Keysight Technologies Inc. · NYSE:KEYS
Keysight Technologies — Technical Presentation (September 2010)
45 slides from Keysight Technologies's technical presentation, Sep 16, 2010
Slide 1: Oscillator Design with Genesys
Oscillator Design with Genesys with techniques from the book by Randy Rhea
Slide 2: I'll Cover
I'll Cover • A unified design process for different active devices, resonators and simulation technologies applicable from low frequency to microwaves • How Genesys aids the design process • What the book offers • Example Genesys workspaces Randy Rhea Susina LLC oscillatordesign@gmail.com Sept 16, 2
Slide 3: Step 1: Perform a linear analysis
Step 1 Perform a linear analysis • This is the design foundation – determines design margins – reveals tuning characteristics – estimates phase noise – fast exploration of topologies • Provides intuitive grasp of the design process • Does not provide level, harmonic or transient data Randy Rhea Susi
Slide 4: Two Linear Analysis Methods
Two Linear Analysis Methods • One-Port Reflection REFLECTION_PORT OSCILLATOR_OUTPUT RESONATOR DEVICE • Open-Loop Cascade LOOP_INPUT LOOP_OUTPUT RESONATOR AMPLIFIER OSCILLATOR_OUTPUT Randy Rhea Susina LLC oscillatordesign@gmail.com Sept 16, 2010 Slide 4
Slide 5: The Open-Loop Cascade
The Open-Loop Cascade 5V J310 R1 CP1 27Ω 0.011A C1 C3 L1 74.7pF 180pF 270nH LoopIn LoopOut OscOut 50Ω C4 .01μF C2 R2 180pF 180Ω Initial 40 MHz Colpitts with a FET device and an L-C resonator Randy Rhea Susina LLC oscillatordesign@gmail.com Sept 16, 2010 Slide 5
Slide 6: Oscillator Starting Conditions
Oscillator Starting Conditions The open-loop response criterion that insure the oscillator will start are • The oscillation frequency is the transmission phase zero-crossing, f0 • The initial linear gain at f0 must be greater than 0 dB • The phase slope at f0 must be negative Additional objectives a
Slide 7: Initial Linear Cascade Response
Initial Linear Cascade Response 18 90 S11 1.0 0 .9 1 .2 0 .8 1 .4 0.7 1 .6 S21 S22 0. 2 0 .6 8 1. .0 15 60 5 2 0. QL 4 0.4 S21 ang 0. 0. 6 0 3. .3 8 0. 0 4 .0 0 12 30 1. 5 .0 0 0 .2 1. .8 40MHz: -0.344dB, 12.728° 0 0 .6 QL, S21 ang 0 1 0 .1 4 0. S21 (dB) 20 9 0 0 .2 40MHz: -12.735dB, -167.772° 50 0.
Slide 8: Randall/Hoch Correction
Randall/Hoch Correction S 21 S12 G 1 S11S 22 S 21S12 2S12 G is the true open-loop complex gain with the loop self- terminated. [1] M. Randall and T. Hoch, "General Oscillator Characterization Using Linear Open- Loop S-Parameters," IEEE Trans. MTT, Vol. 49, June 2001, pp 1094-1100 Randy Rhea
Slide 9: Optimizing the Open-Loop Response
Optimizing the Open-Loop Response Randy Rhea Susina LLC oscillatordesign@gmail.com Sept 16, 2010 Slide 9
Slide 10: Optimized Cascade Design
Optimized Cascade Design 5V 18 90 S11 1.0 0 .9 1 .2 0 .8 S21 1 .4 0.7 QL, S21 ang, StartOpt:QL, StartOpt:S21 ang 1 .6 QL S22 0. 2 0 .6 8 1. J310 .0 5 S21 ang 2 0. 15 60 StartOpt:S11 0.4 R1 CP1 StartOpt:S21 4 StartOpt:S22 0. 0. 6 120Ω 0 StartOpt:QL 3. 0.011A .3 8 0. 0 StartOpt:S21 ang 1. 0 4 .0 S21,
Slide 11: Alechno Technique
Alechno Technique [1] S. Alechno, "The Virtual Ground in Oscillator Analysis – A Practical Example", Applied Microwave & Wireless, July 1999, pp.44-53 Randy Rhea Susina LLC oscillatordesign@gmail.com Sept 16, 2010 Slide 11
Slide 12: Back when…….
