`radio frequency interference (RFI). Here input capacitor, CN,
`filters high frequencies above I 35 kHz before they reach the
`preamp input. In addition, further filtering is provided in the
`second stage by RrC 5, at 241 kHz. Additional RFI filtering can
`take on several options, both in-circuit or on a system level (see
`Box). Within the circuit of Figure 8, separate low-resistance in(cid:173)
`line RF chokes can be used in series with the two inputs, as can a
`single common-mode choke, chosen to reject the expected
`interference. RF bypassing of the SSM2017's input transistors can
`also be used, from pins 1-2 and 3-8 (shown optionally on the
`schematic as CRF1 and CRJ'2).
`This amplifier's performance is quite good over programmed gain
`ranges of 6 to 66 dB (i.e., 2 to 2000 VN). For a typical audio load
`of 600 !l, the THD+N performance (Figure 10) at various gains
`and an oucput level of 10 V rms is consistent and well below 0.01 %,
`for all but the highest gains, where it becomes principally limited
`by noise. Noise performance is quite good; at an operating gain of
`
`20
`
`100
`
`1k
`FREQUENCY - Hz
`
`10k
`
`20k
`
`Figure 10. Low-noise transformerless-input mic preamp,
`THD+N (%)vs.frequency (Hz); V0 ut = 10 V rms, gain= 12 to
`66 dB, RL = 600 n.
`
`1000, it is equivalent to 1 nV/Hz, referred to the preamp input.
`Maximum output is a function of the power supplies, and-as
`shown-can be as high as IO V rms, with higher-voltage supplies.
`The series output resistor, R3, does limit available swing somewhat
`in driving 600 Q, but it should be retained for isolation and shon(cid:173)
`circuit protection. Supply voltages of ± 18 V or more are
`appropriate for highest output into 600 n, but lower supply voltage
`is appropriate for lower maximum output levels.
`
`Toe SSM2017 device architecture is basically similar to that of
`family predecessors, SSM2015 and SSM2016, but the 8-pin
`package rules out access to the internal gain resistors. These
`devices, applied in circuits similar to Figure 10 with phantom
`powering, have individual virtues. For example, the SSM2016 can
`be applied with supplies of up to ± 36 V, and it has a high-current
`output stage (±40 mA minimum). This enables it to drive low(cid:173)
`I]
`impedance audio loads to high levels without a buffer.
`
`REFERENCES
`I. C. Motchcnbacher, F. Pitchen, Low-Nobe Electronic Design, Wiley,
`NewYork, 1973.
`2. Walt Jung," Audio Prearnplifiersn, Chapter 8 within System Appllcadon
`Guide (Walt Kester, ed.), Analog DevicC$, 1993*.
`3. Walt Jung, James Wong, "High Performance ICs in Single-Supply Analog
`Circuits", Analog OWogue, Vol. 27-2, 1993. Circle 7
`4. Jen~n Transfonners, 7135 Hayvenhurst Avenue, Van Nuys, CA, 91406,
`(213) 876--0059.
`5. G. Bore, "Powering Condenser Microphonesn, db, June 1970.
`6. ANSI Standard 268-15 (Revision 1987, amendments 1989, 1990, 1991).
`American National Standards Institute, 11 W. 42nd Sc., New York, NY,
`10036.
`7 . Steve Hogan, "Standard Mic lnpuc Applicationn, Jensen Transformers
`Application note JT99-0003, November, 1992.
`8 . Wale Jung, Audio IC Op Amp Applicadons, 3rd ed., Howard W. Sams,
`1987.
`9. Scott Wurcer, "An O,perational Amplifier Archiceciure with a Single Gain
`Stage and Distortion Cancellation", prc~nted at March, 1992 AES
`Convention, preprint # 323 l.
`10. W. Jung, A. Glircia, "A Low Noise Microphone Preamp with a Phantom
`Power Option", Analog Devices AN-242, November 1992. Circle 8
`
`•Book av:i.ilablc for purchase from Analos Devices. U$( book p111Chase cud.
`
`SYSTEM RF INTERFERENCE PREVENTION
`lengths (read antenna!), as a transformer preamp type. Not only
`As the main text notes, both circuit- and system-level
`does the ttansformerless mic preamp lack up-front filtering, but
`perspectives may be required to prevent potential RFI problems
`with low level preamplifiers (such as mic preamps). System
`the inherent design principles of low noise stages which make
`them attractive for mic preamps also tend to make them
`performance is impacted by both the topology chosen and the
`physical arrangement of components within a given circuit.
`vulnerable to RFI.
