← All course materialsLecture 02 · Mixers & the AD633 multiplierReading notes →

Lecture 02: Mixers, Squarers, and the AD633

1 / 28
PHYS 351 · Lecture 0201

Lecture 02

Mixers, Squarers, and the AD633

Part 1 · what a mixer does · Part 2 · the radio receiver and the NE602 · Part 3 · the AD633 · Part 4 · squaring · Part 5 · errors and offset trim · Part 6 · the square root by feedback

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0202

Part 1 · A mixer moves a signal to a new frequency

Mixer block diagram: the RF signal at f1 and the local oscillator at f2 enter a multiplier; its output passes an IF filter to give f_out. A general nonlinear circuit produces f = |m f1 ± n f2|; an ideal multiplier produces only f1 − f2 and f1 + f2.

RF in, local oscillator (LO) in, a filter picks the output. The AD633 in Lab 2 is a precise mixer for about $20: it multiplies two voltages.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0203

Multiply two cosines: Euler does the work

cos⁡(ω1t+ϕ1)cos⁡(ω2t+ϕ2)\cos(\omega_1 t+\phi_1)\cos(\omega_2 t+\phi_2)

Euler: cos⁡θ=12(eiθ+e−iθ)\cos\theta=\tfrac12\left(e^{i\theta}+e^{-i\theta}\right), with θk=ωkt+ϕk\theta_k=\omega_k t+\phi_k:

=14(eiθ1+e−iθ1)(eiθ2+e−iθ2)=\tfrac14\left(e^{i\theta_1}+e^{-i\theta_1}\right)\left(e^{i\theta_2}+e^{-i\theta_2}\right)=14(ei(θ1+θ2)+e−i(θ1+θ2)+ei(θ1−θ2)+e−i(θ1−θ2))=\tfrac14\left(e^{i(\theta_1+\theta_2)}+e^{-i(\theta_1+\theta_2)}+e^{i(\theta_1-\theta_2)}+e^{-i(\theta_1-\theta_2)}\right)=12cos⁡ ⁣((ω1−ω2)t+ϕ1−ϕ2)=\tfrac12\cos\!\big((\omega_1-\omega_2)t+\phi_1-\phi_2\big)+ 12cos⁡ ⁣((ω1+ω2)t+ϕ1+ϕ2)+\ \tfrac12\cos\!\big((\omega_1+\omega_2)t+\phi_1+\phi_2\big)
inout
f1f_1 = 6 kHzf1−f2f_1-f_2 = 1 kHz
f2f_2 = 5 kHzf1+f2f_1+f_2 = 11 kHz

The inputs are gone; two new frequencies appear.

Phases subtract too: the difference tone carries ϕ1−ϕ2\phi_1-\phi_2. At equal frequencies its DC value measures phase, which is how a multiplier works as a phase detector.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0204

In time: a beat at the difference frequency

Top: two unit cosines at 6 kHz and 5 kHz over 2 ms. Bottom: their product oscillates rapidly inside an envelope, with the dashed difference term, one half cos 2π(1 kHz)t, tracing its slow 1 kHz component.

The product wobbles at 11 kHz inside a 1 kHz envelope. A low-pass filter keeps the dashed 1 kHz term; a band-pass at 11 kHz keeps the sum.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0205

In frequency: two new lines, and a 6 dB loss

Top: input spectrum with lines of amplitude 1 at f2 = 5 kHz and f1 = 6 kHz. Bottom: the product has lines of amplitude 0.5 at f1 − f2 = 1 kHz and f1 + f2 = 11 kHz, with the inputs gone; each tone has half the amplitude and a quarter of the power, 6 dB below a tone of amplitude AB.

Each tone has amplitude AB/2AB/2. Power goes as V2/RV^2/R, so each carries ¼ of the power of a tone of amplitude ABAB: 6 dB down. Keeping one throws the other away.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0206

Any nonlinear circuit mixes

Simulated spectrum of a diode-law nonlinearity driven by 20 mV tones at f1 = 10 kHz and f2 = 3 kHz: besides the inputs, many products appear at 2f2, f1 − f2, f1 + f2, f1 + 2f2, 2f1, 2f1 + f2 and others; the two lines an ideal multiplier would make, 7 and 13 kHz, are highlighted.

