Amplifier Design Using ADS

1. 2. 3. 4. 5. 6. DC Simulation and circuit modeling S-Parameters and Optimization Harmonic Balance Simulations Circuit Envelope Simulation Filters - Trans, Design Guide, Momentum Final Circuit & System simulations

Amplifier Design Using ADS July 2004

Start with some specifications...

AMP with max gain & low noise:
Available voltage: 5 volts Device: Generic BJT (Gummel-Poon) Collector current: about 3.25 mA Frequency: RF = 1900 MHz Gain: > 15 dB (or much more with this model) 50 ohm match: input and output Later labs: matching and testing the AMP for TOI, distortion, noise, compression, GSM & CDMA modulation response, and more.

Filters: also, build 1900 MHz BPF for the input and a LPF for the IF output

Steps in the Design Process

Lab 1:

DC Simulations and modeling the sub-circuit

Start amp_1900 design subckt parasitics Simulate amp DC conditions & bias network Simulate amp S-parameter response Define amp matching topology and tune input Optimize the amp in & out matching networks Filter design lumped 200MHz LPF - use E-Syn Filter design microstrip 1900 MHz BPF Transient and Momentum filter analysis Amp spectrum, delivered power, Zin - HB Test amp comp, distortion, two-tone, TOI CE basics for spectrum and baseband CE for amp_1900 with GSM source Replace amp and filters in rf_sys receiver Test conversion gain, NF, swept LO power Final CDMA system test CE with fancy DDS Co-simulation of behavioral system

You are here:

DC Simulation
You get steady-state DC voltages and currents according to Ohm Law: V= IR s
Capacitors = treated as ideal open circuits Inductors = treated as ideal short circuits Topology check: dc path to ground (if not => error message) Kirchoff Law satisfied: sum of node current = 0 s Convergence simulator algorithms (modes) can be set

DC simulation controller
Palette and editor (dialog box)

Swept Variable in the controller

Sweep: allows you to sweep a parameter but it must be declared as a variable. Note the dialog entry automatically puts quotes on the controller (screen) entry.

VAR
More on DC...

DC Parameters
Available after simulation on schematic.

You can get V, I, and Power! Send to dataset.

Convergence tips: increase Max Delta V or iterations, or change mode if you don converge. t

Back Annotation...

Schematic Annotation of DC values

Immediately after DC simulation, click: Simulate > Annotate DC Solution.
Simulate >

No controller settings necessary!

Clear it here

Minus sign used for current flowing out of a connection. Otherwise, current flows into a connection or device. DC Simulation Controller is required in all simulations if you want DC annotation.
Next, named nodes...

Device with package parasitics

G-P Model Card, BJT symbol, parasitics, and ports.
BJT Symbol G-P Model Card

Port connector numbers: Num= must be set in specific order as shown - this is necessary to correctly use BJT symbol.

Model Card parameters: Bf is a passed parameter. Vaf is changed as shown.

L in pico: pH C in femto: fF

BJT Symbol: Name of model card BJTM1 used for simulation. Library devices do not require this mapping.

Design Parameters for your schematic

Click: File > Design / Parameters

Description for library annotation Component Instance name: Q Symbol: SYM_BJT_NPN Passed parameter for Bf = beta

You can specify a layout: built-in sot23

You can copy parameters from other library models.

Insert the model in a new schematic

Insert the sub-circuit from the library.

Design parameters follow the sub-circuit: Q1, beta, etc.

ICONS: Push into and Pop out of the hierarchy.

Set up a DC curve sweep with a template

This template also has a data display template.
Probe give current current in dataset. Your model (bjt_pkg) with annotation and passed parameter: beta.

Data Display template gives curve tracer results:

Initialized VARs: VCE=0V & IBB=0A

NOTE: DC controller sweeps the X-axis and the Parameter Sweep, sweeps the Y-axis.

Finally, calculate and test a bias network

Setup a new sweep: then use voltage and current to calculate R values for the DC specs for Ib and Ic.

