Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Operational amplifier state of the art.

Status
Not open for further replies.

schmitt trigger

Advanced Member level 5
Joined
Apr 17, 2013
Messages
2,725
Helped
859
Reputation
1,722
Reaction score
850
Trophy points
1,393
Activity points
18,445
This is more a question along the lines of curiosity, and to have a friendly chat with fellow engineers, nothing else.

I come a looong way, over 40 years, using monolithic opamps. I first started with externally compensated, unprotected output, limited CMR devices like the uA709. I remember when I first used the uA741: internally compensated, short circuit proof, an engineering marvel! And nowadays it looks soooooooooo antiquated.
Back then state of the art meant a fully bipolar device.

There were thereafter a major technology leap, with incredibly improved characteristics. These used mixed process devices, either BiCMOS or BiJFet processes. The first exemplified with the CA3140 and the latter with the TL07x series.

I know that nowadays there are many, full CMOS opamps. Some of them with quite impressive specs.
But, what is the state of the art? For a truly high performance opamp, are mixed process devices still required??

Unfortunately the datasheets no longer provide the amount of information that was available in the 70s and 80s, where even an internal device schematic, along with a detailed discussion was provided.
 

Hi,

I don´t know the state of the art of the production process.

Just speaking for the currently available OPAMPs and their specifications.
Maybe I should remark that I´m a very "mathematical" and "physical" designer of electronic circuits. I always try to find the best solution.
Therefore I have a very detailed specification in mind which OPAMP i´d like to use.
Then, when i feed the online selction guides with my specifications ,it is not seldom that there is not a single OPAMP that 100% meets my specifications.

But I always find one that is almost perfect.
--> Some are CMOS, some are bjt, some are BICMOS (I don´t select about process) . Each of them has it´s advantages and drawbacks.
Currently we have a huge count of different OPAMPs to choose from - for me this is good. But I see many users here in the forum that are frightened at this huge count.

Btw: I design electronic circuits for about 30 years now. I think I never used a 741 - it just didn´t fit to my specifications.

Klaus
 

High performance opamps are tailored to their application, with incredible amounts of variety in their characteristics, so there's no single state-of-art technology. Additionally manufacturers are copackaging opamps with other components to replace entire circuits (current sense amplifiers, differential amplifiers, programmable gain/filters, etc) with a single IC.

- - - Updated - - -

Btw: I design electronic circuits for about 30 years now. I think I never used a 741 - it just didn´t fit to my specifications.
In an undergraduate design course we were often explicitly told we had to use old parts like LM741, LM324, and TL082. Sneakily replacing a TL082 with a TLC082 could get you in hot water.
 

Recently doing design for an amplifier V/I feedback which was my first actually challenging analog design I kind of marveled at all the available devices and how they pretty completely span all the different trade-offs in terms of supply voltage, common mode voltage, output voltage, input offset currents, input offset voltage, bandwidth, distortion etc. If you want any one of these things you need to make trade-offs in the other ones.

At the beginning I'd focus on just a couple parameters and would find devices I thought were 'perfect'. Then I'd move farther along in the design and realize that by focusing on just a couple parameters I had inevitably arrived at something that was terrible in a couple others. So, for example my 'perfect' high bandwidth, low voltage offset amplifiers had crap input offset currents.

Prior to this design I had a fuzzy understanding of what the tradoeffs were and why there are so many opamps out there.

Though in the big picture it is remarkable how reasonably ideal opamps actually are. It's only for challenging problems that you need to do a deep dive into datasheets or look into expensive parts. Even the cheapest opamps are quite good when you consider it.
 

The OP product range has diversified at lot since the 70s and became partly confusing for a casual analog designer.

It's rather difficult now to identify state-of-the-art with a few outstanding OP types, instead you'll probably list twenty or more candidates, each with special advantages for a range of applications.

And you have OPs from the 70s that are good standard for many purposes like said TL07x/08x, low offset OP07, low noise OP27. Even class B output LM324 is still useful for control applications without particular linearity requirements.
 

Even class B output LM324 is still useful for control applications without particular linearity requirements.

Of course...since both cost and availability are also VERY IMPORTANT performance parameters.

When some digital designers come back to me with "Analog Electronics are DEAD", I only have to point them to look at the huge selection available from the major manufacturers.
Texas Instruments, all by itself, lists 1463 types, not counting comparators, current sense amps, video amps and other specialized options.
 

When some digital designers come back to me with "Analog Electronics are DEAD"
Unfortunately, digital signal processing is nothing without the right analog front end. And without analog power supply, it doesn't even blink an error LED.
 

It seems that analog performance has increased substantially as well and can still handle high precision high bandwidth applications digital can't touch or can only do so at a much higher cost. ADC prices spiral up in cost quickly beyond a certain point where cheap opamps can still operate just fine. So there is still incentive to keep processing in the analog world.

On the other hand large IC's continue to absorb more and more functionality in both the digital and analog domain.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top