A diode's exponential law makes every ∣mf1±nf2∣|mf_1 \pm nf_2|, not just the two products. A multiplier makes only f1±f2f_1\pm f_2; a real mixer relies on the filter after it.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0207

Part 2 · The superheterodyne receiver

Superheterodyne receiver: antenna, RF band-pass, RF amplifier, mixer driven by a tunable local oscillator, IF band-pass, IF amplifier, demodulator, audio amplifier, speaker. Example: all stations near 1000 kHz, times 1455 kHz from the oscillator, give 455 kHz at the IF, always.

Tuning moves only the LO. Whichever station you tune lands on the same intermediate frequency (IF), where one fixed, sharp filter and amplifier do the work: 455 kHz for AM, 10.7 MHz for FM.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0208

The image: a second station that also lands on the IF

Frequency axis from 0 to 2400 kHz: the wanted station at 1000 kHz inside the RF preselector's passband, the LO at 1455 kHz 455 kHz above it, and the image at 1910 kHz a further 455 kHz above; station and image are 2 f_IF = 910 kHz apart, and the preselector is far down at the image.

fLO=fRF+fIFf_{LO}=f_{RF}+f_{IF} = 1455 kHz. The mixer only sees ∣f−fLO∣|f-f_{LO}|, so 1910 kHz gives 455 kHz too. The RF filter must reject it, 2fIF2f_{IF} away.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0209

The mixer carries AM or FM along unchanged

A slow sine message; an AM signal whose envelope follows the message; an FM signal of constant amplitude whose cycles bunch and spread with the message.

Shifting a carrier in frequency keeps its envelope (AM) and its frequency swings (FM). Only the demodulator after the IF knows the difference.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0210

NE602 (SA602A): mixer and oscillator in one chip

NXP SA602A datasheet Fig. 1, block diagram: a voltage regulator on pin 8 VCC; an oscillator on pins 7 (emitter) and 6 (base) whose buffered output drives a multiplier; an input amplifier from pins 1 and 2 (IN_A, IN_B) feeding the multiplier; mixer outputs on pins 4 and 5; ground on pin 3.
NXP SA602A datasheet Table 3, pin description: IN_A pin 1 RF input A; IN_B pin 2 RF input B; GND pin 3 ground; OUT_A pin 4 mixer output A; OUT_B pin 5 mixer output B; OSC_B pin 6 oscillator input (base); OSC_E pin 7 oscillator output (emitter); VCC pin 8 supply voltage.

RF in on pins 1–2, oscillator on 6–7, mixer out on 4–5, supply 4.5–8 V on pin 8.

NXP SA602A Rev. 3, Fig. 1 and Table 3. Inputs to 500 MHz, oscillator to 200 MHz.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0211

Inside the NE602: a Gilbert cell

NXP SA602A datasheet Fig. 3, equivalent circuit: a Gilbert cell. A lower differential transistor pair driven from pins 1 and 2 with 1.5 kΩ emitter resistors feeds two cross-coupled upper pairs switched by the buffered oscillator from pins 6 and 7; 1.5 kΩ collector loads connect to outputs 4 and 5; bias blocks set the operating currents.
SA602A equivalent circuit (datasheet Fig. 3)

RF (pins 1, 2) drives a differential pair.

The oscillator switches the upper cross-coupled pairs.

Output current ≈ RF × sign(LO): a switching multiplier.

17 dB conversion gain at 45 MHz.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0212

Part 3 · Inside the AD633

AD633 datasheet Rev. K Fig. 1, functional block diagram: inputs X1 and X2 feed one unity-gain differential amplifier and Y1 and Y2 another; their outputs are multiplied and scaled by 1/10 V, summed with the Z input, and buffered by amplifier A to output W.
AD633 functional block diagram (Rev. K, Fig. 1)
W=(X1−X2)(Y1−Y2)10 V+ZW=\frac{(X_1-X_2)(Y_1-Y_2)}{10\ \text{V}}+Z

Differential X and Y inputs.

