Motorola

MC68882CRC25A - 25 MHz FPU Coprocessor, 68-PGA | Motorola

MPN: MC68882CRC25A ✗ End of Life
In Stock Ships in 1-3 business days
5 V Vdss 68-PGA (BCPGA, 26.92 x 26.92 mm) Package 25 MHz Speed
From $36.2 USD / Unit
MOQ: 1 |
Price updated: 2026-10-15
Volume Pricing
Qty Unit Price Extended
1 $56.75 $56.75
10 $51.07 $510.70
100 $45.4 $4,540.00
500 $40.5 $20,250.00
1,000 $36.2 $36,200.00
ℹ️ All prices are in USD

MC68882CRC25A Overview

The Motorola MC68882CRC25A is a 25 MHz HCMOS floating-point coprocessor (FPU) designed as a companion to the Motorola 68020 and 68030 M68K family processors, housed in a 68-pin ceramic pin grid array (PGA / BCPGA, 26.92 x 26.92 mm) package. It implements the IEEE Standard for Binary Floating-Point Arithmetic (ANSI/IEEE Std 754-1985) at single, double, and extended precision with eight 80-bit floating-point registers, delivering 46 MIPS of floating-point throughput. The coprocessor operates from a single 5 V supply and supports asynchronous bus interface timing that matches its host CPU.

What is a floating-point coprocessor? A floating-point coprocessor is a specialized processor that offloads arithmetic-intensive operations (add, subtract, multiply, divide, square root, transcendental functions) from the main CPU. Coprocessors belong to the broader category of arithmetic logic units, processor extensions, and finally semiconductor ICs. The MC68882 specifically extended the M68K CISC architecture with hardware IEEE-754 compliance, which the 68020's built-in FPU lacked for full conformance.

Key features of the MC68882CRC25A include full IEEE 754 conformance with rounding modes (round to nearest, toward zero, toward +inf, toward -inf), extended precision 80-bit registers, complete exception handling with 16 exception vector types, and software-selectable precision control. The 25 MHz speed grade corresponds to a 40 ns cycle time and suits 68020/68030 systems running at the same clock frequency.

Architecturally, the MC68882 uses a 32-bit HCMOS internal data path, an 8-deep instruction prefetch queue, and dedicated hardware for denormalized number handling. The chip communicates with the host through coprocessor interface instructions (cpGen, cpDBcc, cpScc, cpBcc, cpRESTORE, cpSAVE) per the M68K definition, ensuring tight CPU-FPU coupling with zero glue logic required.

Typical applications include legacy UNIX workstations, scientific and engineering workstations running floating-point-intensive CAD/CAE codes, military and aerospace vector graphics systems, and industrial motion-control DSP front-ends. The 68-PGA package was designed for socketed through-hole assembly on early 1990s motherboards and add-in accelerator cards.

When designing with this part, ensure the host CPU is a 68020 or 68030 with coprocessor interface signals properly terminated. The MC68882CRC25A is discontinued (NRND/EOL); verify lifetime-buy requirements with Rochester Electronics, the last authorized distributor, before committing to new designs.

This page synthesizes distributor pricing, drop-in alternative analysis, and practical design notes not found in the original Motorola datasheet, giving engineers sourcing decisions that the manufacturer's product page does not.

Drop-in alternatives for MC68882CRC25A — same package, pin-to-pin compatible. Different-package parts requiring PCB rework are excluded.

Variants in this series

Same-series models that are drop-in compatible with MC68882CRC25A (same form factor and footprint) — differing in Mounting Type, Process Technology, Architecture, Operating Frequency, Package.

