Which Op Amps Do You Use?

Thread Starter

OBW0549

Joined Mar 2, 2015
3,566
I suspect that most of us, whether engineers or hobbyists (or both), have a fairly short list of "preferred" op amps that we tend to use again and again in our designs, selected for specific characteristics that make them good performers in one application or another. I've listed my stable of preferred parts below, arranged in roughly descending order of usage, and summarized their important (to me, anyway) characteristics.

What parts do you use, and why? I'd be especially interested in hearing about any that might be good additions to the list, or might be superior substitutes for one or more parts I currently use.

My list, including links to data sheets:

LMC6482: General-purpose dual/quad CMOS op amp with rail-to-rail inputs and output and EXTREMELY low input bias current (0.02 pA typ). Input offset voltage is 3 mV max with 1 uV/C stability. Operates on single supplies as low as 3 volts. Input voltage noise density is 37 nV/rt-Hz, so not suited to low-noise applications. Low input current makes it useful for interfacing high-impedance signal sources such as piezoelectric sensors. Key features: RRIO and low Ib.

TL071/2/4: General-purpose single/dual/quad JFET input op amp for routine work. Somewhat faster than a 741 (3 MHz GBW, 13 V/us SR). Inputs and output are not rail-to-rail, therefore limited usefulness in single-supply applications. Unsuitable for precision applications due to low open-loop voltage gain, high input offset voltage and offset voltage drift, and noise. Minimum supply voltage +/- 5V. Key feature: very low cost.

LT1490A: Bipolar dual micropower op amp with rail-to-rail inputs and output. 50 uA maximum supply current per amplifier. Operates on supply voltages from +/- 1.5V to +/- 22V. Has a unique input stage with an input common-mode range up to +44V relative to V-, regardless of supply voltage. Input noise and offset voltage drift make it somewhat unsuitable for precision work. Key features: RRIO and low power.

LT1793: JFET input op amp with good input specs: 250 uV Vos, 4 pA Ib, low input voltage noise (6 nV/rt-Hz typ.), low input current noise (0.8 fA/rt-Hz typ.), and low 1/f noise corner frequency make it good for interfacing with high-impedance sensors such as piezoelectric accelerometers. Inputs and outputs are not rail-to-rail and minimum total supply voltage is 10V, therefore limited usefulness in single-supply applications. Key features: low noise and high input impedance.

OP177G: Bipolar precision op amp with very low Vos (20 uV typ.) and Vos drift (0.7 uV/C typ.) Low noise (en = 10 nV/rt-Hz) and low 1/f noise corner frequency (2 Hz). Inputs and output are not rail-to-rail, therefore limited usefulness in single-supply applications. Somewhat slow (GBW = 600 kHz). Good for processing thermocouple signals. Key features: low input offset and offset drift.

NE5532: Bipolar dual low-noise op amp optimized for high-quality audio work. Input noise is 5 nV/rt-Hz. Fast (10 MHz, 9 V/us). Inputs and output are not rail-to-rail, therefore limited usefulness in single-supply applications. Minimum supply voltage is +/- 5 volts. High input bias current (800 nA max), low open-loop voltage gain (2200 min) and low input resistance (30 k min) make it unsuitable for precision applications. Key feature: low noise.

LM6171: Bipolar VERY high-speed op amp for applications working with MHz signals. GBW is 100 MHz, slew rate is 3600 V/us. Inputs and output are not rail-to-rail, therefore limited usefulness in single-supply applications. Minimum supply voltage is +/- 3 volts. High input bias current (3 uA max) makes it unsuitable for precision applications. Key feature: high speed.

MAX44246: Bipolar single/dual/quad precision chopper-stabilized op amp with 5 uV max. Vos and 20 nV/C max Vos drift. Low noise (en = 9 nV/rt-Hz typ.) and ZERO 1/f noise. Fairly fast (GBW = 5 MHz, SR = 3.8 V/us). Output is rail-to-rail, inputs can range from V- to (V+ - 1.5) volts. Not available in DIP package. Key features: extremely low Vos and Vos drift, and complete absence of 1/f noise.

LT1496: Bipolar quad op amp with extremely low supply current: 1 uA per amplifier. Output is rail-to rail, input voltage range at 25C goes from V- to (V- + 36V) regardless of supply voltage. Minimum total supply voltage is 2.1 volts. Very slow (GBW = 2.7 kHz), very noisy (en = 185 nV/rt-Hz), and very limited output drive (Isc = 700 uA). Key features: extremely low supply current, low minimum supply voltage.
 
An interesting thread that I will bookmark. To be clear, I don't have much experience with op-amps, at least not compared to many others that might respond. Played around with a few many years ago (741, LM324, maybe a few others).

