TL081 vs TL071

[Michael A. Covington]

What is the difference between these op-amps? The latter is billed as
"low-noise" but they have the same noise specification (18 nv/sqrt(Hz)), as
well as nearly all the same other specifications ('81 has less offset
error). And the same price (48 cents).

I've been using them both as general-purpose op-amps for a long time and
haven't noticed any difference between them. Are they generally well
regarded? What do people prefer as an all-purpose, default-choice op-amp
these days (in place of the old 741, which I always thought was too noisy)?

[Watson A.Name - "Watt Sun, the Dark Remover"]

Post by Michael A. Covington
What is the difference between these op-amps? The latter is billed as
"low-noise" but they have the same noise specification (18

nv/sqrt(Hz)), as

Post by Michael A. Covington
well as nearly all the same other specifications ('81 has less offset
error). And the same price (48 cents).
I've been using them both as general-purpose op-amps for a long time and
haven't noticed any difference between them. Are they generally well
regarded? What do people prefer as an all-purpose, default-choice op-amp
these days (in place of the old 741, which I always thought was too noisy)?

5532 series for audio apps? The TL082 was sold by Rat Shack but the
package would come with either that or a LF353. I guess they were
interchangeable. I would not use a 741 for decent Hi-Fi audio apps.

[Phil Vossler]

5532 & 5534??? These were the bees knees back in the '80s. Are they
*still* the best for audio? I'd be surprised if there were nothing more
recent that has better specs....

I admit my days of analog design ended years ago - but i have a little
requirement for an audio distribution amp hence recommendations for good
audio op-amps would be very welcome.....

Post by Michael A. Covington

Post by Michael A. Covington
What is the difference between these op-amps? The latter is billed as
"low-noise" but they have the same noise specification (18

nv/sqrt(Hz)), as

Post by Michael A. Covington
well as nearly all the same other specifications ('81 has less offset
error). And the same price (48 cents).
I've been using them both as general-purpose op-amps for a long time

and

Post by Michael A. Covington
haven't noticed any difference between them. Are they generally well
regarded? What do people prefer as an all-purpose, default-choice

op-amp

Post by Michael A. Covington
these days (in place of the old 741, which I always thought was too

noisy)?
5532 series for audio apps? The TL082 was sold by Rat Shack but the
package would come with either that or a LF353. I guess they were
interchangeable. I would not use a 741 for decent Hi-Fi audio apps.

[Watson A.Name - "Watt Sun, the Dark Remover"]

Post by Phil Vossler
5532 & 5534??? These were the bees knees back in the '80s. Are they
*still* the best for audio? I'd be surprised if there were nothing more
recent that has better specs....
I admit my days of analog design ended years ago - but i have a little
requirement for an audio distribution amp hence recommendations for good
audio op-amps would be very welcome.....

You need an Opamp? Y for? A distrbution amp just needs to have a high
input impedance so it doesn't load the source, and low output impedance
so it can drive longish cables without rolling off the high freqs.
Something as simple as an emitter follower would do just fine, if all
you need is unity gain. if you need a bit of voltage gain, then a
single transistor common emitter stage before the drivers with a bit of
NFB will give you enough gain. And with the NFB, the bandwidth could be
enough to handle video - 4 to 6 MHz.
[snip]

[Walter Harley]

Post by Watson A.Name - "Watt Sun, the Dark Remover"
You need an Opamp? Y for? A distrbution amp just needs to have a high
input impedance so it doesn't load the source, and low output impedance
so it can drive longish cables without rolling off the high freqs.
Something as simple as an emitter follower would do just fine, if all
you need is unity gain. if you need a bit of voltage gain, then a
single transistor common emitter stage before the drivers with a bit of
NFB will give you enough gain. And with the NFB, the bandwidth could be
enough to handle video - 4 to 6 MHz.

An emitter follower introduces harmonic distortion on the order of 1/beta,
no? (A quick simulation with a 2N2222 in LTSpice gives about 3% THD.)
Clearly unacceptable for most audio work. You need extra gain to fix that.
An opamp is a nicely packaged, inexpensive way to get it.

[Watson A.Name - "Watt Sun, the Dark Remover"]

Post by Walter Harley

Post by Watson A.Name - "Watt Sun, the Dark Remover"
You need an Opamp? Y for? A distrbution amp just needs to have a high
input impedance so it doesn't load the source, and low output impedance
so it can drive longish cables without rolling off the high freqs.
Something as simple as an emitter follower would do just fine, if all
you need is unity gain. if you need a bit of voltage gain, then a
single transistor common emitter stage before the drivers with a bit of
NFB will give you enough gain. And with the NFB, the bandwidth could be
enough to handle video - 4 to 6 MHz.