A buried Zener sets the 10 V scale.

ZZ adds after the product.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0213

AD633JN pinout

AD633 datasheet Rev. K Fig. 2 and Table 4, 8-lead PDIP AD633JN/AN: pin 1 X1, 2 X2, 3 Y1, 4 Y2, 5 minus VS, 6 Z summing input, 7 W product output, 8 plus VS, with W = (X1 − X2)(Y1 − Y2)/10 V + Z.

Pins 1, 2: X1X_1, X2X_2

Pins 3, 4: Y1Y_1, Y2Y_2

Pin 5: −15 V, pin 8: +15 V

Pin 6: ZZ, pin 7: WW

The SOIC version (AD633JR) is wired differently: check the marking.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0214

Lab 2 wiring: returns grounded, rails decoupled

AD633 datasheet Rev. K Fig. 12, basic multiplier connections: X input across pins 1 and 2, Y input across pins 3 and 4, pin 8 to +15 V and pin 5 to −15 V each with a 0.1 µF capacitor to ground, pin 6 as the optional summing input Z, and output W = (X1 − X2)(Y1 − Y2)/10 V + Z on pin 7.

X2X_2, Y2Y_2 and ZZ go to ground, never left floating. A 0.1 µF capacitor from each rail to ground, at pins 8 and 5.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0215

Why the 0.1 µF sits at the pin

Illustration of impedance to ground against frequency for a 0.1 µF capacitor: at the pin with about 3 nH it stays low to about 10 MHz, dipping at its resonance; 15 cm away with about 150 nH it resonates near 1.3 MHz and rises along the wire's own reactance above that; a line marks the AD633's 1 MHz bandwidth.

A capacitor is only a short to ground below its resonance. 15 cm of wire adds about 150 nH and moves that resonance near 1 MHz, right where the AD633 still has gain and can oscillate.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0216

What the datasheet promises

AD633 datasheet Rev. K Table 1, multiplier performance at 25 °C and ±15 V: total error ±1 % typical, ±2 % maximum of full scale for inputs within ±10 V; scale voltage error ±0.25 %; supply rejection ±0.01 %; nonlinearity X ±0.4 % typical, ±1 % maximum, Y ±0.1 % typical, ±0.4 % maximum; X and Y feedthrough; output offset voltage ±5 mV typical, ±50 mV maximum.

Total error ±2 % of full scale (10 V) at most. Output offset up to ±50 mV, input offset up to ±30 mV, scale error 0.25 %, nonlinearity 0.1–0.4 %. Lab 2 measures each one.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0217

Bandwidth: 1 MHz, so mix in the audio range

AD633 datasheet Rev. K Fig. 4, frequency response: output flat at 0 dB from 10 kHz to about 300 kHz, then falling, about −3 dB near 1 MHz and −20 dB near 3 MHz, with curves for 0.01 µF and 1000 pF loads.

XX = 1 V at 6 kHz, YY = 1 V at 5 kHz, ZZ = 0:

W=0.05cos⁡(2π 1 kHz t)W = 0.05\cos(2\pi\,1\,\text{kHz}\,t)+ 0.05cos⁡(2π 11 kHz t) V+\ 0.05\cos(2\pi\,11\,\text{kHz}\,t)\ \text{V}

Rev. K, Fig. 4. Two 1 V tones give only 50 mV lines; MHz radio signals are past the corner.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0218

Part 4 · Tie X to Y: a squarer

AD633 datasheet Rev. K Fig. 13, connections for squaring: input E drives both X1 and Y1, X2, Y2 and Z are grounded, supplies ±15 V with 0.1 µF bypass capacitors, and the output is W = E²/10 V.
Connections for squaring (Rev. K, Fig. 13)

X1=Y1=VinX_1=Y_1=V_{in}, X2=Y2=Z=0X_2=Y_2=Z=0:

W=Vin210 VW=\frac{V_{in}^2}{10\ \text{V}}

Inverse:

∣Vin∣=10 V⋅W|V_{in}|=\sqrt{10\ \text{V}\cdot W}
© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0219

Predict before you measure

Left: squarer output W = V_in²/10 V against V_in from −10 to 10 V, a parabola through 2.50 V at ±5 V and 10.00 V at ±10 V. Right: an output offset as a percentage of the ideal output against V_in on log axes: 20 mV is 20 % at 1 V and 0.2 % at 10 V; the 50 mV datasheet maximum is 2.5 times worse.