`
`From a system point of view, the basic choice between
`transformer and transformerless preamp circuitry has a major
`impact on RFI sensitivity. This is so because a typical audio(cid:173)
`grade transformer has high CMR and is a natural low-pass filter
`at RF. Thus, to at least a first order, a transformer-input mic
`preamp can be expected to have more RFI immunity than a
`transformerless design. The degree to which these expectations
`can be realized depends on both the relative frequency of the
`interfering signals and the transformer's physical construction.
`
`On the other hand, a transformerless mic preamp would seem
`inherently suscepnble to RFI problems, since it lacks an intrinsic
`low-pass filter (comparable to a transformer) preceding it.Yet,
`for the same application it must work from the same cable
`
`A low noise preamp input stage has high transconductance, and
`typically this means the use of bipolar transistors. While bipolars
`are indeed excellent for internally generated noise
`(transconductance and noise have an inverse relationship), their
`high transconductance at the same time also opens the door to
`greater RFI sensitivity. In fact, a comparison ofbipolars versus
`FETs, in both op-amp and instrumentation-amplifier input
`stages, both in theory and experimentally, shows the bipolar
`stage more susceptible to RFI.[Bl]The key application question
`then becomes what can one do in the circuitry external to a low
`noise mic preamp to obtain best RFI immunity.
`
`While the relatively simple circuit measures of bandwidth
`limiting, such as those shown in Figure 8, can be helpful, they
`
`Analog Dialogue 28-2 119941
`
`17
`
`Xilinx, Inc. and Xilinx Asia Pacific Pte. Ltd. Exhibit 1025 Page 17
`Xilinx, Inc. and Xilinx Asia Pacific Pte. Ltd. v. Analog Devices, Inc. IPR2020-01559
`
`
`
`may not be adequate when the preamp must work directly within
`a strong RF field-for example, in a radio-station studio adjacent
`to a transminer. lbis problem can be most acute (and easiest
`to observe) when the RF is at a relatively low frequency, such
`as the standard AM broadcast band (0.5 to 1.5 MHz). lbis
`frequency span is difficult to filter with high attenuation, since
`it is only one decade (or less) above the approximately I 00-kHz
`bandwidth typically used with mic preamps to avoid bandwidth
`limiting in multistage audio system applications. As a result,
`suppression of this interfering band by 40 dB or more is not at
`all a trivial matter when using conventional filtering. Interference
`at higher frequencies, for example, IO MHz or more, does allow
`relatively simple R-C or L-C filters to be effective, typically
`when they are placed at the point of signal entrance (antenna
`input) to the equipment.
`Fortunately, one filter type, not as widely known as the more
`common R-C/L-C types, can be quite effective for the lower
`band RF noise. It is the common-mode choke, which can be
`designed for effective interference filtering at frequencies below
`100 kHz, but with minimal impact on normal audio signals.
`Shown below in simplified form with a balanced-input preamp,
`a common-mode choke is a tightly coupled and somewhat
`specialized transformer (for audio use); it is typically wound
`with bifilar wire. A detailed technical analysis of CM choke
`operation can be found in [B2); this discussion focuses on the
`practical implications for high-quality audio stages. The trifilar(cid:173)
`wound CM chokes described by Gelbach [B3,B4] operate on
`similar principles, in de and LF instrumentation.
`
`L1
`
`BAUNCEO
`AUDIO
`WITH
`CM RF NOISE
`
`Lt au
`
`DIFF'EAENTW.. INPUT
`MICPREAIIP
`
`Most interfering signals picked up on audio signal lines appear
`to the receiving circuit as CM signals, or at least predominantly
`CM. Therefore, for high rejection of CM signals in cases of
`extreme high level, high frequency inputs or RF contamination,
`a single common-mode choke in series with differential amplifier
`inputs can be a very effective RFI solution.
`Perhaps the most important point to understand about CM
`chokes is that their behavior is distinctly different for signals
`applied in CM fashion, compared to the desired differential(cid:173)
`mode (DM) signals. Because of tight flux coupling in the
`common core, CM signals see the full rated inductance of the
`choke, which works in conjunction with the balanced capacitive
`termination to provide a two-pole low-pass filter for these signals.
`On the other hand, the DM signals do not see this high
`inductance, because their flux cancels. As a result, they only
`see the leakage inductance, and the higher associated bandwidth.
