linearity of Actice-RC filters

[jimito13]

linearity of Active-RC filters

Dear members,

I am studying on continuous-time analog filters and i need some suggestions in my research.I am trying to find some references (book/papers/online notes) that prove the fact that the Active-RC
method of implementation offers better linearity among other available methods (like Gm-C).I would appreciate if somebody could contribute any useful information.

Thanks in advance.

Best Regards,

 

[LvW]

Hi Jimito,

I am afraid, a general answer to your question is not easy because linearity depends on dynamic range and some other parameters of the active device:
Power supply range, input signal (range of values), input voltage/current?, output voltage/current ?, gain requirements (degree of feedback), differential mode yes/no ?, .....
(I am sure that I forgot something).

---------- Post added at 19:43 ---------- Previous post was at 19:41 ----------

Yes, for example: Topology of the filter structure (cascade approach or direct realization), pole-Q of the stages (cascade design),..

 

[jimito13]

Hi LvW,

I absolutely agree with you,but i am looking for a general explanation or some mathematics that prove the claim that Active-RC filters are more linear than other filter implementations.I don't want to
relate the specs with a specific value of linearity that may be achieved.

But,as for the topology i am interested in cascading biquadratic stages to reach the order of the filter.

Thanks in advance.

 

[LvW]

Jimito,

in this context, the following link could be interesting

**broken link removed**

The author of this contribution has published another - more detailed - paper dealing with the same subject.
Please give notice if you are interested.

 

[FvM]

Besides other non-ideal properties, the transfer characteristic of an OTA is basically the characteristic of a differential pair:

ΔI2 = I0 * tanh(ΔV1/2 VT)

As a possible variation, the characteristic can be linearized by adding "degradation" resistors or other linerization means. Part of the tanh() characteristic will still show in the OTA behaviour. In the said filter structures, the OTA is applied as an open loop integrator. Active RC filters are in contrast using feedback amplifiers.

So why are you asking about better linearity of active RC filters?

 

[varunkant2k]

Active-RC filters are more linear

Because their transfer characteristics in the operating range is less susceptible to process / temperature and supply. An ideal RC filter can give good linearity but very tough to get the same in the chip (comes with a lots of parasitic). With this these values have some dependency with temperature and voltage supply.

Another point, with active filter the Gain is increased and hence attenuation region is reduced with increased bandwidth.

 

[jimito13]

To LvW,

I read the article you proposed,it is interesting but i think it has not immediate connection to my concern.In any case,if you can,please provide me the paper you talked about in your previous post.Thanks a lot in advance.

To FvM,

So why are you asking about better linearity of active RC filters?

As part of my research i want to prove that Active-RC filters provide more linearity than other methods.So,i wonder if there is something already available to refer or i should start from scratch to come up
with a conclusion.

Thanks both of you for your time and answers

 

[FvM]

The conclusion may be more or less obvious - depending on the exact problem formulation. LInearity of active filters can be derived from OP respectively feedback circuit theory.

The preferred filter technology will nevertheless depend on the sum of application parameters. You'll most likely find domains of other technologies, e.g. OTA (for integratable filters in RFIC), passive (general RF, power applications), distributed passive (microwave). Also time discrete filters (SC) and full digital filters have their application fields.

 

[jimito13]

FvM,

The conclusion may be more or less obvious - depending on the exact problem formulation. LInearity of active filters can be derived from OP respectively feedback circuit theory.

Is it possible that you can propose to me a good book or another reference for this?

Thanks a lot in advance.

 

[LvW]

As part of my research i want to prove that Active-RC filters provide more linearity than other methods.

Jimito, may I comment to your research activities?

I think, it is always problematic to compare two different circuits to each other because immediately the question arises:
"Which performance parameters are important - and which are less important"?
In the case under discussion you focus on "linearity" - without any consideration of complexity, dynamic range, frequency range,...?
In any case, for a fair comparison it is important that in all investigated alternatives you are using voltages at the signal input as well as output.
That means, for filter structures with current inputs and/or outputs you should include corresponding converters in your designs.
And in all circuits you should use the same input signal range - otherwise you cannot come to a meaningful result.
Good luck
LvW

(By the way: Comparing voltage and current mode, the papers from HP Schmid can be a great help, I think)