An emitter follower introduces harmonic distortion on the order of 1/beta,
no? (A quick simulation with a 2N2222 in LTSpice gives about 3% THD.)
Clearly unacceptable for most audio work. You need extra gain to fix that.
An opamp is a nicely packaged, inexpensive way to get it.

The way I've always understood it is that a common collector (emitter
follower) circuit has 100% negative voltage feedback.

And as I mentioned above, NFB will take care of the bandwidth, and the
distortion, too.

[Norm Dresner]

Post by Michael A. Covington
What is the difference between these op-amps? The latter is billed as
"low-noise" but they have the same noise specification (18 nv/sqrt(Hz)), as
well as nearly all the same other specifications ('81 has less offset
error). And the same price (48 cents).

To make matters even murkier, if you look at the circuit diagrams on the
data sheets, they're identical! But identical circuitry doesn't necessarily
mean that the processing is exactly the same, that the size of the
transistors is identical, ...

But they do look very similar. I've been told that there are circuits in
which one will work and not the other. I don't choose to find out for
myself -- see below.

Post by Michael A. Covington
I've been using them both as general-purpose op-amps for a long time and
haven't noticed any difference between them. Are they generally well
regarded? What do people prefer as an all-purpose, default-choice op-amp
these days (in place of the old 741, which I always thought was too noisy)?

I bought lots of 100 NE5534 single and NE5532 dual op amps on eBay to use in
my own audio designs. There are better choices but at $4-8 each instead of
40-80 cents. The 741 isn't worth anything above a few kilohertz and even
then the noise and offset specs are lousy. But they're cheap.

Norm

[Michael A. Covington]

Thanks. I should add that I'm not looking for the ideal audio amp -- I have
others for that. Rather, I'm trying to settle on something better than the
741 as a cheap, abundant op-amp for a wide range of non-critical uses. Part
of the appeal of the TL081/2/4 (or TL071/2/4) is that it's easy to remember
the numbers (single, dual, quad).

[Tim Shoppa]

Post by Michael A. Covington
Thanks. I should add that I'm not looking for the ideal audio amp -- I have
others for that. Rather, I'm trying to settle on something better than the
741 as a cheap, abundant op-amp for a wide range of non-critical uses.

The 741 has its problems but by definition it's the all-around jellybean
part :-).

See the thread from January of this year "Singing the praises of the LM324",
too. It's often-villified for crossover distortion (there *are* workarounds)
but it works very well for DC-ish applications from rail to rail.

Tim.

[Dave VanHorn]

Post by Tim Shoppa
See the thread from January of this year "Singing the praises of the LM324",
too. It's often-villified for crossover distortion (there *are* workarounds)
but it works very well for DC-ish applications from rail to rail.
Tim.

Well, the Xover distortion can be a really hideous nasty surprise, if you
don't know about it in advance.

[Watson A.Name - "Watt Sun, the Dark Remover"]

Post by Dave VanHorn

Post by Tim Shoppa
See the thread from January of this year "Singing the praises of the LM324",
too. It's often-villified for crossover distortion (there *are* workarounds)
but it works very well for DC-ish applications from rail to rail.
Tim.

Well, the Xover distortion can be a really hideous nasty surprise, if you
don't know about it in advance.

I thought it weas SOP to put a 6.8k resistor from output to ground (or
was it + supply) to alleviate this problem.

[Dave VanHorn]

Post by Watson A.Name - "Watt Sun, the Dark Remover"
I thought it weas SOP to put a 6.8k resistor from output to ground (or
was it + supply) to alleviate this problem.

IIRC it's +, and I remember 2.2k.. OTOH, I rarely use them.
The TL074's are my favorite GP chips.

How many schematics do you see with 324's specified, and how many have
anything there to kill the xover distortion?

[Watson A.Name - "Watt Sun, the Dark Remover"]

Post by Dave VanHorn

Post by Watson A.Name - "Watt Sun, the Dark Remover"
I thought it weas SOP to put a 6.8k resistor from output to ground (or
was it + supply) to alleviate this problem.

IIRC it's +, and I remember 2.2k.. OTOH, I rarely use them.
The TL074's are my favorite GP chips.
How many schematics do you see with 324's specified, and how many have
anything there to kill the xover distortion?

I see the schems often in projects, and the 324 is available at Radio
Snack. The 358 is the same except it's dual, not quad. They're often
used in signal processing and filters, but not Hi-Fi. They work ok with
the inputs at the neg rail.

[Tim Shoppa]

Post by Dave VanHorn

Post by Watson A.Name - "Watt Sun, the Dark Remover"
I thought it weas SOP to put a 6.8k resistor from output to ground (or
was it + supply) to alleviate this problem.

IIRC it's +, and I remember 2.2k.. OTOH, I rarely use them.
The TL074's are my favorite GP chips.
How many schematics do you see with 324's specified, and how many have
anything there to kill the xover distortion?

It's a Floor Wax....AND a Dessert Topping!

