Why We Need New Op Amps Despite Already Having So Many Good Ones

It’s hard to determine how many distinct operational amplifiers (op amps) are available from the dozens of manufacturers, but the number is in the multiple thousands. That’s not even counting those op amps that are variations on a base model but have slightly different specifications, temperature grades, or packaging.

These basic, humble, yet versatile analog building blocks provide critical functions in signal-processing analog front-ends (AFEs), filters, and sensor interfaces. The list of circuit functions needed has always been long, and is getting longer due to the demands of edge artificial intelligence (AI), smart controllers, and countless other applications that require systems to interface with the real world.

Given all the op amps already on the market, from mid-range, good-enough specifications to outstanding, high-performance precision devices, it’s tempting to assume there’s little need for new ones. However, recent introductions demonstrate yet again the naiveté of that assumption.

Latest op amp innovations offer extremely low drift

I have seen dozens of new op amps from major and mid-tier vendors in the past year. While some are highly application-specific, such as high-voltage designs or galvanic isolation, most are fairly standard yet excel in some specifications.

Consider these three recently introduced op amps featuring extremely low drift:

  • Analog Devices’ MAX74810ARMZ-RL, a chopper-stabilized, low-power, dual-channel, zero-drift op amp with low noise, ground-sensing inputs, and rail-to-rail outputs optimized for total accuracy over time, temperature, and voltage conditions
  • STMicroelectronics’ TSZ901IYLT, an AEC-Q100 qualified op amp (from the company’s TSZ line of ultra-precision op amps) that boasts near-zero drift and ultra-low offset, both critical attributes for high-performance sensor signal conditioning
  • Texas Instruments’ TLV4888PWR, a 36 volt, 14 megahertz (MHz) chopper-stabilized zero-drift, quad-channel, multiplexer-friendly CMOS precision op amp

The parallel vendor and designer dilemmas

Both vendors and users of these ubiquitous components have mixed feelings about older op amps. Vendors want to keep making the older parts with their high margins and predictable manufacturing and testing, while users want them for their known performance and subtleties. At the same time, both want the potential benefits of newer products, including immediate and longer-term lock-in (for vendors) and higher performance (for users).

Although op amps are functionally simple blocks, they have many important and often subtle parameters. While the required tests and subsequent data analysis are highly automated, they still take considerable time and effort to initiate, complete, and document.

Among the necessary but difficult-to-test specifications are input bias current (IB) versus temperature, as seen for the MAX74810ARMZ-RL (Figure 1).

Figure 1 : Shown is a graph of IB versus temperature for the MAX74810ARMZ-RL; these graphs are critical to designers implementing a precision AFE. (Image source: Analog Devices)

Op amp datasheets often plot input offset voltage (VIO) distribution across thousands of tested devices. These plots, such as the one for the TSZ901IYLT (Figure 2), assure users that the vendor has the production process under control, making simulations far more robust and providing confidence.

Figure 2 : Plots such as the VIO distribution for the TSZ901IYLT assure designers that the production process is tightly controlled. (Image source: STMicroelectronics)

Vendors must also call out maximum (or minimum) specifications for some parameters, as typical values are fine for first-level estimates but fall short when doing a full analysis using Spice or other tools. Op amp tables, such as the one for the TLV4888PWR (Figure 3), specify both typical and maximum values over the full temperature range to support a conscientious design evaluation. These are often provided in tabular form, including typical values for parameters such as offset voltage and input bias current.

Figure 3 : A conscientious design evaluation of a device such as the TLV4888PWR requires typical and maximum values; op amp vendors provide them in tabular form if appropriate. (Image source: Texas Instruments)

What’s in it for the vendor?

Is all this development effort and vendor expense worth it? Generally, yes. A successful analog product can be a viable, high-profit revenue source for many years. If a vendor puts together the right combination of functions, features, and performance specifications, and the part is matched to successful customer products, there’s a good chance the op amp will be used in both the current product generation and its successors. The result is that the older preferred part is displaced.

But supplanting an older part is not easy. Unlike processors, a good op amp is often chosen again for the circuit rather than being designed out and replaced. So why are designers reluctant to replace an older, potentially inferior op amp with a newer one?

