Talk:Frequency

From Wikipedia, the free encyclopedia

WikiProject Physics This article is within the scope of WikiProject Physics, which collaborates on articles related to physics.
??? This article has not yet received a rating on the assessment scale. [FAQ]
??? This article has not yet received an importance rating within physics.

Help with this template Please rate this article, and then leave comments to explain the ratings and/or to identify its strengths and weaknesses.

Contents

[edit] 440 Hz

some think thisIs it not true the A tuning note is only 440 Hz in the United States, while in fact 442 Hz in Europe? Perhaps this should be changed.

See Pitch (music) — Omegatron 14:36, 20 February 2006 (UTC)

[edit] Changing wavelength

I was just wondering about the comment that when a wave goes from one medium to another, the frequency remains more or less the same, only the wavelength changes -- this doesn't make sense to me. If frequency has an inverse relationship to wavelength, how can one change without the other?

It's correct. Suppose the wave propagates 340 metre per second in the first medium, and 680 m/s in the second, and suppose the frequency is 340 hertz so that the period (i.e. the time it takes for one complete oscillation) is 1/340 seconds. These values fit a sound (a rather deep tone) passing from air into a somewhat harder medium. Then, in air, this sound travels 1 metre per period, so the wavelength is 1 metre. In the harder medium, it will travel 2 metre in the same time, doubling the wavelength.

Put differently: Yes, frequency has an inverse relationship to wavelength in a given medium, but the constant of proportionality depends on the medium.--Niels Ø 09:35, Apr 15, 2005 (UTC)


[edit] Invariance of frequency

Just wondering if any one has thought about why frequency is invariant (apart from doppler effect). ie whatever you do to a signal, you can change its wavelength and or velocity but you cant change its frequency. (I'm not considering mixers here). Any musings from anyone as to why this should be so?? --Light current 06:52, 23 November 2005 (UTC)

Because it's based on time? Theory of relativity and all that?
I'm not sure if it's worth pointing out, but the frequency of a pre-recorded signal can certainly be changed by playing it back at a different speed. Practically, if you sample something with an incorrect sampling frequency (you think it's 48 kHz but the oscillator's actually running at 48.01 kHz), then reproduce it correctly, the frequencies will be shifted.
Plus there are things like pitch shifters to simulate a change in frequency, though that's even less relevant. — Omegatron 21:32, 7 December 2005 (UTC)

Quite correct about pre recorded signals, but Im thinking more of a 'pitch shifting' method but broad band and not using mixers (multipliers). I guess it just can't be done? --Light current 22:07, 12 December 2005 (UTC)

It could probably be done with wormholes. :-) — Omegatron 15:02, 18 February 2006 (UTC)
I've heard that strong gravitation could change the frequency, so wormholes, why not :)
Oh you're right! You could change frequency just by flying around at a different altitude from your observer. Gravity Probe A was exactly that. — Omegatron 02:22, 7 June 2006 (UTC)
This is a Doppler effect. GoldenBoar 16:13, 7 June 2006 (UTC)
I don't think so. Time flows at a different speed depending on your distance from the Earth or another body. If you took a huge loudspeaker and put it in a hot air balloon, time would pass at different rates for you and the speaker, so the frequency would be shifted, since frequency is really just another word for time. — Omegatron 23:21, 7 June 2006 (UTC)
What about the Invariance of Speed, with regard to a constant medium? The above discussion and corresponding paragraph in the article both make some broad assumptions about the source of the Waves. It is also true that if the medium remains the same, speed is invariant with regard to wavelength and frequency... That is, if you have a slinky... and start sending pulses through it at a fixed speed, changing the frequency with which you initiate the pulses will only affect the wavelength, and vice-versa (i.e., frequency is NOT invariant). Representing this fact as well would probably make the article more sensible-seeming to those who really don't know anything about waves. I mean, strictly speaking, since the article is a general description of the quantity Frequency, with respect to other common quantities or terms used to describe the nature or behaviour of waves and signals - e.g., period, phase, amplitude, wavelength, cycle, etc. - this rather random statement about the Invariance of Frequency is not true, at least not without some context. - joe

I removed from the bottom of the section Frequency of waves:

Apart from being modified by the Doppler effect or any other nonlinear process, frequency is an invariant quantity in the universe. 
That is, it cannot be changed by any linearly physical process unlike velocity of propagation or wavelength.