`Thus the DM and CM filter bandwidths of a high quality CM
`choke can indeed be quite different. It is this characteristic that
`allows the CM choke to pass full audio-range bandwidths
`without attenuation, while suppressing CM signals beginning
`just above (or actually within) the audio band. In addition, this
`
`type of L-C filtering introduces little if any noise degradation
`to the signal, as long as the choke's de resistance is of the order
`of 10 Q or less.
`The utility of CM filtering applies not only to microphone and
`other low-level circuits; it can also apply to higher-level line receiver
`circuits andA/D inputs as well. David Birt [B5] and Allen Burdick
`[B6) have described CM choke filtering techniques, like those
`outlined above, aimed specifically at audio circuits. CM filter
`circuitry can be inserted within a preamp proper, and/or used
`externally, as a packaged device, with standard XLR connector
`interfaces ahead of noise-susceptible gear. ff used within a preamp,
`the network values of the filter would be optimized to complement
`the preamp input impedances. For wide bandwidth l.F & HF
`filtering, cascaded chokes can be used. [B4]
`The Benchmark Media Systems CMF-1 is an example of a two(cid:173)
`pole CM filter in pre-packaged form; it is suitable for in-line
`use ahead of any preamp or line-level audio stage. In the figure
`below, its performance is shown in a 200-kHz bandwidth by
`log frequency response plots of: DM input/ DM output ("I"),
`CM input/CM output (" 2"), and CM input/ DM output
`response ("3"), to 1-V rms signals (0 dB), RL = 9.09 kn. The
`normal signal DM passband is in excess of 200 kHz, while the
`CM passband is 26 kHz. The DM response to CM signals is
`below -80 dB or bener at frequencies above 50 Hz, but the
`DI
`filter still has 60 dB of attenuation at I MHz (not shown).
`
`DMIDM
`
`1
`
`I'-
`
`"I..
`CIIICM
`
`I"-
`
`:I
`
`' "Ii
`
`C•IIJM
`I•
`
`a
`r-.
`
`,,,,,
`
`"'
`
`.....:
`RI !slDUAL ~
`I
`I I
`1DOII 200k
`
`10
`0
`
`-10
`
`,,
`CD -20
`I -30
`w
`~-40
`9i-60
`:l!-60
`~ -70
`F=
`j -eo
`1/J a: -eo
`-100
`-110
`
`I'.
`
`r-
`
`~ .....
`
`-120
`20
`
`100
`
`11111
`1k
`FREOUEHCV - Hz
`
`REFERENCES
`Bl. A. Garcia,Jnput-Stoge RFJ Rtetifea,icn Smsinvuysecrion of"Precision
`Sensor Signal Conditioning and Transmission", Chapter 3 within
`System Application Guide (Walt Kester, ed.), Analog Devices,
`1993.*
`82. H. On, "Low Frcqu~cy Analysis of Common-Mode Choke", Chapter
`3 wilhin Noise Reducdon Techniques in Electronic Systems,
`2d ed., Wiley, 1988.
`83. H. Gelbach, "High Frequency Common Mode, the Contaminator of
`Signals", Proceedings oflSA 3Slh lnternationaJ Instnunentadon
`Symposium, May 1993.
`B4. J. Bryant., H. Gelbach, "High Frequency Signal Contamination",
`Analog Dialogue, Vol. 27-2, 1993. Circle 7
`B5. David Birt, "Elecironically Balanced Analogue-Line Interfaces",
`Proceedings oflnsdtute of Acoustics Conference, P.O. Box 320,
`St. Albans ALI IPZ, U.K., Nov. 1990.
`86. Allen Burdick, "A Clean Audio Installation Guide", Benchmark Media
`Systems application note, 1993 (5925 Court Street Rd., Syracuse,
`NY 13206, (315) 437-6300).
`
`•Book available for purchase Crom Analog DC\ices. Use book purchase card.
`
`18
`
`Analog Dialogue 28-2 (19941
`
`Xilinx, Inc. and Xilinx Asia Pacific Pte. Ltd. Exhibit 1025 Page 18
`Xilinx, Inc. and Xilinx Asia Pacific Pte. Ltd. v. Analog Devices, Inc. IPR2020-01559
`
`
`
`(For information use reply card or see back cover) New-Product Briefs
`AID Converters, Multiplexers, Codec
`+5-V Stereo Codec Has Serial Port
`Stereo 16-Bit ADC
`Single-supply L-~ AD1877:
`AD1847 SoundPort® interfaces directly to DSPs
`92-dB SNR; 100 µW "sleep"
`Low-cost device is housed in 44-pin PLCC/TQFP
`PCMCIA card. When used with a DSP, the
`The AD 1877 is a low-cost, +5-V 16-bit dual
`ND converter on a single chip for audio(cid:173)
`AD I 84 7 can support conversions for
`band width applications. It includes a
`advanced audio capabilities, such as
`2.25-V on-chip voltage reference, 4th-order
`compression, spatial sound effects, and
`64x oversampling I-A modulators, and
`music synthesis.
