Existing equipment may need fresh testing even when replacement components appear pin compatible.
For decades, engineers treated the NE5532 as one of audio’s most dependable building blocks. They knew its limits, designed equipment around its behavior, and ordered replacements under the same part number.
Scott Dorsey, Senior Technical Editor at Positive Feedback, discovered that this trust no longer held when newly supplied TI chips caused production boards to fail quality-assurance tests. The name on the package had stayed the same. But the component behind it had changed enough to affect designs that had already been tested and approved.
Here’s how the redesign was uncovered, what TI confirmed, and why existing equipment may need to be checked again.
What Changed in the Revised NE5532 Specifications
Revision K changed several specifications engineers had relied on when treating the NE5532 as a drop-in part. The most important changes were:
- Maximum supply voltage fell from ±22V to ±18V. Professional audio equipment commonly uses ±17V or ±18V rails, as EEVblog forum member Yansi noted. Those designs now have little or no margin below the revised part’s absolute maximum rating and may require requalification.
- Slew rate fell from 9V/µs to 5V/µs. That is a 44% reduction in how quickly the output can change. The lower figure reduces the part’s margin when reproducing fast, high-level signals. Unity-gain bandwidth increased from 10MHz to 12MHz, but that does not restore the lost slew-rate or voltage margin.
- The original input-protection diodes disappeared. TI removed the anti-parallel diodes that had protected the NE5532’s inputs against excessive differential voltage.
- ESD tolerance was cut in half. The Human Body Model rating fell from 2,000V to 1,000V, making the revised part less resistant to electrostatic discharge during handling and assembly.
But beyond the spec sheet, TI also replaced the original PNP input stage with an NPN design, showing that the internal circuit itself had been redesigned.
Even worse, Revision K made the redesign harder to evaluate by replacing the full internal schematic with a minimal block diagram. Plus, it also removed published specifications for output impedance, crosstalk attenuation, and 600-ohm drive capability.
And while those omissions do not prove that the revised part lacks those capabilities, they do mean engineers no longer have the same published guarantees when deciding whether it will behave like the original NE5532 in an existing circuit.
It’s Not Even an NE5532 Anymore
The changes documented in Revision K led engineers to question whether TI was still manufacturing the original NE5532 design or had substituted another chip under the same part number.
That suspicion was later tested directly on TI’s E2E support forum. Asked whether the NE5532, LM833, MC33078, and RC4580 had been consolidated onto a single die, TI confirmed that they had.
An EEVblog forum user known as “The Chump” also reported receiving private confirmation from TI that the revised NE5532 contained an RC4580 die.
These confirmations supported an earlier conclusion from Rich Cabot, co-founder of audio-test-equipment manufacturer Audio Precision.
“It’s not a 5532 on a new process or different fab… they just relabeled a 4580,” Cabot wrote.
Cabot’s identification fit the changes already documented in Revision K. The thing is, RC4580 uses a different internal design from the original NE5532, including NPN inputs and a complementary output stage.
All this matters because engineers had designed and qualified circuits around the original NE5532’s behavior, not merely its pin layout and part number.
Replacing its internal topology can change how the part responds to input conditions, drives loads, and behaves near its operating limits. So even when the revised chip fits the same socket, it can no longer be assumed to perform identically in existing designs.
Quality-Assurance Failures
TI’s Product Change Notification (PCN) stated:
However, this assessment is difficult to reconcile with the revised specifications and internal design.
More importantly, Scott Dorsey reported that boards fitted with the new parts began failing quality-assurance tests.
“I came back severely jet-lagged to discover that in my absence boards had been failing quality assurance tests because Texas Instruments had replaced their NE5532 opamps with a completely different part with different characteristics being sold under the same name.” Dorsey wrote in Positive Feedback.
“Without informing customers, without any warning at all or any opportunity to get a last-time buy of the correct and working part.”
Unfortunately, the issue went beyond updating an old component. Engineers had designed and qualified products around the original NE5532’s published limits and behavior. And because TI retained the same part number, purchasing teams could receive the revised version without realizing that existing designs might need to be tested again.
Industries with strict component-qualification requirements face even higher stakes, though. For instance, a retired German automotive component manager identified as Dr. Frank argued that a redesign of this scale would normally require a new approval process rather than a routine production-change notice. He estimated that requalification could cost from EUR 50,000 to nearly EUR 1 million per affected project.
Same Pattern, Different Part Numbers
The NE5532 was not the only long-running TI component to change substantially while retaining a familiar name.
After December 2024, the OPA134 lost its offset-trim function, removing a feature some engineers used for precision calibration. TI also moved the TL07x family from JFET to CMOS technology. And according to EEVblog forum member santiall, the revised parts carried a much higher published noise figure of 37nV/√Hz instead of the previous 15nV/√Hz.
These cases differ in their details, but they create the same practical problem, as engineers often select legacy components for characteristics that extend well beyond their pin layout.
A circuit may depend on a trim function, a particular noise level, an input topology, or a known operating limit. Keeping the old part number can therefore hide changes significant enough to require new testing.
Aside from this, TI also had a strong manufacturing incentive to consolidate older designs. The company is closing its remaining 150mm wafer fabs in Dallas and Sherman, Texas, while shifting production toward newer 300mm facilities.
That’s why Mithat Konar of Audio by Van Alstine suggested that the older NE5532 production line may no longer have been profitable or practical to maintain.
Basically, moving legacy parts onto newer production lines is understandable. The controversy comes from preserving the old names without making the redesigned parts easy to distinguish from the components engineers had already qualified.
The Part Number That Proves the Point
TI had already shown how a legacy component could be redesigned without leaving engineers to guess what had changed.
When it replaced the LM317M voltage regulator, the redesigned version became the LM317MQ. Engineers could immediately tell that it was not the same part, and the datasheet included separate performance curves for the legacy and replacement designs.
The NE5532 transition gave them no such distinction. The part number stayed the same even though the revised version had different operating limits, protection features, and internal circuitry.
That leaves engineers with a practical choice. Designs already built around the legacy NE5532 may need to be tested again, especially if they run near ±18V, rely on the original input protection, or depend on specifications TI no longer publishes.
For projects that still require the older limits, manufacturers such as onsemi continue to offer NE5532 versions rated to the original ±22V maximum supply voltage.
The point is not that TI had to keep manufacturing the old design forever. It is that the company already knew how to retire or replace a legacy part without making the new silicon look interchangeable with the old one.