Analog Switches and Multiplexers: Key Specs, Selection Guide, and Obsolete Part Sourcing

Analog switches and multiplexers are among the most quietly essential components in electronic design. They appear in data acquisition systems, medical instruments, audio routing, industrial automation, and telecommunications infrastructure — any application where a signal path needs to be opened, closed, or redirected under digital control. Despite that ubiquity, the specifications that actually govern their performance are often underestimated at the selection stage, and their procurement lifecycle is frequently left unmanaged until a shortage or end-of-life notice arrives.

This article serves as a practical reference for both angles. The first half works through the technical specifications that matter most when selecting or evaluating an analog switch or multiplexer — on-resistance, leakage, charge injection, off-isolation, crosstalk, and bandwidth. The second half addresses the procurement reality: many of the most widely designed-in parts from families like CD4051, DG408, and MAX4051 are decades old, and maintaining supply for fielded systems or long-lifecycle products requires deliberate planning well before a part reaches end-of-life status.

What Are Analog Switches and Multiplexers?

An analog switch is an electronically controlled switching element that connects or disconnects a signal path — passing analog voltages and waveforms in either direction with minimal distortion. Unlike a relay, which uses mechanical contacts, an analog switch operates entirely in silicon, offering microsecond or nanosecond switching speeds with no moving parts. A multiplexer (MUX) extends this concept into a multi-channel arrangement: it places several analog switches in an array so that one of multiple input signals can be routed to a shared output under the direction of a digital address bus.

The two components are related but not interchangeable. An analog switch controls a single signal path — one input, one output, digitally controlled on or off. An analog multiplexer consists of multiple switch cells and a decoded address circuit, allowing one-of-N selection. A demultiplexer (DEMUX) inverts that topology, distributing a single input to one of several outputs. Many ICs combine MUX and DEMUX functionality, and some devices are fully bidirectional, functioning as either depending on how the circuit is wired.

These components are classified under analog switches, multiplexers and demultiplexers within the broader interface and analog signal-chain product family. They are manufactured by virtually every major semiconductor vendor, including Texas Instruments, Analog Devices (which absorbed Maxim Integrated), Renesas, onsemi, Vishay Siliconix, and STMicroelectronics.

Analog Switch vs. Multiplexer: Understanding the Difference

The distinction matters most at the design stage. If an application requires one input per output path, a standalone analog switch is the right building block. If the need is to connect several inputs — say, multiple sensors — to a single instrumentation amplifier or ADC, a multiplexer provides the channel selection logic on-chip, reducing component count and simplifying the PCB layout. In practice, a multiplexer is essentially an array of analog switches paired with a binary decoder; the performance specifications of both device types are governed by the same underlying switch physics.

One practical implication of this relationship is that legacy designs sometimes use individual analog switches where a modern engineer might choose a fully integrated MUX, and vice versa. When maintaining or repairing fielded equipment, it is common to find either approach, and understanding the equivalence — and the differences in pinout, logic thresholds, and supply voltage — is essential for any substitution or cross-reference work.

How CMOS Analog Switches Work

Nearly all modern analog switches and multiplexers are built on CMOS technology. The core structure of a CMOS analog switch consists of a P-channel MOSFET and an N-channel MOSFET connected in parallel between the signal terminals. Connecting the two device types in parallel — a complementary pair — reduces the combined on-resistance and improves linearity, producing an on-resistance that varies less with signal voltage across the operating range. This is sometimes referred to as a bilateral switch, because signals can pass in either direction through the closed device.

The digital control signal drives the gates of the two transistors in opposite polarities: when the control input asserts the switch on, the NMOS gate is pulled high and the PMOS gate is pulled low, turning both devices into a conducting path. In the off state, both transistors are held non-conducting, and the resistance between the signal terminals rises to hundreds of megohms. The residual current that flows in the off state — the leakage current — is governed by the drain-to-source and drain-to-substrate junction characteristics of the MOSFETs rather than any intentional conduction.

Key Specifications Explained

Selecting the right analog switch or multiplexer for an application requires a careful look at several interdependent parameters. Optimizing for one specification frequently involves a trade-off with another, so the right choice depends on the signal characteristics and the error budget of the system as a whole.