My motivation is:

  1. The Doppler effect is a linear process and changes frequency.
  2. The statement is contradictory since it states frequency is invariant but it can be changed by the Doppler effect.
  3. The statement depends on the definition of frequency. If using, for instance the Fourier transform, frequency is the new independent variable and will be invariable since it is independent. I do not know any other definitions of frequency, e.g. the time derivative of the phase, which leave it in general as an invariant (apart from trivial cases as periodic oscillations).

Kraaiennest (talk) 17:07, 22 February 2008 (UTC)


[edit] Measurement

We have a rather strange paragraph that tells us how to measure frequency by dividing one number by another which is of course true. But do we have anything about various classes of frequency meter - i.e. how we actually measure frequency? Pcb21 Pete

[edit] radians per second

In my opinion, this is not a unit of frequency, but of angular frequency. --84.159.248.246 17:04, 20 November 2006 (UTC)

[edit] Human heartbeat

I removed the non sequitur "Notes" heading and its entry about the human heartbeat being close to one Hertz. Actually, the human heartbeat varies quite a bit--it can get down to 0.8 Hertz in mellow marathoners and up to 2 Hz in times of high stress activity. Average heart rate is about 72 bpm, or 1.2 Hz. No studies support the statement that the heartbeat is exactly 60 beats per minute which means there's no benefit to the reader by announcing that the heartbeat is approximately 1 Hz. It doesn't shed any light on the concept of frequency. Binksternet (talk) 08:36, 25 January 2008 (UTC)

[edit] Period estimators (for ocean surface waves, copied from Talk:Wave period

Mean zero up-crossing period, TZ or Tm0,2

When recordings were first taken this was onto charts and simple counts could be made. First the charts were zero-meaned (the average and trend calculated and drawn through the plot to provide a new axis for measuring) and then the number of times the wave record crossed the mean going up (or sometimes down) was counted and this gave the number of waves and, as a time measure, the zero-crossing period. The parameter is estimated by taking the mean of these periods for a given wave record. For wave records on paper the mean level is found by eye and Tvisual is estimated from the record length and the number of zero up-crossings counted on the record. This method can also be applied to digitised data using a computer but if the wave records are available in machine readable form it is preferable to estimate from the moments of the spectrum using,

Tz=sqrt(m0 / m2)

where the mi values are spectral moments. The alternative symbol of Tm0,2 is derived from the moment equation. It can be seen that TZ is very dependent on the higher frequency end of the spectrum and although TZ is the most commonly used period estimator it is not very stable.

Significant wave period, Ts

The significant wave period is the mean of the zero up-crossing periods associated with the highest one third of the waves. It is sometimes denoted by Ts. Note that this parameter cannot be obtained directly from the wave spectrum. It is not very useful, but sometimes is used!

Spectral peak period, Tp

The spectral peak period is the inverse of the frequency at which the value of the frequency spectrum is a maximum. It cannot be defined satisfactorily in multi-peaked spectra. fp is very important in characterising spectra.

Period associated with the most likely highest wave, Tmax

The most likely height of the highest wave in a record of duration 3 hours is Hmax and Tmax is its period. It is often obtained indirectly from Tz or Tp using empirical relationships, or from Hmax and steepness assumptions – usually to obtain a range of possible associated periods. These methods should be applied only in the water depth for which the empirical relationships have been found, usually deep water (i.e. depth >1/2 wavelength). It should be possible to use the steepness method in shallow water provided that refraction is minimal and that allowance can be made for shoaling effects. It cannot, however, be derived directly from the wave spectrum.

Energy period, Te

This period is important for power estimation and is used in wave power design as a preferred comparator. The most appropriate way to consider energy period is as the period of the regular wave that has the same significant height and the same power density as the sea-state under consideration. It is defined as,

Te=(m − 1 / m0)

Because of the relationship with power it is worth giving the expression for time averaged power associated with a spectrum,

power = rho g^2 m0 Te / 4Pi

Hence we can find an expression for Te as follows,

Te=(64 Pi Power)/ rho g^2 Hs^2

Average wave period, Tav

This is the inverse of the average frequency calculated from the mean of all component sine waves weighted by the spectral energy. It cannot be measured in the time domain unless the waves are simple sine curves.

Average crest period, Tc

Defined as, Tc=sqrt(m2 / m4)

This is equivalent to dividing the length of the wave record by the number of crests where a crest is any point either side of which the surface elevation decreases. Crests are not necessarily associated with zero up-crossings. Clearly this is very much influenced by the ‘tail’ of the spectrum through the fourth moment.

(copied from Talk:Wave period by Kraaiennest (talk) 21:46, 10 February 2008 (UTC)).