The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Modeling >> Behavioral Models >> Modeling amplifier non-linearity in combination with bandwidth limitation
https://designers-guide.org/forum/YaBB.pl?num=1255544522

Message started by Visjnoe on Oct 14^th, 2009, 11:22am

Title: Modeling amplifier non-linearity in combination with bandwidth limitation
Post by Visjnoe on Oct 14^th, 2009, 11:22am

Dear all,

I'm stuck at trying to model an amplifier's non-linearity in combination with a bandwidth limitation. For the non-linearity, I would use a classical third-order Taylor series approach, while for the bandwidth limitation, I would use a first-order, low-pass s-domain transfer function with a certain cut-off frequency.

My conceptual problem arises when it comes down to ordering these two in the signal flow (block) diagram: first the Taylor series and than the s-domain transfer function or vice versa?

When I put the Taylor series first, the model might generate e.g. a third harmonic with the wanted/correct power, but that third harmonic might subsequently be attenuated by the s-domain (lowpass) function (if it falls beyond the cut-off frequency), lowering its power so it 's not in line anymore with the predicted power by the Taylor series.

If I put the s-domain lowpass function first, it might be so that two tones that are present at the model's input get attenuated first and than pass through the Taylor series...thus not generating e.g. the intermodulation products as predicted by the Taylor series (not the correct power).

Any insights/comments welcome on this modeling problem...

Regards

Peter

Title: Re: Modeling amplifier non-linearity in combination with bandwidth limitation
Post by boe on Oct 16^th, 2009, 8:51am

Visjnoe,
Assuming a simple opamp, I'd expect non-linearity is mainly caused by input stage and BW limitation by the loading of that or later stages.
So I write my models as non-linearity first, then low-pass.
Of course, if your non-linearity comes from clamping in the output stage, the inverse applies.
BOE