Integrator with a current mode OpAmp?

[Turin231]

Integrator with a current mode OpAmp?Also ADS problem

How do you implement a RC integrator with a current mode operational amplifier that has multiple(identical) outputs? Do you use separate Compasitors to feedback the outputs as I suspect or do I have it completely wrong?

Also the simulator in ads when I try to use the current mode OpAmp as a component in a new schematic find an error about "incorrect number of terminals specified for design " and I have no idea what that means.The current mode schematic is this(see below)...Does anyone have any ideas?

 

[LvW]

What in your opinion is a "current mode" opamp?
OTA (VCCS) or Norton amp (CCCS) or something between (CFA, CC)?

 

[Turin231]

A CCCS is what I had in mind....

 

[LvW]

A CCCS is what I had in mind....

OK, in this case: What is the transfer characteristic of the active device? Something like Iout=K[I(in1)-I(in2)] ?

 

[Turin231]

Yeah Exactly...Something in the likes of Iout=RoGm(Iin1-Iin2)....

 

[LvW]

So, what is your problem now?
In principle, each current source (OTA for example) turns into an integrator when it is loaded with a capacitor.

 

[Turin231]

Re: Integrator with a current mode OpAmp?Also ADS problem

I want to recreate a filter done with voltage mode opAmp integrators and I was wondering how do I use the feedback with the multiple (identical)outputs of the CC opAmp that are needed to implement the filter with currents...

 

[LvW]

...multiple (identical)outputs of the CC opAmp

Multiple outputs? What does CC stands for (Current Conveyor) ?

 

[Turin231]

Oops...Sorry I meant CM for Current mode...

Yeah...The omAmp has single output(not differential in other words) but you need to duplicate them for the filter...Check the image in the first post.
I am just not sure how to place the compasitor(s) in the integrator....

 

[LvW]

Oops...Sorry I meant CM for Current mode...

Yeah...The omAmp has single output(not differential in other words) but you need to duplicate them for the filter...Check the image in the first post.
...

I've got the impression that the outputs you have shown in the diagram are at constant VDD level.
Or am I wrong?

 

[Turin231]

Yeah...There should be some current sources there which i neglected...But these are not constant VDD outputs...

 

[LvW]

I am afraid, without a correct and realistic amplifier description and without a clear description of your idea we cannot help you. For example, it is not clear to me why you need "duplicate outputs". Do you intend to design completely symmetrical circuits?

 

[Turin231]

ok this is the circuit with a single output...



You can replicate the output if need to be used in different places in the overall design, since the output is a current and not voltage, with another extra transistor as above. What this implements is an opAmp that instead of a Vout=(gm0*ro)Vin it control currents in the same fashion. Iout = (gm0*r0)Iin...So my question is how would I go to create an integrator to design a filter in which I have the design with voltage mode opAmps, with the current mode opAmps...

 

[LvW]

I am afraid, you have to design by yourself. I am rather familiar with filter classical topologies - however, at the moment I cannot remember any circuit with a current input and a current output.
Perhaps you can gather some ideas from existing circuits:
* Norton amplifier (LM13900) with current input (vltage out)
* OTA with current output (voltage in)
* Current conveyor with current output and one voltage plus one current input.

In principle, I propose to start with one of the classical block diagrams for an integrator-based universal filter structure (KHN or Tow-Thomas or Fleischer-Tow). Then, you can try step-by-step (block-by-block) to realize the structure in the current mode.
For some theoretical background information you can try to find the book "The current mode approach" (Toumazou and Lidgey)

 

[Turin231]

Ok...I'll do that...Thanks for the advice...

 

[LvW]

Hi Turin231,

here is a link that could be interesting for you:

https://web.itu.edu.tr/~kuntman/Y_l...ations/Design_of_a_fully_differential_COA.pdf

 

[Turin231]

It does actually..Thanks for your time my friend....