Discrete Opamp Open Design

[bcarso]

Originally Posted by bear

Brad, sorry I should have said "that I had not seen elsewhere...". Doesn't mean that this stuff was not out there, just that sitting in relative isolation, I had not found it via various pseudo random efforts to locate this sort of information - this was before the internet as we know it today... that meant trips to various university libraries and bookstores - many of the texts were even back then in the >$100 range, which was daunting. I persuaded someone to make me a photocopy of Grebene, because I couldn't swing the bux for even a used copy... whaaaaa!

_-_-bear

I know the feeling. Some out-of-print texts and monographs are holding their value extremely well. That Grebene you mentioned is available as shown in the reissue Classics series, but isn't cheap --- and it's softcover!

Just checked bookfinder.com for Ed Cherry's Amplifying Devices and Low-pass Amplifier Design, a generally reliable book and one with a unified treatment of hollow-state and sand-state, from 1968. A somewhat beaten-up ex-library copy is offered for nearly $200, while a Canadian seller helpfully offers one for a mere ~3500. Of course it can hardly cover everything, and much of it is very dated, but fundamentals are pretty much timeless. They will pry my copy from my cold dead hands

[scott wurcer]

Originally Posted by jneutron

Is the PMI SSM2210 still being made?

jn

Yes, and THAT makes matched NPN's and PNP's too.

[RNMarsh]

Originally Posted by scott wurcer

The crossover is an "S" shaped voltage discontinuity across the output stage. Inside the loop the current needed in the compensation capacitance is the derivative of this voltage or a "hump" at zero crossing which gets larger as the dV/dT increases (higher input frequency). The more bias (also degeneration tricks) will reduce the size of the discontinuity in the first place.

Thank you. Are there any degeneration tricks that you would be allowed to show/teach?

[john curl]

Look everyone, I KNOW how to design audio gain modules, already. I am only trying to help out with some 'new' (typically 40 years old) ideas. Have fun, but in my opinion, you won't get farther at this rate, than what is already out there.

[Wavebourn]

Originally Posted by RNMarsh

Thank you. Are there any degeneration tricks that you would be allowed to show/teach?

As I proposed before, close local feedback before the buffer. If you want to keep high DC gain for better output zero control at the buffer's output when global feedback is closed, use series R-C for this local feedback path.

[scott wurcer]

Originally Posted by RNMarsh

Thank you. Are there any degeneration tricks that you would be allowed to show/teach?

The one I was thinking of is well known (Bernie Oliver?), just the right drop on the emitter degeneration resistors (output devices) will do some first order distortion cancellation. I don't remember the exact numbers.

[bcarso]

Originally Posted by jneutron

Do they still make the LM 194/294/394?

Is the PMI SSM2210 still being made?

jn

I think there is still some NOS on 194 but the word on the street is that you can do as well with existing monolithic duals for about every parameter (the MAT series?).

Don't know about the 2210.

Also not to be forgotten is THAT's arrays, (EDIT: just noticed SW's post crossed in cyberspace) which have emphasized log conformance. I wish they could do something with higher beta though. I spoke to them about this and was told that they had played around with attempts in that direction, to find that breakdown voltages were all over the map.

I wish I could afford to fund some duals based around the BF862.

The obsolete parts problem reminds me of the story about the Arnoux Associates design for Infinity (the Harman company) of a switchmode power amp for automotive aftermarket. The product was announced about three years running at Winter CES, and then never shipped until after that. The late Mack Turner was the designer, and he swore that the only comparator that worked for the pulse-width modulator was a part that had been discontinued by TI. Steve Dove and I joked about this a bit, perhaps somewhat naively, but Infinity took the admonition seriously, and with the financial resources available then, commissioned a run of devices. I think the tab was over 100k

The saddest part of the story was that when the amp was finally shipping, it was somewhat expensive for the output power, and the market, consisting mostly of "kids" pimping out their vehicles, was almost nonexistent: the thing didn't look like it could do serious power, no big heatsinks etc. I heard that there was a final attempt to sell it by simply putting it in a larger chassis! Still no dice.

As I've said before, You can't make this stuff up --- no one will believe you.

[EUVL]

John is my hero.

Love this one :

Discrete Opamp Open Design

and this one :

Discrete Opamp Open Design

But stupid question, where to put the -Vin on the second one ?


Patrick

[Wavebourn]

Originally Posted by bcarso

The saddest part of the story was that when the amp was finally shipping, it was somewhat expensive for the output power, and the market, consisting mostly of "kids" pimping out their vehicles, was almost nonexistent: the thing didn't look like it could do serious power, no big heatsinks etc. I heard that there was a final attempt to sell it by simply putting it in a larger chassis! Still no dice.

I am going to try today in the field Behringer 2x300W per channel amp, class D with SMPS. I paid $185 for it including shipment. No heatsinks (almost), just a fan.

[dirkwright]

Originally Posted by bcarso

I think there is still some NOS on 194 but the word on the street is that you can do as well with existing monolithic duals for about every parameter (the MAT series?).

Don't know about the 2210.

Also not to be forgotten is THAT's arrays, (EDIT: just noticed SW's post crossed in cyberspace) which have emphasized log conformance. I wish they could do something with higher beta though. I spoke to them about this and was told that they had played around with attempts in that direction, to find that breakdown voltages were all over the map.

I wish I could afford to fund some duals based around the BF862.

Linear Systems makes duals:
Linear Integrated Systems - Products