Alpha Industries, Inc. [781] 935-5150 Fax [617] 824-4579 Email [email protected] www.alphaind.

com 1
Specifications subject to change without notice. 6/99A
A Colpitts VCO for Wideband (0.952.15 GHz)
Set-Top TV Tuner Applications
APN1006
Introduction
Modern set-top DBS TV tuners require high performance,
broadband voltage control oscillator (VCO) designs at a
competitive cost. To meet these goals, design engineers
are challenged to create high performance, low-cost
VCOs.
The Colpitts oscillator is a traditional design used for many
VCO applications. Designing a broadband Colpitts
oscillator with coverage from 12 GHz requires the
selection and interaction of an appropriate varactor diode
for its resonator. This application note describes the design
of a broadband Colpitts VCO that incorporates the
SMV1265-011 varactor diode. This varactor diode was
specifically developed at Alpha for this application. The
VCO design, based on Libra Series IV simulation, shows
good correlation between measured and simulated
performance. This application note includes a board layout
and materials list.
VCO Model
Figure 1 shows the VCO model built for open loop analysis
in Libra Series IV.
Figure 1. VCO Model Built for Open Loop Analysis in Libra Series IV
Application Note
2 Alpha Industries, Inc. [781] 935-5150 Fax [617] 824-4579 Email [email protected] www.alphaind.com
Specifications subject to change without notice. 6/99A
A Colpitts VCO for Wideband (0.952.15 GHz) Set-Top TV Tuner Applications APN1006
This circuit schematic, which is a simple Colpitts
structure, uses a series back-to-back connection of two
SMV1265-011 varactors instead of a single varactor. This
connection allows lower capacitance at high voltages,
while maintaining the tuning ratio of a single varactor. The
back-to-back varactor connection also helps reduce
distortion and the effect of fringing and mounting
capacitances. These parasitic capacitances are included
in the model as C
5
, valued at 0.6 pF. This value may
change depending on the layout of the board.
DC bias is provided through resistors R
1
and R
4
, both
3 k, which may affect phase noise, but allows the
exclusion of chokes. This reduces costs and the
possibility of parasitic resonances the common
cause of spurious responses and frequency instability.
The resonator inductance was modeled as a lossy
inductor (with Q = 25 at 100 MHz) in parallel with a
capacitance of 0.25 pF. This is typical for a multilayer
inductor of style 0603 (60 x 30 mil) footprint (TOKO Coils
and Filters catalog). The inductor value of 5.6 nH was
optimized to fit the desired 12 GHz frequency band. The
DC blocking series capacitance (C
SER
) was modeled as
a SRC network, including associated parasitics; it was
selected at 1000 pF to avoid affecting the resonator (Q).
The Colpitts feedback capacitances (C
DIV1
= 1 pF and
C
DIV2
= 1.62 pF) were optimized to provide a flat power
response over the tuning range. These values may also
be re-optimized for phase noise if required.
The NEC NE68533 transistor was selected to fit the
required bandwidth performance. Note: The circuit is very
sensitive to the transistor choice (tuning range and
stability) due to the wide bandwidth requirement. The
output is supplied from the emitter load resistance (RL
1
)
through the 2 pF coupling capacitor, modeled as a series
SLC
1
component.
The microstrip line (TL
1
) simulates the design layout which
may be incorporated in the resonator.
Figure 2 shows the Libra test bench. In the test bench we
define an open loop gain (Ku = V
OUT
/V
IN
) as a ratio of
voltage phasors at input and output ports of an
OSCTEST component. Defining the oscillation point
requires the balancing of input (loop) power to provide zero
gain for a zero loop phase shift. Once the oscillation point
is defined, the frequency and output power can be
measured. Use of the OSCTEST2 component for the
close loop analysis is not recommended, since it may not
converge in some cases, and doesnt allow clear insight
into understanding the VCO behavior. These properties
are considered an advantage of modeling over a purely
experimental study.
Figure 2. Libra Test Bench
Alpha Industries, Inc. [781] 935-5150 Fax [617] 824-4579 Email [email protected] www.alphaind.com 3
Specifications subject to change without notice. 6/99A
A Colpitts VCO for Wideband (0.952.15 GHz) Set-Top TV Tuner Applications APN1006
Figure 3 shows the default bench. The variables used for
more convenient tuning during performance analysis and
optimization are listed in a variables and equations