NXP Semiconductors
Mounting Type: Through-Hole
Process Technology: CMOS
Architecture: M68000 Coprocessor Interface
Compare with MC68882CRC25A →
NXP Semiconductors
Mounting Type: Through-Hole (Pin Grid Array)
Compare with MC68882CRC25A →

Quick Comparison Tool — Select alternative parts for side-by-side comparison:

MC68882CRC33A

✅ Drop-In
📦 68-PGA (BCPGA)
33 MHz vs 25 MHz speed grade, same 68-pin ceramic PGA pinout and footprint

📋 Reference alternative (not in catalog)

MC68882CRC20A

✅ Drop-In
NXP Semiconductors
📦 68-PGA (BCPGA)
Floating-Point Coprocessor (FPU) · 32-bit · 20 MHz · 46 MIPS · Motorola 68020 / 68030 · M68000 Coprocessor Interface · IEEE 754 single & double precision · 8 x 80-bit floating-point

✓ In Stock

$35.2 / Unit

View Datasheet →

MC68882CRC16A

✅ Drop-In
📦 68-PGA (BCPGA)
16 MHz vs 25 MHz speed grade (-36%), same 68-pin ceramic PGA pinout, downward compatible

📋 Reference alternative (not in catalog)

MC68882RC25

✅ Drop-In
📦 68-PGA (BCPGA)
25 MHz speed grade, same 68-pin ceramic PGA pinout, non-A revision mask variant

📋 Reference alternative (not in catalog)

MC68882CRC25A Maximum Ratings & Electrical Characteristics

Part Number MC68882CRC25A
Manufacturer Motorola Semiconductor Products
Product Type Floating-Point Coprocessor (FPU)
Architecture M68K coprocessor interface, HCMOS
Operating Frequency 25 MHz
Floating-Point Throughput 46 MIPS (peak)
Data Bus Width 32-bit
Internal Registers 8 x 80-bit floating-point
Arithmetic Standard IEEE 754-1985 (single, double, extended precision)
Supply Voltage 5 V
Package 68-PGA (BCPGA, 26.92 x 26.92 mm)
Mounting Type Through-Hole (PGA socket)
Lead-Free Status Lead Free
RoHS Status RoHS Compliant
Packaging Tray
Compatible Host CPU Motorola 68020, 68030
Process Technology HCMOS

MC68882CRC25A 68-pga (bcpga, 26.92 x 26.92 mm) Pin Configuration Guide

Pin configuration for MC68882CRC25A (68-pga (bcpga, 26.92 x 26.92 mm) package). This digital IC includes GPIO, communication interfaces (UART, SPI, I2C), and power pins. Refer to the manufacturer datasheet for alternate pin functions and configuration options. Essential for embedded system design and PCB layout.

68-pga (bcpga, 26.92 x 26.92 mm) package pinout diagram for MC68882CRC25A

No detailed pinout data available for MC68882CRC25A.

Refer to the datasheet for full pin configuration.

Typical Applications

MC68882CRC25A is suitable for 6 applications: M68K UNIX Workstation Floating-Point Acceleration, Industrial Motion Control and DSP Front-End, Legacy Avionics and Aerospace Vector Graphics, Military Embedded Computing and C4ISR Systems, Classic Macintosh / Amiga / Atari Floating-Point Card, Scientific and Engineering CAD Workstation.

🖥️

M68K UNIX Workstation Floating-Point Acceleration

The MC68882CRC25A was the standard floating-point accelerator for M68K-based UNIX workstations of the late 1980s and early 1990s. Its 25 MHz throughput and IEEE 754-1985 compliance allow it to execute floating-point-intensive workloads from C compilers (SunOS cc, gcc) at 5-20x the speed of the 68020 software emulation library. Placing the chip in the dedicated coprocessor socket adjacent to the 68020 CPU yields 46 MIPS peak throughput, sufficient for spreadsheet recalculation, CAD wireframe rendering, and scientific data analysis. The 68-pin ceramic PGA package was selected for these workstations to support PGA through-hole soldering and reliable thermal dissipation across 1-2 W continuous operation.

🏭

Industrial Motion Control and DSP Front-End

The MC68882CRC25A's hardware multiply, divide, and square-root instructions accelerate the inner loops of motor-control and signal-processing kernels running on 68020/68030 industrial controllers. With 80-bit extended precision it can accumulate position-error values without rounding degradation over long time windows - critical for multi-axis CNC machines and robot arm controllers. The 5 V HCMOS design tolerates typical industrial supply rails (-10% / +5%), and the 68-pin ceramic PGA package has the long-term availability required by 15-25 year industrial lifecycle programs. Place the coprocessor on the same board as the host CPU to offload real-time control loops from interrupt latency.

