Best perfs/complexity ratios are obtained first by Cherry like designs with nested feedback loops, second is the 741 topology, aka blameless, and third is the full symetrical topology.
Linearity wise the first quoted is about out of reach by the two other topologies at the higher audio frequencies, said topologies can be partly used of course as a basis for a NDFL amp.
Linearity wise the first quoted is about out of reach by the two other topologies at the higher audio frequencies, said topologies can be partly used of course as a basis for a NDFL amp.
Attachments
Last edited:
Hi,
An observation that seems to me to be totally wrong.
You work out which part is wrong, its easy.
rgds, sreten.
Last edited:
The main reason I ask is I want to gauge how important is using complementary by looking a what successful high end amp using it. If a lot of recognized high end amp don't use complementary, then I am not going to do it. I don't care about theory that complementary gives lower distortion, I care about what people actually use in the high end world that are recognized as top quality.
Is Accuphase from Japan good enough for you? Look at their actual power amps, they even publish schematics in each of their brochure, see and learn. 😉
Go to products, power amplifier, select model and on the left side is details, English catalog, download it and open in pdf viewer. It's a brochure with everything, even schematics. By the way they even use power mosfet switches in speaker outputs, etc.
Last edited:
Many rail to rail input opamps have complementary LTPs.
Are there any rail to rail input opamps that don't have complementary input transistors?
Are there any rail to rail input opamps that don't have complementary input transistors?
A low power stage with single differential input, complementary push-pull VAS and very high performances, has been proposed by S. Groner :
A new audio amplifier topology with push-pull transimpedance stage - Part 1: Introduction | EE Times
The primary aim in the design seems to avoid power supply injection by Cdom.
A new audio amplifier topology with push-pull transimpedance stage - Part 1: Introduction | EE Times
The primary aim in the design seems to avoid power supply injection by Cdom.
You probably shouldn't, the high end world is 95% snake oil and BS.
The technical differences between the regular and complementary diff pairs have been covered in this thread already, but either one of them will give a perfectly good sounding amp with enough slew rate to reproduce audio.
I prefer the regular ("Blameless" or "741" topology) as it is much easier to define the DC conditions. The complementary ones can turn into a real pain, with two input stages and two VASs that have to be coordinated to have stable bias currents and operate smoothly in push-pull without fighting each other. Of course this can be done, there are many successful examples of it both commercial and DIY, but why bother when the simple circuit is already good enough?
Also, the regular diff pair input with current mirror load will provide symmetrical push-pull drive to the VAS. If an amp with this input stage has slew rate asymmetry, it's the fault of the VAS stage and its current source.
Why don't you post a sketch of a (discrete) input stage with rail-to-rail input capability? Then other thread participants can tell you whether or not they've seen that in an opamp schematic.
Symmetrical drive of the VAS stage, and symmetrical slew rates are good enough reasons for me to go for a balanced complementary topology (aka Leach et al). In designs that use single ended LTP, symmetrical VAS can be obtained, but at the expense of additional complexity.
I can't understand what the 'fuss' is about. We developed the complementary differential input stage in the late '60's to get more ideal performance with minimum feedback (necessary at the time) in order to compete with the reference triode amps that existed at the time. It is not 'magic' but I almost always use it, and have done so since 1968. Complementary differential bipolar is only a beginning, because complementary differential jfet is even better and more elegant, for added reasons.
For me, I found that the complementary differential input stage could improve the distortion about 2.5 times better than the typical single ended second stage, still popular today.
For me, I found that the complementary differential input stage could improve the distortion about 2.5 times better than the typical single ended second stage, still popular today.
- Status
- Not open for further replies.