component.
Figure 3. Default Test Bench
Figure 4. SMV1265-011 Libra IV SPICE Model
SMV1265-011 SPICE Model
Figure 4 shows a SPICE model for the SMV1265-011
varactor diode, defined for the Libra IV environment, with
a description of the parameters employed.
4 Alpha Industries, Inc. [781] 935-5150 Fax [617] 824-4579 Email [email protected] www.alphaind.com
Specifications subject to change without notice. 6/99A
A Colpitts VCO for Wideband (0.952.15 GHz) Set-Top TV Tuner Applications APN1006
Voltage Range C
JO
M V
J
C
P
(V) (pF) (V) (pF)
02.5 22.5 2.0 4.00 0.00
2.56.5 21.0 25.0 68.00 0.00
6.511 20.0 7.3 14.00 0.90
11up 20.0 1.8 1.85 0.56
usable varactor voltages is segmented into a number of
subranges each with a unique set of the V
J
, M, C
JO
, and
C
P
parameters as given in the Table 2.
Parameter Description Unit Default
IS Saturation current (with N, determine the DC characteristics of the diode) A 1e-14
R
S
Series resistance 0
N Emission coefficient (with IS, determines the DC characteristics of the diode) - 1
TT Transit time S 0
C
JO
Zero-bias junction capacitance (with V
J
and M, defines nonlinear junction capacitance of the diode) F 0
V
J
Junction potential (with V
J
and M, defines nonlinear junction capacitance of the diode) V 1
M Grading coefficient (with V
J
and M, defines nonlinear junction capacitance of the diode) - 0.5
E
G
Energy gap (with XTI, helps define the dependence of IS on temperature) EV 1.11
XTI Saturation current temperature exponent (with E
G
, helps define the dependence of IS on temperature) - 3
KF Flicker noise coefficient - 0
AF Flicker noise exponent - 1
FC Forward-bias depletion capacitance coefficient - 0.5
B
V
Reverse breakdown voltage V Infinity
I
BV
Current at reverse breakdown voltage A 1e-3
ISR Recombination current parameter A 0
NR Emission coefficient for ISR - 2
IKF High-injection knee current A Infinity
NBV Reverse breakdown ideality factor - 1
IBVL Low-level reverse breakdown knee current A 0
NBVL Low-level reverse breakdown ideality factor - 1
T
NOM
Nominal ambient temperature at which these model parameters were derived C 27
FFE Flicker noise frequency exponent 1
Table 1 describes the model parameters. It shows default
values appropriate for silicon varactor diodes which may be
used by the Libra IV simulator.
According to the SPICE model in Figure 4, the varactor
capacitance (C
V
) is a function of the applied reverse DC
voltage (V
R
) and may be expressed as follows:
This equation is a mathematical expression of the
capacitance characteristic. The model is accurate for
abrupt junction varactors (Alphas SMV1400 series);
however, the model is less accurate for hyperabrupt
junction varactors because the coefficients are dependent
on the applied voltage. To make the equation fit the
hyperabrupt performances for the SMV1265-011 a
piece-wise approach was employed. Here the coefficients
(V
J
, M, C
JO
, and C
P
) are made piece-wise functions of the
varactor DC voltage applied. Thus, the whole range of the
These subranges are made to overlap each other. Thus,
if a reasonable RF swing (one that is appropriate in a
practical VCO case) exceeds limits of the subrange, the
C
V
function described by the current subrange will still fit
in the original curve.
Table 1. Silicon Varactor Diode Default Values
Table 2. Varactor Voltages
C
V
= + C
P
1 +
C
JO
M
V
R
V
J
( )
Alpha Industries, Inc. [781] 935-5150 Fax [617] 824-4579 Email [email protected] www.alphaind.com 5
Specifications subject to change without notice. 6/99A
A Colpitts VCO for Wideband (0.952.15 GHz) Set-Top TV Tuner Applications APN1006
Figure 5 demonstrates the quality of the piece-wise fitting
approach.
Special consideration was given to the fit at the lowest
capacitance range (high-voltage area) since it
dramatically affects the upper frequency limit of the VCO.
To incorporate this function into Libra, the pwl() built-in
function was used in the variables component of the
schematic bench.
M = pwl (V
VAR
0 2 2.5 2 2.500009 25 6.5 25
6.50009 7.3 11 7.3 11.0009 1.8 40 1.8)
V
J
= pwl (V
VAR
0 4 2.5 4 2.500009 68 6.5 68
6.50009 14 11 14 11.0009 1.85 40 1.85)
C
P
= pwl (V
VAR
0 0 2.5 0 2.500009 0 6.5 0
6.50009 0.9 11 0.9 11.0009 0.56 40 0.56)
C
JO
= pwl (V
VAR
0 22.5 2.5 22.5 2.500009 21 6.5
21 6.50009 20 11 20 11.0009 20 40 20)*10
12
Note: While C
P
is given in picofarads, C
GO
is given in
farads to comply with the default nominations in Libra. (For
more details regarding pwl() function see Circuit Network
Items, Variables and Equations, Series IV Manuals,
p. 1915).

