Jerry Avins wrote:> Jani Huhtanen wrote: > > ... > >> Phase shift, in general, does not imply time shift. Only if the phase >> shift is _not_ a constant as a function of frequency, it implies time >> shift. Inverter is an example where phase shift does not imply time shift >> ;) > > wt + phi = wt', where t' = t + phi/w. If phi is the phase shift, phi/w > is the time shift. If an inverter is an example of phase shift, does the > phase lead, or lag? > > JerryBy those definitions, I don't know. I thought that by time shift you meant group delay as you were also talking about transients. I'm not sure how "your" time shift relates to transients in question, however, IMO group delay would explain the transient behaviour intuitively (i.e., inverter does not delay nor advance the signal as the phase shift is 180' for all w). -- Jani Huhtanen Tampere University of Technology, Pori
How does an inverter affect phase?
Started by ●June 15, 2006
Reply by ●June 18, 20062006-06-18
Reply by ●June 18, 20062006-06-18
Just a little addition.. I wrote:> Jerry Avins wrote: > >> Jani Huhtanen wrote: >> >> ... >> >>> Phase shift, in general, does not imply time shift. Only if the phase >>> shift is _not_ a constant as a function of frequency, it implies time >>> shift. Inverter is an example where phase shift does not imply time >>> shift ;) >> >> wt + phi = wt', where t' = t + phi/w. If phi is the phase shift, phi/w >> is the time shift. If an inverter is an example of phase shift, does the >> phase lead, or lag? >> >> Jerry > > By those definitions, I don't know. I thought that by time shift you meant > group delay as you were also talking about transients. I'm not sure > how "your" time shift relates to transients in question, however, IMO > group delay would explain the transient behaviour intuitively (i.e., > inverter does not delay nor advance the signal as the phase shift is 180' > for all w). >Wikipedia, our always trustworthy information source, has this to say: "It is common to speak of inverting the polarity of a wave as "flipping the phase" or "shifting the phase by 180 degrees". These are not completely equivalent, though, since a 180 degree phase shift of all signal frequencies would also delay the signal. Inverting the signal is instantaneous." However, I still disagree ;) -- Jani Huhtanen Tampere University of Technology, Pori
Reply by ●June 18, 20062006-06-18
Jani Huhtanen said the following on 18/06/2006 15:03:> Just a little addition.. > > I wrote: >> Jerry Avins wrote: >> >>> Jani Huhtanen wrote: >>> >>> ... >>> >>>> Phase shift, in general, does not imply time shift. Only if the phase >>>> shift is _not_ a constant as a function of frequency, it implies time >>>> shift. Inverter is an example where phase shift does not imply time >>>> shift ;) >>> wt + phi = wt', where t' = t + phi/w. If phi is the phase shift, phi/w >>> is the time shift. If an inverter is an example of phase shift, does the >>> phase lead, or lag? >>> >>> Jerry >> By those definitions, I don't know. I thought that by time shift you meant >> group delay as you were also talking about transients. I'm not sure >> how "your" time shift relates to transients in question, however, IMO >> group delay would explain the transient behaviour intuitively (i.e., >> inverter does not delay nor advance the signal as the phase shift is 180' >> for all w). >> > > Wikipedia, our always trustworthy information source, has this to say: > "It is common to speak of inverting the polarity of a wave as "flipping the > phase" or "shifting the phase by 180 degrees". These are not completely > equivalent, though, since a 180 degree phase shift of all signal > frequencies would also delay the signal. Inverting the signal is > instantaneous." > > However, I still disagree ;)I also disagree. Constant phase-shift gives zero group delay. I think the article is incorrect. As for Jerry's example, let's suppose we had two components: x(t) = sin(w_1 t + phi) + sin(w_2 t + phi) We cannot transform this to x'(t') as we could for the single-frequency example. -- Oli
Reply by ●June 18, 20062006-06-18
