single ended mode and differential mode have similar results if
a square wave is used to drive the OSCin pin. If using a sine
wave, higher frequencies tend to work better due to their
higher slew rates.
fVCO = fPD × N = fOSCin × N / R
In order to the reduce the VCO tuning gain and therefore improve the VCO phase noise performance, the VCO frequency
range is divided into many different frequency bands. This
creates the need for frequency calibration in order to determine the correct frequency band given a desired output frequency. The frequency calibration routine is activated any
time that the R0 register is programmed. It is important that
the OSC_FREQ word is set correctly to have this work correctly. The time that the frequency calibration takes is dependent on fOSCin and the size of the frequency change. As a
general rule of thumb, this time improves for higher OSCin
frequencies and is on the order of 200 us for an OSCin freqency of 100 MHz and a small frequency change. For frequency changes on the order of the entire VCO tuning range,
this time is closer to 300 us.
The VCO also has an internal amplitude calibration algorithm
to optimize the phase noise which is also activated any time
the R0 register is programmed. The optimum internal settings
for this are temperature dependent. If the temperature is allowed to drift too much without being re-calibrated, some
minor phase noise degradation could result. For applications
where this is an issue, the AC_TEMP_COMP word can be
used to sacrifice phase noise at room temperature in order to
improve the VCO phase noise over all temperatures. The
maximum allowable drift for continuous lock, ΔTCL, is stated
1.2 R DIVIDER
The R divider divides the OSCin frequency down to the phase
detector frequency. If the doubler is enabled, then the R divider is bypassed.
1.3 PHASE DETECTOR AND CHARGE PUMP
The phase detector compares the outputs of the R and N dividers and generates a correction current corresponding to
the phase error. This charge pump current is software programmable to 32 different levels.The phase detector frequency, fPD, can be calculated as follows:
fPD = fOSCin / R
1.4 N DIVIDER AND FRACTIONAL CIRCUITRY
The N divider in the LMX2541 includes fractional compensation and can achieve any fractional denominator (PLL_DEN)
from 1 to 4,194,303. The integer portion, PLL_N, is the whole
part of the N divider value and the fractional portion,
PLL_NUM / PLL_DEN, is the remaining fraction. PLL_N,
PLL_NUM, and PLL_DEN are software programable. So in
general, the total N divider value, N, is determined by:
N = PLL_N + PLL_NUM / PLL_DEN
17
www.national.com
LMX2541
The order of the delta sigma modulator is programmable from
integer mode to fourth order. There are also several dithering
modes that are also programmable. In order to make the fractional spurs consistent, the modulator is reset any time that
the R0 register is programmed.
1.0 Functional Description
LMX2541
in the electrical specifications. For this part, a number of +125
C means the part will never lose lock if the part is operated
under recommended operating conditions.
run in the powerdown state. Also, the special programming
for VCO_DIV = 4 or 5 has to be done when the part is powered
up. In order for the CE pin to properly power the device down
when it is held low, the all the registers in the device need to
have been programmed at least one time.
1.7 PROGRAMMABLE VCO DIVIDER
The VCO divider can programmed to any value from 2 to 63
as well as bypass mode if device is in full chip mode. In external VCO mode or divider mode, all values except bypass
mode can be used for the VCO divider. The VCO divider is
not in the feedback path between the VCO and the PLL and
therefore has no impact on the PLL loop dynamics. After this
programmable divider is changed, it may be beneficial to reprogram the R0 register to recallibrate the VCO . The frequency at the RFout pin is related to the VCO frequency and
divider value, VCO_DIV, as follows:
1.10 FASTLOCK
The LMX2541 includes the Fastlock™ feature that can be
used to improve the lock times. When the frequency is
changed, a timeout counter is used to engage the fastlock for
a programmable amount of time. During the time that the device is in Fastlock, the FLout pin changes from high
impedance to low, thus switching in the external resistor R2pLF with R2_LF as well as changing the internal loop filter
values for R3_LF and R4_LF.
fRFout = fVCO / VCO_DIV
When this divider is enabled, there will be some far-out phase
noise contribution to the VCO noise. Also, it may be beneficial
for VCO phase noise to reprogram the R0 register to recalibrate the VCO if the VCO_DIV value is changed from bypass
to divided, or vise-versa.
The duty cycle for this divider is always 50%, even for odd
divide values. Because of the architecture of this divider that
allows it to work to high frequencies and always have a 50%
duty cycle, there are a few extra considerations:
• In divider only mode, there must be 5 clock cycles on the
CLK pin after the divide value is programmed in order to
cause the divide value to properly changed. It is fine to use
more than 5 clock cycles for this purpose.
• For a divde of 4 or 5 ONLY, the R4 register needs to be
programmed one more time after the R0 register is loaded
in order to synchronize the divider. Failure to do so will
cause the wrong divide values. Furthermore, if the VCO
signal ever goes away, as is the case when the part is
powered down, it is necessary to reprogram the R4
register again to re-synchronize the divider.
30073305
The following table showshow the charge pump gain, loop
filter resistors, and FLout pin change between normal operation and Fastlock.
1.8 PROGRAMMABLE RF OUTPUT BUFFER
The output power at the RFout pin can be programmed to
various levels as well as on and off states. The output state
of this pin is controlled by the RFoutEN pin as well as the
RFOUT word. The RF output buffer can be disabled while still
keeping the PLL in lock. In addition to this, the actual output
power level of this pin can be adjusted using the VCOGAIN,
DIVGAIN, and OUTTERM programming words.
CE Pin
Device State
Low
Don't Care
Powered Down
High
0
Powered Up
1
Powered Down
Fastlock
Charge Pump Gain
CPG
FL_CPG
Loop Filter Resistor R3_LF
R3_LF
FL_R3_LF
Loop Filter Resistor R4_LF
R4_LF
FL_R4_LF
FLout Pin
High
Impedance
Low
Parameter
The device can be programmed in the powerdown state.
However, the VCO frequency needs to be changed when the
device is powered up because the VCO calibration does not
www.national.com
Normal
Operation
Once the loop filter values and charge pump gain are known
for normal mode operation, they can be determined for fastlock operation as well. In normal operation, one can not use
the highest charge pump gain and still use fastlock because
there will be no larger current to switch in. If the resistors and
the charge pump current are done simultaneously, then the
phase margin can be preserved while increasing the loop
bandwidth by a factor of K as shown in the following table:
1.9 POWERDOWN MODES
The LMX2541 can be powered up and down using the CE pin
or the POWERDOWN bit. When the device is powered down,
the programming and VCO calibration information is retained,
so it is not necessary to re-program the device when the device comes out of the powered down state (The one exception
is when the VCO_DIV value is 4 or 5, which has already been
discussed.). The following table shows how to use the bit and
pin.
