Question about Opamps

Did some more research on the OP37 and 5532. Have the following corrections to myself:
- The OP37 is as ancient as the 5532. (Is there really nothing newer?)
- I found out Douglas Self (after some (superficial) reading of some literature by Douglas Self (small signal audio design)) took the time to do some proper measuring. According to these measurements the 5532 is in practice less noisy and has less THD as the OP27, which is closely related to the OP37.
He does however never load the thing below 500 ohm during these measurements, which he calls it's 'maximum'. No idea where he got that information.

Does anybody have suggestions?

At one time, Self's web site had a review of all the common opamps, including measurements he took. Unfortunately, last I checked he had taken the information down.

The NE5532 datasheet from Texas Instruments references R (subL) as being 600 ohms.

There's nothing magical about Burr-Brown or Analog devices. They all have their applications where they're good at. Op amps, to work their best, need to be placed in circuits that are correctly designed, so that the intended op amp works sell.

Sorry Burr-Brown and Analog Devices Fan boys.

nashvillebill said:

At one time, Self's web site had a review of all the common opamps, including measurements he took. Unfortunately, last I checked he had taken the information down.

The NE5532 datasheet from Texas Instruments references R (subL) as being 600 ohms.

Click to expand...

I guess you're talking about: http://www.douglas-self.com/ampins/closed.htm ??

Maybe this page comes close?

http://nwavguy.blogspot.nl/2011/08/op-amp-measurements.html

But is there anything newer than the OP37 or 5532 with similar or better specs?

Self originally reviewed about a dozen or so opamps, with a separate page for each one. Very good stuff, and he concluded the 5532 was still pretty much as good as it gets for audio work. Maybe the pages are available on the "wayback machine" or whatever that web archive site is called.

The second link has some very good info and is well worth reading. Especially the part about better specs not necessarily meaning better performance. The 5532 is fast enough for audio, but going faster doesn't make another opamp better: the faster one may have a higher gain bandwidth and go into oscillation when used in a circuit that the 5532 would handle just fine.

I don't claim to have anywhere near the expertise of Self or of some of the other experts who post on here, but I use 5532's for my general audio projects.

When it's all said and done, it's the circuit design that really counts.

NE/SE5532 and 5534 are still great opamps for many/most applications.

Personally I spend a couple bucks more and use the LME49710. Properly applied there's isn't much better, but bypassing and layout are important because it has a GBP of something like 55 MHz. They aren't well suited to just plopping them in some old circuit without careful testing and IMO that's why the "rollers" often don't like them. They drive 600 ohms and even in my phono section front end, are hard to beat for noise and distortion.

I rarely use duals, but note that the NJM4562 is absolutely not the equivalent of the LM4562. If they had completely different part numbers, they'd be treated as completely different opamps.

It's good to look at why an opamp was designed. The OP27 was aimed at test and measurement where excellent DC specs were needed. It's not that fast. OP37 is the uncompensated version and is faster, but was still aimed at applications where high DC precision was important. Oddly, if you sharply filter the OP37 above 20 kHz or so, it's quieter than OP27, though just looking at the unfiltered noise on a scope you'd never know it. There are lots of fans of the OPA627 and 637, but you're paying big bucks for insane low input offset currents and other parameters that have nothing to do with audio.

There have been a number of threads about opamps here in DIY. I followed the one by dr*audio: http://audiokarma.org/forums/showthread.php?t=76689, and installed an OPA2228 to replace an NJM5532 in a Sony CD changer. Sounds great now. I also swapped the 5532 in an Onkyo DX-1800 for the LM4562 for some improvement but I haven't quantified it as that machine gets less use.

Jon_Logan said:

There's nothing magical about Burr-Brown or Analog devices. They all have their applications where they're good at. Op amps, to work their best, need to be placed in circuits that are correctly designed, so that the intended op amp works sell.

Sorry Burr-Brown and Analog Devices Fan boys.

Click to expand...

exactly, some people are swapping op amps and hearing oscillations and confusing that with a clearer high end

What about the AD797? Nobody mentions this by some people highly recommended opamp.

Here's a build using the AD797. Note that the circuit was designed around the opamp, not that the opamp was just thrown in to replace something else:

http://www.audiokarma.org/forums/showthread.php?t=501186&highlight=ad797

nashvillebill said:

Self originally reviewed about a dozen or so opamps, with a separate page for each one. Very good stuff, and he concluded the 5532 was still pretty much as good as it gets for audio work. Maybe the pages are available on the "wayback machine" or whatever that web archive site is called.

