John E. Hadstate wrote:> > "Richard Owlett" <rowlett@atlascomm.net> wrote in message > news:13rho7bd0thgu1d@news.supernews.com... >> John E. Hadstate wrote: >>> >>> "Richard Owlett" <rowlett@atlascomm.net> wrote in message >>> news:13rg6q7a1esbs64@news.supernews.com... >>> >>>> glen herrmannsfeldt wrote: >>> >>> >>>>> I was watching the usual car >>>>> commercial where the (aliased) wheels >>>>> are turning slowly backwards and >>>>> realized that sometimes negative >>>>> frequencies do exist. >>>>> >>>> Never thought of stroboscopic effects >>>> in terms of sampling theory. >>> >>> The flash of a strobe (or the wink of the camera) is "sampling".... >>> >> >> But still haven't really wrapped my mind around aliasing and Nyquist >> being closely related. > > In the analog (continuous) world, there is a related effect called > "heterodyning". When one multiplies two periodic signals, one generates > sum and difference frequencies. Pointing a strobe at a rotating object > causes the reflected light to be a product of the strobe's nearly > "On-Off" illumination (multiplying by 0 or 1) and whatever is going on > in the illuminated scene. When the strobe is synchronized with some > periodic event in the scene, the "beat frequency" will be zero and the > event will appear to be frozen in time. (The beat frequency is the > difference frequency, I don't have a clue as to how to observe the sum > frequency.) > > Another example of this phenomenon is found in the ancient technology of > the "synchronizing lamp". Suppose we have two AC generators, one > on-line and one in standby. We want to bring the standby generator to > the same frequency and phase as the currently on-line generator so we > can connect them in parallel. ("Whenever I connect them together she go > 'boom'" -- you had to be there.) > > If we hook a sufficiently robust light bulb between the generators > (assuming of course that they have a common ground), we will see this > heterodyning take place as the speed of the stand-by generator is > brought up to match the speed of the on-line generator. When the light > bulb goes dark (phase match) and stays that way (frequency match), it's > safe to connect the generators in parallel.*Dark* is the key word. A classmate got the attention of everybody in the machinery lab when he went for maximum brightness. :-) Jerry -- Engineering is the art of making what you want from things you can get. �����������������������������������������������������������������������
Possibly unrelated questions
Started by ●February 15, 2008
Reply by ●February 18, 20082008-02-18
Reply by ●February 18, 20082008-02-18
On Feb 18, 2:36 am, "John E. Hadstate" <jh113...@hotmail.com> wrote:> "Richard Owlett" <rowl...@atlascomm.net> wrote in message > > news:13rho7bd0thgu1d@news.supernews.com... > > > > > John E. Hadstate wrote: > > >> "Richard Owlett" <rowl...@atlascomm.net> wrote in message > >>news:13rg6q7a1esbs64@news.supernews.com... > > >>> glen herrmannsfeldt wrote: > > >>>> I was watching the usual car > >>>> commercial where the (aliased) wheels > >>>> are turning slowly backwards and > >>>> realized that sometimes negative > >>>> frequencies do exist. > > >>> Never thought of stroboscopic effects > >>> in terms of sampling theory. > > >> The flash of a strobe (or the wink of the camera) is > >> "sampling".... > > > But still haven't really wrapped my mind around aliasing and > > Nyquist > > being closely related. > > In the analog (continuous) world, there is a related effect > called "heterodyning". When one multiplies two periodic > signals, one generates sum and difference frequencies. > Pointing a strobe at a rotating object causes the reflected > light to be a product of the strobe's nearly "On-Off" > illumination (multiplying by 0 or 1) and whatever is going on > in the illuminated scene. When the strobe is synchronized with > some periodic event in the scene, the "beat frequency" will be > zero and the event will appear to be frozen in time. (The beat > frequency is the difference frequency, I don't have a clue as > to how to observe the sum frequency.) > > Another example of this phenomenon is found in the ancient > technology of the "synchronizing lamp".A common tool of piano tuners used to be (maybe still is) the strobe tuner. Drop a microphone into the piano, hook the amp up a light bulb on a rotating disk with a precisely controlled rotational rate. Play with the piano's tuning pegs until the light pattern (or multiple patterns caused by the harmonics) stood still, or drifted around by just the right amount for the amount of stretch that made that type of piano sound best. This type of visual feedback allowed more precise tuning that could be done by (an untrained ear) listening for decaying beats. This form of feedback is so good that imitating it was one of the best ways I found for a displaying the output of a digital frequency estimation based algorithm for a guitar tuning application. IMHO. YMMV. -- rhn A.T nicholson d.0.t C-o-M http://www.nicholson.com/rhn/dsp.html
Reply by ●February 18, 20082008-02-18