OPTIONAL: Sweep Temperature

Back Annotation verifies DC bias using Rb and Rc as designed.

Steps in the Design Process

Lab 2: S-parameter Simulation and Optimization

You are here:
Start amp_1900 design subckt parasitics Simulate amp DC conditions & bias network Simulate amp S-parameter response Define amp matching topology and tune input Optimize the amp in & out matching networks Filter design lumped 200MHz LPF - use E-Syn Filter design microstrip 1900 MHz BPF Transient and Momentum filter analysis Amp spectrum, delivered power, Zin - HB Test amp comp, distortion, two-tone, TOI CE basics for spectrum and baseband CE for amp_1900 with GSM source Replace amp and filters in rf_sys receiver Test conversion gain, NF, swept LO power Final CDMA system test CE with fancy DDS Co-simulation of behavioral system

S-parameters are Ratios

Usually given in dB as 20 log of the voltage ratios of the waves at the ports: incident, reflected, or transmitted.

S-parameter ratios: S out / S in

Best viewed on a S11 - Forward Reflection (input match - impedance) Smith chart (next S22 - Reverse Reflection (output match - impedance) slides). S21 - Forward Transmission (gain or loss) These are easier S12 - Reverse Transmission (leakage or isolation) to understand and simply plotted.

Results of an S-Parameter Simulation in ADS

S-matrix with all complex values at each frequency point Read the complex reflection coefficient (Gamma) Change the marker readout for Zo Smith chart plots for impedance matching Results are similar to Network Analyzer measurements

The Impedance Smith Chart simplified...

This is an impedance chart transformed from rectangular Z. Normalized to 50 ohms, the center = R50+J0 or Zo (perfect match). For S11 or S22 (two-port), you get the complex impedance.
Top Half: Inductive Reactance (+jx)
25 50 100

Circles of constant Resistance

SHORT

OPEN Bottom Half: Capacitive Reactance (-jx)

Lines of constant Reactance (+jx above and -jx below)

Zo (characteristic impedance) = 50 + j0

The Smith chart in ADS Data Display

Impedance: Z

Z=0+j1

Reflection Coefficient: gamma

Z = real

/ imaginary

0 to +infinity / -infinity to + infinity

S(1,1): mag / phase 0 to 1 / 0 to +/- 180

Z= 0 + j0 Gamma or S11=1 / 180

Z = 1 + j0

Z=infinity + j infinity Gamma or S11= 1 / 0

ADS marker defaults to: S(1,1) = 0.8/ -65 Z0 * (0.35 - j1.5)

m1 Z=0-j0.5 Z=0-j1

Z=0-j2

but can be changed to give Z in ohms.

S-Parameter Simulation Controller

Default sweep variable = freq
The simulator requires a port termination Num = __

Sweep plan can also be used (see next slide). Either way, simulation data results in an S matrix in the data set for the specified range and points. Next, Other tabs

Parameters and Noise tabs

Parameters Noise

Calculate other parameters. Turn on for SS noise. If more than 2 ports, specify for 2 port NF. If not, leave blank.

Enable Frequency Conversion for a mixer. Leave blank and get all values.

Next, Sweep plan...

S-Parameter measurement equations

All simulation palettes have specific measurement equations - you set the arguments if necessary. Here, S is the matrix, 30 is the value in dB, and 51 points used to draw the circle.
Example: 3 circles for 3 different values of gain.

Arguments explained briefly here.

You will use some of these in the labs...

Next, matching...

First, simulate with ideal components

2 different datasets

Plot the data and compare to ac_sim data. Change Term Z and list the S matrix.

Calculate C and L values and re-simulate

Reactance of 10 pF at 1.9 GHz and a list of L values:

Creating Matching Networks

Various topologies can be used: L, C, R Avoid unwanted oscillations (L-C series/parallel) Yield can be a factor in topology (sensitivity) Use the fewest components (cost + efficient) Sweep or tune component values to see S-parameters Optimization: use to meet S-parameter specs (goals)

In the lab, you will optimize the match for the amplifier.