The reason is that analog components are more prone to subtleties and idiosyncrasies of design, layout, and even manufacturing than their digital counterparts. Experienced analog circuit designers prefer not to change to a new part and its associated learning curve, unless there are compelling reasons to do so, while digital-centric designers don’t want to think about the analog aspects. Their thinking is more like, “If it works well enough, we’ll just leave it alone and move on.”

For the vendor and the designer, there are other long-term benefits:

  • The refined manufacturing and test processes greatly reduce concerns about production and availability.
  • Improved manufacturing expertise and higher yields lead to higher margins for the vendor.
  • The part is on the designer’s approved vendor and component list that many OEMs maintain, so there’s no corporate hesitation to putting it on the bill of materials (BOM).

An example that demonstrates this “don’t change” mindset is the Burr-Brown INA133 instrumentation amplifier (a special op amp topology). This device was introduced around 1998 and is still offered in a variety of packages and grades, such as the INA133UA/2K5 from Texas Instruments (which acquired Burr-Brown in 2000).

Of course, the dilemma with all these new op amps, supplemented by older ones, is how to select the best one for the application. Some designers start with a few preferred vendors, others with suggestions from their peers. Here’s a case where AI may help: you enter the must-have specs, the nice-to-have ones, and minimums/maximums for other parameters, and you get back a ranked list of suitable op amps.

That would be an interesting starting point, but you can’t beat documentation and one-on-one conversations with your vendor contact as you narrow down your choices to truly understand the nuances of these devices.

Conclusion

The continuous wave of new op amps, despite the thousands of very good versions already available, shows that there is always a need for better parts. These devices can be slightly improved in overall performance or be greatly improved in just one or two critical specifications. Either way, a successful new op amp can have a long life, minimizing headaches for the designer, while maximizing yield and margins for the vendor.

Related Content

1: A Guide to Op Amp Selection

https://www.digikey.com/en/blog/a-guide-to-op-amp-selection

2: Choose the Right Precision Op Amp to Simplify Analog Front-End Design

https://www.digikey.com/en/blog/choose-the-right-precision-op-amp

3: How to Choose and Use Precision Op Amps Effectively

https://www.digikey.com/en/articles/how-to-choose-and-use-precision-op-amps-effectively

4: How to Use Zero-Drift Op-amps to Achieve Precise, Accurate, Low-Power Industrial System Control

https://www.digikey.com/en/articles/how-to-use-zero-drift-op-amps-to-achieve-industrial-system-control

5: How to Achieve Both DC Precision and Wide Bandwidth Using Zero-Drift Amplifiers

https://www.digikey.com/en/articles/how-to-achieve-both-dc-precision-and-wide-bandwidth-using-zero-drift-amplifiers

Achtergrondinformatie over deze auteur

Image of Bill Schweber

Bill Schweber is een elektronisch ingenieur die drie boeken over elektronische communicatiesystemen heeft geschreven, alsmede honderden technische artikelen, opiniërende columns en productkenmerken. In voorgaande functies was hij werkzaam als technisch websitemanager voor meerdere onderwerpspecifieke sites van EE Times, alsmede de hoofd- en analoge redacteur van EDN.

Bij Analog Devices, Inc. (een leidende verkoper van geïntegreerde schakelingen met analoog en gemengd-signaal) hield Bill zich bezig met marketingcommunicatie (public relations). Hierdoor heeft hij beide kanten van een technische pr-functie ervaren, door het introduceren van bedrijfsproducten, verhalen en berichten aan de media en ook als ontvanger daarvan.

Voorafgaand aan zijn marketing- en communicatierol bij Analog was Bill meewerkend redacteur van hun gerespecteerde technische tijdschrift en ook werkzaam op hun afdelingen voor productmarketing en applicatie-engineering. Daarvoor was Bill actief bij Instron Corp. door het werken aan analoge en geïntegreerde schakelingen en de systeemintegratie van materiaaltestende machinebesturingen.

Hij houdt een MSEE (Univ. van Mass.) en BSEE (Columbia-universiteit), staat geregistreerd als professioneel ingenieur en heeft een geavanceerde licentie voor amateurradio. Bill heeft ook online cursussen over vele technische onderwerpen georganiseerd, geschreven en gepresenteerd, waaronder MOSFET-basics, ADC-selectie en led-schakelingen.

More posts by Bill Schweber
 TechForum

Have questions or comments? Continue the conversation on TechForum, DigiKey's online community and technical resource.

Visit TechForum