For those that are interested in common-mode distortion in JFET-input op amps, and why it is dependent on source impedance for some op amps, I've written an article on this topic: http://www.ti.com/lit/an/slyt595/slyt595.pdf
Full disclosure: I am an applications engineer for TI Precision Analog in Tucson AZ. The article shows measurements on TI op amps, but the principles are applicable to JFET input op amps from most manufacturers.
Enjoy!
Full disclosure: I am an applications engineer for TI Precision Analog in Tucson AZ. The article shows measurements on TI op amps, but the principles are applicable to JFET input op amps from most manufacturers.
Enjoy!
Udo, the load was the 100kOhm input impedance of the Audio Precision SYS-2722 that I was using.
AndrewT, I like that idea and will add it to my profile. I will probably keep stating my job overtly because some may view my role at TI as a potential source of bias in my responses. I will try to be as non-biased as possible in any contributions I make to this forum. There's a lot of GREAT op amps out there, and a true engineer appreciates good engineering!
AndrewT, I like that idea and will add it to my profile. I will probably keep stating my job overtly because some may view my role at TI as a potential source of bias in my responses. I will try to be as non-biased as possible in any contributions I make to this forum. There's a lot of GREAT op amps out there, and a true engineer appreciates good engineering!
its good to see more on the issue, it is long known but new generations do need refreshers
new product, and process measurements are quite helpful to all
unfortunately we have to rely on 3rd parties to make even reasonably complete op amp measurements
SG-Acoustics · Samuel Groner · IC OpAmps
(see his input impedance linearity data in the op amp distortion pdf)
the older junction isolated process op amps can be improved with composite amplifier techniques
http://www.analog.com/static/imported-files/application_notes/742022599AN232.pdf
Supply Bootstrapping Reduces Distortion In Op-Amp Circuits | New operational amplifiers optimized for high-performance audio and ultrasound applications combine extremely low total harmonic distortion plus noise (THD+N), -130 dB, with large output vo
new product, and process measurements are quite helpful to all
unfortunately we have to rely on 3rd parties to make even reasonably complete op amp measurements
SG-Acoustics · Samuel Groner · IC OpAmps
(see his input impedance linearity data in the op amp distortion pdf)
the older junction isolated process op amps can be improved with composite amplifier techniques
http://www.analog.com/static/imported-files/application_notes/742022599AN232.pdf
Supply Bootstrapping Reduces Distortion In Op-Amp Circuits | New operational amplifiers optimized for high-performance audio and ultrasound applications combine extremely low total harmonic distortion plus noise (THD+N), -130 dB, with large output vo
Last edited:
The substrate bootstrapping technique is another possible solution, I believe this is also the approach Douglas Self took in his Small Signal Audio Design book for dealing with this distortion on the TL072. My one caution to circuit designers evaluating that approach is that in some circuits, bootstrapped supplies can be a recipe for latch-up during start-up and shutdown. I haven't used it for distortion reduction, but I did use it once to build a high compliance voltage Howland Current Pump (http://www.ti.com/lit/ml/slyy054/slyy054.pdf)
I was hoping someone would catch this! A process improvement was made after the original characterization of the OPA164x family of parts. And although other parameters were measured to confirm the effect of the process improvement, THD+N was not re-characterized. So basically, the OPA164x op amps are better than their datasheet would suggest.
This is a valid statement, integration always forces some kind of engineering compromise (it's hilarious in 2014 to put op amps in the category of integration, but it is still an integrated circuit). There are the engineering compromises that come from limiting the design to a single semiconductor process (or a few in the case of multi-chip modules) and the limitations of the process itself. But even worse are the compromises imposed by the market place. It's terrible when money gets in the way of absolutely cutting edge performance.
No disrespect to you Mr. Caldwell, but I don't care for the sound of the OPA1642, either with or without a buffer after it. I tried with and without a TPA6120 after it.
It sounds a little too sibilant and crisp to me.
But...some people may love the sound of it and describe it as very "detailed".
I'll try some of the OPA16xx bipolar-input op-amps.
Thank you for the informative article!
No disrespect to you Mr. Caldwell, but I don't care for the sound of the OPA1642, either with or without a buffer after it. I tried with and without a TPA6120 after it.
It sounds a little too sibilant and crisp to me.
But...some people may love the sound of it and describe it as very "detailed".
I'll try some of the OPA16xx bipolar-input op-amps.
Thank you for the informative article!
Haha, it's not disrespectful to me, sound is a very personal thing. Hopefully you verified that the circuit was stable and there were no overt causes of distortion (excessive loading, clipping on the output or input, high power supply impedance, etc.) or wild deviations in the frequency response. If everything checks out and you just don't like it, to each their own!
There are some very good OPA16xx op amps, although I sit next to the designer of the OPA132/134 and he'd probably claim that his is the best audio op amp BB/TI ever did.
Who is the designer of the OPA132/134 op-amps?
That series is quite good, but overall I prefer the sound of bipolar-input op-amps.
Shoot, even the old LM833 sounds quite good to my ears.
He probably wouldn't like me posting his name on here without his permission. But I will say that if you ask that question on our e2e forum (e2e.ti.com) you'll probably get an answer