NOTE: For a mixer, match S11 @ RF and S22 @ IF.

Matching means: Moving toward the center of the Smith Chart!

Add Series or Parallel (shunt) components. You will do this in the lab.
Adjust the value to move toward open, short, L, C, or center of chart.

L llel ra Pa
Pa ra lle l

Se rie sL

C
Ser ies R
lel R Paral

es eri S C

Design Guide - Smith Chart Utility for matching...

Insert the component in schematic Bring up the utility Set source and load on chart Select components: L, C, R, etc. View response and build circuit

Next, optimization...

Smith Chart Utility for a matching network...

Smart component in schematic becomes a sub-circuit when built:

Define the Load and Source at one frequency. Select the components and move to the center or desired value. Next, Optimize!

Four elements for Optimization setup

1 - Optim controller: set the type, etc. 2 - Goal statement: use valid measurement equation or dataset expression.

3 - Enable component (opt). 4 - Simulation Controller

You will do this in the lab exercise!

ADS Optimization Types Available

= Most commonly used types.

Optimizer Random Gradient Random Minimax Gradient Minimax Quasi-Newton Least Pth Minimax Random Max Discrete Genetic

Search Method random gradient random gradient quasi-Newton quasi-Newton mini-max random discrete genetic

Error Function Formulation least-squares L2 least-squares L2 minimaxL1 MML1 minimaxL1 MML1 least-squares L2 least P-th seqLP mini-max MM worst case negL2 least-squares L2 least-squares L2

Recommendations: Use Random before using Gradient. Minimax is good for filters. Tune before using the optimizer.

Optimization Controller setup

Setup tab: Select type and set iterations. Default setting use all Goals and VARs or select specific ones in OptVar tab.

Parameters tab: type, iterations, etc. All are displayed by default.

NOTE: Avoid saving unwanted data.

Final Analysis: change to SimInstanceName (SP1)

Normalizes multiple goals!

Enable components...

Enabling components for Opt or Stats (yield)

Edit (double click) - Enable and specify continuous or discrete (stepped) variation.

NOTE: If discrete values are not realistic, use file based: DAC PPT is an optimization within a Yield Analysis only. Allows value to be shifted to achieve goal.

noopt = disabled (after optimization):

Discrete lib parts...
NOTE: DOE is an advanced topic (statistical yield analysis).

Add output matching components and Optimize !

Use Random with 2 Goals: Max for S11 and S22 = -10 dB

NOTE: opt function enables components.

Note: If GOAL RangeVar is not set, the optimizer uses all the simulation frequency points.

ADS Discrete Optimization for Library Parts

Inserted library part with listed range of values (like a DAC)

Next, an example of Yield Analysis

After a successful optimization...

EF = O = goals met! Update the values and deactivate the Opt components (noopt).

Lab exercise has final component values for these S-parameters which are good!

Also, gain & noise circles + stability

S-parameter simulation with gain and noise circles, and stability. These pre-defined measurement equations use ADS functions.
You set the arguments if necessary!

NOTE:You must turn on Noise in the simulation controller for NsCircle and list values.

Circle center is source match for optimum NF.

Optional: Read / Write data files

Or, try the yield analysis example!

Write an ADS S dataset as a Touchstone file, then Read it back in... as if it came from a Network Analyzer!

Use a Sweep Plan and compare the data.

Steps in the Design Process

Lab 3:

Harmonic Balance Simulations

Start amp_1900 design subckt parasitics Simulate amp DC conditions & bias network Simulate amp S-parameter response Define amp matching topology and tune input Optimize the amp in & out matching networks Filter design lumped 200MHz LPF - use E-Syn Filter design microstrip 1900 MHz BPF Transient and Momentum filter analysis Amp spectrum, delivered power, Zin - HB Test amp comp, distortion, two-tone, TOI CE basics for spectrum and baseband CE for amp_1900 with GSM source Replace amp and filters in rf_sys receiver Test conversion gain, NF, swept LO power Final CDMA system test CE with fancy DDS Co-simulation of behavioral system