On-Resistance (RON)

RON is the resistance measured between the drain and source terminals when the switch is in the on state. It is one of the most prominent specifications on a datasheet and has a direct effect on signal attenuation and gain error. When a switch with finite RON drives a resistive load, the RON and load resistance form a voltage divider, introducing a gain error proportional to the ratio RON/RLOAD. For precision circuits with high-impedance loads this error is negligible, but in low-impedance drive situations it can be significant. RON also varies with signal voltage — a non-flatness that can introduce distortion that cannot be calibrated away, unlike a fixed gain offset. Modern precision switches achieve RON values below 5 Ω; older CMOS families such as the CD4051 typically exhibit RON in the range of 80–200 Ω depending on supply voltage.

There is also a fundamental trade-off between RON and parasitic capacitance. Reducing RON generally requires a larger transistor geometry, which increases the associated junction capacitances and thereby reduces bandwidth. Designs that need both very low RON and high bandwidth require careful part selection and may need to accept a compromise point.

Leakage Current

Leakage current is the small current that flows through a channel when the switch is nominally off. In a multiplexer with a high-impedance source on an unselected channel, this current flows through the source impedance and creates a small offset voltage error at the output. For applications using operational amplifiers or ADCs that switch high-input-impedance sources, low leakage current is critical to maintaining low offset errors. Precision multiplexers from the current generation achieve leakage currents in the picoampere range at room temperature; the MAX4051A series, for example, specifies off-leakage of 0.1 nA at 25°C. It is worth noting that leakage increases significantly with temperature, which makes the worst-case figure at maximum operating temperature the relevant value for margin calculations.

Charge Injection

Charge injection is the transient charge transferred into the signal path whenever the switch transitions between on and off states. It arises from stray gate-to-drain capacitances in the NMOS and PMOS transistors that form the switch. When the control signal changes state, these capacitances couple a small charge pulse through to the signal node, producing a voltage glitch. In sample-and-hold and precision ADC front-end circuits, this glitch can degrade measurement accuracy, particularly at lower source impedances where the injected charge develops a larger voltage spike. Charge injection is specified in picocoulombs (pC); devices optimized for precision data acquisition aim for values well below 1 pC.

Off-Isolation and Crosstalk

Off-isolation describes how effectively the switch blocks signals on an open channel from reaching the output. It is expressed in decibels (dB) and is frequency-dependent: at 1 MHz, a typical switch achieves approximately 70 dB of off-isolation, but this degrades by 20 dB per decade as frequency rises. At 100 MHz the same device may offer only 30 dB of isolation. Channel-to-channel crosstalk is a related but distinct parameter — it measures the unwanted coupling from an active (on) channel into an adjacent inactive channel within the same IC package. Both parameters are particularly important in multiplexed audio, video, RF, and multi-sensor measurement systems where signal integrity across unused channels must be maintained.

Bandwidth

Bandwidth is the frequency at which a closed switch attenuates the signal by 3 dB relative to its DC value. It is determined primarily by the combination of RON and the parasitic capacitances of the switch, particularly the drain-to-source capacitance (CDS), the off-channel capacitance (COFF), and any external load capacitance. A lower RON and lower CDS both extend usable bandwidth, though as noted above, lower RON tends to come at the cost of higher capacitance. High-bandwidth devices aimed at video and wideband communication routing can achieve bandwidths exceeding 300 MHz; general-purpose precision multiplexers used in low-frequency data acquisition are typically rated from a few megahertz to tens of megahertz.

Supply Voltage and Signal Range

Supply voltage range and the maximum analog signal that a switch can handle are closely linked parameters. Most CMOS analog switches require that the signal voltage remain within the supply rails; if the signal exceeds the supply, the parasitic body diodes of the MOSFETs can conduct, clamping the signal and potentially injecting current into the substrate that affects adjacent channels. The CD4051 family, for instance, supports a single-supply range up to 20 V, but requires that the analog signal stay within the supply limits. The DG408 series (and similar Analog Devices parts) supports dual supplies up to ±20 V, which is more accommodating for bipolar analog signals common in industrial and audio applications. Fault-protected devices add over-voltage tolerance, allowing the signal to exceed the supply without causing damage or latch-up — an important feature in automotive and industrial input stages where transient over-voltages are common.

