Lecture 5 Example Design 1 - University of Toronto
Transcript of Lecture 5 Example Design 1 - University of Toronto
![Page 2: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/2.jpg)
ECE13712
Lecture PlanDate Lecture (Wednesday 2-4pm) Reference Homework
2020-01-07 1 MOD1 & MOD2 PST 2, 3, A 1: Matlab MOD1&22020-01-14 2 MODN + Toolbox PST 4, B
2: Toolbox2020-01-21 3 SC Circuits R 12, CCJM 142020-01-28 4 Comparator & Flash ADC CCJM 10
3: Comparator2020-02-04 5 Example Design 1 PST 7, CCJM 142020-02-11 6 Example Design 2 CCJM 18
4: SC MOD22020-02-18 Reading Week / ISSCC2020-02-25 7 Amplifier Design 1
Project
2020-03-03 8 Amplifier Design 22020-03-10 9 Noise in SC Circuits2020-03-17 10 Nyquist-Rate ADCs CCJM 15, 172020-03-24 11 Mismatch & MM-Shaping PST 62020-03-31 12 Continuous-Time PST 82020-04-07 Exam2020-04-21 Project Presentation (Project Report Due at start of class)
![Page 3: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/3.jpg)
ECE13713
What you will learnβ¦β’ MOD2 implementationβ’ Switched-capacitor circuits
Switched-cap summer + DACβ’ Dynamic-range scalingβ’ kT/C noiseβ’ Verification strategy
![Page 4: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/4.jpg)
ECE13714
Review: A ADC System
![Page 5: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/5.jpg)
ECE13715
Review: MOD2
β’ Doubling OSR improves SQNR by 15 dBPeak SQNR ~ πππ₯π¨π ππ π Β· πΆπΊπΉπ π πβ
![Page 6: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/6.jpg)
ECE13716
Review: Simulated MOD2 PSDβ’ Input at 50% of FullScale
![Page 7: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/7.jpg)
ECE13717
Review: Advantages of β’ ADC: Simplified Anti-Alias Filter
Since the input is oversampled, only very high frequencies alias to the passbandA simple RC filter often sufficesIf a continuous-time loop filter is used, the anti-alias filter can be often eliminated altogether
β’ Inherent LinearitySingle-bit DAC structures can yield very high SNR
β’ Robust Implementation tolerates sizable component errors
![Page 8: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/8.jpg)
ECE13718
Letβs Try Making Oneβ’ Clock at fS = 1 MHz, assume BW = 1 kHz
OSR = fS / 2κBW = 500 ~ 29
β’ MOD1: SQNR ~ 9 dB/octave x 9 octaves = 81 dBβ’ MOD2: SQNR ~ 15 dB/octave x 9 octaves = 135 dB
Actually more like 120 dBβ’ SQNR of MOD1 is not bad, but SQNR of MOD2 is
very goodIn addition to MOD2βs SQNR advantage, MOD2 is usually preferred over MOD1 because MOD2βs quantization noise is more well-behaved
![Page 9: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/9.jpg)
ECE13719
What Do We Need?
1. Summation blocks2. Delaying and non-delaying discrete-time
integrators3. Quantizer (1-bit)4. Feedback DACs (1-bit)5. Decimation filter (not shown)
Digital and therefore assumed to be easy
![Page 10: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/10.jpg)
ECE137110
Switched-Capacitor Integrator
![Page 11: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/11.jpg)
ECE137111
Switched-Capacitor Integrator
![Page 12: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/12.jpg)
ECE137112
Switched-Capacitor Integrator
β’ Since πΈπ πͺππ½ππ and πΈπ πͺππ½πππ
β’ Voltage gain is controlled by a ratio of capacitors
With careful layout, 0.1% accuracy is possible
ππ π π ππ π ππ π
ππΈπ π πΈπ π πΈπ π
πΈπ ππΈπ ππ π
π½πππ ππ½ππ π
πͺπ/πͺππ π
![Page 13: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/13.jpg)
ECE137113
Summation + Integration
β’ Adding an extra input branch accomplishes addition, with weighting
![Page 14: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/14.jpg)
ECE137114
1b DAC + Summation + Integration
![Page 15: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/15.jpg)
ECE137115
Differential Integrator
![Page 16: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/16.jpg)
ECE137116
Non-Delaying Integrator
![Page 17: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/17.jpg)
ECE137117
Non-Delaying Integrator
![Page 18: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/18.jpg)
ECE137118
Non-Delaying Integrator
β’ Delay-free integrator (inverting)
ππ π π ππ π ππ π π
ππΈπ π πΈπ π ππΈπ π
πΈπ πππΈπ ππ π
π½πππ ππ½ππ π
πͺππͺπ
ππ π
![Page 19: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/19.jpg)
ECE137119
Half-Delay Integrator
![Page 20: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/20.jpg)
ECE137120
Half-Delay Integratorβ’ Output is sampled on a different phase than the
input
