Up to now, no reply - perhaps I should give, therefore, some more details (background explanations):

Within the last decade many papers have been published about „New CFOA-based oscillators“ or „Novel CFOA-based filters“. Reading such titles and knowing the meaning of „CFOA“ , one could get the impression that some novel active circuits are described using one of the various amplifier-IC`s based on the Current-Feedback principle.

This is, however, not the case.

All the circuit topologies described in these papers work only for one specific IC from Analog Devices: AD844.

This unit is, indeed, a Current-Feedback Amplifier (CFA) - however, in all of these these papers it is not considered as a CFA but, instead, as a Current Conveyor (CC). This is possible because this unit is the only CFA providing external access to the internal high-impedance node.

Hence, the AD844 is a current conveyor (high resistive current output) with an additional output buffer (low-impedance voltage output) onboard.

With other words: All of the proposed filters and oscillators do not really exploit the current-feedback option of this unit.

And that is the real background for my question: For my opinion, it is misleading to use in this context the shortcut „CFOA“, because

(a) these circuits can work with one specific Current-Feedback IC (AD844) only, and

(b) this unit is operated as a Current Coveyor.

An example paper is attached as a pdf file.

(CC desgnations: X: Low impedance input; Y: High-impedance input; Z: High impedance output)

Dear Lutz,

To me the current feedback operational amplifier otherwise known as CFOA or CFA is a type of electronic amplifier whose inverting input is sensitive to current, rather than to voltage as in a conventional voltage-feedback operational amplifier (VFA).

http://en.wikipedia.org/wiki/Current-feedback_operational_amplifier

best regards

ERIK

Hi Erik - as I have mentioned, I know the working principle of CFA`s and CFOA`s. My question concerns one specific IC (AD844) which in fact is used as a CC - however it is claimed that it would be a CFOA circuit. And that`s quite misleading because it works for one single type only. That was the background of my question.

Thank you for replying.

Regards

Lutz

Current-feedback operational amplifiers are designated, at least, by three acronyms:

*CFA: current-feedback amplifiers; see booktext from Prof. Sergio Franco:Desing with operational ....,McGraw Hill,1998,pag 297

*CFB op amp,as it appears in the book from Prof Sidney Soclof:Desing and Applications of Analog IC, pag 443

*CFOA, in the book from Prof. Ananda Mohan:Current-Mode VLSI Analog FiltersBitkhauser,2002,pag 120.

Also, these acronyms are often used in journal/conference papers about this component;the related systems are typically implemented using the ADD844

Best regards

Pedro A. Martinez

Sr. Martinez - thank you for replying.

As you know, the AD844 can be used also as Current Conveyor (CC). In this case, I think we shouldn`t use the term CFOA in the title because this shortcut implies that ALL CFA`s could be used. But that`s not the case.

As a positive example: In 1995 you (and some co-authors) have published a paper (in Analog Int. circuits and Sign. Proc.)  based on the AD844 - and the title was "Wien-type oscillators using CCII+".

Here, no misinterpretation are possible.  Do you know what I mean?

The IC AD844 is a somehow special CFOA. It is formed by an input stage, which corresponds to a CCII+, connected to a voltage follower. Its distinctive feature is that the output of the conveyor is available and accessible. This allows using the AD844 as a current conveyor, which has been used very often to implement systems with CCII.

Best regads

Yes - that´s what I wrote in my first post; it seems we are moving in a circle.

I`ve got the impression that I am the only one (in this forum!) who is not  happy with the described phenomenon. OK - I have no problem, but some students had a problem recently: They wanted to build a CFOA based oscillator using a Current-Feedback Amplifier (other than AD844) , but they couldn`t because the circuit requires one specific type - the AD844 !

Question to Sr. Martinez:

In most of the papers describing CFOA based oscillators (using the AD844) we can read that amplitude stabilization is achieved due to some „internal non-linearities“ of the device - for my opion, a statement without suitable justification.

As an example, here is an excerpt from your 1995 paper:

„The non-linear internal mechanism of CCII+ ensures the existence of a stable limit cycle.“

And, therefore, my question to you:

1.) Does the AD844 - in contrast to voltage mode opamps - contain any internal non-linearity (other than power supply limitations) which can be exploited for amplitude stabilization purposes?

2.) In which document is this property of the device described or verified?

Thank you.

Good afternoon,

Herein there are two questions.

*Firs, it is evident that there are papers whose title is wrong, leading to confusion. The one you are mentioning does not use CFOA but CCIIs and voltage followers.

*Secondly, the synthesis of sinusoidal oscillators is based on the linearization of the active components. Therefore, we cannot affirm that the derived systems word properly. For this reason, before publishing an architecture obtained following the aforementioned procedure, I experimentally verity its proper working. If it is an oscillator, it is evident that there is a non-linear dominant mechanism that determines the output amplitude. Its particular determination becomes a challenge for the author of the architecture

@Sr. P.A. Martinez

Thank you for your answer.

Quote: "it is evident that there is a non-linear dominant mechanism that determines the output amplitude. Its particular determination becomes a challenge for the author of the architecture"

Yes - of course, I assume that authors convince themselfes accordingly before presenting such statements.

And that was my question:  Where is a document which describes the "internal non-linearity" of the AD844 as mentioned in your paper?

I ask this question because I know (at least) two documents stating exactly the opposite:

Quote: Conclusion: ...It was shown that OTA-based and CCII- or CFOA-based oscillators do not have the same behaviour. For the OTA-based oscillator, it`s tanh-function nonlinearity can always stabilize the magnitude of the oscillation signal, whereas for the CCII- or CFOA-based oscillator, their saturated sinh-function nonlinearity prevents the stabilization phenomenon .

(Proc. ISCIT 2005: „The Describing Function Method and the Analysis of the Magnitude Stabilization Phenomenon in a nonlinear OSC“.)

Similar findings can be found in: Trans. on Circuits and Systems, Fundamental Theory and Applications, vol 49, NO 8, August 2002, page 1231.

To me - Two contradicting findings. That`s the background of my query.

Today - 9 days after my last post - I must admit that I am a bit disappointed (and also surprised). As outlined in my last post, I have found two papers having a content which is in contradiction to some other papers dealing with the same subject (AD844 based oscillators).  Up to now - no response, although some RG members have prepared corresponding papers (about AD844 applications).  For my opinion, this question deserves clarification - am I wrong (or too "sophistic") ?

(Quote RG mission: "Our mission is to connect researchers and make it easy for them to share and access scientific output, knowledge, and expertise.")

Current-feedback operational amplifiers are designated, at least, by three acronyms:

*CFA: current-feedback amplifiers; see booktext from Prof. Sergio Franco:Desing with operational ....,McGraw Hill,1998,pag 297

*CFB op amp,as it appears in the book from Prof Sidney Soclof:Desing and Applications of Analog IC, pag 443

*CFOA, in the book from Prof. Ananda Mohan:Current-Mode VLSI Analog FiltersBitkhauser,2002,pag 120.

Also, these acronyms are often used in journal/conference papers about this component;the related systems are typically implemented using the ADD844

Best regards

Pedro A. Martinez

@ Sr. P.A.Martinez:

If I am not wrong - this text was submitted already 12 days ago.

Submission error?