1 V → 0.10 V, 2 V → 0.40 V, 5 V → 2.50 V, 10 V → 10 V, for either sign. Near zero the output is tiny, so a 20 mV offset is already 20 % at 1 V.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0220

A sine in, twice the frequency out

A 5 V, 1 kHz sine input and the squarer output W = E²/20 (1 − cos 2ωt) over 2 ms: the output stays between 0 and 2.5 V around a dashed dc level of 1.25 V and peaks twice per input cycle, at 2 kHz.

(Esin⁡ωt)2/10=E220(1−cos⁡2ωt)(E\sin\omega t)^2/10 = \frac{E^2}{20}(1-\cos 2\omega t): a DC level plus a tone at 2ω2\omega. A quick scope test that the squarer works.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0221

Fit a quadratic, then plot the residuals

Simulated squarer data from −10 to 10 V with the ideal V²/10 curve and a quadratic fit, a = 0.0985 per volt, b = 0.0047 and c = −17 mV. Below, the residuals, measured minus fit, scatter within about ±15 mV: scatter means noise, a pattern would mean nonlinearity.

Simulated data. Fit W=aV2+bV+cW=aV^2+bV+c (e.g. numpy.polyfit(v, w, 2)). On the curve a good and a poor fit look alike; the residuals show the difference.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0222

Part 5 · Four errors, four signatures

Four exaggerated squarer errors against the dashed ideal parabola: an input offset moves the minimum sideways, giving a linear fit term; an output offset lifts the whole curve, a constant; a scale-factor error makes the curve steeper; nonlinearity leaves a wiggle that shows in the residuals.

Datasheet: "input and output offsets can be eliminated by using the optional trim"; scale-factor error and nonlinearity cannot, so measure and report them.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0223

What the fit coefficients mean

W=a(V+ε)2+c0W=a(V+\varepsilon)^2+c_0=aV2+2aεV+(aε2+c0)=aV^2+2a\varepsilon V+(a\varepsilon^2+c_0)

So ε=b/2a\varepsilon=b/2a, c0=c−aε2c_0=c-a\varepsilon^2.

fitmeans
a = 0.0985 /Vscale 1.5 % low
b = 0.0047ε = 24 mV
c = −17 mVc0c_0 = −17 mV
Zoom near zero of the simulated squarer fit: the fitted parabola sits below and slightly left of the dashed ideal V squared over 10, with data points at −2, −1, 0, 1 and 2 V. Its vertex is marked at minus epsilon and c0, about −24 mV and −17 mV.
© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0224

The datasheet trim network: ±50 mV

AD633 datasheet Rev. K Fig. 11, optional offset trim: a 50 kΩ potentiometer between plus VS and minus VS; its wiper feeds a 300 kΩ resistor to a node with 1 kΩ to ground, giving ±50 mV to the input terminal being trimmed, for example X2, Y2 or Z.
Optional offset trim (Rev. K, Fig. 11)

The pot swings from −15 V to +15 V.

300 kΩ and 1 kΩ divide it down:

15 V×1 k301 k=49.8 mV15\ \text{V}\times\frac{1\ \text{k}}{301\ \text{k}}=49.8\ \text{mV}

One network per pin: X2X_2, Y2Y_2, ZZ.

Offsets trim out this way; scale error and nonlinearity do not (datasheet, Error Sources).

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0225

Trim in order: X, then Y, then Z

Illustration of a typical trim order, output W in millivolts over 2 ms before and after: 1, with X = 0 and a 10 V sine on Y, trim X2 until W stops following Y; 2, with Y = 0 and a sine on X, trim Y2; 3, with X = Y = 0, trim Z until W = 0.

