Some Dual Op Amp Distortion Signatures

Where we discuss new analog design ideas for Pro Audio and modern spins on vintage ones.
Post Reply
User avatar
mediatechnology
Posts: 5444
Joined: Sat Aug 11, 2007 2:34 pm
Location: Oak Cliff, Texas
Contact:

Some Dual Op Amp Distortion Signatures

Post by mediatechnology »

I've been revisiting the low distortion oscillator and wanted to see if "helper" transistors would allow a 5532 to have the same open-loop gain performance as an LME49720. Viktors Mickevics seems to think, and I agree, that the LME49720 is about the only unassisted or non-composite op amp that has enough open loop gain (140dB at DC) for the job. More on that later...

Along the way, while comparing an "assisted" op amp to a "naked" op amp I started doing null testing of a basic op amp gain of -1 inverter. While I had the circuit running and in null I switched op amps for the fun of it.

The DUT op amp had Rfb=10K; Rin=10K; Cfb=0; Rl=2.5KΩ.
Fin was 2 kHz at -6 dBu.

The generator, a DAC output buffered by an op amp for 0Ω source impedance, was passively nulled with the DUT output by a 15T 5K trim pot.
Null testing allows the THD components of the generator, along with the fundamental, to be attenuated to make the generator "distortionless" and expand the dynamic range of the FFT measurement.

The DAC buffer drives both the DUT op amp and the nulling trim pot at one end.
The other end of the pot is connected to the DUT output.
The trim pot wiper, which is the null point containing the THD contribution of the DUT, was connected to a +40 dB gain stage and the output fed to an A/D for FFT.

The FFT was 96 kHz, 24 bit, 32 kpts with a 5 sample average.

In the FFTs below, the generator 2 kHz input fundamental, and its harmonics, are nulled by about -50 to -60 dB.
The displayed harmonics above 2 kHz are the op amp's contribution.
If a peak reads -110 dBFS it's actually about 40 dB less at -150 dBFS due to gain after the null point. (0dBFS = +15 dBu.)

This is not meant to be an exhaustive definitive scientific academic review of op amps but simply a comparative examination of their THD signatures. I grabbed what op amps I had within about 3 feet of me for this experiment.

What was particularly interesting is comparing various vendors' 5532s THD signature versus date code. Keep in mind that the sample sizes are very small. Where I had a significant quantity I did check more than one of the same date code and they appear to be fairly consistent.


I've shown the date code. Some date codes are in YYWW format. Others are not decipherable and their estimated purchase dates were used. Modern parts have been purchased in the last 3-5 years.

The LME49720 and OPA1612 Bipolar Op Amps

TI LME49720
Image

TI OPA1612
Image

NE5532 Varieties

JRC NE5532 1992 Date Code
Image

JRC NE5532 Modern
Image

ON Semiconductor NE5532 2006 Date Code
Image

TI NE5532AP 1980 Vintage
Image

TI NE5532P 1990s Vintage
Image

Signetics NE5532 1989 Date Code
Image

Signetics NE5532 1992 Date Code
Image

Exar NE5532P 1984 Date Code
Image

Raytheon RC5532 1980s Vintage and Unwanted
Image

Other Duals

ST MC33078 Modern (PNP Input similar to an LM833)
Image

NJM2114 "Turbo" NE5532
Image

NJM2068DD Popular with OEMs. Not commonly used in DIY but hits the voltage/current noise "sweet spot" for MM phono carts.
Image

NJM4556AD Point of reference.
Image

FET-input Op Amps

Burr Brown OPA2604 1991 Date Code
Image

TI OPA2604 Modern
Image

TI OPA1642A
Image

TI OPA2134
Image

TI TL072A Point of reference.
Image

Related reading: viewtopic.php?f=12&t=668
Post Reply