You are here:

Harmonic Balance Simulation

Analyze circuits with Linear and Non-linear components: You define the tones, harmonics, and power levels You get the spectrum: Amplitude vs. Frequency Data can be transformed to time domain (ts function) Solutions use Newton-Raphson technique Krylov subspace method also available (large circuits) Use only Frequency domain sources Similar to Spectrum Analyzer

Types of Power Sources for HB

Default power function for these sources is polar, but you can simplify it on the screen as: dmbtow(0) Therefore, dbmtow(0) is the same as polar(dbmtow(0),0)
Notice that these sources are also ports (OK for S-param analysis). Also, they can be considered noiseless like sources in a measurement system.

P_nTone and P_nHarm can have multiple Freqs and Power.

Harmonic Balance convergence & errors

Freq [x] in each source must match Freq [x] in the controller or you get this message:

NOISE TEMP error for all noise simulations: Set Temp=16.85 to eliminate any error message. OPTIONS controller is in all simulation palettes.

Simulator will try to find closest answer, if not it will continue with all remaining valid points.

HB convergence error message: cannot sweep to desired level or arc length continuation error To solve these problems, either loosen the V and I tolerances in the options controller by ten times (for example, set: I_AbsTol= e-11), or reduce the step size for power or frequency sweeps.

16.85

Basic 1 Tone HB simulation setup

Basic HB controller and source setup gives you spectral tones:

Freq[1] is the fundamental tone you want HB to calculate. Freq[1] must match a tone in the circuit or you get a warning message. Numerous builtin sources and measurement equations.

Order [1] = 3 means HB calculates 3 harmonics of Freq [1]

HB gives you a Mix table:

Swept variables in Harmonic Balance

HB Freq tab: specify tones (Freq), harmonics (Order), and mixing products (Max Order).

1) Initialize the VAR to sweep. 2) Specify the variable and range.

3) Be sure the VAR, the source, and simulation controller all have the same information. NOTE: Swept variables always go to the dataset.

First, one tone HB and Meas Eqn

Plot Spectrum

Use ts function

List MeasEqn results Dataset contains node voltages and Mix table. Equation uses Vout[1]

Related harmonic balance controllers ...

Transform HB spectrum into the time domain with ts function: ts(Vout).

You will use HB and XDB in the lab!

XDB and power swept compression

2 ways to simulate gain compression using HB

1 dB

XDB can be set up very quickly for almost any circuit!

Two-tone HB simulation, data, DDS equation

Spacing @ 10 MHz = 1.895 and 1.905 GHz

Eqn tones generates the index values!

Use [ brackets to generate a matrix ] and { curly braces to vector the data from Mix table}

TOI or IP3 Measurement

When the input power drives the non-linear device into saturation or distortion, third order products near the desired frequency can become large. The point at which 3rd order products intercept the linear rise in output power is the intercept point TOI or IP3.
10

TOI
0

-20

ix m
--30 -40

} ,0 {1

mix {2, 1}

Output Power (dBm)

-10

Linear Output of fund(dBm) slope = 1 3rd order intermod product (dBm) at Vout slope = 3.

A measurement equation is used to get the answer...

Lab setup and data

- 30

-20

-10

10

20

Input Power (dBm)

NOTE for mixer 3 tones use: mix{-1,1,0} mix {-1, 2, -1}

TOI simulation setup using IP3 equations

Built-in measurements use functions - you set the arguments.

2-tone Mix HB data

Result of IP3 eqns in DDS:

MIXERS: use this setup for 3 tone TOI.

Steps in the Design Process

Lab 4:

Circuit Envelope Simulations

Start amp_1900 design subckt parasitics Simulate amp DC conditions & bias network Simulate amp S-parameter response Define amp matching topology and tune input Optimize the amp in & out matching networks Filter design lumped 200MHz LPF - use E-Syn Filter design microstrip 1900 MHz BPF Transient and Momentum filter analysis Amp spectrum, delivered power, Zin - HB Test amp comp, distortion, two-tone, TOI CE basics for spectrum and baseband CE for amp_1900 with GSM source Replace amp and filters in rf_sys receiver Test conversion gain, NF, swept LO power Final CDMA system test CE with fancy DDS Co-simulation of behavioral system

You are here:

What is Circuit Envelope ?

