Dont kill me BUT...

[uli]

Hi all,
All those posts about X-Aleph Aleph-X and so on.

I´m missing the X-principle in your circuits

Uli

[UrSv]

Care to clarify? I see it everywhere almost.

[uli]

Originally posted by UrSv
Care to clarify? I see it everywhere almost.

OK,
2 independant amplifiers with identical data(->Supersymmetry)
driven thru the inverting input and loosely coupled with a
defined resistor between the noninverting inputs.
What you call Aleph-X is just a differential pair with both
outputs used!

Uli

PS:X100 backengineered here

[UrSv]

OK. Not quite. And?

[uli]

I´m sure it sounds nice, but it´s no X!

1) X has no feedback from one amp to the other
2) X uses inverting stages
3) X has just one gainstage and a sourcefollower
4) X is easier to control the offset because of loose coupling

Uli

[UrSv]

Originally posted by uli
I´m sure it sounds nice, but it´s no X!

1) X has no feedback from one amp to the other
2) X uses inverting stages
3) X has just one gainstage and a sourcefollower
4) X is easier to control the offset because of loose coupling

Uli

I am no whiz at the X thing obviously but to my knowledge:

1) That is the fundamental of the X topology AFAIK. Without it there is no X.
2) At least one needs to be AFAIK since you want to feed the "error" signal back in opposite phase. The inv. stage has nothing to do with the X as such.
3) Maybe, maybe not. This has nothing to do with the X as such AFAIK.
4) Maybe, maybe not. See 2) and 3).

All of the above replies could naturally be incorrect (although I hope not) but I am sure that some of the X-men (and X-girls) will chip in their 2 cents.

[Nelson Pass]

The confusion revolves around the op-amp analogy.

What appear to be positive inputs are low impedance
connections, thus they can insert current into each other.

At the - inputs, which are high impedance, you theoretically
only see the error signal, and this appears at the + inputs
by virtue of them being the Source/Emitter where the -
input is the Gate/Base, so they follow, and form a low
impedance load.

In this way, the "+ inputs" communicate current to each
other representing the error signal of each side, causing
the circuit to reproduce the errors of each side in phase.

I know it's confusing.....

pass/ - not seeking confusion except among his
economic competitors.

[uli]

Originally posted by UrSv

1) That is the fundamental of the X topology AFAIK. Without it there is no X.

I fear that this is not quite correct.
When you look at NPs writing the 2 amps are NOT connected
except the resistor between the +inputs.

2) At least one needs to be AFAIK since you want to feed the "error" signal back in opposite phase. The inv. stage has nothing to do with the X as such.

Of course it has. the errorsignal is fed thru this particular resistor
into the other stage.

3) Maybe, maybe not. This has nothing to do with the X as such AFAIK.

If you stretch X over 2 stages you would have to feed the feedback into the source of the first fet.

4) Maybe, maybe not. See 2) and 3).

All of the above replies could naturally be incorrect (although I hope not) but I am sure that some of the X-men (and X-girls) will chip in their 2 cents.

I´m waiting for NP´s statement, please don´t kill me

Uli

[UrSv]

Originally posted by uli


I´m waiting for NP´s statement, please don´t kill me

Uli

It is already there....

[uli]

Originally posted by Nelson Pass


In this way, the "+ inputs" communicate current to each
other representing the error signal of each side, causing
the circuit to reproduce the errors of each side in phase.

I know it's confusing.....

pass/ - not seeking confusion except among his
economic competitors.

A little bit...

This is what I ment when talking +inputs (source).
theres no feedback from one side to the other
exept this (47R) resistor!

When looking at SOZ its a VERY simple kind of X
but NOT those circuits called Aleph X.

I´m sorry confusing anybody.

PS: XSOZ = swapping all those 8R power resistors to CCS´s
putting some feedback from the drains to the particular
gates, adding input resistors swap the 1R to 47R and from the 2
sources to ground a 300R each. Add a DC-path from the inputs
to ground.

I´ll post a drawing ASAP.

Uli /-not being any kind of competitor neither economic nor
knowhow wise