Discrete Opamp Open Design

[stinius]

Originally Posted by bcarso

I also should do some sims of settling time after the renewed interest in that topic in another thread. I suspect the compensation needs to be tweaked to instill better single-pole-approximating behavior.

Brad
IMHO no need for that. One of the good things by a discrete design is that the users have access to all the nodes, and can design their own prefered compensation.

BTW: Billy Budd
Did I miss some interesting posts?
Cheers
Stein

[bcarso]

Originally Posted by RNMarsh

What was the s/n and harmonic et al data from the sim? Finding good compliments is a problem and guess it will continue to be.

IF matched jFETs are used without source resistance, the H2, H3 is very, very low and higher harmonics are virtually non-existant. [notice John didnt use source resistors - makes matching more critical but THD and Noise is lowest - purist approach]

But as soon as source resistance is added, you might as well go to bipolars as the distortion isnt lower and H5,7.9th can be made so low as hard to measure with the best equipment available. But bipolars need extra circuitry and size, cost grows and it is harder to do without a dc servo appendage. The end result isnt any better but it is easier to make. -Dick

And there's that prevalent anti-bipolar bias. I was cheered to see that Jim Boyk (with Gerald Sussman, who undoubtedly did most of the math-and-modeling heavy lifting in the article) finally had some good things to say about them in a paper about IM distortion and feedback. Remember that this is the guy who had T shirts made that said "Digital finishes what the transistor began", and in his class at Caltech had students build simple transistor and tube gain stages and judge which one sounded better.

[EDIT] And Boyk and Sussman point out that matched JFETs as, in their example, a push-pull complementary output stage, have distortionless performance. They do acknowledge that their FET models are as ideal square-law devices, with a host of other real attributes ignored. The contact, or as PRR might call it, "dead" resistance, that's not involved in the transistor action, is enough to spoil this "perfection".

[bcarso]

Originally Posted by ticknpop

1971 Petrus? I'll be right over, that will remove the question as to whether to drink it or sell it so you can get on with other things. I'll leave 2SJ74s on the floor so you'll know it was me....

It's safely in secured subterranean storage for some time now, along with a few other treasures, although my collection is doubtless dwarfed by those of some of the other patrons with much higher incomes.

SY knows where it is, more-or-less, but he's not too close and doesn't have a key

[scott wurcer]

Originally Posted by bcarso

[EDIT] And Boyk and Sussman point out that matched JFETs as, in their example, a push-pull complementary output stage, have distortionless performance. They do acknowledge that their FET models are as ideal square-law devices, with a host of other real attributes ignored. The contact, or as PRR might call it, "dead" resistance, that's not involved in the transistor action, is enough to spoil this "perfection".

You're out of context here a little, their example was MOSFET's in extreme class A, gates actually pushed outside the rails by ideal sources, Vgatedrive >> VT at all outputs. JFET's can't do that.

Some other comments about distortion and degeneration made in the last few posts don't take into account input voltage compliance. Boyk and Sussman use .005V for bi-polar vs .150V for JFETs to put them on equal ground for distortion. This difference is the whole basis of JFET vs bi-polar performance with respect to slew rate vs gm and BW.

1948 Graham's Port is my last precious bottle.

[bcarso]

Originally Posted by scott wurcer

You're out of context here a little, their example was MOSFET's in extreme class A, gates actually pushed outside the rails by ideal sources, Vgatedrive >> VT at all outputs. JFET's can't do that.

Some other comments about distortion and degeneration made in the last few posts don't take into account input voltage compliance. Boyk and Sussman use .005V for bi-polar vs .150V for JFETs to put them on equal ground for distortion. This difference is the whole basis of JFET vs bi-polar performance with respect to slew rate vs gm and BW.

1948 Graham's Port is my last precious bottle.

Thanks Scott. I didn't have the paper handy and had forgotten they were using DMOS.

To go with the Petrus if it is ever drunk, I've a second bottle of one drunk years ago, a 1959 Yquem. When I drank the first Yquem, circa 1982, it was preceded by the top wine of my experience to date, a 1945 La Tache, which had no right to be in such good shape. I later noticed that Broadbent accorded both the La Tache and the d'Yquem six stars (five being his conventional maximum).

Three of us drank the La Tache and half of the Yquem, which I siphoned the other half of into a half-bottle for future consumption. Well, that resolve lasted about two more days.

[Wavebourn]

I am not a vine expert, but once in 1986 opened a bottle of 1946 Port from Massandra. I have to admit, the taste was from first to the last sip, as if I would drink a whole bucket of such vine with pleasure.
The bottle was flatter on one side: glass flows.

[ticknpop]

If you store all your precious drops well then your undeserving and unappreciative nephews will have something nice to drink at your wake. Just like the box of jfets their crazy uncle squirled away that they'll toss in the bin........

[RNMarsh]

I am wondering whether to keep and someday use the rare jFETs or just throw them away -- I have a stash of 2SK389, K146, K147, J72, J73, J109 and a whole bunch of 170/74 all grouped by Idss. And, some other types which are great but obsolete now. Cant give them to the undeserving family members. I know what they would do with them. What would you do with them?

[Wavebourn]

Originally Posted by RNMarsh

I am wondering whether to keep and someday use the rare jFETs or just throw them away -- I have a stash of 2SK389, K146, K147, J72, J73, J109 and a whole bunch of 170/74 all grouped by Idss. And, some other types which are great but obsolete now. Cant give them to the undeserving family members. I know what they would do with them. What would you do with them?

I would use them as current sources.

[scott wurcer]

Originally Posted by ticknpop

Just like the box of jfets their crazy uncle squirled away that they'll toss in the bin........

I give my extra stuff to MIT (Jerry Sussman actually) I doubt much gets used by students anymore. In another decade or so who knows.