The second link has some very good info and is well worth reading. Especially the part about better specs not necessarily meaning better performance. The 5532 is fast enough for audio, but going faster doesn't make another opamp better: the faster one may have a higher gain bandwidth and go into oscillation when used in a circuit that the 5532 would handle just fine.

I don't claim to have anywhere near the expertise of Self or of some of the other experts who post on here, but I use 5532's for my general audio projects.

When it's all said and done, it's the circuit design that really counts.

Click to expand...

I continue to find it amusing, after a 25 year career as an engineer in television production (mostly audio production) that people continue to think they hear a difference when they change one op-amp. This signal has already been through literally 100's of NE5532's and countless average quality capacitors in audio mixing consoles and processing equipment. The NE5532 is probably the most common op-amp used in professional audio equipment.

We had several consoles at TBS that had over 1000 NE5532's in the audio paths. (not all in series, though) A typical audio path through the console, which had 48 input channels, probably went through about 40 NE5532's. One more isn't going to make much difference.

And this doesn't even take into consideration the 1000's of feet of plain old copper wire, and countless audio connectors that connect everything together.

Dave

mrdbdigital said:

I continue to find it amusing, after a 25 year career as an engineer in television production (mostly audio production) that people continue to think they hear a difference when they change one op-amp.

Dave

Click to expand...

Well some folks hear different from others. Not right or wrong, just different. Kinda like some folks wear glasses/contacts and others don't need to.

mrdbdigital said:

I continue to find it amusing, after a 25 year career as an engineer in television production (mostly audio production) that people continue to think they hear a difference when they change one op-amp. This signal has already been through literally 100's of NE5532's and countless average quality capacitors in audio mixing consoles and processing equipment. The NE5532 is probably the most common op-amp used in professional audio equipment.

We had several consoles at TBS that had over 1000 NE5532's in the audio paths. (not all in series, though) A typical audio path through the console, which had 48 input channels, probably went through about 40 NE5532's. One more isn't going to make much difference.

And this doesn't even take into consideration the 1000's of feet of plain old copper wire, and countless audio connectors that connect everything together.

Dave

Click to expand...

No offense intended:

By your logic then any old cartridge should do for playing vinyl as the sound has been through countless low cost components prior to the production of a record?

Blue Shadow said:

Well some folks hear different from others. Not right or wrong, just different. Kinda like some folks wear glasses/contacts and others don't need to.

Click to expand...

Yes, there are some people who hear better than others, but that's not the point I'm making. Changing one op-amp in a string of 40 isn't going to make any significant difference. If it did, imagine how much the signal would have already been effected by the other op-amps. I've questioned "golden eared" professional engineers by patching around the console from the source directly to the monitor (bypassing the entire console) with matched levels and not one could ever hear the difference. That's directly putting 40 or so op-amps in and out of the circuit in a blind test.

When I was with Philips Arena in Atlanta, I met a fellow (Elton John's audio guru) who could hear a 0.25 db difference in sound levels. We were setting up the arena's audio system to support the concert while this guy walked the venue and listened. He asked for 0.5db more level in the north end of the seating, so we raised it 0.25db, as we had heard he could detect ultra small level differences. He immediately called us on the radio and said, "Not a quarter db, I want a half db!". We tried 3 times and he caught us all three times.

Try all those people who claim to hear differences in op-amps with a blind test, and see what happens. I'm still waiting to find someone who can actually prove they hear the difference in one of those tests.

Of course, as mentioned previously, the circuit must be properly designed to support the op-amp being used. It's not a fair comparison to just rip out one op-amp and plug in another without taking that into consideration. All op-amps have different circuit requirements.

Dave

mrdbdigital said:

...Of course, as mentioned previously, the circuit must be properly designed to support the op-amp being used. It's not a fair comparison to just rip out one op-amp and plug in another without taking that into consideration. All op-amps have different circuit requirements.

Dave

Click to expand...

Yep, I've heard folks rave about upgrading to an OPA2134 in some music gear--they pulled out a NJM4558 and dropped in the OPA2134. So they've gone from a bipolar to a FET, without changing anything else.

HypnoToad said:

No offense intended:

By your logic then any old cartridge should do for playing vinyl as the sound has been through countless low cost components prior to the production of a record?

Click to expand...

That is not what I said.

Of course phono cartridges sound different. They are electromechanical devices subject to mechanical resonances and other factors.