> "Richard Owlett" <rowl...@atlascomm.net> wrote in message > news:13rho7bd0thgu1d@news.supernews.com... > > John E. Hadstate wrote: > >> "Richard Owlett" <rowl...@atlascomm.net> wrote in message > >>news:13rg6q7a1esbs64@news.supernews.com... > >>> glen herrmannsfeldt wrote:> >>>> I was watching the usual car > >>>> commercial where the (aliased) wheels > >>>> are turning slowly backwards and > >>>> realized that sometimes negative > >>>> frequencies do exist.> >>> Never thought of stroboscopic effects > >>> in terms of sampling theory.> >> The flash of a strobe (or the wink of the camera) is > >> "sampling"....> > But still haven't really wrapped my mind around aliasing and > > Nyquist > > being closely related.> In the analog (continuous) world, there is a related effect > called "heterodyning". When one multiplies two periodic > signals, one generates sum and difference frequencies. > Pointing a strobe at a rotating object causes the reflected > light to be a product of the strobe's nearly "On-Off" > illumination (multiplying by 0 or 1) and whatever is going on > in the illuminated scene. When the strobe is synchronized with > some periodic event in the scene, the "beat frequency" will be > zero and the event will appear to be frozen in time. (The beat > frequency is the difference frequency, I don't have a clue as > to how to observe the sum frequency.)> Another example of this phenomenon is found in the ancient > technology of the "synchronizing lamp".A common tool of piano tuners used to be (maybe still is) the strobe tuner. Drop a microphone into the piano, hook the amp up to a light bulb on a rotating disk with a precisely controlled rotational rate. Play with the piano's tuning pegs until the rotating light pattern (or multiple patterns caused by the harmonics) stood still, or drifted around by just the right amount for the amount of stretch that made that type of piano sound best. This type of visual feedback allowed more precise tuning than could be done by (an untrained ear) listening for decaying beats. This form of feedback is so good that imitating it was one of the best ways I found for a displaying the output of a digital frequency estimation based algorithm for a guitar tuning application. IMHO. YMMV. -- rhn A.T nicholson d.0.t C-o-M http://www.nicholson.com/rhn/dsp.html
Reply by ●February 19, 20082008-02-19
Ron N. wrote: (snip)> A common tool of piano tuners used to be (maybe still > is) the strobe tuner. Drop a microphone into the piano, > hook the amp up to a light bulb on a rotating disk with > a precisely controlled rotational rate. Play with the > piano's tuning pegs until the rotating light pattern (or > multiple patterns caused by the harmonics) stood still, > or drifted around by just the right amount for the amount > of stretch that made that type of piano sound best.Last I saw a piano being tuned, the tuner used a palmtop computer with an attached microphone. The visual display in some cases was a pattern of dots that would rotate in just the way you say. I believe it could also give the frequency or frequency ratio. Piano tuners measure frequency shifts in cents, 100th of a semitone or 1200 cents per octave. That is, 1200*log2(frequency ratio), where the ratio might be the actual frequency relative to the desired frequency. -- glen
Reply by ●February 19, 20082008-02-19
On 17 Feb, 13:46, "John E. Hadstate" <jh113...@hotmail.com> wrote:> "Richard Owlett" <rowl...@atlascomm.net> wrote in message > > news:13rg6q7a1esbs64@news.supernews.com... > > > glen herrmannsfeldt wrote: > >> I was watching the usual car > >> commercial where the (aliased) wheels > >> are turning slowly backwards and > >> realized that sometimes negative > >> frequencies do exist. > > > Never thought of stroboscopic effects in terms of sampling > > theory. Are wheels going "backwards" and stop/slow > > motion effects generally really an aliasing effect or > > artifact of not > > meeting a Nyquist related criterion? > > I slightly disagree with Rune's answer, "sort of". �Assuming > your question is posed to need a Yes/No answer, I think the > answer is "absolutely".Well, technically speaking you are right. However, sampling and aliasing in more than one dimension is not at all trivial to understand. It is common to use severe spatial undersampling in seismic applications. However, because of the way the spatial spectra combine with the time dimension, a unique 2D spectrum can be obtained. The movie in question is a 3D signal (time + two spatial dimensions), so I would prefer not to get too involved with the interpretation of the spectrum with coresponding aliasing effects. Rune
Reply by ●February 19, 20082008-02-19
Rune Allnor wrote:> On 17 Feb, 13:46, "John E. Hadstate" <jh113...@hotmail.com> wrote: > >>"Richard Owlett" <rowl...@atlascomm.net> wrote in message >> >>news:13rg6q7a1esbs64@news.supernews.com... >> >> >>>glen herrmannsfeldt wrote: >>> >>>>I was watching the usual car >>>>commercial where the (aliased) wheels >>>>are turning slowly backwards and >>>>realized that sometimes negative >>>>frequencies do exist. >> >>>Never thought of stroboscopic effects in terms of sampling >>>theory. Are wheels going "backwards" and stop/slow >>>motion effects generally really an aliasing effect or >>>artifact of not >>>meeting a Nyquist related criterion? >> >>I slightly disagree with Rune's answer, "sort of". Assuming >>your question is posed to need a Yes/No answer, I think the >>answer is "absolutely". > > > Well, technically speaking you are right. However, sampling > and aliasing in more than one dimension is not at all trivial > to understand. It is common to use severe spatial undersampling > in seismic applications. However, because of the way the spatial > spectra combine with the time dimension, a unique 2D spectrum > can be obtained. > > The movie in question is a 3D signal (time + two spatial > dimensions), so I would prefer not to get too involved with > the interpretation of the spectrum with coresponding > aliasing effects. > > RuneI'd think it would be time + 1 spatial as the effect is seen if camera is paned to keep the axle in the same position in the frame. The spatial dimension would then be the angular position of the wheel.