Jerry: Well trolling or not... [smile] if you are referring to modulation theory, in particular to phase modulation, then there may in fact be a difference in the effects on phase by "inserting" an inversion. If one is only interested in the waveform at sample instants, then phase transitions of + or - pi don't matter. Often however what happens everywhere on the waveform [other than at sample points] is used to "recover" stuff like timing information and then, depending upon the implementation of the timing algorithms, one might care whether the transitions are + or - pi. In continuous and "oversampled" systems there is a difference [often with significant side effects] in how the phase of a carrier wave is directed [by the modulator] or detected by a de-modulator, while making "transitions" from zero to pi radian phase. i.e. a modulated sinusoidal-like waveform may make transitions in either the positive [counter-clockwise] or negative [clockwise] direction. Such transitions may be forced digitally, or become the result of passing a sampled wave through a linear system [e.g. filter], which then adds its own "effects" to phase transitions. How can one distinguish the direction of rotation of the complex [analytic] signal? Does it matter? Thoughts, comments? -- Pete Indialantic By-the-Sea, FL "Jerry Avins" <jya@ieee.org> wrote in message news:mMCdnbE7gOsqTQzZnZ2dnUVZ_rKdnZ2d@rcn.net...> For a pure sinusoid, the effects of an inverter and of a 180 degree phase > shift are indistinguishable. I lay the burden upon anyone who maintains > that "indistinguishable" means "the same thing" to explain whether the > phase shift is positive or negative, and to reconcile the behavior of > transients in inverters and diode ring modulators. > > I apologize to all who may believe that this is a troll. It may be > indistinguishable, but it's not the same thing. :-) > > Jerry > -- > Engineering is the art of making what you want from things you can get. > �����������������������������������������������������������������������
Reply by ●June 19, 20062006-06-19
Jani Huhtanen wrote:> Jerry Avins wrote: > >> Jani Huhtanen wrote: >> >> ... >> >>> Phase shift, in general, does not imply time shift. Only if the phase >>> shift is _not_ a constant as a function of frequency, it implies time >>> shift. Inverter is an example where phase shift does not imply time shift >>> ;) >> wt + phi = wt', where t' = t + phi/w. If phi is the phase shift, phi/w >> is the time shift. If an inverter is an example of phase shift, does the >> phase lead, or lag? >> >> Jerry > > By those definitions, I don't know. I thought that by time shift you meant > group delay as you were also talking about transients. I'm not sure > how "your" time shift relates to transients in question, however, IMO group > delay would explain the transient behaviour intuitively (i.e., inverter > does not delay nor advance the signal as the phase shift is 180' for all > w).180 degrees lead, or 180 degrees lead? If inverted twice, is that 360 degrees? An inverter works just fine with non-periodic waveforms, but phase is a hard concept to force upon them. Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●June 19, 20062006-06-19
Oli Filth wrote: ...> As for Jerry's example, let's suppose we had two components: > > x(t) = sin(w_1 t + phi) + sin(w_2 t + phi) > > We cannot transform this to x'(t') as we could for the single-frequency > example.Sure we can. sin[w_1(t + phi/w_1)] + sin[w_2(t + phi/w_2)] Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●June 19, 20062006-06-19
Jerry Avins said the following on 19/06/2006 16:13:> Oli Filth wrote: > > ... > >> As for Jerry's example, let's suppose we had two components: >> >> x(t) = sin(w_1 t + phi) + sin(w_2 t + phi) >> >> We cannot transform this to x'(t') as we could for the >> single-frequency example. > > Sure we can. sin[w_1(t + phi/w_1)] + sin[w_2(t + phi/w_2)]t' = ??? -- Oli
Reply by ●June 19, 20062006-06-19
Oli Filth wrote:> Jerry Avins said the following on 19/06/2006 16:13: >> Oli Filth wrote: >> >> ... >> >>> As for Jerry's example, let's suppose we had two components: >>> >>> x(t) = sin(w_1 t + phi) + sin(w_2 t + phi) >>> >>> We cannot transform this to x'(t') as we could for the >>> single-frequency example. >> >> Sure we can. sin[w_1(t + phi/w_1)] + sin[w_2(t + phi/w_2)] > > > t' = ???t_1' = t + phi/w_1 t_2' = t + phi/w_2 Naturally, equal phase shifts at different frequencies implies different time shifts. Show me a phase shifting scheme that inverts at DC, and I'll agree that inversion and phase shift are equal. Until then, stick with Wikipedia. Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Reply by ●June 19, 20062006-06-19
