This is an amp with another kind of symmetric design.
Has anyone ever seen this idea?
Has anyone ever seen this idea?
An externally hosted image should be here but it was not working when we last tested it.
It has a diamond input and another diamond for feedback and a triple darlingotn output. I say it is a great circuit and works well.
Yes, I have seen it before, it's buffered Sansui DD (diamond drive), but I think it will not work outside simulator. Bases of Q9-11 needs higher then 1Vbe + voltage over 10R resistor, something like 2Vbe will be more appropriate in this schematic.
See posts for inspiration
http://www.diyaudio.com/forums/anal...screte-opamp-open-design-199.html#post3229102
http://www.diyaudio.com/forums/anal...screte-opamp-open-design-199.html#post3229421
See posts for inspiration
http://www.diyaudio.com/forums/anal...screte-opamp-open-design-199.html#post3229102
http://www.diyaudio.com/forums/anal...screte-opamp-open-design-199.html#post3229421
Last edited:
It is patented by comlinear or is it National circa 1991 ?? I will have to take a look again. Used by Marantz in their top of the range.
It has excellent performance, wide bandwith with very high slewrate. Btw it is voltage VFB topology but it easily outperforms 99 % of all CFB topologies.
See Lm6172 as a example.
It has excellent performance, wide bandwith with very high slewrate. Btw it is voltage VFB topology but it easily outperforms 99 % of all CFB topologies.
See Lm6172 as a example.
Last edited:
Had sketched this circuit on a rough piece of paper last night as I was having dinner and wanted to open a thread to discuss it and why it may be overkill 🙂 . The focus of this input stage involves Cob compensation, so cascodes are required on Q9 and Q11. Usually R1 is greater than R2 for valid reasons. You have also made it voltage feedback. Very nice. 🙂
Why would cascodes be required ?? They would then also be required on Q11 and Q12, the circuit is symmetric.
The bandwith and slewrate achievable with this circuit makes cascoding of little relevance.
The bandwith and slewrate achievable with this circuit makes cascoding of little relevance.
Why would cascodes be required ?? They would then also be required on Q11 and Q12, the circuit is symmetric.
The bandwith and slewrate achievable with this circuit makes cascoding of little relevance.
Q10 and Q12 yes. was talking about its native form. You will be surprised that this circuit if used as an audio amplifier rather than a video amplifier is easily beaten by a more elegant solution 🙂
Yes, I is voltage feedback design. But I use a differential amplifier for controlling the current control and thus unite the advantages of both principles.
I've never seen a differential amplifier in which the signal was picked up at the emitter.
All circuits that you have shown me so far are similar except in the details of the differential amplifier.
But nice discussion so far 🙂
I've never seen a differential amplifier in which the signal was picked up at the emitter.
All circuits that you have shown me so far are similar except in the details of the differential amplifier.
But nice discussion so far 🙂
Q10 and Q12 yes. was talking about its native form. You will be surprised that this circuit if used as an audio amplifier rather than a video amplifier is easily beaten by a more elegant solution 🙂
Beaten in what why, performance ?? I very much doubt it.
If you are going to say the way it sounds, well Ill repeat a old Arab phrase (thanks to Wahab) here you wont like too much 😀
Beaten in what why, performance ?? I very much doubt it.
If you are going to say the way it sounds, well Ill repeat a old Arab phrase (thanks to Wahab) here you wont like too much 😀
For audio above a certain threshold for slew rate bandwidth and THD is moot
There is no question about symmetry. I find the symmetric solutions to be the best. I have no doubt the circuit as shown in post #1 is a good one.
This is the simulated bandwith, open and closed loop:
An externally hosted image should be here but it was not working when we last tested it.
Moschfet
I really like it, have something very similar in the works but at my rate I suspect 2014.
Elvee's thread "Is the CFB topology superior, and why?" led me to it but that thread has been somewhat up-rooted.
I have cascoded input transistors but for reasons of power dissipation and component selection flexibility.
Not that you need more gain but have you considered using those symmetrical collector currents from the feedback buffer?
Thanks and good Luck
-Antonio
I really like it, have something very similar in the works but at my rate I suspect 2014.
Elvee's thread "Is the CFB topology superior, and why?" led me to it but that thread has been somewhat up-rooted.
I have cascoded input transistors but for reasons of power dissipation and component selection flexibility.
Not that you need more gain but have you considered using those symmetrical collector currents from the feedback buffer?
Thanks and good Luck
-Antonio
Last edited:
Hi Antonio,
I only want to publish my idea. Maybe someone can take parts of it or has a continuing idea.
It's only a buffer. There is no voltage gain, so I don't need it. And it is going to grow too complicated with so many cascoded stuff...Not that you need more gain but have you considered using those symmetrical collectors from the feedback buffer?
I only want to publish my idea. Maybe someone can take parts of it or has a continuing idea.
Moschfet
Indeed, again I like your idea.
(Edited my post as I meant to say collector currents).
Thanks
-Antonio
Indeed, again I like your idea.
(Edited my post as I meant to say collector currents).
Thanks
-Antonio
Harrison
I can't tell if you like this topology or not, or just dont favor cascode stages.
What would you recommend to further this approach (with it's given linearity and potential for a lot of gain)?
Thanks
-Antonio
I can't tell if you like this topology or not, or just dont favor cascode stages.
What would you recommend to further this approach (with it's given linearity and potential for a lot of gain)?
Thanks
-Antonio
I have already build a similar circuits and it works well (only the drive don't like clipping.)
The slewrate is not too bad:
Green is output at 8ohm
Blue is input(1.5V), 1nsec rise an fall.
The slewrate is not too bad:
An externally hosted image should be here but it was not working when we last tested it.
Green is output at 8ohm
Blue is input(1.5V), 1nsec rise an fall.
- Status
- Not open for further replies.
- Home
- Amplifiers
- Solid State
- alternativ symmetric amp