Ballistic Theory, Progress report...Suitable for 5yo Kids
in [Relativity]
|
Prev: OWLS is not equal to c
Next: Mathematical Inconsistencies in Einstein's Derivation of the Lorentz Transformation
From: bz on 13 Jul 2005 06:24 "George Dishman" <george(a)briar.demon.co.uk> wrote in news:db2e6u$f1u$1 @news.freedom2surf.net: > Frequency is a measure of momentum so an > accurately known momentum implies a single > frequency, but the bandwidth of a tone burst > is inversely proportional to the duration. > The uncertainty of the value of the momentum > therefore relates inversely to the 'length' > of the photon but it is hard to say where a > 'long' burst of sine wave is located. This > is basically another way of looking at the > uncertainty principle, dx * dp has a minimum > value. > Photons are not tone bursts. That we might have difficulty accurately measuring the frequency/wavelength/energy of a single photon would not seem to require that those values are broadened by our uncertanty. Femto and even atto second laser pulses have been produced that are less than two periods of the wavelength involved. This would seem to set an upper limit on the number of cycles in a photon. Logic says that a pulse can not be shorter than the time it takes to create a single photon. It would also seem to say that a single photon can not "be longer" than the shortest laser pulse. This review of techniques will give you a bit of an overview of the field. <http://phys.strath.ac.uk/alpha-x/Assets/articles/Reid-time-resolved- spectroscopy-2003.pdf> -- bz please pardon my infinite ignorance, the set-of-things-I-do-not-know is an infinite set. bz+sp(a)ch100-5.chem.lsu.edu remove ch100-5 to avoid spam trap
From: George Dishman on 13 Jul 2005 08:27 "bz" <bz+sp(a)ch100-5.chem.lsu.edu> wrote in message news:Xns9692372FE82F4WQAHBGMXSZHVspammote(a)130.39.198.139... > "George Dishman" <george(a)briar.demon.co.uk> wrote in news:db2e6u$f1u$1 > @news.freedom2surf.net: > >> Frequency is a measure of momentum so an >> accurately known momentum implies a single >> frequency, but the bandwidth of a tone burst >> is inversely proportional to the duration. >> The uncertainty of the value of the momentum >> therefore relates inversely to the 'length' >> of the photon but it is hard to say where a >> 'long' burst of sine wave is located. This >> is basically another way of looking at the >> uncertainty principle, dx * dp has a minimum >> value. >> > > Photons are not tone bursts. I'm not suggesting they are, AFAIK they are point particles, but those particles seem to be subject to Heisenberg and there appear to be parallels. > That we might have difficulty accurately measuring the > frequency/wavelength/energy of a single photon would not seem to require > that those values are broadened by our uncertanty. QM seems to differ with that view, or you are getting into 'hidden variable' territory. I'm not sufficiently familiar with QM these days to argue the point though. > Femto and even atto second laser pulses have been produced that are less > than two periods of the wavelength involved. > > This would seem to set an upper limit on the number of cycles in a photon. Certainly, but from the paper you cite "The large bandwidth of femtosecond pulses causes experimental difficulties." Chopping a pure sinewave creates sidebands hence increases the bandwidth. Think of a Fourier analysis of the chopping waveform. Now I would think a single photon cannot have a bandwidth but if you take a single photon from a stream with a wide bandwidth, then that would translate into uncertainty about the energy of the particular photon. > Logic says that a pulse can not be shorter than the time it takes to > create > a single photon. It would also seem to say that a single photon can not > "be > longer" than the shortest laser pulse. I put length in quotes because IMHO a photon is a particle, but I think this is another aspect of duality. > This review of techniques will give you a bit of an overview of the field. > <http://phys.strath.ac.uk/alpha-x/Assets/articles/Reid-time-resolved- > spectroscopy-2003.pdf> Excellent stuff, it will take me some time to read that but thanks! George
From: bz on 13 Jul 2005 11:26 "George Dishman" <george(a)briar.demon.co.uk> wrote in news:db3123$lv1$1(a)news.freedom2surf.net: > > "bz" <bz+sp(a)ch100-5.chem.lsu.edu> wrote in message > news:Xns9692372FE82F4WQAHBGMXSZHVspammote(a)130.39.198.139... >> "George Dishman" <george(a)briar.demon.co.uk> wrote in news:db2e6u$f1u$1 >> @news.freedom2surf.net: >> >>> Frequency is a measure of momentum so an >>> accurately known momentum implies a single >>> frequency, but the bandwidth of a tone burst >>> is inversely proportional to the duration. >>> The uncertainty of the value of the momentum >>> therefore relates inversely to the 'length' >>> of the photon but it is hard to say where a >>> 'long' burst of sine wave is located. This >>> is basically another way of looking at the >>> uncertainty principle, dx * dp