As long as you can keep the output current from passing through -50 uA
then you're OK.

Tim.

[Jamie]

i could be wrong but i think the 71 is a bipolar input type
and the 81 is the cmos high gain input with a much better
linearity to it.

Post by Michael A. Covington
What is the difference between these op-amps? The latter is billed as
"low-noise" but they have the same noise specification (18 nv/sqrt(Hz)), as
well as nearly all the same other specifications ('81 has less offset
error). And the same price (48 cents).
I've been using them both as general-purpose op-amps for a long time and
haven't noticed any difference between them. Are they generally well
regarded? What do people prefer as an all-purpose, default-choice op-amp
these days (in place of the old 741, which I always thought was too noisy)?

[Michael A. Covington]

Post by Jamie
i could be wrong but i think the 71 is a bipolar input type
and the 81 is the cmos high gain input with a much better
linearity to it.

They seem to have the same circuit diagram.

[Jamie]

yes, they have the same lay out, they are pin for pin compatible.
but that does not mean they are the same in electrical ratings
like gain, BandWidth etc.

Post by Michael A. Covington

Post by Jamie
i could be wrong but i think the 71 is a bipolar input type
and the 81 is the cmos high gain input with a much better
linearity to it.

They seem to have the same circuit diagram.

[Michael A. Covington]

Post by Jamie
yes, they have the same lay out, they are pin for pin compatible.
but that does not mean they are the same in electrical ratings
like gain, BandWidth etc.

I mean the schematic of the chip itself is published in the data sheets and
is the same for both. Also, the numbers that indicate performance are
practically identical.

[Graham W]

Post by Michael A. Covington
I mean the schematic of the chip itself is published in the data sheets and
is the same for both. Also, the numbers that indicate performance are
practically identical.

I've just found my copy of "The BiFET Design Manual" 2nd Ed. 1982 and
I'll take a look later. I like the range and also the M'Rola MC33079P for
audio filter applications. I think the MC33079 has gone obsolete now.


--
Graham W http://www.gcw.org.uk/ PGM-FI page updated, Graphics Tutorial
WIMBORNE http://www.wessex-astro-society.freeserve.co.uk/ Wessex
Dorset UK Astro Society's Web pages, Info, Meeting Dates, Sites & Maps
Change 'news' to 'sewn' in my Reply address to avoid my spam filter.

[Dave VanHorn]

Post by Jamie
i could be wrong but i think the 71 is a bipolar input type
and the 81 is the cmos high gain input with a much better
linearity to it.

They are all fet input.

The 61 was low current (or low offset, I forget..) the 71 was low noise, and
the 81 was low cost.

Same for 62/72/82 and 64/74/84

[Walter Harley]

Post by Dave VanHorn
They are all fet input.
The 61 was low current (or low offset, I forget..) the 71 was low noise,
and the 81 was low cost.
Same for 62/72/82 and 64/74/84

Yes. But as the OP noted, and as was noted in an earlier thread on exactly
the same topic, the specs for the 7x and 8x are virtually identical these
days. "Low noise" and "low cost" seem to no longer be differentiators.

My own theory - unsupported by any evidence other than the existence of
multiple part numbers for seemingly the same part, along with historical
anecdotes about the parts behaving differently - is that they used to be
different, but at some point the mfrs decided that the "low noise" process
had gotten cheap enough or the "low cost" process had gotten quiet enough
that they just started using the same innards for both.

Again, that's just a theory; would be nice to hear from someone at TI.

[Michael A. Covington]

Post by Walter Harley
My own theory - unsupported by any evidence other than the existence of
multiple part numbers for seemingly the same part, along with historical
anecdotes about the parts behaving differently - is that they used to be
different, but at some point the mfrs decided that the "low noise" process
had gotten cheap enough or the "low cost" process had gotten quiet enough
that they just started using the same innards for both.
Again, that's just a theory; would be nice to hear from someone at TI.

I need to look at my old data books! My dim recollection is that when I
settled on the TL081 back in the 1980s, it was quite distinct from the
TL071. (Maybe it just cost less.) But Horowitz and Hill 1989 already show
the same numbers for the specs of both chips (actually appreciably less
offset on the '81; no visible advantage to the '71 though they tag it "low
noise").

[Watson A.Name - "Watt Sun, the Dark Remover"]

Post by Walter Harley

Post by Dave VanHorn
They are all fet input.
The 61 was low current (or low offset, I forget..) the 71 was low noise,
and the 81 was low cost.
Same for 62/72/82 and 64/74/84

Yes. But as the OP noted, and as was noted in an earlier thread on exactly
the same topic, the specs for the 7x and 8x are virtually identical these
days. "Low noise" and "low cost" seem to no longer be

differentiators.