POWERDOWN
Bit
Parameter
18
Symbol
Calculation
Charge pump
gain in Fastlock
FL_CPG
Typically choose to be the
largest value.
Loop Bandwidth
Multiplier
K
K=
sqrt (FL_CPG/CPG)
Internal Loop
Filter Resistor
FL_R3_LF
FL_R3_LF =
R3_LF / K
Internal Loop
Filter Resistor
FL_R4_LF
FL_R4_LF =
R4_LF / K
External Loop
Filter Resistor
R2pLF
R2pLF =
R2_LF / (K - 1)
LMX2541
1.11 LOCK DETECT
The Ftest/LD pin of the LMX2541 can be configured to support both analog and digital lock detect. The analog lock
detect is generally more of a legacy feature and requires an
external RC. When configured for push-pull analog lock detect, the Ftest/LD is high with narrow pulses which corresponds to when the charge pump is on. This waveform can
be integrated with an RC filter to generate a lock detect signal.
The open drain lock detect is similar pin puts out short low
pulses when the charge pump comes on. An external RC filter
can be used to integrate this information. Open drain analog
lock detect is typically implemented with an RC filter followed
by a pull-up resistor. The pull-up resistor can be much larger
than the resistor in the RC filter in order to make unbalanced
time constants for improved sensitivity.
The digital lock detect function can be selected for the Ftest/
LD pin. The digital lock detect circuitry compares the difference between the phase of the inputs to the phase detector
with a RC generated delay of ε. To indicate a locked state
(Lock = HIGH) the phase error must be less than ε for 5 consecutive phase detector cycles. Once in lock (Lock = HIGH),
the RC delay is changed to δ. To indicate an out of lock state
(Lock = LOW), the phase error must become greater than δ.
The values of ε and δ are programmable with the DLOCK
word.
30073306
19
www.national.com
LMX2541
2.0 General Programming Information
The LMX2541 is programmed using several 32-bit registers used to control the LMX2541 operation. A 32-bit shift register is used
as a temporary register to indirectly program the on-chip registers. The shift register consists of a data field and an address field.
The last 4 register bits, CTRL[3:0] form the address field, which is used to decode the internal register address. The remaining 28
bits form the data field DATA[27:0]. While LE is low, serial data is clocked into the shift register upon the rising edge of clock (data
is programmed MSB first). When LE goes high, data is transferred from the data field into the selected register bank. For initial
device programming the register programming sequence must be done in the order as shown in the register map. The action of
programming register R7 resets all the registers to default values, including hidden registers. The programming of register R1 and
R0 is also special for the device when operating in full chip mode because the action of programming either one of these registers
activates the VCO calibration. For changes after this initial setup, see the Applications Information section.
www.national.com
20
0
0
0
1
R13
R8
R6
R5
FSK
0
0
R2
R1
R0
0
0
0
0
0
30
1
0
0
0
0
29
0
0
0
0
0
0
C4_LF[3:0]
R3
R4
0
R7
31
0
0
0
0
26
0
0
0
0
0
0
0
23
0
0
0
0
22
DITH
[1:0]
1
1
0
0
20
ORDER[2:0]
PLL_NUM[15:0]
1
0
0
0
19
18
1
0
0
0
OSC
_2X
PLL_N[17:12]
FDM
17
1
0
0
16
1
0
0
15
CPP
DEN[21:0]
MUX[3:0]
VCO_DIV[5:0]
1
0
0
14
VCOGAIN[3:0]
0
0
0
DATA[27:0]
R3_LF[2:0]
FL_R3_LF[2:0]
0
0
0
0
21
R4_LF[2:0]
FL_R4_LF
[2:0]
0
1
0
0
24
PLL_NUM[21:16]
1
CPT
C3_LF[3:0]
0
0
0
0
25
FL_CPG[4:0]
0
0
0
0
27
DLOCK[2:0]
0
0
0
0
28
0
0
0
12
1
0
0
11
1
0
0
10
1
0
0
9
0
5
PLL_N[11:0]
PWDN
OSC_FREQ[7:0]
XO
1
4
VCO_DIV_OPT[2:0]
0
6
MODE
[1:0]
RFOUT
[1:0]
AC_TEMP_COMP[4:0]
1
0
7
DIVGAIN[3:0]
0
0
8
PLL_R[11:0]
CPG[4:0]
FL_TOC[13:0]
OUTTERM[3:0]
0
0
0
13
0
0
0
0
0
0
0
1
1
0
C3
3
0
0
0
0
1
1
1
0
1
1
C2
2
0
0
1
1
0
0
1
0
0
1
C1
1
0
1
0
1
0
1
0
0
1
1
C0
0
The following table lists the registers as well as the order that they should be programmed. Register 7 is programmed first and the action of programming register R7 resets all the
registers after the LE pin is pulled to a low state. Register R0 is programmed last because it activates the VCO calibration. The one exception to this is when the VCO_DIV value is 4
or 5. Consult the programming section on VCO_DIV for more details.
2.01 Register Map
LMX2541
21
www.national.com
LMX2541
2.1 REGISTER R7
Although Register 7 has no elective bits to program, it is very important to program this register because the action of doing so
with the bit sequence shown in the register map resets all the registers, including hidden registers with test bits that are not
disclosed. Register 7 should always be programmed first, because it will clear out all other programming information. The register
reset occurs only after the LE signal has transitioned from low to high and back to low again.
2.1.1 REGISTER R13
This register needs to be programmed only in the event that the RFout pin is being used and VCO_DIV = 1.
VCO_DIV_OPT[2:0]
This word optimizes the RFout power level based on the VCO divider in the case that VCO_DIV = 1. In this case, this word should
be programmed to 15. Otherwise, this word does not need to be programmed.
Condition
VCO_DIV_OPT
VCO_DIV = 1,
RFout pin not disabled
7
All other conditions
0, but does not need to be
programmed.
2.1.2 REGISTER R8
AC_TEMP_COMP[4:0]
This word optimizes the VCO phase noise for possible temperature drift. When the VCO frequency is changed, the internal tuning
algorithm optimizes the phase noise for the current temperature. In fixed frequency applications, temperature drift may lead to suboptimal phase noise over time. In dynamic frequency applictions, the re-tuning of the VCO frequency overcomes this problem
because the phase noise is re-optimized each time the VCO frequency is changed. The AC_TEMP_COMP word can be used to
optimize the VCO phase noise for temerature drift for these different scenarios. The following table indicates which values of this
word should be used for each scenario.
AC_TEMP_COMP
Application Type
5
Dynamic Frequency
24
Fixed Frequency
All Other States
Invalid
2.2 REGISTER R6
Register R6 has words that impact the output power of the RFout pin.