With X = 0, any output that follows Y is X's offset times Y: null it at X2X_2. Swap for Y. Then, with both at zero, set W=0W=0 at ZZ. Task 3 needs every offset under ±50 mV.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0226

Part 6 · Feedback inverts a function

Square root by feedback: a negative input E feeds the op-amp's inverting summing node through R; the op-amp output passes a diode to V_out, which drives an AD633 squarer giving W = V_out²/10; W returns through a second R to the summing node. The loop forces W = −E, so V_out = √(−10E).

The summing node is at 0 V, so the currents E/RE/R and W/RW/R cancel: W=−EW=-E. The op-amp drives VoutV_{out} until Vout2/10=−EV_{out}^2/10=-E, so Vout=−10EV_{out}=\sqrt{-10E}.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0227

The datasheet square-root circuit

AD633 datasheet Rev. K Fig. 15, connections for square rooting: a negative input E drives the inverting input of an AD711 op-amp through 10 kΩ; the op-amp output passes through a 1N4148 diode to both X1 and Y1 of the AD633; the AD633 output W returns through 10 kΩ to the op-amp input; the output is W = square root of −(10 V)E.

Rev. K, Fig. 15: output (after the diode) =−(10 V)E=\sqrt{-(10\ \text{V})E} for E<0E<0. The 1N4148 blocks the wrong polarity; without it, a moment of E>0E>0 latches the loop.

© Ran Yang, Ph.D.Advanced Instrumentation
PHYS 351 · Lecture 0228

Where the square root struggles

Left: V_out = √(−10E) against E from −10 to 0 V, through 10.00 V at −10 V, 3.16 V at −1 V and 1.00 V at −0.1 V; for E above 0 there is no real root and the loop latches. Right: output error against |E| for 50 mV and 10 mV input offsets, rising as E approaches zero; a 50 mV offset gives 22 % error at |E| = 0.1 V.

Near E=0E=0 the square has no slope, so the loop gain collapses and any offset dominates: 50 mV is a 22 % error at ∣E∣|E| = 0.1 V. Fit Vout=k−EV_{out}=k\sqrt{-E} and expect the low end to fit worst.

© Ran Yang, Ph.D.Advanced Instrumentation

Use ← → to move, Home / End to jump, and F for fullscreen.

Figure descriptions

Slide 2 · Part 1 · A mixer moves a signal to a new frequency

Mixer block diagram: the RF signal at f1 and the local oscillator at f2 enter a multiplier; its output passes an IF filter to give f_out. A general nonlinear circuit produces f = |m f1 ± n f2|; an ideal multiplier produces only f1 − f2 and f1 + f2.

Slide 4 · In time: a beat at the difference frequency

Top: two unit cosines at 6 kHz and 5 kHz over 2 ms. Bottom: their product oscillates rapidly inside an envelope, with the dashed difference term, one half cos 2π(1 kHz)t, tracing its slow 1 kHz component.

Slide 5 · In frequency: two new lines, and a 6 dB loss

Top: input spectrum with lines of amplitude 1 at f2 = 5 kHz and f1 = 6 kHz. Bottom: the product has lines of amplitude 0.5 at f1 − f2 = 1 kHz and f1 + f2 = 11 kHz, with the inputs gone; each tone has half the amplitude and a quarter of the power, 6 dB below a tone of amplitude AB.

Slide 6 · Any nonlinear circuit mixes

Simulated spectrum of a diode-law nonlinearity driven by 20 mV tones at f1 = 10 kHz and f2 = 3 kHz: besides the inputs, many products appear at 2f2, f1 − f2, f1 + f2, f1 + 2f2, 2f1, 2f1 + f2 and others; the two lines an ideal multiplier would make, 7 and 13 kHz, are highlighted.