`3-stage linear-phase decimation filters; its
`outputs are available via a serial output
`interface having 8 user-defined modes, both
`master and slave.
`
`The AD1847 is a monolithic low-cost 16-
`bit stereo codec for audio. Its flexible serial
`port interfaces directly to a DSP or system
`VO chip. The device comprises key audio
`data-conversion and control functions,
`including stereo pairs of :f.-tJ. ND and DI A
`converters, programmable gain/attenuate/
`mute functions, µ/A-law compression and
`expansion, filtering, and a voltage reference.
`Two pairs of stereo line inputs are available
`at the input via a multiplexer; and two
`auxiliary pairs can be muxed to the input
`or mixed with the audio output.
`
`Supporting the Microsoft Windows sound
`system, the AD 184 7 is a key element of a
`low-cost solution for business, game audio,
`and multimedia applications requiring
`operation from a single +5-V supply. The
`serial interface allows implementation on
`a computer mother-board, add-in, or
`
`little off-chip support circuitry is needed.
`Anti-imaging output filters are included on(cid:173)
`chip. Dynamic range exceeds 70 dB over the
`20-kHz audio band. An on-chip oscillator,
`with external crystals, provides sample rates
`from 5.5 kHz to 48 kHz.The 16-bit output
`pair from the ADCs is available over a serial
`interface chat also supports 16-bit digital
`input to the DACs, plus concrol/scatus
`information.
`
`Specified for operation from 0 to +70°C,
`the device is housed in a 44-pin PLCC.
`With a +5-V supply, max power dissipation is
`750 mW operating, 4 mW power-down. Price
`in I 0,000s is well below SI 0. Circle 9 ll
`
`Perfonnance features include 92-dB dynamic
`range (90-dB min), 20 Hz to" 20 kHz (without
`A-weighted filtering) and 90-dB signal-to(cid:173)
`noise-plus-distortion (88 dB min). Input
`overrange indication is provided, and power
`dissipation is low-315 mW max operating,
`325 µW max power-down while clocked (and
`a mere 5 µW unclocked).
`
`It is housed in a 28-pin SOIC, and operates
`at temperatures from o to +70°C. Price is
`ll
`SIO in 1000s. Circle 10
`
`Dual I,-~ Modulator
`5th-order ADMOD79 has
`103-dB dynamic range
`The ADMOD79 is a 2-channel differential(cid:173)
`input modulator for high-perfonnance :f.-tJ.
`stereo AID converters of up to 18-bit
`resolution. Its single-bit monotonic outputs are
`produced by fifth-order modulators with
`proprietary, patented noise shaping and
`an on-chip de voltage reference. With a 3.072-
`MHz modulator clock and 64x oversampling,
`it offers I 03 dB of dynamic range over 20 kHz.
`Sophisticated system designers use this key
`building block
`to construct high(cid:173)
`performance AID functions. Performance
`can be tailored to the application using an
`application-specific digital decimation
`filter-implemented either as a fixed
`function or by a programmable DSP-and
`by an analog front end to match the input
`source to the modulator. Using ± 5-V
`supplies, and housed in a 28-pin cerdip, the
`ADMOD79 operates from Oto +70.,C. Price
`ll
`is S25 in I 000s. Circle 11
`
`10-Bit 20-MSPS ADC
`MIL-qualified AD773A
`S/(N+D) ~51-dB@ 10 MHz
`The AD773A A/D converter is an
`improved direct replacement for the IO(cid:173)
`bit, 18-MSPS AD773, introduced here
`3 years ago (A-D 25-2, p. 14). It is
`available for both commercial and military
`(/883) temperature ranges, requires
`300-mW less power (I .2 W max), has 8-
`dB less THD at 10 MHz (-55 dB max,
`K version). Its full-power input bandwidth
`is I 00 MHz. Typical applications include
`infrared imaging, radar and com(cid:173)
`munications, and applications in space/
`radiation environments.
`
`Other features include no-missing-codes,
`guaranteed; high-impedance (2!5 0 kO)
`reference input (2. 5 V); out-of-range output
`flag; output available in binary and twos
`complement coding. The AD77 3A is housed
`in a 28-pin ceramic DIP and is available for
`0 co 70°C and -55 to +125°C operation.