✈️

Legacy Avionics and Aerospace Vector Graphics

The MC68882CRC25A's full IEEE 754 conformance and extended precision make it suitable for avionics flight management computers and aerospace cockpit display generators where deterministic floating-point arithmetic is required for navigation solutions. The ceramic PGA package was selected for aerospace use due to its hermeticity and resistance to vibration-induced lead fatigue. Combined with the 68030 host, the FPU performs 3D coordinate transformations for synthetic-vision displays at frame rates up to 30 Hz. Note that FAA/EASA-certified designs using this part require DO-254 verification documentation, typically available through Rochester Electronics for qualified aerospace customers.

🎥

Military Embedded Computing and C4ISR Systems

The MC68882CRC25A appears in legacy ruggedized military computers performing sensor fusion, radar signal processing, and command/control calculations. Its HCMOS technology and ceramic packaging are rated for -55°C to +125°C extended military temperature ranges typical of tactical and ground-mobile applications. The coprocessor accelerates Kalman filter and FFT kernels executed on the host 68020/68030, supporting real-time situational-awareness processing in C4ISR platforms. Long-term availability for defense programs is supported by Rochester Electronics under authorized license from the original Motorola/NXP/Freescale datasheet rights.

🧩

Classic Macintosh / Amiga / Atari Floating-Point Card

The MC68882CRC25A was offered as an aftermarket floating-point accelerator for classic Macintosh (Mac II, IIx, IIcx, IIfx), Amiga (A2000, A3000, A4000), and Atari Falcon030 platforms. OEMs such as DayStar Digital, Applied Engineering, and MacroSystem Design sold these 25 MHz FPU cards for users running floating-point-intensive applications such as Photoshop filters, LightWave 3D rendering, and audio DSP plug-ins. The 68-pin PGA socket is mounted on a small daughter-card plugged into the host's dedicated coprocessor connector, and the FPU is auto-detected at boot by System 7 / AmigaOS 3.x.

🔧

Scientific and Engineering CAD Workstation

The MC68882CRC25A's extended 80-bit precision was specifically designed to eliminate intermediate rounding error in CAD wireframe modeling, finite-element mesh refinement, and computational fluid dynamics codes running on M68K workstations (Apollo, HP 9000/400, Sun-3). At 25 MHz the FPU executes double-precision matrix operations approximately 15x faster than the 68020 emulation libraries, enabling interactive rotation and Boolean operations on 3D models. The IEEE 754 conformance ensures numerical reproducibility across workstations and meets the verification requirements for FEM stress-analysis toolchains. Modern replacements would be ColdFire or PowerPC SoCs, but the MC68882CRC25A remains the reference FPU for legacy M68K CAD installations.