Jerry Avins wrote:> Jani Huhtanen wrote: >> Jerry Avins wrote: >> >>> Jani Huhtanen wrote: >>> >>> ... >>> >>>> Phase shift, in general, does not imply time shift. Only if the phase >>>> shift is _not_ a constant as a function of frequency, it implies time >>>> shift. Inverter is an example where phase shift does not imply time >>>> shift ;) >>> wt + phi = wt', where t' = t + phi/w. If phi is the phase shift, phi/w >>> is the time shift. If an inverter is an example of phase shift, does the >>> phase lead, or lag? >>> >>> Jerry >> >> By those definitions, I don't know. I thought that by time shift you >> meant group delay as you were also talking about transients. I'm not sure >> how "your" time shift relates to transients in question, however, IMO >> group delay would explain the transient behaviour intuitively (i.e., >> inverter does not delay nor advance the signal as the phase shift is 180' >> for all w). >Firstly, I have a feeling that I have somehow completely missed your point. Just to make sure, is this discussion purely on theoretic basis, or do the "problems" arise only when an analog inverter is actually implemented? I assume that this is a theoretic discussion ;)> 180 degrees lead, or 180 degrees lead? If inverted twice, is that 360 > degrees?Why would it matter? If I would have to answer that I would say that both, 180' and -180', but also any phase of form 360'*n+180'. And if inverted twice it would be just n*360'. By you definition of a timeshift it seems as if 370' phase shift would cause larger timeshift than 10'. Perhaps this is true for a sinusoid (however the results are identical), but it certainly does not tell that the delay would be more than in case of 10'. In general, single sinusoid of a more complex waveform does not carry information by which one could say anything about how much signal has been delayed.> An inverter works just fine with non-periodic waveforms, but > phase is a hard concept to force upon them.As far as I care, phase is a property of *a* sinusoid. More complex waveforms -periodic or not- are composed (or can be composed) of sinusoids. I don't force a concept of phase upon any other waveform (at least in the context of Fourier theory).> > Jerry-- Jani Huhtanen Tampere University of Technology, Pori
Reply by ●June 19, 20062006-06-19
Jerry Avins wrote:> Jani Huhtanen wrote: >> Jerry Avins wrote: >> >>> Jani Huhtanen wrote: >>> >>> ... >>> >>>> Phase shift, in general, does not imply time shift. Only if the phase >>>> shift is _not_ a constant as a function of frequency, it implies time >>>> shift. Inverter is an example where phase shift does not imply time >>>> shift ;) >>> wt + phi = wt', where t' = t + phi/w. If phi is the phase shift, phi/w >>> is the time shift. If an inverter is an example of phase shift, does the >>> phase lead, or lag? >>> >>> Jerry >> >> By those definitions, I don't know. I thought that by time shift you >> meant group delay as you were also talking about transients. I'm not sure >> how "your" time shift relates to transients in question, however, IMO >> group delay would explain the transient behaviour intuitively (i.e., >> inverter does not delay nor advance the signal as the phase shift is 180' >> for all w). >Firstly, I have a feeling that I have somehow completely missed your point. Just to make sure, is this discussion purely on theoretic basis, or do the "problems" arise only when an analog inverter is actually implemented? I assume that this is a theoretic discussion ;)> 180 degrees lead, or 180 degrees lead? If inverted twice, is that 360 > degrees?Why would it matter? If I would have to answer that I would say that both, 180' and -180', but also any phase of form 360'*n+180'. And if inverted twice it would be just n*360'. By you definition of a timeshift it seems as if 370' phase shift would cause larger timeshift than 10'. Perhaps this is true for a sinusoid (however the results are identical), but it certainly does not tell that the delay would be more than in case of 10'. In general, single sinusoid of a more complex waveform does not carry information by which one could say anything about how much signal has been delayed.> An inverter works just fine with non-periodic waveforms, but > phase is a hard concept to force upon them.As far as I care, phase is a property of *a* sinusoid. More complex waveforms -periodic or not- are composed (or can be composed) of sinusoids. I don't force a concept of phase upon any other waveform (at least in the context of Fourier theory).> > Jerry-- Jani Huhtanen Tampere University of Technology, Pori