RFOUT[1:0] - RFout enable pin
This word works in combination with the EN_RFout Pin to control the state of the RFout pin.
RFOUT
EN_RFout Pin
RFout Pin State
0
Don't Care
Disabled
2
Don't Care
Enabled
Low
Disabled
High
Enabled
1 or 3
DIVGAIN[3:0], VCOGAIN[3:0], and OUTTERM[3:0] - Power Controls for RFout
These three words may be programmed in a value from 0 to 15 and work in conjunction to control the output power level of the
RFout pin. Increasing any of these values increases the output power at the expense of higher current consumption of the buffer.
Although there may be more than one way to get the same output power, some combinations may have lower current. The typical
performance characteristics show these trade-offs. The default setting for all these bits is 12. The value of VCO_DIV determines
which two of these three words have an impact.
www.national.com
VCO_DIV
Bits that Impact Power
1 (Bypass)
OUTTERM, VCOGAIN
>1 (Not Bypass)
OUTTERM, DIVGAIN
22
This register controls the fastlock mode which enables a wider loop bandwidth when the device is changing frequencies.
FL_TOC[13:0] -- Time Out Counter for FastLock
When the value of this word is 3 or less, FastLock time out counter is disabled, and the FLout pin can be used for general purpose
I/O. When this value is 4 or greater, the time out counter is engaged for the amount of phase detector cycles shown in the table
below.
TOC Value
FLout Pin State
0
High Impedance
Disabled
1
Low
Always Engaged
2
Low
Disabled
3
High
Disabled
4
Low
Engaged for
4 × 2 Phase DetectorCycles
.
.
.
Low
Engaged for
16383 × 2 Phase Detector Cycles
16383
Fastlock Engagement Time
When this count is active, the FLout Pin is grounded, the FastLock current is engaged, and the resistors R3 and R4 are also
potentially changed. The table below summarizes the bits that control various values in and out of FastLock.
FastLock State
FLout
Charge Pump Current
R3_LF Value
Not Engaged
High Impedance
CPG
R3_LF
R4_LF Value
R4_LF
Engaged
Grounded
FL_CPG
FL_R3_LF
FL_R4_LF
FL_R3_LF[2:0] -- Value for Internal Loop Filter Resistor R3 During Fastlock
FL_R3_LF Value
R3 Resistor During Fastlock (kΩ)
0
Low ( 200 Ω )
1
1
2
2
3
4
4
16
5-7
Reserved
FL_R4_LF[2:0] -- Value for Internal Loop Filter Resistor R4 During Fastlock
FL_R4_LF Value
R3 Resistor During Fastlock (kΩ)
0
Low ( 200 Ω )
1
1
2
2
3
4
4
16
5-7
Reserved
23
www.national.com
LMX2541
2.3 REGISTER R5
LMX2541
FL_CPG[4:0] -- Charge Pump Current for Fastlock
When FastLock is enabled, this is the charge pump current that is used for faster lock time.
www.national.com
FL_CPG
Fastlock Charge Pump State
Typical Fastlock Charge Pump Current
at 3.3 Volts (µA)
0
1X
100
1
2X
200
2
3X
300
3
4X
400
...
...
...
31
32X
3200
24
This register controls miscellaneous functions of the device. The action of programming the R4 register also synchronizes the VCO
divider, which is necessary when VCO_DIV = 4 or 5.
OSC_FREQ [7:0] -- OSCin Frequency for VCO Calibration Clocking
This word is used for the VCO frequency calibration. This word should be set to the OSCin frequency rounded to the nearest MHz.
OSC_FREQ
OSCin Frequency
0
Illegal State
1
1 MHz
2
2 MHz
...
...
255
255 MHz
VCO_DIV[5:0] - VCO Divider
The output of the VCO is divided by the value of VCO_DIV, which can range from 1 (Bypass Mode) to 63 and all values in between.
There are a few special considerations. The VCO divider can only be set to bypass mode when the device is operating in full chip
mode. Also, there is one extra programming step required to synchronize the VCO divider when it has a value of 4 or 5. This extra
programming step is to load all registers as normal and then re-load register R4 with the same value. . When VCO_DIV=4 or 5, it
is also necessary to re-synchronize the divider in Fulll chip mode whenever the R0 or R1 registers are reprogrammed, or in External
VCO or Divider Only mode whenever the VCO signal goes away temporarily. This re-synchronization is ONLY for VCO_DIV values
of 4 and 5.
When VCO_DIV is 4 or 5 ONLY, the R4 register needs to be programmed one additional time (after the R0 register is loaded) with
the same value after the VCO signal (or ExtVCOin) signal is applied. If the VCO signal ever goes away, as is the case when the
chip is powered down, or the VCO signal is taken away in External VCO or Divider Only mode, the R4 register needs to be reloaded again to re-synchronize the divider.
VCO_DIV
VCO Output Divide
Comments
0
n/a
Illegal State
1
Bypass Mode
This state only available for MODE=Full Chip Mode
2
Divide by 2
3
Divide by 3
4
Divide by 4
5
Divide by 5
6
Divide by 6
...
...
62
Divide by 62
63
Divide by 63
Extra programming is required for divide by 4 and divide by 5 only.
Refer to the functional description for more details.
R3_LF[2:0] -- Value for Internal Loop Filter Resistor R3
This word controls the state of the internal loop filter resistor R3_LF when the device is in Full Chip Mode and Fastlock is not active.
R3_LF Value
R3 Resistor During Fastlock (kΩ)
0
Low ( 200 Ω )
1
1
2
2
3
4
4
16
5-7
Reserved
R4_LF[2:0] -- Value for Internal Loop Filter Resistor R4
This word controls the state of the internal loop filter resistor R4_LF when the device is in Full Chip Mode and Fastlock is not active.
25
www.national.com
LMX2541
2.4 REGISTER R4
LMX2541
R4_LF Value
R3 Resistor During Fastlock (kΩ)
0
Low ( 200 Ω )
1
1
2
2
3
4
4
16
5-7
Reserved
C3_LF[3:0] -- VALUE FOR C3 IN THE INTERNAL LOOP FILTER
This word controls the state of the internal loop filter resistor C3_LF when the device is Full Chip Mode.
C3_LF
C3 (pF)
0
0
1
1
2
5
3
6
4
10
5
11
6
15
7
16
8
20
9
21
10
25
11
26
12
30
13
31
14
35
15
26
C4_LF[3:0] -- VALUE FOR C4 IN THE INTERNAL LOOP FILTER
This word controls the state of the internal loop filter resistor C4_LF when the device is Full Chip Mode.
www.national.com
C4_LF
C4 (pF)
0
0
1
5
2
20
3
25
4
40
5
45
6
60
7
65
8
100
9
105
10
120
11
125
12
140
13
145
14
160
15
165
26
LMX2541
2.5 REGISTER R3
This register controls miscellaneous features of the device.