0.1
0 5 10 15 20 25 30
1
10
C
a
p
a
c
i
t
a
n
c
e

(
p
F
)
100
0
0.2
0.6
1.0
0.4
0.8
Varactor Voltage (V)
Approximation
Measured
0 30 25 20 15 10 5
Varactor Voltage (V)
1.2
1.4
1.6
1.8
2.0
2.2
2.4
2.6
S
e
r
i
e
s

R
e
s
i
s
t
a
n
c
e

(

)

RS_PWL
RS Measured
Since the epitaxial layer for this kind of hyperabrupt
varactor has relatively high resistivity the series resistance
is strongly dependent on the reverse voltage applied to
varactor junction. The value of series resistance (R
S
)
measured at 500 MHz is shown in Figure 6, with a piece-
wise approximation of R
S
also given.
The piece-wise function may be used as follows:
R
S
= pwl (V
VAR
0 2.4 3 2.4 4 2.3 5 2.2 6 2 7 1.85
8 1.76 9 1.7 10 1.65 11 1.61 12 1.5 40 1.5)
Note: The pwl() function in Libra IV is defined for the
evaluation of harmonic balance parameters rather than
variables. Therefore, although series resistance was
defined as dependent on reverse voltage, for harmonic
balance it remains parametric and linear. The same
applies to capacitance, which remains the same as in the
original diode model, but its coefficients (V
J
, M, C
JO
, and
C
P
) become parametric functions of the reverse voltage.
Figure 5. SMV1265 Capacitance vs. Voltage Figure 6. SMV1265 Resistance vs. Voltage
6 Alpha Industries, Inc. [781] 935-5150 Fax [617] 824-4579 Email [email protected] www.alphaind.com
Specifications subject to change without notice. 6/99A
Figure 8 shows the PCB layout. The board is made of
standard FR4 material 60 mils thick.
A Colpitts VCO for Wideband (0.952.15 GHz) Set-Top TV Tuner Applications APN1006
Table 3 shows the bill of materials used.
Table 3. Bill of Materials
Designator Part Type Footprint
C
1
0603AU561JAT9 (AVX) 0603
C
2
0603AU2R0JAT9 (AVX) 0603
C
3
0603AU561JAT9 (AVX) 0603
C
4
0603AU201JAT9 (AVX) 0603
C
5
0603AU1R0JAT9 (AVX) 0603
C
6
0603AU1R6JAT9 (AVX) 0603
D
1
NE68519 (NEC) SOT-419
L
1
LL1608-F5N6S (TOKO) 0603
R
1
CR10-332J-T (AVX) 0603
R
2
CR10-912J-T (AVX) 0603
R
3
CR10-201J-T (AVX) 0603
R
4
CR10-302J-T (AVX) 0603
R
5
CR10-302J-T (AVX) 0603
V
1
SMV1265-011 (Alpha Ind.) SOD-323
V
2
SMV1265-011 (Alpha Ind.) SOD-323
3 k
320 x 30 mils
300 p
5.6 nH
SMV1265-011
SMV1265-011
560 p
3 k
9.1 k
3.3 k
V
CC = 5 V
I
cc = 9 mA
NE68519
1 p
1.62 p
200
2 p
RF Output
V
TUNE
Figure 7. VCO Circuit Diagram
VCO Design Materials, Layout, and
Performance
Figure 7 shows the VCO circuit diagram.
720 MIL
7
2
0

M
I
L
Figure 8. PCB Layout
Alpha Industries, Inc. [781] 935-5150 Fax [617] 824-4579 Email [email protected] www.alphaind.com 7
Specifications subject to change without notice. 6/99A
A Colpitts VCO for Wideband (0.952.15 GHz) Set-Top TV Tuner Applications APN1006
Table 4 shows tabulated measurement data. In voltage
ranges of 127 V, the usable frequency coverage was
estimated from 0.982.15 GHz.
0 30 25 20 15 10 5
Varactor Voltage (V)
0.9
1.0
1.1
1.2
1.3
1.4
1.5
1.6
1.7
1.8
1.9
2.0
2.1
2.2
F
r
e
q
u
e
n
c
y

(
G
H
z
)
0
1
2
3
4
5
6
7
P
O
U
T

(
d
B
m
)
F
EXP
F
MODEL
P
OUT_EXP
P
OUT_MODEL
Figure 9. Measured and Simulated
Frequency vs. Varactor Voltage
Table 4. Tabulated Measurement Data
Figure 9 shows both the measured performance of this
circuit and the simulated results, obtained with the above
model. The simulated tuning curve (frequency vs. voltage)
is in excellent agreement with measured data, proving the
effectiveness of the piece-wise approximation technique.
The measured power response, shows some differences
from its simulation, but is within the same range.
A possible reason for the discrepancy could be the effect
of higher harmonics. To simulate this would require
significantly more complicated modeling of the
components, board parasitics, and discontinuities.
However, for most engineering purposes the circuit
performance prediction indicated here should be
satisfactory.
V
VAR
Frequency P
OUT
(V) (GHz) (dBm)
0.5 0.950 5.7
1.0 0.974 5.5
2.0 1.018 5.4
4.0 1.184 4.7
8.0 1.680 3.2
12.0 1.886 5.2
14.0 1.932 4.9
18.0 2.008 5.0
22.0 2.076 3.9
25.0 2.120 3.5
30.0 2.188 2.2
List of Available Documents
1. Colpitts Wideband VCO Simulation Project Files
for Libra IV.
2. Colpitts Wideband VCO Circuit Schematic and PCB
Layout for Protel EDA Client, 1998 version.
3. Colpitts Wideband VCO Gerber Photo-plot Files
4. A Colpitts VCO for Wideband (0.952.15 GHz)
Set-Top TV Tuner Applications. (Current Document).
5. Detailed measurement and simulation data.
For the availability of the listed materials, please call our
applications engineering staff.
Alpha Industries, Inc., 1999. All rights reserved.