Series peaking Active Inductor

[SpanishGuyOC]

Hello there people,
I'm looking forward to the possibility of a Differential TIA (for BW expansion, I've seen some papers use it in for that purpose) in an AFE involving CTLE + VGA, my main motivation (at least now) is to reduce the use of another CTLE stage because that would end damaging the BW, also if well designed it might consume less power (I may be wrong with this), however the Differential TIA usually involves a series peaking inductor on the Feedback Loop, a real inductor, that consumes a lot of area..., in comparison to shunt peaking at the CTLE load.

Anyways the thing is that I've done some research in the IEEE data base, books, even Lectures (Sam Palermo, etc.), and haven't found any active inductor technique applied on series, almost every of them is applied for shunt peaking, am I missing on some explanation?, while this is true, there are some TIAs that've used shunt peaking (Gyrator, Cherry-Hooper type, etc.), but far from that they're used for Single ended purposes, everything that I've said actually makes me wonder if I should maybe consider discarding the possibility.

 

[LvW]

I have to admit that it's not entirely clear to me what your problem is.
Are you looking for a circuit for an active inductor? And can you also use a classic opamp or does it have to be a TIA?

 

[KlausST]

Hi,

isn´t it an IC design topic? (wrong forum section)

And can you also use a classic opamp or does it have to be a TIA?

Isn´t it the wiring of resistors around an OPAMP that makes it a TIA?

Klaus

 

[LvW]

Hi,

isn´t it an IC design topic? (wrong forum section)


Isn´t it the wiring of resistors around an OPAMP that makes it a TIA?

Klaus

Klaus, I know what you mean - and I agree.
But I think, that with "TIA" the TO means "CFA" (Current Feedback Amplifier).
I remember that in the first days the CFA was introduced also the term TIA was used for such devices.
This is because the "gain" of a CFA is in fact a "transimpedance".

 

[KlausST]

Hi,

current feedback amplifiers - as far as I know - have a narrow range for the feedback resistor.
The benefit of CMF amplifiers usually is higher bandwidth.

But schematic wise a curreent feedback amplifier circuit looks the same as a voltage feedback amplifier circuit.
So you can have a non inverting amplifier and an inverting amplifier .. both containing two resistors typically for gain setting.
And in both cases the (uint_less) gain is V_out/V_in
https://www.ti.com/lit/pdf/slva051

But a TIA only needs one resistor .. at the feedback. Wide range of feedback resistor. Where the feedback resistor determines the transfer ratio V_out/I_in.

Transimpedance amplifier - Wikipedia

en.wikipedia.org

Klaus

 

[LvW]

So you can have a non inverting amplifier and an inverting amplifier .. both containing two resistors typically for gain setting.
And in both cases the (uint_less) gain is V_out/V_in
https://www.ti.com/lit/pdf/slva051

Yes.
In post#4 I have mentioned the "gain of a CFA" andthe corresonding unit "transimpedance".
Of course, I was referring to the "open-loop transfer function".

 

[KlausST]

Hi,

In post#4 I have mentioned the "gain of a CFA" andthe corresonding unit "transimpedance".
Of course, I was referring to the "open-loop transfer function".

I´m confused.
For a TIA it is impossible to have no feedback ... while open loop always is without feedback.
So in my understanding it is impossible to have a TIA and open_loop at the same time.

What do I miss?

Klaus

 

[LvW]

Hi,



I´m confused.
For a TIA it is impossible to have no feedback ... while open loop always is without feedback.
So in my understanding it is impossible to have a TIA and open_loop at the same time.

What do I miss?

Klaus

Quote "Design with Operational Amplifiers....", Sergio Franco, McGraw Hill, 1998, page 297-298:
Chapter 6.7 CURRENT FEEDBACK AMPLIFIERS
"...the gain...is in volts per ampere rather than volts per volts. For this reason CFAs are also called transimpedance amplifiers."

I know that this is a - more or less - oldfashioned name.
However, after introduction of the first integrated CFAs, it was quite common to also use the designation TIA for these ICs,

 

[KlausST]

This might all be correct. (Or maybe was decades ago), I can´t deny.

however .. I personally vote for using the least confusing terminology.
And with this I comply with Wikipedia and TI documents.

Klaus