New Amplifier Technology

[EddyF]

A small introduction:
Back in those days there was the tube, only tube, and amplifier-technology was developing around tube-properties.
But then the transistor was invented. Didn't get so hot, had a longer lifespan and was virtually unbreakeble.
Since there allready was a amplifier-technology, the tube got merely replaced by the transistor.
Tube-amplifiers with transistors. There was no real transistor-amplifier design! Untill now.
The acompanying circuit shows the schematic of the poweramplifier.

Some feature's;
Completely ballanced.
Inherently stable, no frequency compensating capacitor needed.
Frequency range from DC to 150Meghz.
Shortcircuit proof.
No dynamic load on the powersupply, so no need for low impedance powersupply. I even use a unregulated, transformer rectifier elco, powersuply.
Pure classe A.
As musical as a tube-amplifier but without the coloration.
So rich in detail that the need to turn up the volume is no longer needed.

A discription;
The inputstage is a differential stage converting the input voltage to a differential current.
Loaded by a high impedance ballanced currentmirror. Followed by the outputstage wich is again differential loaded by a high impedance ballanced currentmirror.

The outputstage draws a constant current of about 2Amp. wich is also the peak current deliverable to the loudspeaker, therefore the shortcircuitproof.

Also drawn are two possible inputs, one ballenced filtered, one single-ended input filtered, Dc and to about 90khz.

And it is easy adaptable to different, lower or higher power. Different voltage-powersuply needs a different voltage zener to get the output around the middle.
And the output transistors need heavy cooling.

The whole is my design completely and anyone interrested can build one for his own domestic use.
It is not on the market, I yet need to find a company who´s willing to market it.

I am currently working on a MC-MM phono-stage using the same technology offcourse, and planning on a line-stage with tone-controll as wel as a single-ended to ballanced converter.
Eddy

[jechentau]

nice design, i like it i plan to incorporate and try balanced current mirrors in my following amp (that amp i build with many components in very short sockets, so i can easily try out various topologies here and there)

[ingenieus]

Nice design.

Not efficient though. LTspice suggests 8W RMS into 8 ohm with the PSU pumping out 75W.

[MagicBox]

It looks interesting. Have you done any linearity measurements? It seems you're going by local gain settings in the current mirrors only seeing there's no global feedback as far as I can tell.

[Nelson Pass]

At the least, it is creative.

[DF96]

Originally Posted by EddyF

There was no real transistor-amplifier design! Untill now.

I assume you don't actually mean this. There are plenty of SS designs which don't look anything like "valve designs with transistors in". For a start, where would you find complementary polarity valves?

[EddyF]

Originally Posted by DF96

I assume you don't actually mean this. There are plenty of SS designs which don't look anything like "valve designs with transistors in". For a start, where would you find complementary polarity valves?

With that I mean it's all Op-Amp like. And an Op-Amp is the ultimate meant to be a tube. High impedance input for starters.

I was mostly inspired by M. Hawksford by his description of how the transistor works and from there I started designing, to this and continuing.

[DF96]

I think some people might wish to differ from your view that op-amps and valves are almost the same thing. There is more to life than input impedance. Open-loop gain, transfer characteristics, frequency response are all very different.

Many SS amps look a bit like the innards of a basic op-amp, for quite sensible reasons. However, there are SS designers (such as Nelson Pass) who do things differently, and have done so for quite a while.

[Conrad Hoffman]

This is actually a bridged version of the totem pole circuit used in early TTL logic, then adapted to high voltage amplifiers made by Burleigh Instruments and subsequently published by Horowitz & Hill. Much more evolved, but conceptually operating in a very similar manner. The downside to this technique is horrendous efficiency, so it probably doesn't have any commercial application in an energy conscious world. But I've been wrong before- just ask my wife!

[Elvee]

Originally Posted by Conrad Hoffman

This is actually a bridged version of the totem pole circuit used in early TTL logic, then adapted to high voltage amplifiers made by Burleigh Instruments and subsequently published by Horowitz & Hill.

I don't think so: the totem-pole I know is more JLH-like (but there might be earlier versions unknown to me).

This circuit is a bridged SE amplifier.

The important part is the bottom one: it is a V-I converter based on a CFP, and must for that reason have a good linearity.
The top half is merely decorative and just has to provide a sufficient source current to keep the CFP's biased.
The top-down symmetry is pleasing to the eye and does no harm, it can even cancel some of the Early effect, but it isn't essential.
The CFP + lateral symmetry certainly contributes to cancel many even order non-linearities.

Apart from energetic efficiency, I would be concerned by the damping factor: I guess it is pretty low, perhaps even <1.