MODE[1:0] -- Operational Mode
The LMX2541 can be run in several operational modes as listed in the table below:
MODE
Name
Divider
PLL
VCO
0
Full
Enabled
Enabled
Enabled
1
External VCO
Enabled
Enabled
Disabled
2
Divider Only
Enabled
Disabled
Disabled
3
Test (Reserved)
Enabled
Enabled
Enabled
PWDN -- Powerdown Bit
Enabling this bit powers down the entire device, although register and VCO calibration information is retained.
XO - Crystal Oscillator Mode Select
When this bit is enabled, a crystal with appropriate load capacitors can be attached between the OSCin and OSCin* pins in order
to form a crystal oscillator.
CPG[4:0] -- Charge Pump Current
This word programs the charge pump current gain. The current is programmable between 100 uA and 3.2 mA in 100 uA steps.
CPG
Charge Pump State
Typical Charge Pump Current (µA)
0
1X
100
1
2X
200
2
3X
300
...
3
4X
...
...
31
32X
27
3200
www.national.com
LMX2541
MUX[3:0] -- Multiplexed Output for Ftest/LD Pin
The MUX[3:0] word is used to program the output of the Ftest/LD Pin. This pin can be used for a general purpose I/O pin, a lock
detect pin, and for diagnostic purposes. When programmed to the digital lock detect state, the output of the Ftest/LD pin will be
high when the device is in lock, and low otherwise. The output voltage level of the Ftest/LD is not equal to the supply voltage of
the device, but rather is given by VOH and VOL in the electrical characteristics specification.
Because the Ftest/LD pin is close to the OSCin pin, the state of this pin can have an impact on the performance of the device. If
any of the diagnostic modes (8-13) are used, the OSCin sensitivity can be severely degraded, so these should only be used for
diagnoistic purposes. The fractional spurs can also be impacted a little by fractional spurs. The Push-Pull digital lock detect modes,
like mode 3, tend to have the best fractional spurs, so these states are recommended, even if the digital lock detect function is not
needed.
MUX
Output Type
Function
0
High Impedance
Disabled
Comments
1
Push-Pull
Logical High State
2
Push-Pull
Logical Low State
3
Push-Pull
Digital Lock Detect
4
Push-Pull
Inverse Digital Lock Detect
5
Open Drain
Digital Lock Detect
6
Open Drain
Analog Lock Detect
7
Push-Pull
Analog Lock Detect
8
Push-Pull
N Divider
N Divider / 2
9
Push-Pull
10
Push-Pull
R Divider
11
Push-Pull
R Divider / 2
12
Push-Pull
PFD Up
13
Push-Pull
PFD Down
14-15
N/A
Reserved
General Purpose I/O Modes
Lock Detect Modes
Consult Functional Description for more details
State 3 is recommended for optimal spurious
performance.
Diagnostic Modes
These allow the user to view the outputs of the N
divider, R divider, and phase frequency detector (PFD)
and are intended only for diagnostic purposes.
Typically, the output is narrow pulses, but when the
output is divded by 2, there is a 50% duty cycle. The
use of these modes (including R Divider) can degrade
the OSCin sensitivity.
CPP - Charge Pump Polarity
This bit sets the polarity of the phase detector.
CPP
Charge Pump Polarity
Typical Applications
0
Positive
External VCO Mode
1
Negative
Full Chip Mode
External VCO Mode with an inverting active filter.
OSC2X-- OSCin Frequency Doubler
Enabling this bit doubles the OSCin frequency. This is useful in achieving a higher phase detector frequency to improve PLL phase
noise, push out noise from the delta sigma modulator, and sometimes reduce fractional spurs . Note that when this bit is enabled,
the R divider is bypassed.
OSC_2X
State
0
Normal
1
OSCin frequency is doubled
FDM - Extended Fractional Denomoinator Mode Enable
Enabling this bit allows the fractional numerator and denominator to be expanded from 10 bits to 22 bits. In 10-bit mode, only the
first 10 bits of the fractional numerator and denominator are considered. Disabling this saves about 0.5 mA of current. When using
FSK mode, this bit has to be disabled.
www.national.com
FDM
Fractional Mode
0
10-bit
1 (Default)
22-bit
28
This word determines the order of the delta sigma modulator in the PLL. In general, higher order fractional modulators tend to
reduce the primary fractional spurs that occur at increments of the channel spacing, but can also create spurs that are at a fraction
of the channel spacing. The optimal choice of modulator order is very application specific, however, a third order modulator is a
good starting point. The first order modulator has no analog compensation or dithering
Delta Sigma
Modulator
Mode
0
Disabled
Integer
1
First Order
2
Second Order
3
Third Order
4
Fourth Order
5-7
Illegal States
ORDER
Comments
Allows larger N Counter
This has no analog compensation or dithering
Fractional
Traditional Delta Sigma Operation
n/a
n/a
DITH[1:0] -- Dithering
Dithering randomizes the delta sigma modulator output. This reduces sub-fractional spurs at the expense of adding phase noise.
In general, it is recommended to keep the dithering strength at None or Weak for most applications. Dithering should never be
used when the device is used in integer mode or a first order modulator. When using dithering with the other delta sigma modulator
orders, it is beneficial to disable it in the case where the fractional numerator is zero, since it can actually create sub-fractional
spurs.
DITH
Dithering Strength
0
Weak
1
Medium
2
Strong
3
Disabled
CPT - Charge Pump Tri-state
When this bit is enabled, the charge pump is tri-stated. The Tri-state mode could be useful for open loop modulation applications
or as diagnostic tool for measuring the VCO noise, but is generally not used.
CPT
Charge Pump
0
Normal Operation
1
Tri-state
DLOCK[2:0] - Controls for Digital Lock detect
This word is controls operation of the digital lock detect function through selection of the window sizes (ε and δ). In order to indicate
the PLL is locked, there must be 5 consecutive phase detector output cycles in which the time offset between the R and N counter
outputs is less than ε. This will cause the Ftest/LD pin output to go high. Once lock is indicated, it will remain in this state until the
time offset between the R and N counter outputs exceeds δ. If the OSCin signal goes away, the digital lock detect circuit will reliably
indicate an unlocked condition. Consult the functional description for more details. A larger window size makes the lock detect
circuit less sensitive. The window size is limited by the phase detector frequency, fPD.
