OpAmp Measurement Series

[Larrchild]

I'm bummed you omitted the Philbrick GAP/R K2 SERIES, but other than that, Samuel, your comprehensive paper is the talk of many forums.
Great work!

 

[bcarso]

About the highest compliment I can bestow: I'm already frequently referring to the data.

I found the quite-revealing data on the new National Semi parts particularly important to a high-end design I'm in the midst of---in fact the data drove me to do a driven-rail design, a la Funasaka/Kondou, while using a similar National part (the LME49710).

 

[Underthebigtree]

Phenomenal work, Sam.

Already useful for me. I recently bought a Soundcraft Ghost, and have been wading through the internet for weeks trying to figure out what to do to upgrade it sonically. The "upchip" market is a big thing, and finding unbiased information to help me in my real-life problem, knowing the author does not have a vested interest in a particular opamp, helps me greatly.

The Ghost is loaded with TL072's and NE5532's. What I took away from your paper:

1)The TL072 stinks. The OPA 2132 is a possible substitute, though it devours twice the current and would thus require substantially upgrading the console's power supply.

2)The NE5532 is good for the money, though there are better options out there. An excellent subsitute would be the Linear Technology LT1469.

Thank you very, very much. And if anyone reading this has another suggestion for a TL072 substitute that has similar quiescent current specs, I'm all ears.

Best,

Nick Peck

 

[pstamler]

I took my copy into the class I teach, "Understanding Audio Specifications". By great good fortune, it had arrived right when I was going to explain to them what distortion was about and why a THD figure wasn't necessarily going to tell you whether something would sound good. They were blown away that they were working with material published only nine days before. Thanks, again, for sharing this with the world.

Peace,
Paul

 

[Samuel Groner]

Thanks for your comments--I'm glad the work finds practical use!

And if anyone reading this has another suggestion for a TL072 substitute that has similar quiescent current specs, I'm all ears.

I don't have a real answer to that yet; LT1057 might be a possible option but I haven't tested this amplifier yet. In any case you could use OPA2134 instead of OPA2132 as they appear to be identical except for tighter DC specs for the later. Does nothing about current draw but at least should be cheaper--and where a TL072 worked there hardly can be a demand for good DC precision.

Samuel

 

[tv]

For 071 (single), check out this one (AD820):
http://www.analog.com/en/amplifiers-and-comparators/operational-amplifiers-op-amps/ad820/products/product.html

but slewrate is slightly worse than 061!! (not as in 071 by far)

-edited for clarity-

 

[ClaudeTucker]

Walt Jung wrote:

2) These families of distortion plots can give a clever circuit designer lots of info as to what to do (or *not* to do) inside. Like for example, use JFETs, but watch out for CM distortion due to nonlinear C, a killer mechanism with high-Z drive.

Bcarso wrote:

I found the quite-revealing data on the new National Semi parts particularly important to a high-end design I'm in the midst of---in fact the data drove me to do a driven-rail design, a la Funasaka/Kondou, while using a similar National part (the LME49710).

Has anyone here previously cited this Electronic Design Ideas for Design article from 9.11.08?

http://electronicdesign.com/Articles/Print.cfm?ArticleID=19609

He cites Walt twice.

 

[bcarso]

Hadn't seen that Claude.

The author seems unaware of Funasaka/Kondou, although he has plenty of company.

 

[ClaudeTucker]

The author seems unaware of Funasaka/Kondou, although he has plenty of company.

Yes, they made a 4558 into a 300 V/us high slew rate op amp in 1982.

 

[bcarso]

But it looks like maybe the Sandman ref. predates F&K, although I don't have it at hand...

 

[ClaudeTucker]

Anybody got:

A.M. Sandman, “Reducing Amplifier Distortion,” Wireless World, Vol. 80, No. 1466, pp. 367-371, Oct. 1974.

Here's another cite which points to Sandman:

http://www.angelfire.com/az3/dimitri/images/riaa.pdf

In it they attribute Funasaka to Sandman:

The method [19] for substantial improving of series feedback connection performance is realized in JVC phonograph preamps [20]. It is called FFPS and applies a floating power supply with feedforward and bootstrapping.

 

[bcarso]

So now two different groups have stolen my idea years before I had it! (see my ill-informed 1997 AES paper).

The nerve of some people!!

 

[JohnRoberts]

Definitive proof of time travel methinks,,,

JR

 

[ClaudeTucker]

Computer-Controlled Attenuator for Subjective Evaluation Systems

This one?