Back when……. • A random combination of a device and resonator may not satisfy the starting criterion • Historically, successful combinations were named for the discoverer and became standards • Copying designs was (and is!) common, often resulting in nonoptimum performance • Modern linear, nonlinear
Slide 13: Unified Approach
Unified Approach • These techniques are general • All resonator or device types are treated in a like manner • In Genesys, the same schematic is used for linear, nonlinear, transient and noise simulation • Ideal or real models are treated in a like manner Randy Rhea Susina LLC oscillatordesign@gmail
Slide 14: Step 2: Perform a harmonic-balance simulation
Step 2 Perform a harmonic-balance simulation • This determines the output level and harmonics • Predicts noise more accurately • Predicts the frequency more accurately Requires nonlinear device models • Numerous models are provided with Genesys • Genesys imports Spice models • Devices can also be ch
Slide 15: 330 MHz Colpitts & Spectrum
330 MHz Colpitts & Spectrum CP1 0.02A 8.321 dBm R1 2700Ω BFT92 10 1*330.028 330.028 MHz C3 47pF Pout 0 C1 20pF Csimulation 270pF FB L1 27Ω 16.16nH Port_1 C2 120 R2 -10 36pF 4700Ω Port_2 Pout (dBm) R3 C4 150Ω 10pF -20 Port_3 -30 9V -40 Open_Loop 1 2 3 -50 0 330 660 990 1320 1650 1980 2310 Frequency (
Slide 16: Nonlinear Noise Analysis
Nonlinear Noise Analysis Randy Rhea Susina LLC oscillatordesign@gmail.com Sept 16, 2010 Slide 16
Slide 17: Nonlinear Noise Analysis Plot
Nonlinear Noise Analysis Plot -70 -80 Q1 Base Output Thermal -90 Q1 Base, Output Thermal, Q1 Collector, FB (dB) Q1 Collector FB -100 -110 -120 -130 -140 -150 -160 -170 -180 -190 100e-6 1e-3 0.01 0.1 1 10 100 Frequency (MHz) 4 most significant noise contributors of the 330 MHz Colpitts bipolar oscill
Slide 18: Step 3: Perform a time-step transient analysis
Step 3 Perform a time-step transient analysis • This determines the starting characteristics • This helps discover transient spurious modes • Can be used as a substitute to HB to find output level and harmonics if nonlinear noise estimates are not required Uses the same models as HB simulation Data_
Slide 19: Output Waveform
Output Waveform 2 1.6 1.2 0.8 0.4 VPORT (V) 0 -0.4 -0.8 CP1 0.02A -1.2 R1 BFT92 2700Ω C3 -1.6 47pF -2 0 20 40 60 80 100 120 140 160 180 200 C1 Time (ns) 20pF Csimulation 270pF FB L1 27Ω 16.16nH Port_1 C2 120 R2 36pF 4700Ω Port_2 R3 C4 150Ω 10pF Port_3 9V 330 MHz Colpitts transient starting waveform
Slide 20: Genesys Oscillator Analysis
Genesys Oscillator Analysis Randy Rhea Susina LLC oscillatordesign@gmail.com Sept 16, 2010 Slide 20
Slide 21: Genesys is
Genesys is • An integrated linear, nonlinear, transient, noise simulation environment with schematic and layout tools • A process for creating new oscillator designs or repairing under performing designs • A learning tool that provides intuitive insight into all aspects of oscillator behavior Randy
Slide 22: Oscillator Topics Not Covered in this Webinar
Oscillator Topics Not Covered in this Webinar • Genesys oscillator synthesis • Integrated layout tools • Electromagnetic simulation of the layout • Instrument control for data acquisition • System simulation Randy Rhea Susina LLC oscillatordesign@gmail.com Sept 16, 2010 Slide 22
Slide 23: Why Did I Write the Book?
Why Did I Write the Book? • My previous books only covered nonlinear and transient theory qualitatively • I gained knowledge while teaching an oscillator design course to over a thousand engineers • Both oscillator and simulation technologies have advanced significantly in the last decade • I needed
Slide 24: What's in the Book?
What's in the Book ? • Detailed descriptions of linear, nonlinear, transient and noise techniques for practical oscillator design • 350 illustrations • 200 applicable equations • 60 example oscillators covering bipolar, FET, and MMIC devices with R-C, L-C, ceramic, crystal, SAW and distributed reson
Slide 25: Genesys Workspaces
Genesys Workspaces 140 Genesys workspaces were used to create the illustrations in the book. Genesys is a powerful documentation tool. • Agilent EEsof EDA has established a website for downloading 73 of the more important workspaces • These workspaces may be used with either full or trial Genesys li
Slide 26: The Available Workspaces
The Available Workspaces • The workspace template.wsx may be used to start a design. Simply add your schematic to the workspace • There are 13 utility and tutorial workspaces for amplifier and resonator design • There are 27 workspaces for general purpose oscillators • There are 13 workspaces of VCO
Slide 27: Summary
Summary • Genesys is an integrated tool, ideally suited for a unified approach to oscillator design • Discrete Oscillator Design covers practical oscillator design for a wide variety of oscillator types • Example Genesys workspaces from the book are available from Agilent EEsof Randy Rhea Susina LLC
Slide 28: For More Info: Just Google "Agilent Genesys"
For More Info: Just Google “Agilent Genesys” Email Randy Rhea at oscillatordesign@gmail.com • for questions about this presentation • for questions about the book • for a Microsoft Excel spreadsheet of equations in the book • for an errata sheet for the book, as published About Genesys: • Genesys pr
Slide 29: Question and Answer
Question and Answer
Slide 30: Questions regarding the open-loop match
“This is a list of questions posed after my Agilent webinar presentation: Oscillator Design with Genesys. I must say, responding to these questions brought back many memories for me, and it was a lot of fun”, Randy Rhea 16 Sept 2010 1. QUESTIONS REGARDING THE OPEN-LOOP MATCH 1) S11 and s22 would not
Slide 31: Point 3: Why 50 ohms?
