Samuel Groner's super opamp

AX tech editor
Joined 2002
Paid Member
In 2017 Samuel Groner, with Martin Polak, published a new composite opamp design: Journal of the Audio Engineering Society ENGINEERING REPORTS, Vol. 65, No. 5, May 2017: Low-Distortion, Low-Noise Composite Operational Amplifier.

Late last year I was looking for a very low distortion opamp and thought I'd give this a try. I laid out a small adapter PCB for a DIL08 foortprint, see pics. The result was all I expected, an opamp with distortion specs literally out of this world. I was not able to verify the claimed THD < 180dB (yes you read that correctly) but got numers at -160dB and better.
I'll post measurements in my next post.

Anybody here build this as well, experiences?

Jan
 

Attachments

  • groner composite opamp.PNG
    groner composite opamp.PNG
    14.5 KB · Views: 3,968
  • 20190506_093359.jpg
    20190506_093359.jpg
    262.5 KB · Views: 3,723
  • 20190506_095319.jpg
    20190506_095319.jpg
    262.9 KB · Views: 3,642
Last edited:
AX tech editor
Joined 2002
Paid Member
Here are some measurements with the AP SYS2722. The AP cannot directly measure that low, so I put the opamp in an inverting configuration with a 12 ohms resistor between the two inputs, giving a (measured) noise gain of 50.3dB.

So for all these measurements, you need to subtract another 50.3dB from the shown values. As I said, exceptional!

Jan
 

Attachments

  • G-P THD 1kHz 1k load 50.3dB NG 3.8Vo 32 avg.pdf
    101.1 KB · Views: 749
  • G-P THD 1kHz 1k load 50.3dB NG 7.5Vo 32 avg.pdf
    101.1 KB · Views: 340
  • G-P THD 1kHz 375R load 50.3dB NG 3.8Vo 32 avg.pdf
    101.2 KB · Views: 309
  • G-P THD 1kHz 375R load 50.3dB NG 7.5Vo 32 avg.pdf
    101.1 KB · Views: 367
Last edited:
Member
Joined 2011
Paid Member
Jan, thanks very much for this. It is extremely impressive.

If I remember correctly, Samuel Groner's website used to contain schematics for four or five different all-discrete opamps. One of them was a composite opamp made with all discretes having circa 180dB dc gain. But using Google and the Internet Wayback Machine, I can only find three of his discrete opamps today: "simple", "low noise", and "high voltage" (SOP, LOA, HOA abbrevs by Groner). Does anyone have a copy of the schematics and written descriptions of the others, especially the composite opamp? Would you please post them on diyAudio or, if not, please send me a PM.

Again congratulations and thank you.
 
Last edited:
AX tech editor
Joined 2002
Paid Member
Jan, thanks very much for this. It is extremely impressive.

If I remember correctly, Samuel Groner's website used to contain schematics for four or five different all-discrete opamps. One of them was a composite opamp made with all discretes having circa 180dB dc gain. But using Google and the Internet Wayback Machine, I can only find three of his discrete opamps today: "simple", "low noise", and "high voltage" (SOP, LOA, HOA abbrevs by Groner). Does anyone have a copy of the schematics and written descriptions of the others, especially the composite opamp? Would you please post them on diyAudio or, if not, please send me a PM.

Again congratulations and thank you.

Mark, I am not familiar with the others. If you have access to the AES library it's worthwhile to get this paper, but I shouldn't post it here, its (c)-ted.
Note that my measurements put the 2nd at about -165dB with 7.5V out on a 375 ohms load, making it even more fantastic.

Attached is Samuels' own measurement. Boring ;-)

Jan
 

Attachments

  • groner-copa.PNG
    groner-copa.PNG
    16.4 KB · Views: 3,097
Last edited:
In 2017 Samuel Groner, with Martin Polak, published a new composite opamp design: Journal of the Audio Engineering Society ENGINEERING REPORTS, Vol. 65, No. 5, May 2017: Low-Distortion, Low-Noise Composite Operational Amplifier.

Late last year I was looking for a very low distortion opamp and thought I'd give this a try. I laid out a small adapter PCB for a DIL08 foortprint, see pics. The result was all I expected, an opamp with distortion specs literally out of this world. I was not able to verify the claimed THD < 180dB (yes you read that correctly) but got numers at -160dB and better.
I'll post measurements in my next post.

Anybody here build this as well, experiences?

Jan

Interesting.

Both OPA211 and THS4031 are well known for very low distortions, but it is obviously the huge loop gain that is responsible for -180dB distortions.

Since I don't have access to the JAES paper, could you post a few details about:

a) Is this composite op amp unity gain stable?
b) Any results about the open loop gain and phase?
c) Any details about the response in capacitive loads?
d) How was -180dB distortion measured? Is -180dB @ 10KHz as suggested by the graph?
e) Is the -180dB number in inverting or non inverting configuration? That is, are the common mode distortions included or not?
 
