Yes, I have always understand what you have said.

My problem with your previous posts, was that it did not satisfactorily show (at least to me) why you needed a HP passband so far below the square-wave fundamental to pass the square-wave without distortion since, even with a passband below 1/10 of the fundamental where the fundamental frequency is only slight attenuated, there was still large distortion of the square-wave.

My continuing quibble with your explanation is, it would appear the filter's phase-shift for HP filters with a corner below the fundamental has the most effect on the square-wave shape, and the amplitude change has only a smaller second-order effect, as Figures 6 and 8 in post #38, and my post #25 would seem to show.
Do you not agree with that?

Hi,

Almost, but the effect depends on different factors, and you changed your stance two times now yet you are talking like I was somehow wrong all along. Maybe you do not realize that. I'll try to explain.

First, you said that the square wave has frequencies lower than the fundamental, which is not true and I pointed that out and now I think you agree with that.
Second, if we are dealing ONLY with a HP filter then it could be that the AC analysis phase shift has more of an effect on the distortion than the AC amplitude. Note the use of "AC" here, and that is because a time domain analysis shows it is the instantaneous amplitude that affects the entire waveform and thus the distortion, and you never really acknowledged the effect of even the AC amplitude until now as you said it was the (AC) phase shift, so you can see why I had a problem with that.

I think we both agree now that in the frequency domain the AC amplitude and the AC phase shift play a part in the HP square wave distortion, and that for the filters we looked at the AC phase shift had a more profound effect on the distortion than the AC amplitude.
I think we both agree now that in the time domain the instantaneous amplitude summations reveal the reason for any distortion or really for any waveshape at all.

What I am not sure about though is if you understand me about what I was saying about the fundamental alone and it's play in the distortion effect on the square wave. That's the reason I gave for needing a lower filter set point (lower -3db point setting for the filter) and I think that is the answer you were looking for. The reasoning is very simple I think and I'll state that here...

"If the fundamental is the lowest frequency and the distortion is affected by the fundamental almost exclusively, and since a high pass filter set to have a lower cutoff frequency will pass the fundamental with less of an effect, then the distortion will decrease if we lower the cutoff frequency of the HP filter."

Since we agree now that the fundamental is the lowest frequency, and I had proved that the distortion is affected mostly by the fundamental, then it is reasonable to believe that lowering the cutoff frequency will decrease the distortion because the fundamental is affected less by the filter then.

Here's one more little point we don't have to get into if you do not want to because I realize you were talking about one particular filter.
My point above is about the change in the fundamental and how it affects the distortion. To me, it does not matter that much if it is the AC phase or the AC amplitude, because either way (or both) it will cause distortion. We happen to see a particular waveshape (with a particular distortion) because of the particular filter, but that could change if we change the filter even if it remains a high pass. If we could design a high pass filter with no phase shift, then the distortion would be due only to the AC amplitude decrease. This would result in a different type of distortion but it could get pretty bad
We do not really have to get into this, I was just hoping to point out the strong effect we see because of the alteration of the fundamental.

 

you changed your stance two times now yet you are talking like I was somehow wrong all along.

How did I change my stance two times, other than agreeing that there are no Fourier frequencies below the fundamental in a square-wave?

My problem was that you initially only emphasized the amplitude of the fundamental as causing the distortion from a HP filter, which didn't make good sense to me, whereas it's actually mostly the phase-shift as shown by the time-domain simulations between an HP filter and an all-pass filter.
If you could design a magical filter that only affects amplitude and not phase, than of course, it would only be amplitude affecting the waveform.

But I think we are now mostly beating a dead horse here, so no further discussion is really needed.