`Prices start at S55 in I 00s. Circle 12 ll
`
`Analog MUXes
`16-channel ADG406/426,
`Dual-8-channel ADG407
`The ADG406 and ADG426 are I 6-
`channel CMOS multiplexers; the ADG407
`is differential with 8-channels. The ADG426
`has on-chip address and control latches; the
`ADG406/407 require external latches for
`microprocessor interfacing.
`The devices will work with single- or dual
`supplies. The signal range for all three
`extends to the supply rails, and switching
`action is "break-before-make." RoN is 80 n
`max, and switching is fast ( < 160 ns toN,
`< 150 ns toFF)- Power requirements are 5 µA
`max1 00,VEN = 0;500 µAmax,VEN = 2.4 V.
`They are available in DIPs and PLCCs;
`and the ADG426 is the first latchable 16-
`channel MUX to be available in an SSOP
`package.They are currently available for the
`extended industrial (-40 co +85°C) tem(cid:173)
`perature range. Prices (I 00s) start at $6.25
`(ADG406/407) and S6.65 (ADG426).
`ll
`Circle 13
`
`All brand or product names mentioned are trademarks or registered trademarks of their respective holders.
`
`Analog Dialogue 28-2 119941
`
`19
`
`Xilinx, Inc. and Xilinx Asia Pacific Pte. Ltd. Exhibit 1025 Page 19
`Xilinx, Inc. and Xilinx Asia Pacific Pte. Ltd. v. Analog Devices, Inc. IPR2020-01559
`
`
`
`New Product Briefs (For information use reply card or see back cover)
`D/A Converters, Analog Switches
`Fast 16-Bit DAC
`8 x 8-Bit Multiplying TrimDAC®
`DAC16 has cunent output &
`AD8842 has 8 independent gain-control channels,
`500•ns settling, is monotonic
`Serial input, flexible 4-quadrant multiplication
`The DACJ6 is a high-speed parallel-input,
`current-output 16-bit DIA converter
`characterized by low noise (0.3 LSB max),
`low nonlinearity (± 1-LSB max differential
`nonlinearity), and low drift (0.025 ppm/°C
`zero, 5 ppm/°C gain) . Its combination of fast
`(500-ns) settling and high de accuracy make
`it well-suited for generating precise levels at
`high speed for control and test.
`
`A TrimDAC® provides low-cost digitally
`controlled adjustment of parameters in
`analog circuits, permitting designs that
`supplant manual or servoed pots. The
`AD8842 is a low-cost TrimDAC offering
`eight independent channels of four(cid:173)
`quadrant-multiplying gain control. "Four
`quadrant" means that the controlled ac or
`de signals can be unipolar or bipolar, and
`the range of gain adjustment includes both
`positive and negative gains.
`
`The AD8842 contains eight independent
`vohage-input, voltage-output multiplying
`DIA converters, each having a bandwidth
`from de to 50 kHz. Typical applicacions
`include vertical amplitude adjustment in
`CRT-based computer graphic displays,
`de setpoint control of video amplifiers,
`automatic calibration, and waveform
`generation and modulation.
`
`Typical high-speed, high-accuracy
`applications include waveform synthesis and
`modulation in equipment for communi(cid:173)
`cations, control, instrumentation, automatic
`test, and graphic displays.
`
`TfL and CMOS-compatible, the DAC-16
`is available in 24-pin plastic DIPs & SOL
`for the -40 to +85°C range, and ceramic
`DIPS & LC Cs for -55 co + I 25°C; an
`evaluation board is available. DAC 16 prices
`ll
`start at S25 in I 00s. Circle 14
`
`Voo
`
`LO
`
`so,
`
`Ct.K
`
`The digital gain-control input arrives on a
`standard 3-wire serial interface as a 12-bit
`serial word; 8-bic data is latched into one of
`eight DAC registers selected by the first four
`bits. A serial data output pin allows multiple
`DACs to be daisy-chained.
`
`The AD8842 consumes 95 mW from ±5-V
`power supplies. It is available in 24-pin
`plastic DIP and SOL-24 packages, for
`operation at temperatures from -40 co
`+85°C. Prices Start at SS.69 in 1000s.
`ll
`Circle IS
`
`Dual 12-Bit DAC
`Single-supply AD8582
`Has rail•to-rail outputs
`The AD8582, operating on a +5-V supply,
`is a complece, adjustment-free parallel-input,
`dual 12-bit DAC with a rail-to-rail voltage
`output. It includes an on-chip laser-trimmed
`2. 5-V bandgap vol rage reference and
`double-buffered data interface logic. le is
`essentially a dual-channel version of the
`DAC8562, introduced here last year ( Analog
`Dialogue 27-1, p. 25) . Typical applications
`include digitally controlled calibration,
`portable equipment, digitized servo and
`process controls, and power-level
`adjustment in communication equipment.
