UAS - Universal Amplification Stage

[Stee]

Hello to All

I will present my recent idea - the UAS

This amplifier has the ability to govern themselves alone
thanks to local feedback on output resistors

it's esoteric because ultra simple
it's complementary
can build in chain with different power

need matching devices
do not need adjustment
have gain in voltage and in current

can use LED for biasing
between pre and final stages better put a potentiometer
its natural evolution provides the drive with mosfets

do you think it's intresting?
never tested

 

[jan.didden]

Is this supposed to be used as a power buffer because of the gain of one? As the output stage of a power amp?

One potential problem would be the output DC offset.

Jan Didden

 

[Stee]

Better this?

Regards

 

[Stee]

optimization

now you can see
at the same time there is :

VOLTAGE GAIN

DC offset control
Thermal drift biasing
Negative feedback subtraction

 

[Guthorst]

It will have very high THD because of the full class B bias setting.
And there is NO negative feedback.

regards,
Guthorst

 

[Stee]

Guthorst see now

I have edited with double input diode

please write on schematic your modification

 

[lineup]

you can add one film capacitor
across those 2 input diodes
... from base to base of input transistors

will act as a sort of BOOTSTRAP
something like 1 uF would maybe be suitable

otherwise I think your amplifier is a very nice idea
more creative and interesting
than most I can discover at this board from other posters

 

[jan.didden]

It will have very high THD because of the full class B bias setting.
And there is NO negative feedback.

regards,
Guthorst

... but there is a LOT of nfb. And the biasing is probably class AB. But that's easy to set with some resistor values.

My question is more, what's 'universal' about this?

Jan Didden

 

[Stee]

Hi Janneman

I called Universal
because as for OPA
is a model that you can repeat the same at every level of power

does not require settings - so it is configured with a standard, perhaps an integrated

also evades the tradition that wants to always be the mirror of current in input
bias control in VAS
and the output stage only in current gain

also has the ambition to become globally esoteric
because extremely simple (perhaps sounds good)
you may have to hold a conference to AES in New York...

the only new idea (I think)
is to use a shunt to monitor the current on each branch
in very short rings

I hope also become Universal with your contribution
maybe a good idea use FET/Mosfet in input...

Regards and Thanks

Hi lineup
thanks a lot for your support
this 1uF film capacitor from base to base it's for RF filter?

 

[Eva]

This circuit has been in use for the past 30 years. There is nothing innovative in it. Its main drawback is cross-conduction at high frequencies.

 

[lineup]

Eva .. or some one else:
what is cross-conduction?

thanks for explain to me

 

[Eva]

Cross-conduction means current flowing from the positive supply rail to the negative supply rail because both banks of output devices are conducting at the same time.

It happens because in bipolar transistors collector current does not stop immediately when base current is removed, it takes a few dozen microsedonds to fall to 0. A negative base current is required in order to stop conduction quickly, which this circuit can't provide.

 

[lineup]

thanks Eva

how do we deal with this?

I mean we sometimes see one capacitor used 'to suck' out base current
from output bipolar devices
connected base-to-base in complementary follower output stages

is that one of tricks?

 

[syn08]

thanks Eva

how do we deal with this?

I mean we sometimes see one capacitor used 'to suck' out base current
from output bipolar devices
connected base-to-base in complementary follower output stages

is that one of tricks?

Yes and it works pretty well for EF output stages.

For CFP, it's more complicated. Here's a possible solution that I am looking into, for a MOSFET CFP stage (take this as a concept only). Note the bipolar drivers that can both source and sink current from the MOSFET gates.

 

[Lumba Ogir]

lineup,

how do we deal with this?

Class A biasing is the only way to avoid the highly disagreeable esoteric UAS cross-over distortion.

 

[Stee]

dimensioning

Eva you are a small scientist!

If serves the class A is not a problem
need is a giant dissipator for each channel (300x160mm)

there is some good soul can assign values of resistance to this Universal Pre? (Class A please)

 

[Stee]

complementary pair

the problem of ideal amplifier are the devices
are not so complementary

at last the solution of minimal stage is a passive unit
of course in class A

 

[Lumba Ogir]

Stee,
you are making progress!

 

[Stee]

Yes thanks

now I'm making full complementary
(I don't like passive females)

then will try the difference between last project...
see you soon

 

[lumanauw]

Similiar output stage (Eva should know this one )