 

[Samuel Groner]

This one?

Yes.

I've seen the Danyuk article before but I have my doubts whether it is a particularly clever implementation. Funasaka/Kondou seems to be much better.

Samuel

 

[mikep]

thanks for posting that, claude

there is a lurking problem with almost every version of modulated rail or bootstraped cascode distortion reducing mechanism I have studied, ugly overload characteristics. Even if care is taken to keep the circuit from going into oscillation or outright blowing transistors, you still seem to always end up with at least spikes of current on the supply rails as protection diodes start to conduct. and there is the sudden shift in conditions caused when a rail driver clips and the voltage across the active device starts to change with the signal. imagine a capacitance multiplier for each rail, with the shunt capacitors connected to the output of a common mode voltage follower instead of ground. When the followers clip It doesnt cause gross distortion but it far from ideal. this is not easy to indicate with a clip LED and changes with gain.

I am of the belief that, for audio circuitry, a well defined clipping behavior is more than a good housekeeping type of little design detail. It really influences people's perception of the gear as a whole.

mike p

 

[pstamler]

Hi folks:

Last night I had the measurement gear set up, so I did a comparison of two 1/3-octave graphic EQs I had borrowed for tomorrow's concert gig. They were a Yamaha GQ1031BII and a Behringer Ultragraph Pro. I believe the Behringer uses something like 4560 opamps; I have no idea what the Yamaha uses. I've always thought the Behringer had something of a crispy-crunchy sound.

First I did a midband harmonic distortion test, 400Hz, +19dBu in and out. (For both units, this was about 0.5dB below where the clip light came on.) I used an oscillator I built a couple of decades ago, from one of Walt Jung's designs; it's quite clean. I ran each output through a 20dB pad into my CardDeluxe soundcard, recorded the results, normalized to -0.2dBFS, and viewed the spectrum in Adobe Audition. I did the screen grabs with "PrintScreen", a useful, free utility.

The Yamaha looked like this (you can see the image a lot more clearly if you click on it):

http://twin-x.com/groupdiy/displayimage.php?pos=-2677

And the Behringer looked like this:

http://twin-x.com/groupdiy/displayimage.php?pos=-2676

Note that the Yamaha's distortion products go up to 6th harmonic, while the Behr's go all the way to the edge of the audible band.

Next I did an intermodulation distortion test using 19kHz and 19.5kHz, mixed at equal levels. This signal came from the computer; since the CardDeluxe is set for nominal -10dBV levels in and out, the maximum output level for this signal was +10dBu. Since I didn't want to hassle with going into the computer and changing the switch (the CardDeluxe doesn't let you change the output level in software), I stuck with that. As it turned out, +10dBu was plenty.

The output and analysis hookup was the same. Here's what the Yamaha's spectrum looked like:

http://twin-x.com/groupdiy/displayimage.php?pos=-2675

And here's what the Behringer's looked like:

http://twin-x.com/groupdiy/displayimage.php?pos=-2674

I think we have the beginnings of an explanation for the crispy-crunchies.

Samuel, you mentioned that you were interested in possibly exploring the use of HF intermodulation distortion tests as a supplement to the incredibly valuable work you've already done. I'm posting these pics to suggest that I think this would be a most fruitful path.

Peace,
Paul

 

[skipwave]

Whoa. That's pretty striking. Thanks for sharing Paul.

Intriguing how the 400Hz distortion profile seems to show the Behringer circuit has some even order nulling going on. The second is a touch below the third, and the fourth is way down from the fifth. In contrast, the Yamaha has mostly third and falls off steadily up to the sixth.

 

[pstamler]

[quote author="skipwave"]Whoa. That's pretty striking. Thanks for sharing Paul.

Intriguing how the 400Hz distortion profile seems to show the Behringer circuit has some even order nulling going on. The second is a touch below the third, and the fourth is way down from the fifth. In contrast, the Yamaha has mostly third and falls off steadily up to the sixth.[/quote]

You're right about the Behringer. But the Yamaha is mainly 2nd, not 3rd. Also, each harmonic except the 2nd is significantly lower than the equivalent harmonic on the Behringer.

Peace,
Paul

 

[JohnRoberts]

+1 for multi-tone IMD, I like the real world aspect of in band stimulus and in band distortion artifacts.

The power of Sam's effort is that he has applied a common or standardized test criteria across so many devices, so we can compare apples to apples. These discussions don't diminish that effort.

JR