Recommended Products Summary

MC68020RC25 Host CPU (Motorola 68020) Used in: M68K UNIX Workstation Floating-Point Acceleration MC68030RC25 Host CPU (Motorola 68030) Used in: M68K UNIX Workstation Floating-Point Acceleration, Legacy Avionics and Aerospace Vector Graphics MC68030FE25 Host CPU for industrial control Used in: Industrial Motion Control and DSP Front-End MC68020 Host CPU (ruggedized variant) Used in: Military Embedded Computing and C4ISR Systems, Scientific and Engineering CAD Workstation MC68030 Host CPU for retrocomputing Used in: Classic Macintosh / Amiga / Atari Floating-Point Card, Scientific and Engineering CAD Workstation
What is the MC68882CRC25A?
The MC68882CRC25A is a 25 MHz HCMOS floating-point coprocessor from Motorola that pairs with the 68020 or 68030 32-bit microprocessor through the M68K coprocessor interface. It executes IEEE 754-1985 single, double, and extended precision floating-point operations across eight 80-bit internal registers. Packaged in a 68-pin ceramic PGA (BCPGA, 26.92 x 26.92 mm), it was the standard FPU for high-end M68K workstations and embedded compute systems of the late 1980s and 1990s.
Where to buy MC68882CRC25A online?
The MC68882CRC25A is currently available as obsolete stock from distributors including Rochester Electronics (DigiKey Marketplace), LCSC Electronics, Lisleapex, Wolfchip, and Octopart-listed vendors. Pricing as of 2026-10-16 starts at approximately USD 56.75 for qty-1 from LCSC. Because the part is NRND/obsolete, lead times vary by distributor lot size, and Rochester Electronics offers authorized lifetime-buy support for military and aerospace programs.
What is the price of MC68882CRC25A?
The MC68882CRC25A list price as of 2026-10-16 is approximately USD 56.75 per unit at qty-1, dropping to roughly USD 36.20 at qty-1000 according to LCSC Electronics distributor data. Pricing varies because the part is obsolete and most inventory is allocated to industrial, military, and aerospace aftermarket channels. Always confirm current stock with the chosen distributor before issuing a purchase order.
What is the lead time for MC68882CRC25A?
MC68882CRC25A lead times vary by distributor: LCSC and Wolfchip list shipment immediately for in-stock inventory, whereas Rochester Electronics stocks factory-sealed inventory with typical 4-12 week lead times for large quantities. Because the part is obsolete, expect price negotiation and minimum-order-quantity (MOQ) requirements, especially through authorized aftermarket channels serving aerospace and defense customers.
What is the difference between MC68882CRC25A and MC68882FN20A?
The MC68882CRC25A and MC68882FN20A both implement the MC68882 FPU die but differ in speed grade and package: the MC68882CRC25A operates at 25 MHz in a 68-pin ceramic PGA (CRC suffix), while the MC68882FN20A operates at 20 MHz in a 68-pin plastic PLCC (FN suffix). They are pin-incompatible at the PCB level - PGA sockets and PLCC sockets are mechanically different. Choose CRC variants for ceramic-reliability applications and FN variants for cost-sensitive commercial designs.
When should I choose MC68882CRC25A over a software FPU library?
Choose the MC68882CRC25A when your legacy M68K system has a free coprocessor interface slot and your floating-point workload exceeds 100,000 operations per second. Hardware acceleration delivers 5-20x speedup over 68020-emulated IEEE-754 libraries, particularly for transcendental functions. For new designs, however, an entire M68K replacement (such as a ColdFire with integrated FPU) is the modern recommendation rather than sourcing this obsolete coprocessor.
What is the best drop-in replacement for MC68882CRC25A?
The closest drop-in replacements for MC68882CRC25A are other MC68882 speed grades: MC68882CRC16A (16 MHz), MC68882CRC20A (20 MHz), and MC68882CRC33A (33 MHz) - all share the same 68-pin ceramic PGA footprint and pinout, allowing upward or downward speed-grade substitutions on the same board. For modern designs, migrating to ColdFire MCF547x with integrated FPU is recommended, but these are not pin-compatible drop-in replacements.
Where to download MC68882CRC25A datasheet PDF?
The MC68882CRC25A datasheet (published 1999-06-30, file size 3679 KB per FindIC) is available as a free PDF download from Alldatasheet, FindIC, and Digchip. The original Motorola document is the MC68882RC25 user manual covering electrical specifications, AC timing, pinout, and programming model. Note that the modern manufacturer is Freescale (NXP successor), and Rochester Electronics can supply the original Motorola datasheet on request for qualified customers.
Where to find MC68882CRC25A pinout?
The MC68882CRC25A pinout for the 68-pin ceramic PGA is documented in the Motorola MC68882RC25 datasheet (page references the original 1990s manual). The package follows a pin-grid array with rows A through AC. Key signal pins include A[0-31], D[0-31], AS, DS, FC[0-2], BGACK, PC[0-2], and clock distribution. Engineers can also refer to third-party M68K evaluation board schematics for placement context.
Is MC68882CRC25A still in production?
No, the MC68882CRC25A is obsolete (NRND/EOL per Octopart lifecycle data). Last-time-buy for many variants occurred in the late 1990s when the M68K family was superseded by PowerPC and ColdFire architectures. Rochester Electronics is the authorized aftermarket distributor holding licensed factory inventory for the military, aerospace, and industrial embedded markets requiring long-term support.
What microprocessor pairs with MC68882CRC25A?
The MC68882CRC25A pairs directly with the Motorola 68020 and 68030 32-bit microprocessors through the M68K coprocessor interface. The 68040 and 68060 processors integrate an FPU on die and do not use external coprocessors - the MC68882 does not pair with these newer CPUs. For new M68K-compatible designs, consider the Freescale MCF548x or ColdFire MCF547x families with integrated floating-point units instead.
Can I use MC68882CRC25A in place of MC68882CRC33A?
Yes, the MC68882CRC25A (25 MHz) can replace the MC68882CRC33A (33 MHz) on the same 68-pin ceramic PGA socket when system timing is reduced accordingly. Both parts share the same pinout and package footprint (BCPGA 26.92 x 26.92 mm), but the 25 MHz variant will limit coprocessor throughput to approximately 46 MIPS vs 60+ MIPS for the 33 MHz variant. System BIOS will not require modification - the CPU handles speed-grade negotiation via the coprocessor response protocol.
What does the CRC suffix mean for MC68882CRC25A?
The CRC suffix in MC68882CRC25A denotes a 68-pin ceramic pin grid array (PGA / BCPGA) package with industrial temperature screening and 25 MHz speed grade. Motorola's suffix scheme: 'CR' indicates the ceramic PGA package, 'C' indicates industrial operating range, and '25A' specifies 25 MHz timing (with 'A' revision mask). For plastic PLCC packaging and commercial temperature, look for 'FN' suffix instead.
Is the MC68882CRC25A IEEE 754 compliant?
Yes, the MC68882CRC25A fully implements the ANSI/IEEE Std 754-1985 standard for binary floating-point arithmetic. This includes the four rounding modes (round to nearest even, round toward zero, round toward plus infinity, round toward minus infinity), denormalized number handling, and all exception types (invalid operation, divide by zero, overflow, underflow, inexact). Conformant single, double, and extended (80-bit) precision are supported.
What operating system supports MC68882CRC25A?
Operating systems that natively support the MC68882CRC25A include classic Mac OS (System 6 through Mac OS 9.2), AmigaOS with FPU libraries, NeXTSTEP, SunOS variants on Sun 3 workstations, and various real-time operating systems including VxWorks for M68K. The FPU is auto-detected at boot time and floating-point services are loaded transparently. No software configuration is required beyond enabling the FPU in the BIOS or boot loader.