Post by Walter Harley
My own theory - unsupported by any evidence other than the existence of
multiple part numbers for seemingly the same part, along with

historical

Post by Walter Harley
anecdotes about the parts behaving differently - is that they used to be
different, but at some point the mfrs decided that the "low noise" process
had gotten cheap enough or the "low cost" process had gotten quiet enough
that they just started using the same innards for both.
Again, that's just a theory; would be nice to hear from someone at TI.

Well, TI has pages and pages of opamps in the catalogs of the likes of
Digi-key and Mouser. Makes choosing an appropriate opamp totally
bewildering.

And that's from just one manufacturer!

[Dave VanHorn]

Post by Walter Harley
Yes. But as the OP noted, and as was noted in an earlier thread on exactly
the same topic, the specs for the 7x and 8x are virtually identical
these days. "Low noise" and "low cost" seem to no longer be
differentiators.

This was originally done back in the late 70's or early 80's and that was
the official word from TI.

[Michael Black]

Post by Michael A. Covington
What is the difference between these op-amps? The latter is billed as
"low-noise" but they have the same noise specification (18 nv/sqrt(Hz)), as
well as nearly all the same other specifications ('81 has less offset
error). And the same price (48 cents).

It took a bit of digging, but I pulled out the brochure TI sent out when
these devices were just introduced, dated 1977. Well, I knew where it was,
but the specific box was under some others.

TL081 = 47nV
TL071 = 18nV

input offset voltage
TL081 = 15 to 20mV
TL071 = 10 to 13mV

input offset current
TL082 = 0.2 to 5nA
TL071 = .05 to 2nA

Icc max
TL082 = 2.8mA
TL071 = 2.5mA

Prices in 100 unit quantities
TL082 = .33
TL071 = .47

This is from their comparison chart. I've not dug into the separate
data sheets later in the brochure. The internal diagrams look the same,
but not all values are given.

It doesn't seem to say much about why the two devices differ, ie if
one is a hand selected from the larger pool.

There clearly was a stated difference from the starting gate, but who
knows how long that held true.

Michael

Post by Michael A. Covington
I've been using them both as general-purpose op-amps for a long time and
haven't noticed any difference between them. Are they generally well
regarded? What do people prefer as an all-purpose, default-choice op-amp
these days (in place of the old 741, which I always thought was too noisy)?

[Michael A. Covington]

Post by Michael Black

Post by Michael A. Covington
What is the difference between these op-amps? The latter is billed as
"low-noise" but they have the same noise specification (18 nv/sqrt(Hz)), as
well as nearly all the same other specifications ('81 has less offset
error). And the same price (48 cents).

It took a bit of digging, but I pulled out the brochure TI sent out when
these devices were just introduced, dated 1977. Well, I knew where it was,
but the specific box was under some others.
TL081 = 47nV
TL071 = 18nV

AHA! Nowadays both are specified as 18 nV/sqrt(Hz).

I'll bet nowadays they are the same chip! Originally they weren't.

[Michael Black]

Post by Michael A. Covington

Post by Michael Black

Post by Michael A. Covington
What is the difference between these op-amps? The latter is billed as
"low-noise" but they have the same noise specification (18 nv/sqrt(Hz)), as
well as nearly all the same other specifications ('81 has less offset
error). And the same price (48 cents).

It took a bit of digging, but I pulled out the brochure TI sent out when
these devices were just introduced, dated 1977. Well, I knew where it was,
but the specific box was under some others.
TL081 = 47nV
TL071 = 18nV

AHA! Nowadays both are specified as 18 nV/sqrt(Hz).
I'll bet nowadays they are the same chip! Originally they weren't.

Of course, that now leads to the question of when this occurred. Do
we all have the low noise ones in our junkboxes, or is a relatively recent
change?

I don't have a complete sequence of TI literature (or any company's literature
for that matter) in order to check year by year, or even decade by decade.

Michael

[Michael A. Covington]

Post by Michael Black

Post by Michael A. Covington

Post by Michael Black
TL081 = 47nV
TL071 = 18nV

AHA! Nowadays both are specified as 18 nV/sqrt(Hz).
I'll bet nowadays they are the same chip! Originally they weren't.

Of course, that now leads to the question of when this occurred. Do
we all have the low noise ones in our junkboxes, or is a relatively recent
change?

The chart in The Art of Electronics, 1989, lists them both as 18 nV/sqrt(Hz)
but describes the '71 as "low noise." Does anybody have the earlier edition
of The Art of Electronics?

[Watson A.Name - "Watt Sun, the Dark Remover"]

"Michael Black" <***@FreeNet.Carleton.CA> wrote in message news:cfr208$vb$***@freenet9.carleton.ca...
[snip]

Post by Michael Black
It doesn't seem to say much about why the two devices differ, ie if
one is a hand selected from the larger pool.

Or if the opamp chip was laser trimmed or whatever had to be done to get
the tighter specs.

Post by Michael Black
Michael