DLOCK
Maximum fPD
0
(Default)
Window Size (ns)
ε
δ
All
3
3
1
TBD
5.5
5.5
2
TBD
8
8
3
TBD
10.5
10.5
4
TBD
13
13
5
TBD
15.5
15.5
6 -7
Reserved
Reserved
Reserved
29
www.national.com
LMX2541
ORDER[2:0] -- Delta Sigma Modulator Order
LMX2541
FSK - Frequency Shift Keying
This bit enables a binary FSK modulation mode using the PLL N counter. Consult the applications section for more details.
FSK
FSK Mode
0
Disabled
1
Enabled
2.6 REGISTER R2
This word contains all the bits of the fractional denominator. These bits apply if the device is being used fractional mode.
PLL_DEN[21:0] -- Fractional Denominator
These bits determine the fractional denominator.
PLL_DEN[21:0]
Fractional
Denominator
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
...
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
4194303
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
www.national.com
30
Both registers R1 and R0 contain information for the PLL R counter, N counter, and fractional numerator. The action of programming
either one of these registers, even to the same value, runs the VCO calibration when the device has the internal VCO operating.
There are some programming words that are split across these two registers.
PLL_R[11:0] -- PLL R Divider Value
The R divider divides the OSCin signal. Note that if the doubler is enabled, the R divider is bypassed.
PLL_R[11:0]
0
Illegal State
1
0
0
0
0
0
0
0
0
0
0
0
1
2
0
0
0
0
0
0
0
0
0
0
1
0
3
0
0
0
0
0
0
0
0
0
0
1
1
...
...
...
...
...
...
...
...
...
...
...
...
...
4095
1
1
1
1
1
1
1
1
1
1
1
1
PLL_N[17:0] PLL N Divider Value
When using integer mode, the PLL N divider value is split up into two different locations. In fractional mode, only the 12 LSB bits
of the N counter are used. Based on the order of the modulator, the range is shown in the table below.
PLL_N[17:12]
PLL_N[11:0]
12
13
...
Divide Values below 12 are prohibited
Integer
Mode
<12
262143
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
1
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
12
Possible with first order modulator only
Possible with first or second order modulator
15-16
Possible with first, second, or third order modulators only
17
18
...
4090
4091
-4092
Fractional Mode
13-14
x
x
x
x
x
x
0
0
0
0
0
0
0
1
0
0
0
1
x
x
x
x
x
x
0
0
0
0
0
0
0
1
0
0
1
0
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
x
x
x
x
x
x
1
1
1
1
1
1
1
1
1
0
1
0
Possible with a first, second, or third order modulator only
4093
Possible with a first or second order modulator only
4094
Possible with a first order modulator only
PLL_NUM[21:0] -- Fractional Numerator
The fractional numerator is formed by the NUM word that is split between two registers and applies in fractional mode only.
PLL_NUM[21:15]
PLL_NUM[14:0]
Fractional
Numerator
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
...
4194303
31
www.national.com
LMX2541
2.7 REGISTERS R1 AND R0
LMX2541
Single-Ended Operation
3.0 Applications Information
3.1 TYPICAL CONNECTIONS
The components used at various pins can have an impact on
the performance of this part, particularly fractional spurs. The
following table contains guidelines for pin components and
configurations for using the device in full chip mode with the
fractional engine running.
Pin(s)
Connection Recommendations
RFout
Input should be AC coupled to this pin. If
using external VCO mode, this pin can be left
open.
RFoutEN
If not using this pin, leave it open and
program the device accordingly.
L1
L2
Lmid
Do not connect these pins, but include the
pads so the device will solder correctly.
VccOSC
VccDIG
VccFRAC
VccPLL1
For the best fractional spurs, bypass to
ground with a ferrite bead and shunt 0.1 uF
capacitor.
VregRFout
Attach a series 10 Ω resistor and 1 uF
capacitor to ground
VregVCO
Attach a shunt 4.7 uF capacitor to ground.
VregFrac
Attach a series 10 Ω and 1 uF capacitor to
ground.
VrefVCO
Attach a shunt 0.1 uF capacitor to ground.
VccDiv
VccRFout
VccCP1
VccCP2
It is acceptable to short these directly to the
power plane.
GND
GND DAP
30073323
For differential operation, as is the case when using an LVDS
or LVPECL driver, a 100 Ω resistor is placed across the OSCin/OSCin* traces
Differential Operation
Each pin should have a
connection to the ground plane.
30073324
A third way to configure the device is in crystal mode (XO =
1). For this, the crystal is placed across the OSCin/OSCin*
pins. Crystals are specified for a specific load capacitance,
CLoad. .The load capacitors shown in the figure each have a
value of CLoad/2.
dedicated
Crystal Mode Operation
This should be grounded and not connected
to other ground pins.
3.1.1 OSCin/OSCin* Connections
For single-ended operation, the signal is driven into the OSCin pin. The OSCin* pin is terminated the same as the OSCin
pin. This is a typical case if the device is driven by a TCXO.
For both single-ended and differential operation, the input is
AC coupled because the OSCin/OSCin* pins self-bias to an
optimal DC operating point. Better performance for both
phase noise and fractional spurs is obtained for signals with
a higher slew rate, such as a square wave. This is especially
important for lower frequency signals, since slower frequency
sine waves have lower slew rates. Fractional spurs are typically a several dB better when running in differential mode as
opposed to single-ended mode.
www.national.com
30073325
3.2 CURRENT CONSUMPTION
The current consumption of the LMX2541 has many factors
that influence it. Determining the current consumption for the
entire device involves knowing which blocks are powered up
and adding their currents together. For instance, the output
buffer current can be impacted by the software controllable
settings. Various blocks such as the VCO divider can be powered up and down in many different combinations. The typical
current consumption for default setup conditions for each of
the blocks is shown in the following table.
32
Current (mA)
RF Output Buffer
40
VCO Divider
32
Internal PLL
(Includes Counters, Charge
Pump, and OSCin Buffer)
(Excludes Fractional Engine)
78
PLL Fractional Engine
5
1525.05 MHz, but the spur will be at a much farther offset that
can easily be filtered by the loop filter of 25.05 MHz (1525.05
MHz mod 50 MHz).
3.4 PLL PHASE NOISE
Disregarding the impact of reference oscillator noise, loop filter resistor thermal noise, and loop filter shaping, the phase
noise of the PLL can be decomposed into three components:
flicker noise, flat noise, and fractional noise. These noise
sources add in an RMS sense to produce the total PLL noise.
In other words:.
LPLL(f) =
10·log(10(LPLL_flat(f) / 10 ) + 10(LPLL_flicker (f) / 10 )+ 10(LPLL_fractional(f) / 10 )
3.3 FRACTIONAL SPURS
Primary Fractional Spurs
The primary fractional spurs occur at multiples of the channel
spacing and can change based on the fraction. For instance,
if the phase detector frequency is 10 MHz, and the channel
spacing is 100 kHz, then this could be achieved using a fraction of 1/100. The fractional spurs would be at offsets that are
multiples 100 kHz.