Slide 7 · Part 2 · The superheterodyne receiver

Superheterodyne receiver: antenna, RF band-pass, RF amplifier, mixer driven by a tunable local oscillator, IF band-pass, IF amplifier, demodulator, audio amplifier, speaker. Example: all stations near 1000 kHz, times 1455 kHz from the oscillator, give 455 kHz at the IF, always.

Slide 8 · The image: a second station that also lands on the IF

Frequency axis from 0 to 2400 kHz: the wanted station at 1000 kHz inside the RF preselector's passband, the LO at 1455 kHz 455 kHz above it, and the image at 1910 kHz a further 455 kHz above; station and image are 2 f_IF = 910 kHz apart, and the preselector is far down at the image.

Slide 9 · The mixer carries AM or FM along unchanged

A slow sine message; an AM signal whose envelope follows the message; an FM signal of constant amplitude whose cycles bunch and spread with the message.

Slide 10 · NE602 (SA602A): mixer and oscillator in one chip

NXP SA602A datasheet Fig. 1, block diagram: a voltage regulator on pin 8 VCC; an oscillator on pins 7 (emitter) and 6 (base) whose buffered output drives a multiplier; an input amplifier from pins 1 and 2 (IN_A, IN_B) feeding the multiplier; mixer outputs on pins 4 and 5; ground on pin 3.

NXP SA602A datasheet Table 3, pin description: IN_A pin 1 RF input A; IN_B pin 2 RF input B; GND pin 3 ground; OUT_A pin 4 mixer output A; OUT_B pin 5 mixer output B; OSC_B pin 6 oscillator input (base); OSC_E pin 7 oscillator output (emitter); VCC pin 8 supply voltage.

Slide 11 · Inside the NE602: a Gilbert cell

NXP SA602A datasheet Fig. 3, equivalent circuit: a Gilbert cell. A lower differential transistor pair driven from pins 1 and 2 with 1.5 kΩ emitter resistors feeds two cross-coupled upper pairs switched by the buffered oscillator from pins 6 and 7; 1.5 kΩ collector loads connect to outputs 4 and 5; bias blocks set the operating currents.

Slide 12 · Part 3 · Inside the AD633

AD633 datasheet Rev. K Fig. 1, functional block diagram: inputs X1 and X2 feed one unity-gain differential amplifier and Y1 and Y2 another; their outputs are multiplied and scaled by 1/10 V, summed with the Z input, and buffered by amplifier A to output W.

Slide 13 · AD633JN pinout

AD633 datasheet Rev. K Fig. 2 and Table 4, 8-lead PDIP AD633JN/AN: pin 1 X1, 2 X2, 3 Y1, 4 Y2, 5 minus VS, 6 Z summing input, 7 W product output, 8 plus VS, with W = (X1 − X2)(Y1 − Y2)/10 V + Z.

Slide 14 · Lab 2 wiring: returns grounded, rails decoupled

AD633 datasheet Rev. K Fig. 12, basic multiplier connections: X input across pins 1 and 2, Y input across pins 3 and 4, pin 8 to +15 V and pin 5 to −15 V each with a 0.1 µF capacitor to ground, pin 6 as the optional summing input Z, and output W = (X1 − X2)(Y1 − Y2)/10 V + Z on pin 7.

Slide 15 · Why the 0.1 µF sits at the pin

Illustration of impedance to ground against frequency for a 0.1 µF capacitor: at the pin with about 3 nH it stays low to about 10 MHz, dipping at its resonance; 15 cm away with about 150 nH it resonates near 1.3 MHz and rises along the wire's own reactance above that; a line marks the AD633's 1 MHz bandwidth.

Slide 16 · What the datasheet promises

AD633 datasheet Rev. K Table 1, multiplier performance at 25 °C and ±15 V: total error ±1 % typical, ±2 % maximum of full scale for inputs within ±10 V; scale voltage error ±0.25 %; supply rejection ±0.01 %; nonlinearity X ±0.4 % typical, ±1 % maximum, Y ±0.1 % typical, ±0.4 % maximum; X and Y feedthrough; output offset voltage ±5 mV typical, ±50 mV maximum.