Time samples the modulation envelope (not carrier) Compute the spectrum at each time sample
Output a time-varying spectrum Use equations on the data Faster than HB or Spice in many cases Integrates with System Simulation & HP Ptolemy

Use realistic Signals with CE

GSM, CDMA, GMSK, pi/4DQPSK, QPSK, etc.
Simulations can include:

Example CE results:

32.8 kHz BW for NADC

Adjacent Channel Power Ratio Noise Power Ratio Error Vector Magnitude Power Added Efficiency Bit Error Rate

2-tone tests and linearized models do not predict this behavior as easily!

890 MHz

carrier

Finally, use AMP_1900 with the GSM source

variable t_step = 1 / (5X BW of GSM)

Plot the GSM spectrum

Verify baseband integrity using an equation to demodulate the signal:

Optional - channel power calculation

On a new page in DDS, write two equations: limits defines the bandwidth and channel_pwr calculates power in the channel.

No need to resimulate, simply use Vout[1] which is 1900 MHz!

Steps in the Design Process

Lab 5: Filters: Design Guide,Momentum,

Transient Simulation
Start amp_1900 design subckt parasitics Simulate amp DC conditions & bias network Simulate amp S-parameter response Define amp matching topology and tune input Optimize the amp in & out matching networks Filter design lumped 200MHz LPF - use E-Syn Filter design microstrip 1900 MHz BPF Transient and Momentum filter analysis Amp spectrum, delivered power, Zin - HB Test amp comp, distortion, two-tone, TOI CE basics for spectrum and baseband CE for amp_1900 with GSM source Replace amp and filters in rf_sys receiver Test conversion gain, NF, swept LO power Final CDMA system test CE with fancy DDS Co-simulation of behavioral system

You are here:

ADS Filter
1) Place a component.

DesignGuide

Schematic

Specify and build low / high pass or band pass / stop filters...

2) Specify the response, Zo, etc., by entry or spec lines.

3) View the design and simulate!

Momentum

Design a 200 MHz LPF with Filter DesignGuide

Design and simulate!

Filter will be used at the IF output (100 MHz) of the receiver.

Next, BPF...

Using Momentum

MOM engine gives S-parameter results

What is Momentum? E-M (electro-magnetic) solver using Method of Moments

technique and Green functions to compute the current in planar structures, s including vias and the coupling between surfaces. Antenna patterns!

Why use Momentum? You have no accurate model for a passive layout. You want to know the coupling effects between structures. You want to optimize the layout real-estate, performance, etc. Your other structure simulator takes too long to simulate! You want to use the results in ADS simulations.

Example spiral meshed as a strip geometry. Hole in ground plane is meshed as a slot, which is more efficient than meshing the entire ground plane.

Create a Microstrip BP Filter: 1900 MHz

Microstrip coupled line filter with substrate and VtSine source.

This filter will be used on the input to the RF system. It will also be used to show the value of ADS Momentum.

Generate the Layout of 1900 MHz BPF

Remove controllers and Terms - insert Port connectors

Use MOM menus ...

Set up Momentum: substrate, mesh, simulation

Compare to circuit S21:

Use MOM to display coupling effects

Use the same layout but add a rectangle alongside the filter:
Coupled Line Filter

draw a trace in close proximity

Momentum simulation shows resonance from coupling.

Transient simulation
Analysis performed in the Time Domain Use any Source Solutions use Newton_Raphson iterations You get Amplitude vs. Time Time Domain data can be transformed: FS
NOTE on Convolution: Frequency domain models (microstrip) can be brought into the time domain and converted to the time domain then convolved with the time-domain input signal to obtain the time-domain output signal. The convolution tab in the transient simulator allows you to define methods and settings.