Sub-Fractional Spurs
Sub-fractional spurs spurs occur at sub-multiples of the channel spacing, Fch. For instance, in the above example, there
could be a sub-fractional spur at 50 kHz. The occurrence of
these spurs is dependent on the modulator order. Integer
mode and the first order modulator never have sub-fractional
spurs. If the fractional denominator can be chosen to avoid
factors of 2 or 3, then there will also be no sub-fractional
spurs. Sub-fractional spurs get worse for higher order modulators. Dithering tends to reduce sub-fractional spurs at the
expense of increasing PLL phase noise. The following table
provides guidance on predicting sub-fractional spur offset frequencies.
Potential Influencing Factors
Symbol
f
fVCO
fPD
KPD
FRAC
LPLL_flat(f)
No
Yes
Yes
Yes
No
LPLL_flicker(f)
Yes
Yes
No
Yes
No
LPLL_fractional(f)
Yes
No
Yes
No
Yes
The preceding table shows which factors of offset frequency
(f), VCO frequency (fVCO), phase detector frequency (fPD),
charge pump gain (KPD), and the fractional settings (FRAC)
can potentially influence each phase noise component. The
fractional settings include the fraction, modulator order, and
dithering.
For the flat noise and flicker noise, it is possible to normalize
each of these noise sources into a single index. By normalizing these noise sources to an index, it makes it possible to
calculate the flicker and flat noise for an arbitrary condition.
These indices are reported in the electrical characteristics
section and in the typical performance curves.
Sub-Fractional Spur Offset
Frequencies vs. Modulator Order
and Fractional Denominator Factors
Noise
Component
Index
Relationship
LPLL_flat(f)
LNPLL_flat
(1 Hz)
LPLL_flat(f) =
LNPLL_flat(1 Hz)
+ 20·log(N) + 10·log(fPD)
LPLL_flicker(f)
LNPLL_flicker
(10 kHz)
LPLL_flicker(f) =
LNPLL_flicker(10 kHz)
- 10·log(10 kHz / f) + 20·log
( fVCO / 1 GHz )
Fractional Denominator Factors
ORDER
No Factor Factor of Factor of Factor of
of 2 or 3 2 but not 3 3 but not 2 2 and 3
Integer
Mode
None
1st Order
Modulator
None
None
None
None
2nd Order
Modulator
None
Fch/2
None
Fch/2
3rd Order
Modulator
None
Fch/2
Fch/3
Fch/6
4th Order
Modulator
None
Fch/4
Fch/3
Fch/12
None
None
None
The flat noise is dependent on the PLL N divider value (N)
and the phase detector frequency (fPD) and the 1 Hz Normalized phase noise ( LNPLL_flat(1 Hz) ). The 1 Hz normalized phase
noise can also depend on the charge pump gain as well. In
order to make an accurate measurement of just the flat noise
component, the offset frequency must be chosen sufficiently
smaller then the loop bandwidth of the PLL, and yet large
enough to avoid a substantial noise contribution from the reference and PLL flicker noise. This becomes easier to measure for lower phase detector frequencies.
The flicker noise, also known as 1/f noise, can be normalized
to 1 GHz carrier frequency and 10 kHz offset, LNPLL_flicker(10
kHz). Flicker noise can dominate at low offsets from the carrier and has a 10 dB/decade slope and improves with higher
charge pump currents and at higher offset frequencies . To
accurately measure the flicker noise it is important to use a
high phase detector frequency and a clean crystal to make it
such that this measurement is on the 10 dB/decade slope
close to the carrier. LPLL_flicker(f) can be masked by the refer-
Impact of VCO_DIV on Fractional spurs
Because the fractional and sub-fractional spur levels do not
depend on output frequency, there is a big benefit to division.
In general, every factor of 2 gives a 6 dB improvement to
fractional spurs. Also, since the spur offset frequency is not
divided, the channel spacing at the VCO can be also increased to improve the spurs. However, if the on-chip VCO is
used, crosstalk can cause spurs at a frequency of fRFout mod
fPD. Consider the following example of a 50 MHz phase detector frequency and VCO_DIV = 2. If the VCO is at 3000.1
MHz and divided by 2 to get 1500.05 MHz, there will be a spur
at an offset of 50 kHz (1500.05 MHz mod 50 MHz). However,
if the VCO frequency is at 3050.1 MHz, the output will be at
33
www.national.com
LMX2541
Block
3.6 MODULATOR ORDER
In general, the fractional mode of the PLL enables the use of
a higher phase detector frequency relative to the channel
spacing, which enables the in-band noise of the PLL to be
lower. The choice of modulator order to be used in fractional
mode is based on how much higher fPD can be made relative
to the channel spacing and the acceptable spur levels. The
LMX2541 has a programmable modulator order which allows
the user to make a trade-off between PLL noise and primary
and sub-fractional spur performance. The following table provides some general guidelines for choosing modulator order:
Note that the spurs due to crosstalk will not be impacted by
modulator order.
3.5 IMPACT OF MODULATOR ORDER, DITHERING, AND
LARGER EQUIVALENT FRACTIONS ON SPURS AND
PHASE NOISE
To achieve a fractional N value, an integer N divider is modulated between different values. This gives rise to three main
degrees of freedom with the LMX2541 delta sigma engine:
the modulator order, dithering, and the way that the fractional
portion is expressed. The first degree of freedom, the modulator order, can be selected as zero (integer mode), one, two,
three, or four. One simple technique to better understand the
impact of the delta sigma fractional engine on noise and spurs
is to tune the VCO to an integer channel and observe the impact of changing the modulator order from integer mode to a
higher order. A higher fractional modulator order in theory
yields lower primary fractional spurs. However, this can also
give rise to sub-fractional spurs in some applications. The
second degree of freedom is dithering. Dithering seeks to improve the sub-fractional spurs by randomizing the sequence
of N divider values. In theory, a perfectly randomized sequence would eliminate all sub-fractional spurs, but add
phase noise by spreading the energy that would otherwise be
contained in the spurs. The third degree of freedom is the way
that the fraction is expressed. For example, 1/10 can be expressed as a larger equivalent fraction of 100000/1000000.
Using larger equivalent fractions tends to increase randomization similar to dithering. In general, the very low phase
noise of the LMX2541 exposes the modulator noise when
dithering and large fractions are used, so use these with caution. The avid reader is highly encouraged to read application
note 1879 for more details on fractional spurs. The following
table summarizes the relationships between spur types,
phase noise, modulator order, dithering and fractional expression.
1st Order Modulator Integer Mode
ORDER
2nd Order Modulator
3rd Order Modulator
4th Order Modulator
LMX2541
ence oscillator performance if a low power or noisy source is
used.