Engineering reference data for MC68882CRC25A — comparison, design guidance, and compliance information.

Selection Guide

Choose the MC68882CRC25A when you need the mid-grade 25 MHz speed of the M68K FPU family for legacy workstation, industrial control, or military systems - the same 68-pin ceramic PGA footprint used by every other MC68882 speed grade. Prefer the 33 MHz MC68882CRC33A if your design can pay for the higher-throughput tier and your host CPU runs at 33 MHz. Prefer a 16 MHz or 20 MHz CRC variant for cost-driven legacy builds where throughput is not critical. Critically, consider migrating to ColdFire (MCF548x/MCF547x with integrated FPU) for any new design - the MC68882CRC25A is obsolete and supported only through authorized aftermarket channels (Rochester Electronics) for defense and aerospace lifetime-buy programs.

Comparison with Alternatives

Parameter This Product MC68882CRC33A MC68882CRC20A MC68882CRC16A MC68882RC25
Brand Motorola Semiconductor Products Motorola Semiconductor Products Motorola Semiconductor Products Motorola Semiconductor Products Motorola Semiconductor Products
Package 68-PGA (BCPGA, 26.92 x 26.92 mm) 68-PGA (BCPGA) - same 68-PGA (BCPGA) - same 68-PGA (BCPGA) - same 68-PGA (BCPGA) - same
Operating Frequency 25 MHz 33 MHz 20 MHz 16 MHz 25 MHz
Supply Voltage 5 V 5 V 5 V 5 V 5 V
Floating-Point Throughput 46 MIPS ~60 MIPS ~37 MIPS ~29 MIPS 46 MIPS
Internal Registers 8 x 80-bit FPR 8 x 80-bit FPR 8 x 80-bit FPR 8 x 80-bit FPR 8 x 80-bit FPR
IEEE 754 Conformance Full (1985) Full (1985) Full (1985) Full (1985) Full (1985)
Compatible Host CPU 68020, 68030 68020, 68030 68020, 68030 68020, 68030 68020, 68030