β’ Some use the notation π― π ππ/π
π πto denote the
shift in sampling timeβ’ An alternative method is to declare that the
border between time n and n+1 occurs at the end of a specific phase, say 2
β’ A circuit which samples on 1 and updates on 2is non-delaying (ie, π― π π/ π π ) whereas a circuit which samples on 2 and updates on 1 is delaying (ie, π― π π/ π π )
![Page 21: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/21.jpg)
ECE137121
Timing in a ADCβ’ The safest way to deal with timing is to construct
a timing diagram and verify that the circuit implements the desired difference equations
β’ E.g. MOD2:
![Page 22: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/22.jpg)
ECE137122
Switched-Capacitor Realization
![Page 23: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/23.jpg)
ECE137123
Signal Swingβ’ So far, we have not paid any attention to how
much swing the op amps can support, or to the magnitudes of u, Vref, x1 and x2
β’ For simplicity, assume:The full-scale range of u is +/- 1 VThe op-amp swing is also +/- 1 VVref = 1 V
β’ We still need to know the ranges of x1 and x2 in order to accomplish dynamic-range scaling
![Page 24: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/24.jpg)
ECE137124
Dynamic-Range Scalingβ’ In a linear system with known state bounds, the
states can be scaled to occupy any desired range
![Page 25: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/25.jpg)
ECE137125
State Swings in MOD2β’ Linear Theory
β’ If π is constant and π is white with power πππ π/π, then ππ π, πππ
π ππππ π/π, ππ πand πππ
π ππππ π/π
![Page 26: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/26.jpg)
ECE137126
Simulated Histograms
![Page 27: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/27.jpg)
ECE137127
Observationsβ’ The match between simulations and our linear
theory is fair for x1, but poor for x2x1βs mean and standard deviation match theory, although x1βs distribution does not have the triangular form that would result if e were white and uniformly distributed in [-1, 1]x2βs mean is 50-100% high, its standard deviation is ~25% low, and the distribution is weird
β’ Our linear theory is not adequate for determining signal swings in MOD2
No real surprise since linear theory does not handle overload (ie, where x1 or x2 go to infinity when u > 1)
![Page 28: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/28.jpg)
ECE137128
MOD2 Simulated State Bounds
![Page 29: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/29.jpg)
ECE137129
Scaled MOD2β’ Take ππ π and ππ π
The first integrator should not saturateThe second integrator will not saturate for dc inputs up to -3 dBFS and possibly as high as -1 dBFS
β’ Our scaled version of MOD2 is
![Page 30: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/30.jpg)
ECE137130
First Integrator (INT1)β’ Shared Input/Reference Caps
β’ How do we determine C?
![Page 31: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/31.jpg)
ECE137131
kT/C Noise
β’ Regardless of the value of R, the mean-square value of the voltage on C is
where k = 1.38 x 10-23 J/K is Boltzmannβs constant and T is the temperature in Kelvin
The mean square noise charge is πππ πͺππππ ππ»πͺ
πππππ»πͺ
![Page 32: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/32.jpg)
ECE137132
Derivation of kT/C Noiseβ’ Noise in RC switch network
Noise PSD from R: πΊπΉ π πππ»πΉTime constant of R-C filter: π πΉπͺNoise voltage across C:
π½πͺπ πΊπΉ ππ
π ππ
ππ π
π
πππ»πΉπ
π ππ
ππ π
π
πππ»πΉππ
ππ»πͺ
![Page 33: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/33.jpg)
ECE137133
Implications for a SC Integratorβ’ Each charge/discharge operation has a random
componentThe amplifier plays a role during phase 2, but weβll assume the noise in both phases is kT/C (we will revisit this in βNoise in SC Circuitsβ Lecture)
β’ For a given cap these random components are essentially uncorrelated so the noise is white
![Page 34: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/34.jpg)
ECE137134
Implications for a SC Integratorβ’ The noise charge is equivalent to a noise voltage
with mean square value πππ πππ»/πͺπ added to the input of the integrator
β’ This noise power is spread uniformly over all frequencies from 0 to fS/2
β’ The power in the band from 0 to fB is πππ/πΆπΊπΉ
![Page 35: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/35.jpg)
ECE137135
Differential Noise
β’ Twice as many switched caps twice as much noise power