Common Configurations and Part Families

Analog switches and multiplexers are available in a wide range of channel configurations. Single-pole, single-throw (SPST) switches are the simplest building blocks. Single-pole, double-throw (SPDT) devices add a break-before-make or make-before-break channel-switching capability. Multiplexers come in standard configurations of 2:1, 4:1, 8:1, and 16:1, with the address bus width scaled accordingly (1, 2, 3, and 4 select lines respectively). Some devices combine multiple configurations on a single die — for example, a dual 4:1 MUX, or three independent SPDT switches, as seen in the CD4052 and CD4053 families.

Several part families have become deeply embedded in the industry over the decades and are worth knowing by name, especially for procurement and engineering teams working with legacy designs:

• CD4051 / CD4052 / CD4053 — The original CMOS multiplexer family, introduced in the 4000-series logic era. The CD4051 is an 8:1 single-channel MUX with three binary address inputs. Wide supply range (3–20 V single supply), bidirectional, and still actively manufactured by TI and others, though many second-source variants have been discontinued.
• DG408 / DG409 — Precision 8-channel and differential 4-channel MUXes from the original Siliconix (now Vishay Siliconix). Known for improved RON flatness and dual-supply capability. The DG408 remains popular in modular synthesizers and industrial data acquisition; some package variants are hard to find.
• MAX4051 / MAX4052 / MAX4053 — Low-voltage CMOS versions from Maxim Integrated (now Analog Devices). The A-suffix variants add tighter specifications for RON flatness and leakage. Designed for +2.0 V to +12 V single-supply or ±2.7 V to ±6 V dual-supply operation, making them well suited for battery-powered and mixed-supply applications.
• ADG408 / ADG409 / ADG528 — Analog Devices precision MUX family with low RON and low charge injection; widely used in data acquisition front ends. Check the current lifecycle status of each package variant against Analog Devices’ product pages before designing one in.
• HI-508 / HI-509 — Harris/Renesas 8-channel and 16-channel precision MUXes used extensively in industrial and military instrumentation. Check each variant’s current status against Renesas’ product change notices before relying on it.

Applications by Industry

Analog switches and multiplexers appear across virtually every electronics sector. In data acquisition and instrumentation, they allow a single high-quality ADC to sample signals from multiple sensors in sequence, reducing cost and board area. In medical electronics — ECG, EEG, blood analyzers — multiplexed front ends acquire physiological signals from multiple electrodes through a single precision amplifier chain, where the low leakage and charge injection requirements are especially strict. In audio and broadcast systems, analog switches route signals between mixing stages without the contact noise and mechanical wear of relay-based designs. Automotive body-control modules use MUXes for sensor scanning, relying on fault-protected variants that survive transient over-voltages on the 12 V bus. Industrial programmable logic controllers (PLCs) and process control equipment use them in analog I/O modules that scan multiple field inputs.

The military, aerospace, and space sectors present a particular sourcing challenge. These applications require components that meet enhanced screening, extended temperature range (often –55°C to +125°C), and radiation tolerance. Many MIL-grade analog switch variants were produced in small volumes by a narrow set of suppliers and are no longer in active production — yet the systems they were designed into have service lives measured in decades.

Obsolete and EOL Analog Switches: The Sourcing Challenge

Component obsolescence is a normal outcome of the semiconductor lifecycle. Manufacturers discontinue parts when the production economics no longer justify maintaining a process node or a specialized packaging line, when design wins slow, or when a product line is consolidated after an acquisition. For analog switches and multiplexers, obsolescence tends to hit in waves: a foundry transition makes an older process unavailable, or an acquirer rationalizes the combined portfolio and marks legacy part numbers as NRND or EOL. The functional devices these parts represent remain designed into equipment that may be in service for ten, twenty, or thirty more years.

The practical impact is that engineers maintaining legacy systems — or procurement teams managing long-lifecycle production programs — will periodically encounter situations where a required part number is past last-time-ship; purchasable stock can still exist across a number of supply channels, and locating it becomes a sourcing exercise rather than a catalog order. The challenge is compounded in the analog switch category by the fact that many of these devices are highly application-specific: on-resistance matching, charge injection, and leakage specifications may have been tightly characterized during the original design qualification, and an unvalidated substitute can introduce measurement errors or signal integrity degradation that is difficult to trace.

Counterfeit risk is a legitimate concern in this market segment. Parts that are no longer in active production attract re-marking and re-packaging from bad actors, particularly when demand pressure is high. What decides whether that stock is safe to buy is how the supplier manages quality, and the questions are the same whoever is quoting: documented inspection and testing protocols, quality management certifications, and membership in industry verification programs such as ERAI.