has a minimum >>> value. >>> >> >> Photons are not tone bursts. > > I'm not suggesting they are, AFAIK they > are point particles, but those particles > seem to be subject to Heisenberg and there > appear to be parallels. I agree. > >> That we might have difficulty accurately measuring the >> frequency/wavelength/energy of a single photon would not seem to >> require that those values are broadened by our uncertanty. > > QM seems to differ with that view, or you > are getting into 'hidden variable' territory. > I'm not sufficiently familiar with QM these > days to argue the point though. likewise. > >> Femto and even atto second laser pulses have been produced that are >> less than two periods of the wavelength involved. >> >> This would seem to set an upper limit on the number of cycles in a >> photon. > > Certainly, but from the paper you cite > > "The large bandwidth of femtosecond pulses > causes experimental difficulties." I am not surprised. Rapidly keying a radio transmitter also creates difficulties. Part of the problem is that a high Q circuit element tends to 'ring'. > > Chopping a pure sinewave creates sidebands > hence increases the bandwidth. Quite true.... especially if the chopping isn't done at the time of zero crossing. The antenna would also need to be low Q and non reactive so that current and voltage would be in phase. > Think of a > Fourier analysis of the chopping waveform. > Now I would think a single photon cannot have > a bandwidth I agree. > but if you take a single photon > from a stream with a wide bandwidth, then > that would translate into uncertainty about > the energy of the particular photon. right. On the other hand, if you have a narrow bandwidth beam of photons and you 'chop' it, into small slices, mechanically, I am NOT sure that we would generate sidebands, like 'normal' modulation would. [how does one photon know that those ahead of it or behind it have been absorbed?] If we chopped it fine enough, we should have a single photon, of known energy/wavelength/frequency. We would almost certainly NOT know its exact position, however. I think time would be the expresion of uncertanty. >> Logic says that a pulse can not be shorter than the time it takes to >> create >> a single photon. It would also seem to say that a single photon can not >> "be >> longer" than the shortest laser pulse. > > I put length in quotes because IMHO a photon > is a particle, but I think this is another > aspect of duality. > >> This review of techniques will give you a bit of an overview of the >> field. >> <http://phys.strath.ac.uk/alpha-x/Assets/articles/Reid-time-resolved- >> spectroscopy-2003.pdf> > > Excellent stuff, it will take me some time > to read that but thanks! Quite welcome. -- bz please pardon my infinite ignorance, the set-of-things-I-do-not-know is an infinite set. bz+sp(a)ch100-5.chem.lsu.edu remove ch100-5 to avoid spam trap
From: George Dishman on 15 Jul 2005 15:50 "bz" <bz+sp(a)ch100-5.chem.lsu.edu> wrote in message news:Xns96926A4D6D049WQAHBGMXSZHVspammote(a)130.39.198.139... > "George Dishman" <george(a)briar.demon.co.uk> wrote in > news:db3123$lv1$1(a)news.freedom2surf.net: .... > I agree. .... > likewise. .... >> "The large bandwidth of femtosecond pulses >> causes experimental difficulties." > > I am not surprised. Rapidly keying a radio transmitter also creates > difficulties. Part of the problem is that a high Q circuit element tends > to 'ring'. See below. >> Chopping a pure sinewave creates sidebands >> hence increases the bandwidth. > > Quite true.... especially if the chopping isn't done at the time of zero > crossing. Chopping is a severe form of ampitude modulation so you can think of the process as if you were multiplying the sine wave with a digital waveform. That is the same as heterodyning the Fourier Transform of the chopping waveform with the sine wave which acts as a carrier. > The antenna would also need to be low Q and non reactive so that current > and voltage would be in phase. The chopping creates sidebands. The Q of the circuit then acts as a filter which reduces the sidebands. You might think you could get arbitrarily narrow bandwidth and short bursts but the ringing of the circuit will extend the burst and the higher the Q, the longer it rings. In fact you can imagine that the early part of a burst starts the circuit ringing and the interference between that and the latter part of the pulse is what cancels it out and creates the filter action. >> Think of a >> Fourier analysis of the chopping waveform. >> Now I would think a single photon cannot have >> a bandwidth > > I agree. > >> but if you take a single photon >> from a stream with a wide bandwidth, then >> that would translate into uncertainty about >> the energy of the particular photon. > > right. > > On the other hand, if you have a narrow bandwidth beam of photons and you > 'chop' it, into small slices, mechanically, I am NOT sure that we would > generate sidebands, like 'normal' modulation