Reply by ●February 19, 20082008-02-19
On 19 Feb, 21:12, Richard Owlett <rowl...@atlascomm.net> wrote:> Rune Allnor wrote: > > On 17 Feb, 13:46, "John E. Hadstate" <jh113...@hotmail.com> wrote: > > >>"Richard Owlett" <rowl...@atlascomm.net> wrote in message > > >>news:13rg6q7a1esbs64@news.supernews.com... > > >>>glen herrmannsfeldt wrote: > > >>>>I was watching the usual car > >>>>commercial where the (aliased) wheels > >>>>are turning slowly backwards and > >>>>realized that sometimes negative > >>>>frequencies do exist....> I'd think it would be time + 1 spatial as the effect is seen if camera > is paned to keep the axle in the same position in the frame. The spatial > dimension would then be the angular position of the wheel.Yes, it's the combined effect of temporal and spatial aliasing that causes the illusion. Which is why I was reluctant to get into too detailed discussions about this effect. Rune
Reply by ●February 19, 20082008-02-19
glen herrmannsfeldt <gah@ugcs.caltech.edu> writes:> Ron N. wrote: > (snip) > >> A common tool of piano tuners used to be (maybe still >> is) the strobe tuner. Drop a microphone into the piano, >> hook the amp up to a light bulb on a rotating disk with >> a precisely controlled rotational rate. Play with the >> piano's tuning pegs until the rotating light pattern (or >> multiple patterns caused by the harmonics) stood still, >> or drifted around by just the right amount for the amount >> of stretch that made that type of piano sound best. > > Last I saw a piano being tuned, the tuner used a palmtop > computer with an attached microphone. The visual display > in some cases was a pattern of dots that would rotate in > just the way you say. I believe it could also give > the frequency or frequency ratio. > > Piano tuners measure frequency shifts in cents, 100th of a > semitone or 1200 cents per octave. That is, > 1200*log2(frequency ratio), where the ratio might be the > actual frequency relative to the desired frequency.One of my tuners uses a laptop. I believe that the software he uses doesn't just tune to a perfect equal temperament but also accounts for inharmonicity: http://en.wikipedia.org/wiki/Piano_tuning -- % Randy Yates % "Rollin' and riding and slippin' and %% Fuquay-Varina, NC % sliding, it's magic." %%% 919-577-9882 % %%%% <yates@ieee.org> % 'Living' Thing', *A New World Record*, ELO http://www.digitalsignallabs.com
Reply by ●February 19, 20082008-02-19
Randy Yates wrote: (snip)> One of my tuners uses a laptop. I believe that the software he uses > doesn't just tune to a perfect equal temperament but also accounts for > inharmonicity:> http://en.wikipedia.org/wiki/Piano_tuningPiano strings (being wire) have some stiffness to bending which increases the higher harmonics slightly. What that article doesn't describe is that most notes on a piano have two or three strings, and those strings are not tuned to exactly the same frequency. -- glen
Reply by ●February 19, 20082008-02-19
glen herrmannsfeldt <gah@ugcs.caltech.edu> writes:> [...] > What that article doesn't describe is that most notes on a piano > have two or three strings, and those strings are not tuned to exactly > the same frequency.On a real piano, even if you tuned the unisons at 8:00am on Monday morning to the exact same frequency, at 9:00am on Monday they would be ever-so-slightly off. So, yes and no. I think the goal is to tune them exactly, with the foreknowledge that they will be out and "chorus" within a short period. Also, a tuner is not going to spend the time necessary to make them "exactly" the same - maybe 0.1 Hz close - which is probably far enough off anyway to get a chorusing effect (i.e., so that the tone won't be so "straight" as a perfect tuning would produce). NB: This is my conjecture; I'm not a piano tuner (at least not a very good one - I did tune my old Yamaha CP70 back in the 80's, but that's a different story...) -- % Randy Yates % "She's sweet on Wagner-I think she'd die for Beethoven. %% Fuquay-Varina, NC % She love the way Puccini lays down a tune, and %%% 919-577-9882 % Verdi's always creepin' from her room." %%%% <yates@ieee.org> % "Rockaria", *A New World Record*, ELO http://www.digitalsignallabs.com