For integer mode or a first order modulator, there is no fractional noise (disregarding fractional spurs). For higher order
modulators, the fractional engine may or may not add significant phase noise depending on the fraction and choice of
dithering.
Action
Noise/Spur
Type
Increase
Modulator
Order
WORSE
(But only for
larger
Phase Noise
fractions or
more
dithering)
Increase
Dithering
Using
Larger
Equivalent
Fractions
WORSE
WORSE
Primary
Fractional
Spur
BETTER
NO IMPACT
NO IMPACT
SubFractional
Spurs
WORSE
(Creates
more subfractional
spurs)
BETTER
BETTER
www.national.com
Guidelines for use
— Use if fPD can be made very high without using
a fractional N value.
— Use if it is not desired to make fPD higher using
a fractional N value. This could be the case if
the loop bandwidth is very narrow and smaller
loop filter capacitors are desired.
— Use 1st order if fPD can be increased by at least
a factor of four over the integer case and
fractional spur frequencies and levels are
acceptable.
— If the channel spacing is 5 MHz or greater, the
1st order modulator may provide better spur
performance than integer mode.
— If the spurs of the 1st order modulator are
unacceptable, use a higher order modulator. If
the spurious components are due to crosstalk
they will not be improved by increasing
modulator order. In this case , use the lowest
order modulator that gives acceptable
performance.
— Use if the spurs of the 1st order modulator are
unacceptable.
— In general, use the lowest order modulator
unless a higher order modulator yields an
improvement in primary fractional spurs. If the
spurious components are due to crosstalk,
they will not be improved by increasing the
modulator order.
3.7 PROGRAMMABLE OUTPUT POWER WITH ON/OFF
The RFoutEN pin an RFOUT word can be used to turn the
RFout pin on and off while still keeping the VCO running and
in lock. In addition to being able to turn the output buffer on
and off, it can also be programmed in various steps using the
VCOGAIN, DIVGAIN, and OUTTERM programming words.
There are tables in the typical performance characteristics
section that discuss the impact of these words on the output
power. In addition to impacting the output power, these words
also impact the current consumption of the device. This data
was obtained as an average over all frequencies. In general,
it is desirable to find the combination of programming words
that gives the lowest current consumption for a given output
power level. All numbers reported are relative to the case of
VCOGAIN = OUTTERM = 12. According to this data, using a
VCOGAIN or OUTTERM value of 12 or greater yields only a
small increase in output power, but a large increase in current
consumption.
34
6
3
-26.0
-22.3
-18.6
-15.1
-11.8
6
-18.5
-15.5
-12.6
-9.7
-6.9
9
-11.1
-9.0
-6.9
-4.7
-2.5
3
6
DIVGAIN
9
12
15
3
-24.4
-16.2
-8.3
-0.5
+7.1
6
-21.7
-14.6
-7.6
-0.7
+6.0
9
-18.7
-12.6
-6.8
-0.7
+5.2
12
-15.9
-10.1
-5.0
0.0
+4.9
15
-13.3
-8.0
-3.2
+1.3
+5.6
15
12
-3.8
-2.6
-1.4
0.0
+1.5
15
+3.3
+3.7
+4.0
+4.5
+5.3
OUTTERM
OUTTERM
VCOGAIN
9
12
3
Change in Current Consumption in
Divided Mode as a Function of
DIVGAIN and OUTTERM
35
www.national.com
LMX2541
Change in Current Consumption in
Bypass Mode as a Function of
VCOGAIN and OUTTERM
LMX2541
3.8 LOOP FILTER
Loop filter design can be rather complicated, but there are
design tools and references available at www.national.com.
The loop bandwidth can impact the size of loop filter capacitors and also how the phase noise is filtered. For optimal
integrated phase noise, choose the bandwidth to be about
20% wider than the frequency where the in-band PLL phase
noise (as described in 3.4 PLL PHASE NOISE) and open loop
VCO noise cross. This optimal loop bandwidth may need adjustment depending on the application requirements. Reduction of spurs can be achieved by reducing the loop bandwidth.
On the other hand, a wider loop bandwidth may be required
for faster lock time. Note that using the integrated loop filter
components can lead to a significant restriction on the loop
bandwidth and should be used with care. 2 kΩ for R3_LF and
R4_LF is a good starting point. If the integrated loop filter restricts the loop bandwidth, then first try to relieve this restriction by reducing the integrated loop filter resistors and then
reduce the capacitors only if necessary.
4.
5.
6.
3.9 EXTERNAL VCO MODE
The LMX2541 also has provisions to be driven with an external VCO as well. In this mode, the user has the option of using
the RFout pin output, although if this pin is used, the VCO
input frequency is restricted to 4 GHz. If not used, the RFout
pin should be left open. The VCO input is connected to the
ExtVCOin pin. Because the internal VCO is not being used,
the part option that is being used does not have a large impact
on phase noise or spur performance. It is also possible to
switch between both Full Chip mode and External VCO mode.
3.9 CONFIGURING THE LMX2541 FOR OPTIMAL
PERFORMANCE
1. Determine the Channel Spacing (fCH)
For a system that has a VCO that tunes over several
frequencies, the channel spacing is the tuning increment.
In the case that the VCO frequency is fixed, this channel
spacing is the greatest number that divides both the VCO
frequency and the OSCin frequency.
2. Determine OSCin Frequency (fOSCin)
If the OSCin frequency is not already determined, then
there are several considerations. A higher frequency is
generally, but not always, preferable. One reason for this
is that it has a higher slew rate if it is a sine wave. Another
reason is that the clock for the VCO frequency calibration
is based on the OSCin frequency and in general will run
faster for higher OSCin frequencies.
Although a higher OSCin frequency is desirable, there
are also reasons to use a lower frequency. If the OSCin
frequency is strategically chosen, the worst case
fractional spur channels might fall out of band. Also, if the
OSCin frequency can be chosen such that the fractional
denominator can avoid factors of 2 and/or 3, the subfractional spurs can be reduced.
3. Determine the Phase Detector Frequency (fPD) ,
Charge Pump Gain (KPD) and Fractional Denominator
(FDEN)
www.national.com
In general, choose the highest phase detector frequency
and charge pump gain, unless it leads to loop filter
capacitor values that are unrealistically large for a given
loop bandwidth. In this case, reducing either the phase
detector frequency or the charge pump gain can yield
more feasible capacitor values. Other reasons for not
using the highest charge pump gain is to allow some
adjustment margin to compensate for changes in the
VCO gain or allow the use of Fastlock.
For choosing the fractional denominator, start with FDEN
= fPD/fCH. As discussed previously, there might be
reasons to choose larger equivalent fractions.
