CIRCUIT ANALYSIS EXAMPLES
2.75 GHz Negative Resistance Oscillator
Let's turn our attention to a
different kind of
circuit, a negative resistance common-base oscillator. Here, we'll use
the
SER command to insert an inductor into the base
lead of a bipolar transistor. Then, we'll
tweak
it to maximize the input reflection coefficient.
The base lead inductance, when
properly chosen,
will
raise the magnitude of either the S11 or S22
scattering parameter, (or both), to a value
greater
than one. Then, when that happens, a
negative
resistance will be created. Typically, you would tune the port with the
highest reflection coefficient, and output match the
other
port.
To begin, choose "
(1) Sceptre" from the
Main
Menu. either by pressing "
1" or by using
the "
F1" function key, you'll enter the
Sceptre
Module. You'll see the following
familiar
screen display:
Sceptre
Module
Filename
[<Enter>=Quit] ? _
Enter "
CBOSC" to load the
common-base oscillator circuit file stored in the \mwdata2\
sub-directory, and then press
<Enter>.
The basic model used to evaluate the negative resistance of this
particular active device is shown below. Its followed by the .CKT file
listing.
TWO
|
AA
|
S1
|
50
|
|
|
|
|
IND
|
BB
|
PA
|
1.0
|
|
|
|
|
SER
|
AA
|
BB
|
|
|
|
|
|
TWO
|
DD
|
S1
|
50
|
|
|
|
|
PRI
|
AA
|
S1
|
50
|
|
|
|
|
| END |
|
|
|
|
|
|
|
2750
|
|
|
|
|
|
|
|
0.9
|
150
|
1.7
|
-80
|
0.07
|
120
|
1.08
|
-56
|
Study the .CKT file listing shown above. The
first
two lines describe the circuit elements, and
the
SER command, (please see
Appendix B), is
used to
series connect them in line 3. Note that
the
new two-port
always
takes on the name of the
first element,
AA
in this case.
After you enter the filename above,
Sceptre asks:
Analysis or Tweak
(A/T) [<Enter>=Quit] ? _
In
response, type "
T," and then
press the
<Enter> key to use the
tweak
mode.
Sceptre will analyze the circuit and give the
following on-screen output. It will then
wait for
you to tell it what to do:
Sceptre Circuit Analysis on 07-18-2006 at
08:11:45
Filename: CBOSC
Freq
|
S11 Mag/Ang
|
S21 Mag/Ang
|
S12 Mag/Ang
|
S22 Mag/Ang
|
S21(dB)
K B1
|
|
|
|
|
|
|
2750
|
1.502/
119.0
|
2.071/ -114.7
|
0.492/
111.2
|
1.217/ -84.3
|
6.32
-0.63 0.32
|
Task Completed
This first analysis shows that the
emitter port, as
represented by S11, is probably the
best place to
place a resonant tank since it offers the
highest
negative resistance.
The analysis concludes with the following question:
Tweak Variable
(Y/N) [<Enter>=No] ? _
If you respond with "
Y," (as you should here),
you'll see the following:
Line,Column,New
Value (L,C,NV) ? _
What does this
mean? If you look above at the "
CBOSC"
.CKT file, you'll see that in line 2 we specified a 1.0 nH parallel
connected inductor to start with. The
size
of this inductor is specified in column 4, so, if you wanted to
tweak
this element, and change its value from 1.0 to 1.5 nH, you would
respond to the question as follows:
2,4,1.5
Is this
clear? You've now entered
new
values for the inductor by entering
L,
C,
and
NV. The new values are: L=2, C=4, and NV=1.5.
After pressing
<Enter>, you should
see the following:
2750
|
1.737/ 98.0
|
2.073/ -138.9
|
0.735/ 92.9
|
1.149/ -104.6
|
6.33 -0.57 1.09
|
Note that S11, the input port reflection coefficient magnitude has
increased from 1.502 to 1.737. This means that the negative resistive is
increasing, which is what you want for an oscillator.
O.K., so far so good. Now, let's
tweak
the inductor value
upwards in increments of 0.1 nH,
starting with 1.6 nH, until we reach, or pass, a
peak
in the magnitude of S11M.
We'll show you below what happened when we did this. If you like,
repeat our entries for practice.
Tweak Variable
(Y/N) [<Enter>=No] ? Y
Line,Column,New
Value (L,C,NV) ? 2,4,1.6
2750
|
1.761/ 93.6
|
2.044/ -143.9
|
0.776/ 89.0
|
1.118/ -108.8
|
6.21
-0.56 1.26
|
Tweak Variable
(Y/N) [<Enter>=No] ? Y
Line,Column,New
Value (L,C,NV) ? 2,4,1.7
2750
|
1.777/ 89.3
|
2.006/ -149.0
|
0.813/ 85.1
|
1.081/ -113.1
|
6.05
-0.54 1.43
|
Tweak Variable
(Y/N) [<Enter>=No] ? Y
Line,Column,New
Value (L,C,NV) ? 2,4,1.8
2750
|
1.783/ 85.1
|
1.959/ -154.0
|
0.845/ 81.3
|
1.040/ -117.2
|
5.84
-0.53 1.58
|
Enter the f
Column,New
Value (L,C,NV) ? 2,4,1.9
2750
|
1.781/ 81.0
|
1.906/ -158.9
|
0.872/ 77.5
|
0.995/ -121.3
|
5.60
-0.51 1.72
|
Tweak Variable
(Y/N) [<Enter>=No] ? N
As you can see, the magnitude of S11 peaked at
about 1.8 nH. Choose "
N" to stop tweaking, and
return to the
SmithMatch Module opening screen.
Now, if you would like to see the impedance equivalent of S11, you may
edit
the "
CBOSC" .CKT file using the
File
Utility. To do this, first press
<Enter>
to return to the
Main Menu, and then enter the
File
Utility. If you would rather not,
skip
the rest of this section and go on to design Example 4.
You
edit the "CBOSC" file by using the "Replace"
and "Save" commands. You want to change
line 2 and line 4 so that they read:
IND BB PA 1.8
PRI AA IR 50
Make sure
you make the changes ONE line at
a time. First edit "CBOSC" and "Replace"
line 2, "Save" it using the new name "CBOSCIR,"
then edit the new "CBOSCIR," replace line 4, and
save again as "CBOSCIR."
Keep in mind, that in Sceptre you don't enter any
"three-letter extensions" or "path" commands. Just type the name
you want for a file and press <Enter>.
Return to the Sceptre Module and run the file "CBOSCIR."
Sceptre
Module
Filename
[<Enter>=Quit] ? _
Enter the filename as "
CBOSCIR"
and then press
<Enter>. Choose the
analysis
option, and choose
not to print.
Here is what the
IR output format looks like:
Sceptre Circuit Analysis
07-18-2006
at 08:20:11
Filename: CBOSCIR
Input Reflection Coefficient and VSWR in a 50 ohm system
Freq
|
Rho Mag/Ang
|
VSWR
|
R.L. (dB)
|
Z(R+jX) ohm
|
|
|
|
|
|
2750
|
1.783/
85.1
|
|
5.023
|
-28.118+j 45.849
|
Create SmithMatch .IMP
File (Y/N) [<Enter>=No] ? _
Please enter "
N" to the above question if you do not have our
SmithMatch software installed, or do not wish to create a .IMP file, otherwise an error will occur.
Task Completed
Note that, since the input resistive component
is
negative, VSWR is undefined..
We think you should know how to
Quit
by now?
<
BACK 1 2
3 4 5
6 7 8
9 10 11 12 13 14 15 16 17 18 19
20 NEXT >
You are on Page 10