John Curl's Blowtorch preamplifier part II

[Bob Cordell]

I repeat again another point, low open-loop BW is meaningless. You can use positive/negative feedback to the null pins on many op-amps to make the Aol and 1/gmC break point almost anything you want (even make Aol switch sign!). I have posted pictures here on doing this. Try it PMA, you can make the 797 have 10^5 gain rather than 10^7 and lower the OLB 2 orders of magnitude. Unfortunately null pins are deprecated these days or you could take the AD825 and tweek it up to 10^7 Aol and see for your self it makes no difference.

Amen, Scott.

Cheers,
Bob

 

[Bob Cordell]

It should help with PIM, I agree. And that Stanford professor who vetted Ron Quan's paper came to his defense from people on your side. Did you know that?

Most things that help with PIM will help with AIM even more. Been down this dusty old road before.

Who was the Stanford Prof?

 

[scott wurcer]

Scott, you should have a talk with Dick Sequerra about this. He calls an op amp a 'reactance modulator'. Perhaps you can talk him out of it.

This of course has nothing to do with it, I gave you a simple experiment where anyone can adjust the Aol break point to say 10Hz on just about any op-amp with classic null pins. The open loop BW is an artifact of process, beta, Early voltage, etc.

Try it and listen, you could make the 797 have an open loop BW of 20kHz I'm surprised you're not running to the bench to try it.

 

[john curl]

Dick Sequerra says hi to you, Scott. And by the way, he agrees with me, and not you. '-)

 

[PMA]

Try it PMA, you can make the 797 have 10^5 gain rather than 10^7 and lower the OLB 2 orders of magnitude.

Thanks for the recommendation, I prefer to swap the AD797 directly for the ADA4898 and to compare them. I have already done it several times during the last year.

 

[scott wurcer]

Dick Sequerra says hi to you, Scott. And by the way, he agrees with me, and not you. '-)

On everything, wow?!

 

[scott wurcer]

Amen, Scott.

Cheers,
Bob

Right, I'll be at the ISSCC on my own coin this year, that's how bad things are, are you going?

 

[john curl]

He might agree with you about wine quality.

 

[scott wurcer]

He might agree with you about wine quality.

I doubt it. I'm just a contrarian.

 

[john curl]

Scott, I really don't know why we bicker like this. Why don't we both grow up, and leave each other alone? I realize that you don't believe in high open loop bandwidth as being important. Fine with me. However, I have Matti Otala, Mitch Cotter, Dick Sequerra, Ron Quan and my experience with audio designs on my side. Still, I don't have any solid proof that even a confirmed non-believer would accept, but my opinion is genuine and it has proven useful to me as a rough guide to audio design.
When I talk to Dick Sequerra about this, he is adamant about a number of things of this nature, and I respect him for opening up to me. He is a very private man. Still, he and I have tried to make quality audio, all our lives, and we hope to continue to do so. We have both been fairly successful at it as well. I have three of his designs, I have one of Matti Otala's, and I have a weekly conversation with Mitch Cotter, that is so far beyond what I can present here as to think that this is like high school and he is a post graduate of a university (which he is, of course). Yet I continue here, hoping to convey my opinions to interested parties, and debate with those who disagree, hoping to learn even more.

 

[Steve Eddy]

I doubt it. I'm just a contrarian.

There's a church for that now.

Tom's NU Heaven Trailer - YouTube

se

 

[john curl]

Well, Scott, at least you will attend the ISSCC conference. I could, but it is just too much money for too little that I can use. It wasn't always the case, of course.

 

[ThorstenL]

Scott,

I repeat again another point, low open-loop BW is meaningless.

OLBW alone is, but once we include gain it is very meaningful.

For example, I have this circuit here where I need a gain of 1000 and a bandwidth of 100KHz.

In addition I would like to satisfy Bruno's dictum of "no less than 30dB Feedback".

So I need a GBWP of how much? Well I make it 31.6GHz...

You can use positive/negative feedback to the null pins on many op-amps to make the Aol and 1/gmC break point almost anything you want (even make Aol switch sign!). I have posted pictures here on doing this. Try it PMA, you can make the 797 have 10^5 gain rather than 10^7 and lower the OLB 2 orders of magnitude.

So, can we tweak the AD797 to give me 31.6GHz GBWP?

Ciao T

 

[PMA]

You probably might like to use something else than the dominant pole compensation

 

[john curl]

2 pole compensation will work, but it is difficult to implement, and it probably has some other issues.

 

[PMA]

I would vote for CFB amp in Thorsten's case.

 

[zinsula]

I would vote for CFB amp in Thorsten's case.

Gain 1000x
I would fear the noise in that case.

 

[ThorstenL]

Hi,

I would vote for CFB amp in Thorsten's case.

Funny, that is PRECISELY what I used in this case. Noise was a little iffy but not bad. FOr fun, the PSU regulators for the CFB Amp where als CFB Amp's...

Ciao T

 

[PMA]

PSU regulators for the CFB Amp where als CFB Amp's...

Ciao T

Ich verstehe

The CFB amps are great, though a bit more difficult to design for low noise.

 

[ThorstenL]

Hi,

The CFB amps are great, though a bit more difficult to design for low noise.

Incidentally, I used IC's, less than 6dB noisier than a 2SK147. They where adequate for the job. With a 1K Feedback resistor the gain set resistor was only 1 Ohm, so it's contribution to noise was basically negligible...

Ciao T

PS, that Phono Stage did sound uncommonly good...