Johan,
Its not a variation of the "Blowtorch Principle" (when I did this
circuit, I didnt even know Mr. Curl
). It is more inspired by Mr.
Pass´ Supersymmetry Patent and the work of Erno Borbely. This
particular circuit (with kind of discrete servo around the 2SJ109)
is just an idea. Jam influenced me to redesign a working circuit,
similar but not complementary, with opamp servo, folded
cascode, X Feedback, active CCSes instead of a mirror resistor in
the folded cascode.
There are lots of variations of the idea to use both legs of a diff
amp: Pass, Curl, Borbely, Hansen et al.
Uli
Its not a variation of the "Blowtorch Principle" (when I did this
circuit, I didnt even know Mr. Curl
). It is more inspired by Mr. Pass´ Supersymmetry Patent and the work of Erno Borbely. This
particular circuit (with kind of discrete servo around the 2SJ109)
is just an idea. Jam influenced me to redesign a working circuit,
similar but not complementary, with opamp servo, folded
cascode, X Feedback, active CCSes instead of a mirror resistor in
the folded cascode.
There are lots of variations of the idea to use both legs of a diff
amp: Pass, Curl, Borbely, Hansen et al.
Uli
You are correct Uli. There are only so many ways to make a circuit, and many will look similar. It is the same with auto engines. Could you tell the important differences between them with just a picture?
The important thing is to make the most linear circuitry possible, so that feedback is not very important or even can be reduced to almost nothing. This seems to make audio circuits sound better. This is true with tubes, transistors, or fets.
In general, tubes are the most linear individual active devices, followed by fets, and then bipolar transistors. The problem with tube circuits is that they have to be coupled between stages with caps or transformers. Solid state circuits can be direct coupled, but they are always prone to more nonlinearity, both static (from Gm variation) and dynamic (from nonlinear capacitance with voltage). This is where solid state circuits can be optimized more than is typically employed, by consideration of the non-linear capacitance and its minimization.
The important thing is to make the most linear circuitry possible, so that feedback is not very important or even can be reduced to almost nothing. This seems to make audio circuits sound better. This is true with tubes, transistors, or fets.
In general, tubes are the most linear individual active devices, followed by fets, and then bipolar transistors. The problem with tube circuits is that they have to be coupled between stages with caps or transformers. Solid state circuits can be direct coupled, but they are always prone to more nonlinearity, both static (from Gm variation) and dynamic (from nonlinear capacitance with voltage). This is where solid state circuits can be optimized more than is typically employed, by consideration of the non-linear capacitance and its minimization.
Whats wrong with feedback in general?
It can be proven mathematically that feedback improves stability, bandwidth and linearity of the system, its one of the corner stones of automatic control theory, why doesnt it work with audio ? Sure inhereted linearity is better but why? opamps use feedback at their max and it doesnt do them any worse.
It can be proven mathematically that feedback improves stability, bandwidth and linearity of the system, its one of the corner stones of automatic control theory, why doesnt it work with audio ? Sure inhereted linearity is better but why? opamps use feedback at their max and it doesnt do them any worse.
Negative feedback causes a number of secondary problems, such as: TIM, IIM, FIM distortions, that the ear appears able to easily detect. Please understand: I have designed with negative feedback for the last 40 years. It is just that I have found that, all else being equal, less negative feedback is better sounding than more negative feedback.
- Status
- Not open for further replies.