For more recent projects, where I need some relatively simple functions (comparator or unity gain for impedance matching), I have settled on the MCP601.

My reasoning is:
  • Single supply operation (for 3V3 and 5V).
  • Rail to rail output.
  • DIP package
  • Relatively cheap (~0.46 US).

Frankly, I don't know a whole lot about some of the finer specifications and I would be interested in learning (at least to some extent).
I bought a handful and they, so far, fit the bill.
 

Thread Starter

OBW0549

Joined Mar 2, 2015
3,566
That's a good part. If you want something that has rail-to-rail inputs and can operate off a higher supply voltage (+15V max.), you might consider the LMC6482. It's a bit more expensive, but still not horrible.

Frankly, I don't know a whole lot about some of the finer specifications and I would be interested in learning (at least to some extent).
That gives me an idea for something useful I could do around here: write up a guide to op amp data sheet specifications, what they mean, and which ones to pay attention to when designing something. One of these days...
 
Last edited:

MrChips

Joined Oct 2, 2009
30,708
That's a good part. If you want something that has rail-to-rail inputs and can operate off a higher supply voltage (+15V max.), you might consider the LMC6482. It's a bit more expensive, but still not horrible.


That gives me an idea for something useful I could do around here: write up a guide to op amp data sheet specifications, what they mean, and which ones to pay attention to when designing something. One of these days...
Sounds like a useful idea. I'm on to it. (just a spread sheet of the most popular opamps and specs summary).

Edit: I just realized we already have something like that here:

https://forum.allaboutcircuits.com/threads/components-selection-guide.65137/#post-448192

Perhaps it needs revision.
 

Thread Starter

OBW0549

Joined Mar 2, 2015
3,566
That's different from what I'm contemplating, which is more a guide to what the various op amp data sheet specs mean, how they impact a circuit design, and what "gotchas" can be avoided by understanding them. All in beginner-friendly terms, insofar as that's possible.

How many times have we had someone come in here asking for help figuring out why his op amp circuit doesn't work, and it turns out he's ignored the part's input common-mode voltage range, or the minimum supply voltage requirement, or the output voltage range, or the output drive current limits, or used an op amp with inadequate GBW or slew rate for the signal frequencies he's working with? Hundreds, if not thousands of times.

And he wasn't aware of these things, because he didn't read the op amp's data sheet.

And he didn't RTFDS because he tried to once, but became intimidated by all the strange terms that weren't explained.

That is what I'm aiming to alleviate, if only a little.
 

cmartinez

Joined Jan 17, 2007
8,218
And he didn't RTFDS because he tried to once, but became intimidated by all the strange terms that weren't explained.

That is what I'm aiming to alleviate, if only a little.
You hit the nail in the head with that observation. It applies 100% in my case. I'm looking forward to seeing an example of what you have in mind.

I lack the knowledge or experience to participate in this little venture in a meaningful way. But please let me know if there's anything I can do to help ... maybe fix you a cup of coffee while you do all the work? ;):D
 

dl324

Joined Mar 30, 2015
16,845
I started using opamps a long time ago, so I prefer what I used first.

uA741, LM307, LM308, LM318, LF356, CA3080, LM358, LM324, NE5534, LM3900.

Recently bought some TLV2272 and MC33204 to find out what all this rail to rail I/O or CMOS stuff is about.

My opamp needs these days are modest and my go to is LM358. I have a lifetime supply of them.
 

Thread Starter

OBW0549

Joined Mar 2, 2015
3,566
You hit the nail in the head with that observation. It applies 100% in my case. I'm looking forward to seeing an example of what you have in mind.
I've made a tentative list of the op amp parameters I plan to cover:

Absolute Maximum Ratings
Input Offset Voltage
Input Offset Voltage Drift
Input Bias Current
Input Offset Current
Input Resistance
Input Noise Voltage
Input Voltage Noise Density
Input Voltage Range
Input Capacitance
Open-Loop Voltage Gain
Gain-Bandwidth Product
Slew Rate
Phase Margin
Output Voltage Swing (or Output Voltage High and Output Voltage Low)
Open-Loop Output Resistance
Capacitive Loading
Output Short-Circuit Current
Power Supply Rejection Ratio
Common Mode Rejection Ratio
Minimum Operating Supply Voltage
Supply Current

What I plan on doing is to define each of these in reasonably beginner-friendly language; explain how, and in what kinds of designs, each might be important; and explain when, and why, each might be unimportant in a particular design. Some of these parameters (e.g., Absolute Maximum Ratings) are never to be ignored, in any design; others (like Phase Margin) are almost never used except by engineers doing certain kinds of circuit analysis. Others are sometimes important, sometimes not so important. And a few can be either vital to the success of a design or completely irrelevant, depending on the application.