Point 3: Why 50 ohms? First of all, don’t become hung up on this idea. From the simulator’s standpoint, you can chose any equal and resistive termination impedance: 1 ohm, 50 ohms, 1Mohm. You simply choose the resistance that best matches the source and then use the capacitive tap to match the gate
Slide 32: Question 2: Optimizing at a given frequency
2. QUESTION: To optimize at a given frequency it is possible to put the same value for maximum and minimum. I used it in ADS and maybe this could be useful for Genesys too? ANSWER: The attendee is remarking about the fact that the frequency range set in the optimization goals displayed in slide 9 wa
Slide 33: Question 5: Time domain simulator Cayenne
5. QUESTION: kayan? a time domain simulator used within genesis? ANSWER: Yes. In English it is spelled Cayenne. It is a hot spice named after the city in French Guiana. This is obviously a name play on the popular simulator family SPICE, originally from UC Berkeley. Cayenne is not a derivative of SP
Slide 34: Questions 7: Simulation frequency and EM accuracy
7. The next two questions are related and are answered as a group. QUESTIONS: 1) Up to what frequencies will Genesys simulate to? 2) During simulation, how accurate is just laying the copper (coplanar WG) on the schematic compare vs running the EM simulation? ANSWER: The short answer is up to any fr
Slide 35: Questions 8: Nonlinear model accuracy and libraries
8. The next two questions are related and are answered as a group QUESTIONS: 1) What is the highest frequency accuracy of the nonlinear models? Are they compatible with ADS? 2) Are active device libraries provided? (transistors, varactors, etc.) ANSWERS: The nonlinear active device models are provid
Slide 36: Questions 9: Phase noise design
9. The next three questions are related and are answered as a group. QUESTION: 1) Does the presenter have recommendations for designing for good phase noise besides have a good resonator Q? 2) Can you describe criteria for selecting active device used in oscillators for optimized phase noise? 3) How
Slide 37: Questions 10: Phase noise topologies
10. The next two questions are related and I will answer them as a group. QUESTION: 1) A more generic question. We have focused mainly on Colpitts oscillator. When the main concern is phase-noise, how can it be minimized and how this topology compares to Pierce and Clapp, in terms of phase-noise ? D
Slide 38: Questions 11-12: Noise impulse function
11. QUESTION: Does Genesis provide an option for computing the noise impulse function of an oscillator circuit for predicting phase noise? ANSWER: The noise impulse function is what I referred to previously as the Impulse Sensitivity Function (ISF). Genesys harmonic balance simulator, Harbec, can di
Slide 39: Question 13: DC supply regulators
13. QUESTION: From phase noise perspective, what do you think of using DC supply voltage regulators? When the supply voltage changes, the bias and phase shift of the amplifier will change. If the supply voltage change is noise, then noise is modulated into the oscillator. This is easily quantifiable
Slide 40: Questions 16-18: FET stability and DROs
16. QUESTION: FET is stable at first, when it starts oscillate would it go to unstable region? ANSWER: This is the first time I recall being asked this excellent question. The answer is yes, for both FETs and bipolars. This is one reason why excessive loop gain, negative resistance or negative condu
Slide 41: Question 19: Crystal characteristics
19. QUESTION: How do crystal characteristics define the oscillation frequency? Does it have to do with the difference in frequency between the series and parallel resonances? Is this the region where a steep phase characteristic is present? Is this also the region where a negative resistance is pres
Slide 42: Question 21: Crystal vs TCXO
21. QUESTION: If an oscillator is successfully designed with a Xtal (Crystal) for the resonator source, would replacing the Xtal with a TCXO (for temp stabilty) cause a maor change to resonator performance? ANSWER: I presume when the attendee is referring to a Temperature Compensated Xtal Oscillator
Slide 43: Question 24: Designing a VCO
24. QUESTION: To do a VCO, can you simply assume that one of the capacitors in the feedback divider is variable? What else do you need to consider? ANSWER: Yes, any reactor in the resonator can be variable and tune the oscillator. With a Colpitts, the “lower” capacitor is generally much larger than
Slide 44: Question 26: Obtaining slides and book
26. QUESTION: How do I obtain a copy of the presentation slides and a copy of the book? ANSWER: The presentation slides and the workspaces used in the book may be downloaded from http://www.agilent.com/find/eesof-genesys-osc-workspaces. The book Discrete Oscillator Design is published by Artech Hous
Slide 45: www.agilent.com
www.agilent.com www.agilent.com/find/eesof-genesys For more information about Contact Agilent at: Europe & Middle East Agilent EEsof EDA, visit: Americas Austria 01 36027 71571 www.agilent.com/find/eesof Canada (877) 894-4414 Belgium 32 (0) 2 404 93 40 Brazil (11) 4197 3500 Denmark 45 70 13 1515 For
Keysight Technologies — Technical Presentation (September 2010) on 6ix