Slide 17 · Bandwidth: 1 MHz, so mix in the audio range

AD633 datasheet Rev. K Fig. 4, frequency response: output flat at 0 dB from 10 kHz to about 300 kHz, then falling, about −3 dB near 1 MHz and −20 dB near 3 MHz, with curves for 0.01 µF and 1000 pF loads.

Slide 18 · Part 4 · Tie X to Y: a squarer

AD633 datasheet Rev. K Fig. 13, connections for squaring: input E drives both X1 and Y1, X2, Y2 and Z are grounded, supplies ±15 V with 0.1 µF bypass capacitors, and the output is W = E²/10 V.

Slide 19 · Predict before you measure

Left: squarer output W = V_in²/10 V against V_in from −10 to 10 V, a parabola through 2.50 V at ±5 V and 10.00 V at ±10 V. Right: an output offset as a percentage of the ideal output against V_in on log axes: 20 mV is 20 % at 1 V and 0.2 % at 10 V; the 50 mV datasheet maximum is 2.5 times worse.

Slide 20 · A sine in, twice the frequency out

A 5 V, 1 kHz sine input and the squarer output W = E²/20 (1 − cos 2ωt) over 2 ms: the output stays between 0 and 2.5 V around a dashed dc level of 1.25 V and peaks twice per input cycle, at 2 kHz.

Slide 21 · Fit a quadratic, then plot the residuals

Simulated squarer data from −10 to 10 V with the ideal V²/10 curve and a quadratic fit, a = 0.0985 per volt, b = 0.0047 and c = −17 mV. Below, the residuals, measured minus fit, scatter within about ±15 mV: scatter means noise, a pattern would mean nonlinearity.

Slide 22 · Part 5 · Four errors, four signatures

Four exaggerated squarer errors against the dashed ideal parabola: an input offset moves the minimum sideways, giving a linear fit term; an output offset lifts the whole curve, a constant; a scale-factor error makes the curve steeper; nonlinearity leaves a wiggle that shows in the residuals.

Slide 23 · What the fit coefficients mean

Zoom near zero of the simulated squarer fit: the fitted parabola sits below and slightly left of the dashed ideal V squared over 10, with data points at −2, −1, 0, 1 and 2 V. Its vertex is marked at minus epsilon and c0, about −24 mV and −17 mV.

Slide 24 · The datasheet trim network: ±50 mV

AD633 datasheet Rev. K Fig. 11, optional offset trim: a 50 kΩ potentiometer between plus VS and minus VS; its wiper feeds a 300 kΩ resistor to a node with 1 kΩ to ground, giving ±50 mV to the input terminal being trimmed, for example X2, Y2 or Z.

Slide 25 · Trim in order: X, then Y, then Z

Illustration of a typical trim order, output W in millivolts over 2 ms before and after: 1, with X = 0 and a 10 V sine on Y, trim X2 until W stops following Y; 2, with Y = 0 and a sine on X, trim Y2; 3, with X = Y = 0, trim Z until W = 0.

Slide 26 · Part 6 · Feedback inverts a function

Square root by feedback: a negative input E feeds the op-amp's inverting summing node through R; the op-amp output passes a diode to V_out, which drives an AD633 squarer giving W = V_out²/10; W returns through a second R to the summing node. The loop forces W = −E, so V_out = √(−10E).

Slide 27 · The datasheet square-root circuit

AD633 datasheet Rev. K Fig. 15, connections for square rooting: a negative input E drives the inverting input of an AD711 op-amp through 10 kΩ; the op-amp output passes through a 1N4148 diode to both X1 and Y1 of the AD633; the AD633 output W returns through 10 kΩ to the op-amp input; the output is W = square root of −(10 V)E.

Slide 28 · Where the square root struggles

Left: V_out = √(−10E) against E from −10 to 0 V, through 10.00 V at −10 V, 3.16 V at −1 V and 1.00 V at −0.1 V; for E above 0 there is no real root and the loop latches. Right: output error against |E| for 50 mV and 10 mV input offsets, rising as E approaches zero; a 50 mV offset gives 22 % error at |E| = 0.1 V.