Design the Loop Filter
Determine the Modulator Order
Determine Dithering and Potential Larger Equivalent
Fractional Value
3.10 DIGITAL FSK MODE
The LMX2541 supports 2-level digital frequency shift keying
(FSK) modulation. The bit rate is limited by the loop bandwidth
of the PLL loop. As a general rule of thumb, it is desirable to
have the loop bandwidth at least twice the bit rate. This is
achieved by changing the N counter rapidly between two
states. The fractional numerator and denominator are restricted to a length of 12 bits. The 12 LSB’s of the numerator and
denominator set the center frequency, Fcenter, and the 10
MSB’s of the numerator set the frequency deviation, Fdev.
The LMX2541 has the ability to switch between two different
numerator values based on the voltage at the DATA pin.
When DATA is low, the output frequency will be Fcenter –
Fdev and when the DATA pin is high the output frequency will
be Fcenter + Fdev.A limitation of the FSK mode is that it does
not function on integer channels and the frequency deviation
can not cause the N counter to cross integer boundaries.
When using FSK mode, the FDM bit needs to be set to zero.
36
LMX2541
Ordering Information
Device
Marking
Package
LMX2541SQ2060E
412060E
SQA36A
LMX2541SQ2380E
412380E
SQA36A
LMX2541SQ2690E
412690E
SQA36A
LMX2541SQ3030E
413030E
SQA36A
LMX2541SQ3320E
413320E
SQA36A
LMX2541SQ3740E
413740E
SQA36A
37
www.national.com
LMX2541
Physical Dimensions inches (millimeters) unless otherwise noted
Consult www.national.com/analog/packaging ->LLP footprints in gerber footprint for more complete information on soldering this device reliably.
Leadless Leadframe Package (NS Package Number SQA36A), (Bottom View)
www.national.com
38
LMX2541
Notes
39
www.national.com
LMX2541 Ultra-Low Noise PLLatinum Frequency Synthesizer with Integrated VCO
Notes
For more National Semiconductor product information and proven design tools, visit the following Web sites at:
Products
Design Support
Amplifiers
www.national.com/amplifiers
WEBENCH® Tools
www.national.com/webench
Audio
www.national.com/audio
App Notes
www.national.com/appnotes
Clock and Timing
www.national.com/timing
Reference Designs
www.national.com/refdesigns
Data Converters
www.national.com/adc
Samples
www.national.com/samples
Interface
www.national.com/interface
Eval Boards
www.national.com/evalboards
LVDS
www.national.com/lvds
Packaging
www.national.com/packaging
Power Management
www.national.com/power
Green Compliance
www.national.com/quality/green
Switching Regulators
www.national.com/switchers
Distributors
www.national.com/contacts
LDOs
www.national.com/ldo
Quality and Reliability
www.national.com/quality
LED Lighting
www.national.com/led
Feedback/Support
www.national.com/feedback
Voltage Reference
www.national.com/vref
Design Made Easy
www.national.com/easy
www.national.com/powerwise
Solutions
www.national.com/solutions
Mil/Aero
www.national.com/milaero
PowerWise® Solutions
Serial Digital Interface (SDI) www.national.com/sdi
Temperature Sensors
www.national.com/tempsensors SolarMagic™
www.national.com/solarmagic
Wireless (PLL/VCO)
www.national.com/wireless
www.national.com/training
PowerWise® Design
University
THE CONTENTS OF THIS DOCUMENT ARE PROVIDED IN CONNECTION WITH NATIONAL SEMICONDUCTOR CORPORATION
(“NATIONAL”) PRODUCTS. NATIONAL MAKES NO REPRESENTATIONS OR WARRANTIES WITH RESPECT TO THE ACCURACY
OR COMPLETENESS OF THE CONTENTS OF THIS PUBLICATION AND RESERVES THE RIGHT TO MAKE CHANGES TO
SPECIFICATIONS AND PRODUCT DESCRIPTIONS AT ANY TIME WITHOUT NOTICE. NO LICENSE, WHETHER EXPRESS,
IMPLIED, ARISING BY ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS
DOCUMENT.
TESTING AND OTHER QUALITY CONTROLS ARE USED TO THE EXTENT NATIONAL DEEMS NECESSARY TO SUPPORT
NATIONAL’S PRODUCT WARRANTY. EXCEPT WHERE MANDATED BY GOVERNMENT REQUIREMENTS, TESTING OF ALL
PARAMETERS OF EACH PRODUCT IS NOT NECESSARILY PERFORMED. NATIONAL ASSUMES NO LIABILITY FOR
APPLICATIONS ASSISTANCE OR BUYER PRODUCT DESIGN. BUYERS ARE RESPONSIBLE FOR THEIR PRODUCTS AND
APPLICATIONS USING NATIONAL COMPONENTS. PRIOR TO USING OR DISTRIBUTING ANY PRODUCTS THAT INCLUDE
NATIONAL COMPONENTS, BUYERS SHOULD PROVIDE ADEQUATE DESIGN, TESTING AND OPERATING SAFEGUARDS.
EXCEPT AS PROVIDED IN NATIONAL’S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, NATIONAL ASSUMES NO
LIABILITY WHATSOEVER, AND NATIONAL DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY RELATING TO THE SALE
AND/OR USE OF NATIONAL PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR
PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY
RIGHT.
LIFE SUPPORT POLICY
NATIONAL’S PRODUCTS ARE NOT AUTHORIZED FOR USE AS CRITICAL COMPONENTS IN LIFE SUPPORT DEVICES OR
SYSTEMS WITHOUT THE EXPRESS PRIOR WRITTEN APPROVAL OF THE CHIEF EXECUTIVE OFFICER AND GENERAL
COUNSEL OF NATIONAL SEMICONDUCTOR CORPORATION. As used herein:
Life support devices or systems are devices which (a) are intended for surgical implant into the body, or (b) support or sustain life and
whose failure to perform when properly used in accordance with instructions for use provided in the labeling can be reasonably expected
to result in a significant injury to the user. A critical component is any component in a life support device or system whose failure to perform
can be reasonably expected to cause the failure of the life support device or system or to affect its safety or effectiveness.
National Semiconductor and the National Semiconductor logo are registered trademarks of National Semiconductor Corporation. All other
brand or product names may be trademarks or registered trademarks of their respective holders.
Copyright© 2009 National Semiconductor Corporation
For the most current product information visit us at www.national.com
National Semiconductor
Americas Technical
Support Center
Email: [email protected]
Tel: 1-800-272-9959
www.national.com
National Semiconductor Europe
Technical Support Center
Email: [email protected]
National Semiconductor Asia
Pacific Technical Support Center
Email: [email protected]
National Semiconductor Japan
Technical Support Center
Email: [email protected]