AX tech editor
Joined 2002
Paid Member
Yes, those are the important questions, right. A text excerpt: "First and foremost, it should be appreciated that the novel composite operational amplifier is conditionally stable only. Sufficient phase margin is (except at very low frequencies) only available near the unity open-loop gain frequency. The composite operational amplifier must thus be operated with a unity loop gain frequency close to the unity open-loop gain frequency. In many cases this implies the use of an additional feedback capacitor instead of a purely resistive feedback network." I had no stability issues with 22p across Rf. There is also a diode clamp on output of the 1st opamp to avoid it slamming into the rails at switch-on and other transients.

Jan
 

Attachments

  • groner-4.PNG
    groner-4.PNG
    19.9 KB · Views: 2,858
Last edited:
Disabled Account
Joined 2004
A true opamp has no Gnd connection.
So how did you solve that ?

His discrete op-amps had ground connections, I would assume there is some care in layout here.

Reminds me of Jim William's Super Gain Block.

Analog Circuit Design Volume 2: Immersion in the Black Art of Analog Design
Fig. 36.95

Our modular DAC/AD/isolator guys were making those with TL072's 40yr. ago.
 
AX tech editor
Joined 2002
Paid Member
A true opamp has no Gnd connection.
So how did you solve that ?
To the negative rail ?


Patrick

Patrick, the adapter I made has the gnd routed to pin 1 of the DIL08, which I use in the app circuit.
I included some supply decoupling on the adapter between V+ and V- but not to ground to avoid even tiny ground loops.
At -180dB, every electron counts ;-)

And I got pretty good at manually soldering 0805 size parts!

Jan
 
Last edited:
If I remember correctly, Samuel Groner's website used to contain schematics for four or five different all-discrete opamps. One of them was a composite opamp made with all discretes having circa 180dB dc gain.
Here is attached Groner's NDFL DOA (Nested Differential Feedback Loops Discrete Operational Amplifier) schematics and the list of his discrete opamps in 2011.
 

Attachments

  • Groner_NDFL-DOA-1_r1.pdf
    50.1 KB · Views: 1,175
  • Groner_DiscreteOpAmps.pdf
    49.5 KB · Views: 1,090
Last edited:
Member
Joined 2009
Paid Member
Anybody here build this as well, experiences?

Jan

Yes, I have built a version of it. Actually I made a dual version of it on a 10 x 15 mm PCB. Unfortunately I have not found the time to test it yet. Too many projects in parallel I guess :eek:

There are three additional terminals. One must be soldered to ground. The other two are for the clamp circuits. The connection points for these will depend on the external circuit.

I had to use 0402 for some of the components. I prefer not to, but in this case it was needed to keep it small. There are some components on the other side of the PCB as well. The PCB is a 4-layer PCB.
 

Attachments

  • Samuel Groners super opamp_190513.png
    Samuel Groners super opamp_190513.png
    393.3 KB · Views: 1,604
Account Closed
Joined 2010
In 2017 Samuel Groner, with Martin Polak, published a new composite opamp design: Journal of the Audio Engineering Society ENGINEERING REPORTS, Vol. 65, No. 5, May 2017: Low-Distortion, Low-Noise Composite Operational Amplifier.

Anybody here build this as well, experiences?

Jan
Common guys, you can't be real!
It's 1991 Burr Brown app:
http://www.ti.com/lit/an/sboa002/sboa002.pdf


I have it working in my current headphones amplifier based on this design:
TPA6120 Headphone Amplifier - COMPOSITE TOPOLOGY - DIY AUDIO BLOG, AUDIO WORKSHOP
and modified by a friend of mine for smd input opamp.
I use tpa6120 and opa1642.Of course that opa211 is better at anything just it was too expensive for me.
 

Attachments

  • IMG_0388.jpg
    IMG_0388.jpg
    966.7 KB · Views: 1,423
  • IMG_0391.jpg
    IMG_0391.jpg
    1,002 KB · Views: 1,377
AX tech editor
Joined 2002
Paid Member
Yes, I have built a version of it. Actually I made a dual version of it on a 10 x 15 mm PCB. Unfortunately I have not found the time to test it yet. Too many projects in parallel I guess :eek:

There are three additional terminals. One must be soldered to ground. The other two are for the clamp circuits. The connection points for these will depend on the external circuit.

I had to use 0402 for some of the components. I prefer not to, but in this case it was needed to keep it small. There are some components on the other side of the PCB as well. The PCB is a 4-layer PCB.

I have included two BAF199 diodes on the PCB (2 sided) for the clamps. They are really needed, without them the amp starts squegging (remember squegging?).

Jan