Transient simulation of 1900 MHz BPF

Microstrip coupled line filter with substrate and VtSine source. Stop time after 15 periods and step using Nyquist for 15 harmonics.

Transient results show input vs output waveforms: Write an equation using markers and the indep function to calculate delay:

Steps in the Design Process

Lab 6:

Final System Simulation using 1900MHz amp.

Start amp_1900 design subckt parasitics Simulate amp DC conditions & bias network Simulate amp S-parameter response Define amp matching topology and tune input Optimize the amp in & out matching networks Filter design lumped 200MHz LPF use E-Syn Filter design microstrip 1900 MHz BPF Transient and Momentum filter analysis Amp spectrum, delivered power, Zin - HB Test amp comp, distortion, two-tone, TOI CE basics for spectrum and baseband CE for amp_1900 with GSM source Replace amp and filters in rf_sys receiver Test conversion gain, NF, swept LO power Final CDMA system test CE with fancy DDS Co-simulation of behavioral system

You are here:

What is the final topic in this class?

Simulation of your amp_1900 and filters in the receiver system to verify analog performance.
Gain (S-21), HB with swept LO, and CE with a CDMA source The co-simulation is optional if you have time - otherwise, take the CommSys class.

Co-simulation: simulation of the entire system using digital circuits and analog circuits together.
Constellation, spectrum, etc...

What is Co-simulation?
Integrated Circuit, System Simulation & HP Ptolemy

Co-simulation is simulating an A/RF schematic design with a DSP schematic design.

Top Level System Design:
DSP network A/RF network DSP network

The simulator in the A/RF design must use either Transient or Circuit Envelope

DSP Schematic Window

The A/RF schematic can be any kind of design: amplifier, mixer, PLL, etc...

The DSP components library

Many palette selections and additional libraries:

The use of these tools is covered in the DSP course and also in part of the CommSys course. In this class, it is only briefly introduced!

First, set up the sub-circuits & system

Use port connectors and the File > Design Parameters

Later on, you will replace the behavioral models with the circuits you built.

Next, set up and run HB with swept LO power

Plot: dBm_out vs LO_pwr. Write and list the IF_gain.

NF and conversion gain when status level = 4.

Final system CE simulation with CDMA

data

ACPR and Power calculations: example DDS

1. 2. 3. Copy the example DDS file. Save it in your system_prj. Pass your data into the display. YOU GET TWO PAGES OF DATA!

Change the default dataset name and the equation. All of your data will pass into the plots and equations!

Vfund = IF_out

Marker Slider for CDMA spectrum

Set up a marker slider using equations: m1 slides to a frequency and the spectrum of that carrier frequency is displayed: RF, LO, IF, etc.

m1 is at the IF

NOTE: If you finish this step in the lab, you have achieved all the goals for this class!

OPTIONAL - Co-simulation of rf_sys

This process requires several steps: First step: Modify the system to become a sub-circuit (bottom CE simulation level) as shown here. Co-simulation requires a Transient or CE simulation setup.

In this class, it is easier and faster to set up and run co-simulation using the behavioral system. The Extra exercise shows even more cosimulations you can try after the course!

Co-simulation continued...
Next step: Open a top level DSP network to access the Ptolemy / DSP palettes in schematic. Then build the system shown here: All the steps are in the lab, including the settings for the data components, filters, etc. The t_step and t_stop are now set for symbol rate and time.

Bottom level system.

Data flow controller runs showing filtered bits and IF signal: TkXY plots

Data Flow simulation - TK plots are active!

Quit the DF simulation and connect a SpectrumAnalyzer sink to collect the data. Results of this co-simulation show spectrum of the behavioral system. To use amp_1900 and your filters, replace them in the system and setup a new simulation (requires more time).

Spectrum Analyzer sink:

Contact...
EEsof Local Homepage www.adsuser.com EEsof External Homepage http://eesof.tm.agilent.com Customer Q & A =>Response Center 02-2004-5004