My goals are to 1) de-mystify these parameters so reading an op amp data sheet won't be so confusing and intimidating; 2) show how to use them, with examples in some cases; and 3) give beginners knowledge they can use to choose the best op amp for whatever design task is at hand.

I lack the knowledge or experience to participate in this little venture in a meaningful way. But please let me know if there's anything I can do to help ... maybe fix you a cup of coffee while you do all the work? ;):D
Actually, there is something you can do that would be a big help: read through some of the op amp data sheets you have on hand and flag any other data sheet items (not just parameters) you think deserve explanation, and any other op amp-related questions that might be good to answer.
 
Last edited:

cmartinez

Joined Jan 17, 2007
8,218
Actually, there is something you can do that would be a big help: read through some of the op amp data sheets you have on hand and flag any other data sheet items (not just parameters) you think deserve explanation, and any other op amp-related questions that might be good to answer.
Thanks, I'll take you at your word...

Let's start with this. I have a hard time understanding the meaning of "common mode", I've seen it applied to input range, noise, rejection, etc. But I have yet to read a description of this concept in plain Spanish English...

I've read this application note, and I'm not sure I fully understand it. The statement "Technically, a common-mode voltage is one-half the vector sum of the voltages from each conductor of a balanced circuit to local ground or common." is plain and forward, but it's far from being simple. It's too dense and succinct. The rest of the article does go into explaining things into more detail, but I think maybe it could be explained in simpler terms.
 

Thread Starter

OBW0549

Joined Mar 2, 2015
3,566
Your difficulty is caused simply by the fact that in the Maxim application note you referenced, they are using the term common mode in the context of data transmission in wire pairs.

For op amps, its different-- and much simpler. Considering that in any op amp circuit employing negative feedback and operating the op amp in its linear region (i.e., not saturated against the + or - supply rail), the differential input voltage is essentially zero; thus, its two inputs are at essentially the same voltage with respect to ground. That voltage is called the common-mode voltage. That voltage has an allowable range the inputs must stay within if the op amp is to function normally (or at all). For many op amps (e.g., the 741), the common-mode voltage range extends from a few volts above the negative supply to a few volts below the positive supply. For so-called "single supply" op amps powered by a single positive supply and with the negative supply pin grounded (such as the LM324 or LM358), the input common mode voltage range extends from ground up to within a few volts of the positive supply. For op amps featuring "rail-to-rail" inputs, the input common mode voltage range extends over the entire supply voltage span.
 
Thank you OBW0549 for this interesting post.
I used uA741 , LM307 , LM308 and some other...

But now I'm working on a project (medical device) with LMC6462 Dual/LMC6464 and/or with LMP7731.

I need a low noise and precision amplifier, then I'm more interested on the second solution!
But I will try both the same, since the LMC6462 have a very good ultra low power supply current.

LMC6462/LMC6464 (dual or quad)
Ultra Low Supply Current 20 μA/Amplifier
Rail-to-Rail Output Swing
Low Input Current 150 fA
slew rate 28 V/ms
input referred voltage noise 80 nV/rt(Hz)

LMP7731
Supply Voltage Range 1.8V to 5.5V
RRIO
Input Bias Current ±1.5 nA
GBW 22 MHz
Input Voltage Noise: f = 1 kHz 2.9 nV/√Hz
Slew Rate 2.4 V/µs
Supply Current per Channel 2.2 mA

Do you use this amplifiers?
What do you think about?
 

Thread Starter

OBW0549

Joined Mar 2, 2015
3,566
Excellent explanation, thank you very much... funny you mentioned "rail-to-rail", as it was going to be my next question. Would you please elaborate?
Nothing magical; "rail-to-rail inputs" simply means that the input common mode range extends all the way from V- to V+ and the op amp will work properly provided the circuit puts its inputs anywhere within that range. "Rail-to-rail output," similarly, means that the op amp's output can range from within a few millivolts of V- all the way up to within a few millivolts of V+. To get a better feel, take a look at the data sheet for a typical RRIO op amp, the LMC6482, especially the material beginning on page 18.
 

cmartinez

Joined Jan 17, 2007
8,218
"rail-to-rail inputs" simply means that the input common mode range extends all the way from V- to V+ and the op amp will work properly provided the circuit puts its inputs anywhere within that range.
So I'm guessing that no opamps exist out there with the ability of accepting inputs beyond the limits of its power sources?
 

Thread Starter

OBW0549

Joined Mar 2, 2015
3,566
So I'm guessing that no opamps exist out there with the ability of accepting inputs beyond the limits of its power sources?
Ha! See my description in post #1 above of the LT1490A and the LT1496. These are from Linear Technology's "Over the Top" series of parts, which can function even when the inputs are at a higher voltage than the V+ supply voltage.
 
Top