would. [how does one photon > know that those ahead of it or behind it have been absorbed?] You were talking of pulse lengths of a couple of cycles. Compared to laser coherence lengths of metres, we are talking of letting through a tiny sample of one photon :-) OK, since they are particles, the way I expect that to work is that you get a fractional probability that the photon makes it through the shutter. > If we chopped it fine enough, we should have a single photon, of known > energy/wavelength/frequency. We would almost certainly NOT know its exact > position, however. I think time would be the expresion of uncertanty. The relevant factors are dE*dt or dp*dx of course. If a shutter is used and is open for a very short time then you know t and x very accurately so dE and dp become poorly defined. Of course both depend on the frequency of the photon so I expect a side effect of the shutter operation would be to scatter the photons that get through in some way that adds a random factor to the energy/momentum and hence broadens the linewidth. However, I haven't used lasers in thirty years and never worked with very short pulses so I'm guessing. Perhaps the paper will clue me in a bit when I get a chance to read it. George
From: bz on 15 Jul 2005 17:37
"George Dishman" <george(a)briar.demon.co.uk> wrote in news:db93ro$ogf$1 @news.freedom2surf.net: >> On the other hand, if you have a narrow bandwidth beam of photons and you >> 'chop' it, into small slices, mechanically, I am NOT sure that we would >> generate sidebands, like 'normal' modulation would. [how does one photon >> know that those ahead of it or behind it have been absorbed?] > > You were talking of pulse lengths of a couple > of cycles. Of LESS than a couple of cycles. > Compared to laser coherence lengths > of metres, we are talking of letting through > a tiny sample of one photon :-) I see no reason for a photon to be longer than one cycle. Coherence length isn't the length of the photons, it tells us how big a chunk of light is 'phase, frequency, polarization, coherent'. It is in some sense 'the length of the cavity' or at least it is clearly related to the length of the laser cavity. The longer the cavity, the longer the coherence length. My impression is that it usually represents the stimulated emission from a single pass through the laser cavity. > OK, since they are particles, the way I expect > that to work is that you get a fractional > probability that the photon makes it through > the shutter. Why do you think that a single photon must be longer than one cycle? >> If we chopped it fine enough, we should have a single photon, of known >> energy/wavelength/frequency. We would almost certainly NOT know its >> exact position, however. I think time would be the expresion of >> uncertanty. > > The relevant factors are dE*dt or dp*dx of course. > > If a shutter is used and is open for a very short > time then you know t and x very accurately If I don't know, within a small fraction of a cycle, when the photon is, then I don't know 't' very accurately. > so dE > and dp become poorly defined. Of course both depend > on the frequency of the photon so I expect a side > effect of the shutter operation would be to scatter > the photons that get through in some way that adds > a random factor to the energy/momentum and hence > broadens the linewidth. However, I haven't used > lasers in thirty years and never worked with very > short pulses so I'm guessing. Perhaps the paper will > clue me in a bit when I get a chance to read it. In the work we did with the optogalvanic effect induced by dye laser pulses in plasma, we were not working with single photons, nor with extremely short pulses. That was in the early 90s. I also worked with YAG and CO2 lasers in the early 70s, using them to cut aluminum oxide and to adjust resistors to value. One CO2 laser was 50 W, CW, the other was 500 W, CW. The yags were much lower average power and pulsed. There is a paper http://jchemed.chem.wisc.edu/JCEWWW/Articles/DynaPub/DynaPub.html#ref16 That I disagree with. They appear to believe that photons consist of wavetrains that are millions of cycles long. I see no reason for Radio Frequency Photons to be any different from light photons as to the number of cycles per photon. If that is true, *and* IF they were right THEN there would be no way for me to key a 1.8 MHz transmitter at 30 wpm [where keying rate is about 12 dots per second]. I know for a fact that transmitters opperating at much lower frequencies (in the long wave marine band between 200 and 500 kHz) have been operated with keying speed much higher than 30 wpm. Since transmitters operating at much lower frequencies are regularly keyed at much higher switching rates, their claims of millions of cycles per photon [if RF and Light photons are similar] are clearly false. -- bz please pardon my infinite ignorance, the set-of-things-I-do-not-know is an infinite set. bz+sp(a)ch100-5.chem.lsu.edu remove ch100-5 to avoid spam trap |