`Its output is coded for convenient I mVlbit
`scaling (4,095 V for all-Is). Power dissipation
`is very low, typically 5 mW, ideal for battery(cid:173)
`operated systems. Housed in PDIP-24 and
`SOL-24 packages, it has an operating
`temperature range of -40 co +85°C. Prices
`ll
`stare at S9.44 in 1000s.. Circle 16
`
`8 x 12-Bit DACPORT
`AD75089 is small, accurate
`Has readback and O reset
`The monolithic AD75089 DACPORT®
`comprises 8 channels of I 2-bit voltage(cid:173)
`output D/ A conversion, complete with a + 5-
`vo l c reference, output amp I ifiers, and
`double-buffered 12-bit parallel input. The
`buffer structure provides the ability to read
`back the value stored in a selected DAC latch
`and co reset all DAC latches to the most(cid:173)
`negative value . Typical applications for
`multiple analog outputs include control
`systems, automatic test, and robotics.
`
`With± 12-volt supplies, che analog output
`range is ± 5 V, with offset-binary coding.
`Max differential and integral nonlinearity
`specs are± 1/2 LSB and± I LSB at 25°C.
`The output slew rate is 3 V/µs, with± full(cid:173)
`scale settling time of 8 µs co± 1/2 LSB. The
`device is housed in a 44-pin PLCC and
`has an opera ring temperature range of 0 to
`+70°C . Price is S76 in I D0s. Circle 17 Cl
`
`Quad SPST Switches
`Latchup-proof ADG441 /2/4
`-40 to +85°C in plastic
`The monolithic ADG44 I /42/44 comprise folJJ'
`channels of independently selec1able single·
`pole, single-throw (SPST) CMOS switches.
`A trench isolation process provides latchup(cid:173)
`proof rail-to-rail operation with single- or dual
`supplies. Compatible with existing 441/42/44
`switches, they can improve performance of
`circuits using the ADG201A/02A/l IA.
`
`The ADG44 l and ADG444 s,vilches tum on
`with logic Low; che ADG442 switches with
`logic High.The ADG441/42 generate internal
`logic levels; the lower-cost ADG444 requires
`an external VL. Switches are break-before(cid:173)
`make and conduct equally well in both
`directions. RoN is <70 n, with 3-D. matching;
`toN is < 110 ns and to FF <60 ns. They are
`available for -40 to +85°C in 16-pin plastic
`DlPs and SOIC, and for -55 to + 125°C in
`ccrdip. Prices (100s) start at S2.05/S2 .05/
`Cl
`S1.33. Circle 18
`
`All brand or product names mentioned are trademarks or registered trademarks of their respective holders.
`
`20
`
`Analog Dialogue 28-2 119941
`
`Xilinx, Inc. and Xilinx Asia Pacific Pte. Ltd. Exhibit 1025 Page 20
`Xilinx, Inc. and Xilinx Asia Pacific Pte. Ltd. v. Analog Devices, Inc. IPR2020-01559
`
`
`
`(For information use reply card or see back cover) New-Product Briefs
`Op Amps & Analog Multiplier
`Rail-to-Rail Dual and Quad Op Amps
`250-MHz Multiplier
`Vout, 4-quadrant AD835
`OP291 and OP491: +3 to +10-V single supply
`Inputs swing v. ± 5 V without phase reversal
`8-pin miniDIP or SOIC
`
`The OP291 and OP491 are single-supply
`dual and quad op amps that handle rail-to(cid:173)
`rai! inputs and furnish rail-to-rail outputs.
`As amplifiers tend to be used in applications
`requiring both lower supply voltage and less
`quiescent power, good dynamic range
`requires that the input and output signal
`swing over the whole supply-voltage range.
`These amplifiers will swing to within 50 mV
`of the+ rail and 10 mV of the - rail (100-
`kn load).
`
`When handling mil-to-rail signals, it is also
`important that che input can be driven at or
`beyond the supply voltage with simple
`clipping, i.e., without reversal of phase or
`latchup. The OP29 l & OP491 's amplifiers
`can be overdriven by up to 5 V beyond the
`supplies without latchup or phase reversal.
`This allows che designer co design with
`minimal-if any-input protection for
`almost all applications.