The point I'm making is that if "op-amp sound" is so critical as to be heard by changing only one in an audio path, why do virtually all makers of professional audio equipment, like consoles, use so many of them in series in the audio path? Maybe because in this application they are virtually transparent, and provide the necessary gain with minimalist performance issues?

As far as capacitors, look in most professional equipment and you'll find thousands of good quality standard capacitors, not all the esoteric ones.

Some specific areas of a console, like microphone preamps, use higher quality components to help lower the noise floor. The AMS Calrec console we had at TBS for live events had discrete transistors in the mic preamps in Valley People modules, elsewhere the console was filled with probably 1000 NE5532's, and that console had very impressive specs. The console was acclaimed to be incredibly "transparent and open" by all the professional engineers that operated it, especially while doing live studio sessions.

Dave

mrdbdigital said:

That is not what I said.

Of course phono cartridges sound different. They are electromechanical devices subject to mechanical resonances and other factors.

The point I'm making is that if "op-amp sound" is so critical as to be heard by changing only one in an audio path, why do virtually all makers of professional audio equipment, like consoles, use so many of them in series in the audio path? Maybe because in this application they are virtually transparent, and provide the necessary gain with minimalist performance issues?

As far as capacitors, look in most professional equipment and you'll find thousands of good quality standard capacitors, not all the esoteric ones.

Some specific areas of a console, like microphone preamps, use higher quality components to help lower the noise floor. The AMS Calrec console we had at TBS for live events had discrete transistors in the mic preamps in Valley People modules, elsewhere the console was filled with probably 1000 NE5532's, and that console had very impressive specs. The console was acclaimed to be incredibly "transparent and open" by all the professional engineers that operated it, especially while doing live studio sessions.

Dave

Click to expand...

This is what other people that have professional audio engineering experience have said as well.

If you think about it you don't want an op-amp to add or subtract anything to the signal. You just want what appears at output B to be an exact replica of what is present at input A amplified by factor X.

If there is a mysterious shift in tone that was not calculated then something has probably gone out of balance in the circuit.

mrdbdigital said:

I continue to find it amusing, after a 25 year career as an engineer in television production (mostly audio production) that people continue to think they hear a difference when they change one op-amp. This signal has already been through literally 100's of NE5532's and countless average quality capacitors in audio mixing consoles and processing equipment.

Click to expand...

I'm truly sorry if you've only heard mediocrity. Not all recording engineers are deaf and utterly clueless.

There has been a thread on this topic over in diyaudio where samples have been put up, and a survey indicates preferences. The results were interesting and suggest that opamps might be the least of our worries.

The problem with judgments made under uncontrolled conditions is that it doesn't work, yet everybody continues to do it. There are biases that cannot be defeated. Even if you could fight the ones you know about, what about the ones you don't?

E-Stat said:

I'm truly sorry if you've only heard mediocrity. Not all recording engineers are deaf and utterly clueless.

Click to expand...

Perhaps you should read up on AMS Calrec, and some of the PEOPLE who design and build them. Here's an article on Chris Gaunt, main engineer, had some nice experience in HiFi before moving on to Electrical Engineering and production. Recording desks sounded good for Abba to install in their studios and in use around the world.

CHRIS GAUNT HAS SPENT his entire
working life in the audio industry with his
interest stretching back to ‘doing the sound’
for dramas while at school and a Saturday job in a hifi
shop where he was introduced to the wonders of the
Quad Electrostatic. He studied Electronic Engineering
at university and developed skills in organising a
range of social events, which sometimes included
setting up the lighting and sound. After graduating
he installed and serviced sound equipment in the
ballrooms and nightclubs of the MECCA group before
joining Calrec as the test department supervisor in
1978, where he worked on the J-Series, L-Series and
M-Series broadcast consoles and moved from testing
into system design and project management. He was
one of the driving forces behind Calrec’s UA8000
music console, project managing the design and
build, and taking responsibility for all schematics and
backplane design, and even installed the first desk at
Abba’s Polar Studios.
Following the purchase of Calrec by AMS in 1986,
Chris spent more than ten years with AMS in several
roles and contributed significantly to its digital console
projects before rejoining Calrec in 1998 after it had
spun off again on its own. He has overseen the
development of Calrec’s entire range of digital desks
and is responsible for the continued development
of the Alpha, Sigma and Zeta System Plus range of
consoles. He is also involved in Standards activities
with the AES, IEC and SMPTE while his other passions
include cricket, photography and folk music.