β’ The input-referred noise power in our differential integrator is
πππ πππ»/πͺπ
![Page 36: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/36.jpg)
ECE137136
INT1 Absolute Capacitor Sizesβ’ For SNR = 100 dB @ -3 dBFS input
The signal power is
Therefore we want ππ,π’π§πππ§ππ π.ππ ππ ππ π½π
Since ππ,π’π§πππ§ππ πππ/πΆπΊπΉ
β’ If we want 10 dB more SNR, we need 10x caps
πͺππππ»πππ
π.ππ π©π
πππππ Β·
π π½ π
π π.ππ π½π
π¨π/ππ πππ π
![Page 37: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/37.jpg)
ECE137137
Second Integrator (INT2)β’ Separate Input and Feedback Caps
![Page 38: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/38.jpg)
ECE137138
INT2 Absolute Capacitor Sizesβ’ In-band noise of second integrator is greatly
attenuated
Capacitor sizes not dictated by thermal noise
β’ Charge injection errors and desired ratio accuracy set absolute size
A reasonable size for a small cap is ~10 fF
![Page 39: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/39.jpg)
ECE137139
Behavioral Schematic
![Page 40: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/40.jpg)
ECE137140
Verificationβ’ Open-loop verification
1) Loop filter2) Comparator
Since MOD2 is a 1-bit system, all that can go wrong is the polarity and the timingUsually the timing is checked by (1) so this verification check is not needed
β’ Closed-loop verification3) Swing of internal states4) Spectrum: SQNR, STF gain5) Sensitivity, start-up, overload recovery, β¦
![Page 41: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/41.jpg)
ECE137141
Loop-Filter Checkβ’ Open the feedback loop, set u = 0 and drive an
impulse through the feedback path
β’ If x2 is as predicted then the loop filter is correctAt least for the feedback signal, which implies that the NTF will be as designed
![Page 42: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/42.jpg)
ECE137142
Loop-Filter Check
![Page 43: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/43.jpg)
ECE137143
Impulse Responseβ’ An impulse is {1,0,0,β¦}, but a binary DAC can
only output +/- 1 and cannot produce a 0
Q. How can we determine the impulse response of the loop filter through simulation?
A. Do two simulations: one with v = {-1,-1,-1,β¦} and one with v = {+1,-1,-1,β¦}, take the difference, divide by 2
With superposition the difference is v = {2,0,0,β¦} and the result is divided by 2To keep the integrator states from growing too quickly, you could also use v = {-1,-1,+1,-1,β¦} and then v = {+1,-1,+1,-1,β¦}
![Page 44: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/44.jpg)
ECE137144
Simulated State Swingsβ’ -3 dBFS, ~300 Hz sine wave
![Page 45: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/45.jpg)
ECE137145
Unclear Spectrum
![Page 46: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/46.jpg)
ECE137146
Better Spectrum
β’ SQNR dominated by -108 dBFS 3rd harmonic
![Page 47: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/47.jpg)
ECE137147
Implementation Summary1) Choose a viable SC topology and manually
verify timing2) Dynamic Range scaling
You now have a set of capacitor ratiosVerify operation: loop filter, timing, swing, spectrum
3) Determine absolute capacitor sizesVerify noise
4) Determine op-amp specs and construct a transistor-level schematicVerify everything
5) Layout, fab, debug, documentβ¦
![Page 48: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/48.jpg)
ECE137148
Differential vs Single-Endedβ’ Differential is more complicated and has more
caps and more noise single-ended is better?
β’ Same capacitor area same SNRDifferential is generally preferred due to rejection of even-order distortion and common-mode noise/interference.
![Page 49: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/49.jpg)
ECE137149
Shared vs Separate Input Capsβ’ Separate caps more noise
β’ Using separate caps allows input CM and ref CM to be different and is often preferred
Also improves linearity as reference feeds into separate node and wonβt introduce signal dependent error
![Page 50: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/50.jpg)
ECE137150
Single-Ended Inputβ’ Shared Caps
![Page 51: Lecture 5 Example Design 1 - University of Toronto](https://reader031.fdocuments.in/reader031/viewer/2022011815/61d507dd94288d0a3d7b68ef/html5/thumbnails/51.jpg)
ECE137151
What You Learned Todayβ’ MOD2 implementationβ’ Switched-capacitor integrator
Switched-cap summer & DAC tooβ’ Dynamic-range scalingβ’ kT/C noiseβ’ Verification strategy