`
`The OP291 and OP491 can operate at
`supply voltages from +2. 7 V to + 12 V.
`They require a maximum of 500 µA per
`channel over the en tire -40 to + 12 5 ° C
`temperature range. DC specifications
`include 25 V/mV minimum open-loop gain,
`1.5 mV max offset voltage and 60 nA
`max bias current over temperature. Noise is
`2 mV p-p (0.1 co 10 Hz), 35 nV/filz at
`1 kHz, and 0.8 pA/../Hz.
`
`Dynamically, typical slew rate is ±0.4 V/JJS,
`small-signal gain-bandwidth is 3 MHz, full(cid:173)
`power bandwidth is 1.2 kHz, and senling
`time to 0.01 % is 22 µs. Channel separation
`is 145 dB. They will drive a minimum shore(cid:173)
`circuit current of ±9 mA.
`The OP291 is available in 8-lead mini-DIP
`and SO, and che OP491 is housed in a
`14-lead epoxy DIP or 14-lead SO. Prices in
`I 000s start ac S2. I 6 for the O P29 l and
`Cl
`S3.23 for the OP491. Circle 19
`
`200 MHz TransZAmp
`AD8015 meets 155-Mbps(cid:173)
`fiber BW, noise, sensitivity
`The AD8015 is a wideband (200 MHz)
`cransimpedance amplifier optimized for use
`in fiberoptic receiver circuits with NRZ data
`rates up co 300 Mbps. It can be used in
`FDDI receivers and for SONET/ SDH data
`rates up to 155. 52 Mbps. An alternative to
`GaAs-based transZ amps, it is a complete
`solution for converting photodiode current
`input into a voltage output; its differential
`outputs can interface directly to ECL
`comparators or post-amps.
`
`Specifications include I . 5-ns rise- and fall
`times; noise of 2 pA/iliz at I 00 MHz, (total
`rms of 20 nA in 100-MHz BW); optical
`sensitivity of - 36 dBm at 1 5 5. 5 2 Mbps, with
`peak input current of± 200 µA. The AD8015
`uses a single +5-V supply, and needs only
`25 mA. Housed in an 8-lead plastic SOIC,
`ic operates from --40 to +85°C. Price (I 000s)
`Cl
`is S3.59. Phone (617) 937-1124
`
`OP176 Audio Op Amp
`Dynamic range of bipolars,
`JFET speed & sound quality
`The OP! 76 is a low-noise (6 nVJ../Hz), high(cid:173)
`output-drive op amp with a Butler front end,
`efficiently combining the low noise and
`distortion of bipolar transistors with the
`speed and sound quality of JFETs. With its
`wide output range and short-circuit
`protection, it is an excellent choice for output
`sections of audio systems, driving lines and
`cables. The output is capable of driving 600-
`D. loads to IO V rms, while maintaining low
`noise and distortion (0.00 I% at 3 V rms).
`
`The OP176 draws only 2.5 mA of supply
`current. Specifications include low de offset
`(1.25 mV max, -40 to +85°C)-which
`permits adjustment-free circuits in many
`applications; wide bandwidth (IO MHz) and
`high slew rate (15 V/µs); and it is unity-gain
`stable. It is packaged in 8-pin PDIPs and
`SOI Cs. Price starts at $0.88 (88¢) in 1000s.
`Cl
`Circle 21
`
`The AD835 is the industry's first monolithic
`2 5 0-MHz, 4-quadran t vol tage-ou tpu t
`analog multiplier. It multiplies two single(cid:173)
`polarity or bipolar differential input voltages
`(X and Y) and provides the true inscanc-by(cid:173)
`instant product, with correct polarity, at the
`output. An additional summing input (Z)
`allows the output to be offset or compared
`with a reference (i.e., XY + Z).
`
`It has low output impedance and high input
`impedance (I 00 k.nll2 p F). With ± 5-volt
`supplies, its output range is ±2.0 V min into
`150 D. over the --40 to +85°C temperature
`range; and it can drive load resistance as low
`as 25 D.. It has a bandwidth of de to
`250 MHz (-3 dB), settles co within 0.1 % of
`FS in 20 ns, and has a rise/fall time of 2. 5 ns
`( l ns for small signals). Jes gain is flat to
`within -0.1 dB for frequencies up to
`15 MHz. Nonlinearity (relative error) is
`within ± 1 % of full scale and noise is only
`50 n V /../Hz at 10 MHz.