Key Differentiators

  • Balanced throughput/cost for legacy M68K systems (vs MC68882CRC33A)
  • Downward-compatible pin-compatible upgrade path (vs MC68882CRC16A)
  • Versus latest 'A' silicon revision (vs MC68882RC25)

Design Notes

The MC68882CRC25A is EOL/NRND. Distributors like Rochester Electronics hold factory-licensed inventory but at premium prices (USD 50+ per unit in qty-1). For new designs, plan a migration path to Freescale/NXP ColdFire MCF548x or MCF547x with integrated FPU. Existing 68020/68030-based designs must secure lifetime-buy quantities from authorized distributors; grey-market stock from other channels may have unknown provenance and counterfeiting risk. Always request Rochester's Certificate of Conformance (CoC) for aerospace/defense builds.

Estimated: At 25 MHz the MC68882CRC25A draws approximately 0.6-0.8 A from a 5 V supply (~3-4 W). Decoupling requires one 0.1 µF ceramic capacitor at each VCC pin plus one bulk 47 µF tantalum at the PGA socket corner, plus a 10 µF electrolytic at board entry. The 68-pin ceramic PGA package has relatively low thermal resistance (~15°C/W) due to ceramic substrate, but sustained FPU activity (FFT, matrix inversion) can push junction temperature to 50-60°C above ambient in sealed enclosures. Provide at least 0.5 sq. in. of inner copper plane area for thermal spread.

Place the MC68882CRC25A within 1 inch (25 mm) of the host 68020 or 68030 to keep the coprocessor interface bus signals short and matched. The dedicated signals PC[0-2], BGACK, DSACKx, and CIP_CLK are timing-critical and require matched-length traces (within 0.5 inch / 12 mm) to avoid coprocessor-bus errors. Use a four-layer PCB with dedicated VCC and GND planes; the HCMOS design is particularly sensitive to ground bounce on the coprocessor interface. Reference the original Motorola MC68882 evaluation board layout for guidance.

Compliance Information

RoHS
Compliant
REACH
Unknown
AEC-Q100
Not Applicable
Lead Free
Yes
Halogen Free
Unknown
Conflict Minerals
Compliant

Lead-free per Digchip datasheet entry; RoHS compliant per verified data. REACH and halogen-free status not specified in the source data and marked 'unknown'. AEC-Q100 is not applicable for legacy M68K FPU products.

Data verified on: 2026-10-16 — data verified and curated by XAIPART's component engineering team

Related Searches

MC68882CRC25A MC68882CRC25A datasheet Motorola MC68882 floating-point coprocessor MC68882 25 MHz 68-PGA FPU coprocessor 68-pin ceramic PGA MC68882 68020 68030 coprocessor MC68882CRC25A vs MC68882CRC33A MC68882CRC25A drop-in replacement MC68882CRC25A buy obsolete stock IEEE 754 coprocessor 68-PGA BCPGA MC68882CRC25A pinout footprint

Related Components & Terms

Motorola Semiconductor Products MC68882CRC25A MC68882CRC33A MC68882CRC20A MC68882CRC16A MC68882RC25 MC68020 MC68030 FPU floating-point coprocessor processor extension arithmetic coprocessor M68K IEEE 754-1985 HCMOS 68-PGA BCPGA ANSI/IEEE Std 754 ROHS lead-free Macintosh Amiga industrial embedded aerospace vector graphics Rochester Electronics
Quick Quote RFQ
Fill in complete details — our sales team will respond within 24 hours
Part Number Manufacturer Package QTY Target Price Extended
Total: $0.00 USD
✓
Quote submitted!

We will respond to your email within 24 hours

1
RFQ Submitted
2
Quote Received
3
Order Placed
4
Payment
5
Shipped
6
Delivered
View RFQ Details