From 2005:
http://www.resolutionmag.com/pdfs/MAKER/chrisgaunt.pdf

Here's a 2007 article which details the history of AMS & Calrec mailnly. Seems they were involved with the BBC during the 70's and 80's, pushing through some great products with BBC's sound and broadcast demands of the time. I seem to remember that the BBC was pretty obsessed with sound quality in its recordings and FM/TV broadcasts. They even went so far as to design their own speakers (LS3/5a, LS5/8, LS5/9, LS3/6, LS5/12) because they felt there were no commercial types good enough and consistent enough for their stringent requirements.

Gosh, I just remembered - at the time, 1977-80's, those desks, if using IC op-amp technology, were previously using single or dual op-amps when the Signetics NE5532/5534 were introduced. Seems they were able to build a damn good product then, and now, and seems they have met with critical acceptance at the producer and recording engineer levels for the past 40 years.

http://www.ips.org.uk/files/07_Calrec_Enters_The_Next_Phase.pdf

Here's a gearslutz thread about Calrec consoles, and their use by recording professionals. Seems they had and have a good reputation, compared favorably to NEVE and the like. The commenters are working PROFESSIONALS in the recording and broadcast business, I suspect they do hear pretty well.

http://www.gearslutz.com/board/high-end/779-calrec-desks.html

Here's an ad for a fully refurbished 1989 Calrec console, I've copied a few things that make up the way the console is built and featured. What mrdbdigital is trying to say is that recording consoles are quite complicated and sophisticated tools, built up from literally thousands of active and passive component, each used to shape and create the final sound that the engineer, artist, and producer want on their recording and broadcast. At original cost of £100k to 300K new, they put everything they knew into the design and production of these consoles, and they had to sound good (even with miles of signal length and scads of signal devices) at the end. These don't get bought unless the end user (private or commercial) thinks they sound good and do the job that's required.

And remember, every recording out there is run through consoles like these, or newer fully digital desks. Sad to say also a lot of stuff is done strictly through computers and fairly rudimentary mixers/consoles at much lower price points (indie production for example) that may not have "audiophile" quality op-amps or devices.

http://www.cmlaudio.com/calrec-48-channel-m-series/

Control Area
From left to right, the console has 4 distinct control areas:
24 x PQ1895 Mono Mic / Line input channels (with improved microphony).
Main section with 2 Main stereo outputs, 8 x Stereo Groups, Aux Masters, Monitor Selector and Monitor Control
12 x PQ2588 Stereo Line Input channels.
12 x PQ1895 Mono Mic / Line input channels.
Channel modules have 10dB trim, Mic Gain +10 to -70dB in 5 dB steps. 32dB input headroom. Phase reverse. Parametric High and low pass filters. 4 band parametric EQ. 15 dB cut and boost, high and low Q settings. Bell or shelf setting on high and low bands.
7 Aux sends configured as 4 FB and 1 PA each with +5dB, on/off button, selectable pre or post fader.
PA muting by end fader end switch.
2 echo post fader.
Channel Separate output and insert selectable Pre EQ, Pre fader and post fader. 22dB output headroom.
AFL switch and PFL on fader overpress.
P&G faders (4220 Mono, 4222 Stereo – normal, push for on).
2 Echo return modules with bargraph meter.
Direct inputs with level control to Aux and main busses.
2 DL1656 compressor modules.
1 Oscillator module – variable tone and noise generator, ±12dB gain trim | +8dB to -60dB switched.
1 Calrec condenser talk back microphone.
3 dual needle Sifam type 74 PPMs with A / B, M / S and S+20 switching.
Comprehensive monitor panel with 24 Stereo and 24 monitor selector inputs, independent Control room, PFL, AFL and dim level controls. Independent cut left, cut right, cut both and dim. Mono to left, Mono to both. Phase reverse Right and Loudspeaker change over (L – R swap).
Reverse Talkback Loudspeaker.
Power
Power Supplies comprise:
2 x ZN1998 Audio, fan-assisted PSUs (3U).
1 x ZN1997 Logic, fan-assisted PSU (3U).
1 x ZN1705 Convection cooled phantom power PSU (2U).
Miscellaneous
On stand with deep script area.
Connectivity is via EDAC connectors. All balanced connections.
Full service manual and interface schedules.
► Additional information can be found in the Calrec M Series Outside Broadcast Consoles Sales Brochure » http://www.cmlaudio.com/data/Calrec/Calrec-M-Series-Outside-Broadcast-Consoles-Sales-Brochure.pdf