Repair this S.E. to diff converter

[Elvee]

Hi,

I am trying to find the correct version of this circuit (fig1):
http://www.edn.com/article/CA6339252.html
The circuit shown doesn't work properly, and it is easy to understand why: there is no way the circuit can reference itself to a ground, which means the common mode and differential voltages are interdependent.
The circuit hides this fact behind a uselessly convoluted feedback scheme, that can be simplified to the equivalent shown in the attached pic:

It is now obvious that V4, the supposedly common mode voltage, cannot influence Q+.

I know the proper version exists, with resistors arranged somewhat differently, and both voltages referenced to the ground, but I am unable to find it, all application notes seem to show either this version, or the version having simple +1 and -1 gain paths, which I don't want. I would like the version with mutual cross feedback between the amps.
Thanks

[bosedtobosed]

heres one i almost prepaired earlier, the resistors R2 = 4k7, resistors R1 = 10k for 6db gain 15k for 10db gain, connect 2 resistors across the o\p and adj to a null at mid point (around 1k 1%) at 1khz

[Elvee]

Originally posted by bosedtobosed
heres one i almost prepaired earlier, the resistors R2 = 4k7, resistors R1 = 10k for 6db gain 15k for 10db gain, connect 2 resistors across the o\p and adj to a null at mid point (around 1k 1%) at 1khz

Thanks for the schematic, it begins to look like what I am looking for.

It is not yet exactly right though: the output is AC coupled, and there is no specific input allowing the common mode output level to be set.
This circuit more or less emulates a floating transformer winding.
I am looking for the equivalent of a center-tapped winding, having its center tap set at a definite level.
Thanks anyway

[GK]

Sounds to me like you may be thinking about a fully differential opamp:

http://focus.ti.com/lit/an/sloa099/sloa099.pdf

[Elvee]

Originally posted by G.Kleinschmidt
Sounds to me like you may be thinking about a fully differential opamp:

Not exactly. I want to emulate one (having a common mode pin) using two regular opamps.

[bosedtobosed]

the o\p centre point will be 0v ?, i dont know what you are looking for

[theAnonymous1]

What's the motivation to emulate?

TI's THS4130 and OPA1632 are excellent all-in-one solutions.

[Elvee]

the o\p centre point will be 0v ?, i dont know what you are looking for

The centre point will be some arbitrary, non-zero voltage

TI's THS4130 and OPA1632 are excellent all-in-one solutions.

They are indeed, and to such an extent they represent a massive overkill for this application.
The circuit is not in the signal path, and takes care of near-DC servoing signals, and a 324 is already extravagantly overcapable in this case.

[theAnonymous1]

Originally posted by Elvee
They are indeed, and to such an extent they represent a massive overkill for this application.
The circuit is not in the signal path, and takes care of near-DC servoing signals, and a 324 is already extravagantly overcapable in this case.

But, if you had either of those IC's would you be interested in using them? If so, drop me an e-mail; I have an extreme excess of both.

I would also like to see an op-amp based equivalent circuit of these just out of curiosity though.

[Elvee]

Originally posted by theAnonymous1


But, if you had either of those IC's would you be interested in using them?

Well, I might let my natural laziness take advantage of such a kind an offer.
But on the other hand, I hate to use good, sophisticated parts for menial tasks.
And anyway, the problem would still be nagging me afterwards... it's like having a word on the tip of the tongue. I cannot sleep properly until I've found it.
But anyway, thanks for your generous offer. When I'll really need such an amplifier, I'll come back to you.

I would also like to see an op-amp based equivalent circuit of these just out of curiosity though.

See? You too have been contaminated