`
`Though its speed is state-of-the-arc, the
`AD835 is easy to use; and its differential
`inputs and post-multiplication Z input
`provide added versatility. Besides very fast
`multiplication, division, squaring, &
`frequency doubling, it also provides
`wideband modulation and demodulation,
`and is useful in phase detection &
`measurement, video gain control & keying,
`and voltage-controlled amplifiers & filters.
`
`Packages include 8-pin PDIP and SOIC.
`Cl
`Price (100s) is $8.95. Circle 20
`
`All brand or product names mentioned are trademarks or registered trademarks of their respective holders.
`
`Analog Dialogue 28-2 119941
`
`21
`
`Xilinx, Inc. and Xilinx Asia Pacific Pte. Ltd. Exhibit 1025 Page 21
`Xilinx, Inc. and Xilinx Asia Pacific Pte. Ltd. v. Analog Devices, Inc. IPR2020-01559
`
`
`
`New Product Briefs (For information use reply card or see back cover)
`Interfaces, Regulator, Temperature Sensor
`Vout Temp Sensor
`+5-V Linear Voltage Regulators
`AD22100: ~Oto +150°C
`ADM663 and ADM666 have existing 2nd sources
`On-chip signal conditioning Provide fixed +5 V or adjustable + 1.3 V to + 16 V
`The AD22100 is a monolithic 3-terminal
`The ADM663 and ADM666 are complete
`precision voltage regulators that provide a
`single-supply temperature sensor with on(cid:173)
`fixed +5-volt output.* With two additional
`chip signal conditioning and low-impedance
`voltage output proportional to the supply
`external resistors, they provide a range of
`precision voltages from + 1.3 to + I 6.0 V.
`(22.5 mVf'C with a +5-V supply). Its 200°
`Low quiescent current ( 12 µA max) makes
`span is useful in many applications,
`including
`automotive,
`computer,
`them especially suitable for battery-powered
`instrumentation, and HVAC.
`systems. They will operate over an input
`range of +2 V to + 16.5 V and can provide
`40 rnA of output current (more, if needed,
`using the AD663, via an external pass
`transistor). The AD6 63 also has a
`temperature-proportional output for use in
`powering liquid-crystal displays.
`The AD666 features battery-monitoring
`circuitry to detect low battery voltage. These
`devices are pin-compatible replacements for
`existing 663 and 666 regulators. Typical
`applications include battery-operated
`
`v,.
`
`r----------OVou,
`AOM666
`
`'---->---ov..,
`,....,..----ULBI
`
`L80
`
`equipment such as notebook PCs,pagers,
`etc. Both devices are available in 8- pin
`plastic DIPs and narrow surface-mount
`sore packages.Their operating temperature
`range is -40 to +85°C. Prices ( I 000s) start
`at Sl .40 for the ADM663 and S 1.65 for the
`Cl
`ADM666. Circle 23
`
`• 3-vo It vcrsio os are 3 lso avai I abl c. Ask about
`ADM6b3A and ADM666A.
`
`Its patented circuit measures and
`conditions the voltage developed across a
`tern peratu re-sensitive re sis tor with
`constant-current excitation. Its ratiometric,
`rail-to-rail output provides a cost-effective
`solution with an AID convener, using the
`ADC's reference as the AD22100's supply.
`Accuracy and linearity are bener than ±2%
`& ± 1 % of full scale. Housed in a plastic
`TO92 package, it offers 3 ranges of
`measurement, S: -50 to + l S0°C, A: -40 to
`+85°C, and K: 0 to +100°C. Price (]00s)
`Cl
`scares at SI. 21. Circle 22
`
`SIR Isolated Interface
`AD2S75: Synchro/Resolver(cid:173)
`to-system, X'former-coupled
`The AD2S75 is a universal transformer(cid:173)
`isolated input interface between synchros
`& resolvers and conversion & conditioning
`circuitry. The AD2S75 can be pin(cid:173)
`programmed to interface with all standard
`synchro- and resolver-format signals (90,
`26, and 11.8 V rms) and produces outputs
`in sin-cos resolver format at a standard 2 V
`rms level. le operates over a range of
`reference frequencies from 60 Hz to
`20 kHz.
`
`The AD2S75's miniature transformers
`provide true galvanic isolation for up to
`1000 volts de. Applications include military
`systems & equipment, avionics, factory
`automation. Power input range is ± 5 to ± 15
`V de. It is housed in a hermetically sealed
`24-pin, 0.9" DIP package and is available in
`industrial and military temperature ranges.
`Prices scare at S270 in I 00s. Circle 24 Cl
`
`RS-232 Dual Transceivers, C = 0.1 µF
`ADM2x2 series have exi