Finding Replacements and Alternatives

When a specific part number is no longer available, the options break down into three categories: direct form-fit-function replacements from a second source that manufactured the same device; functional equivalents with compatible pinout and electrical specifications but a different part number; and design-level alternatives that require circuit modification. The first path is always preferred from a risk standpoint but is not always possible. The second path requires a careful datasheet comparison against the original qualification parameters — RON, RON flatness, leakage, charge injection, and supply voltage range must all be within the original design margins.

For teams managing multiple obsolescence events across a bill of materials, the process of identifying and qualifying alternatives benefits from systematic tools and supplier relationships established before a shortage becomes critical. Proactive lifecycle monitoring — tracking PCN (Product Change Notice) and EOL notices from manufacturers — allows design-in of alternatives or buffer stocking while parts are still available at normal commercial quantities and pricing. Engaging supply chain support early, rather than reactively, is what separates manageable obsolescence from a production stoppage.

WIN SOURCE is a global independent distributor of electronic components, stocking analog switches and multiplexers across the families discussed in this article — current-production part numbers as well as parts that are hard to find — and supporting shortage sourcing, lifecycle management and alternative solutions for parts that cannot be directly sourced, with final approval of any substitute resting with the customer’s engineering team. For teams carrying surplus stock of discontinued analog switch parts, it also operates an excess inventory solution. It holds AS9120, ISO 9001:2015 and ISO 14001 certifications and is a member of ERAI.

Conclusion

Analog switches and multiplexers are deceptively simple components whose datasheets contain a dense set of interdependent specifications — RON, RON flatness, leakage current, charge injection, off-isolation, crosstalk, and bandwidth — each of which matters in different ways depending on the application. Getting the selection right at the design stage prevents signal integrity problems that are hard to diagnose later. Understanding these parameters is equally important when evaluating cross-reference alternatives for parts that have gone out of production.

Obsolescence is not a peripheral concern for this component family. Many of the most widely deployed analog switch and multiplexer ICs come from families that are thirty to forty years old, and fielded systems incorporating them will remain in service long after some variants are discontinued. Managing that lifecycle proactively — monitoring PCNs, qualifying alternatives early, and building supplier relationships before shortages occur — is the approach that keeps production lines and service programs running without disruption.

Frequently Asked Questions

What is the difference between an analog switch and an analog multiplexer?

An analog switch controls a single signal path — one input, one output, on or off under digital control. A multiplexer is an array of such switches plus a decoder, so one of N inputs can be routed to a shared output by a digital address; a demultiplexer does the reverse. Both are built on the same CMOS switch physics, so the same specifications govern them.

Which specifications matter most when selecting an analog switch?

On-resistance and its flatness (gain error and distortion), leakage current (offset error into high-impedance sources), charge injection (glitches into sample-and-hold and ADC nodes), off-isolation and crosstalk (both frequency-dependent), bandwidth, and the supply and signal voltage range. Lower on-resistance usually means more capacitance and less bandwidth, so the right part is a compromise set by the application’s error budget.

Why does off-isolation get worse at high frequency?

Because the open switch still has parasitic capacitance across it, which passes more signal as frequency rises. A typical device offering about 70 dB of isolation at 1 MHz loses roughly 20 dB per decade, leaving around 30 dB at 100 MHz — which matters in RF, video and multi-sensor systems.

Can I substitute a CD4051 with a modern multiplexer?

Often, but not blindly. Check the pinout, logic thresholds, supply range (the CD4051 runs from 3–20 V single supply), on-resistance (80–200 Ω for the CD4051 versus a few ohms for modern parts), leakage and charge injection against the original design’s margins, and confirm the signal stays within the new part’s rails. A lower on-resistance can change the behavior of a circuit that was designed around the higher value, so the substitute needs bench validation and your engineering team’s approval.

How should I plan for obsolete analog switch and multiplexer parts?

Monitor PCN and EOL notices for every part on the BOM, document the original qualification parameters so cross-references can be compared quickly, qualify alternatives early, and consider buffer stock for long-lifecycle programs. Once a part is past last-time-ship, purchasable stock can still exist across a number of supply channels — ask whoever is quoting about inspection protocols, certifications and traceability.

© 2026 Win Source Electronics. All rights reserved. This content is protected by copyright and may not be reproduced, distributed, transmitted, cached, or otherwise used without the prior written permission of Win Source Electronics.