Another possible explanation is that the signal going to the speaker at any moment in time is an instantaneous composite of all the separate waveforms superimposed on each other. If you look at full audio on an oscilloscope, you see a very complex signal. The speaker cone simply tries to reproduce the composite as best as its frequency response permits. That may be why there is no Doppler shift with the speaker.
I think a good test for doppler would be to introduce two tones to the amplifier simultaneously, one at 2KHz and the other at 25Hz.
Listening should reveal whether or not the 2KHz tone is stable.
Listening should reveal whether or not the 2KHz tone is stable.
Bill Fitzpatrick said:
I meantioned this in a previous post using different words,
flew over everyones head
It's most audible on woofers trying to play low frequency
notes (high cone excursion) simulatenously trying to
play the delicate upper frequencies, typically the upper
frequencies will modulate, ie the voices are most noticable
like someone speaking into a fan.
People don't talk too much about this because who runs
high excursion full bandwidth systems using a single
driver? /hehe
thylantyr said:
Indeed you did. I do think that a single tone would be more revealing than vocals though.
thylantyr said:
True, but the common two way with a 6"-7" woofer/mid ought to show the problem if, indeed, one exists.
subwo1 said:
There IS doppler distortion in speakers. The theory predicts it, and the measurements confirm it. Think about the example of the ambulance passing (or the train): coming toward you you hear a higher freq tone then when it moves away from you. The fact that there are other freq at the same time is irrelevant.
It is the same with amplifiiers: yoyr statement that at any time the amp is just doing it best to reproduce the complex waveforem doesn't mean that we cannot take the signal apart in components and see distortion, signals that were not in the original signal.
As far as I remember Klipsh' papers is that he pushed horns as a way to battle doppler. Because of the very high resistive loading of the horn diafragma by the throat impedance, it needs relatively little excursion for a given loudness, which means less accelaration, which means less doppler distortion. I myself have never heard doppler distortion, but then again I was not in Woodstock. Or maybe I heard it but didn't recognise it.
Jan Didden
Re: Paper chase
You can get all the articles and conference preprints on cdrom (17 of them). I think I paid $440. Not exactly free, but definitely a bargain.
John
purplepeople said:
You can get all the articles and conference preprints on cdrom (17 of them). I think I paid $440. Not exactly free, but definitely a bargain.
John
janneman said:
I have listened to a single full-range Bandor 50mm unit and compared this with the four full-range units in my current design.
According to theory, the four drivers should have considerably lower excursion than the single for the same loudness and the single driver should therefore have considerably more Doppler distortion. However, I can't detect anything that sounds like Doppler distortion, even with the single driver.
I don't doubt that Doppler distortion exists and certainly don't question Klipsch's theory. I just doubt that this distortion has any audible effects when listening to music.
This is something Ive thougt about:
when recording something even the microphone is subject
to the "Doppler effect" so this is almost a "True Phase" issue.
Keld
when recording something even the microphone is subject
to the "Doppler effect" so this is almost a "True Phase" issue.
Keld
Distortion is caused by many things in loudspeakers, which are not linear devices in (they are only approximately linear at low excursion).
If you look at a DUMAX spec sheet you will see that the Compliance (Cms - stiffness of the suspension) and the Bl factor vary with excursion, which means that the T/S parameters (these are small signal linear parameters) vary with drive level.
One of the effects of these nonlinearities is power compression - you don't get 3dB for a 3dB increase in signal. This is manifested as a gradual rounding of the signal tops and not as "clipping" as stated previously. Clipping is better used to refer to amplifiers, which remain linear, up to the voltage rail limits, when overdriven and then cut off anything above that.
Even if you had a perfectly linear suspension and Bl product, the magnetic field created by the coil can modulate the magnetic field created by the magnet system - causing distortion. That is why things like shorting rings and T-shaped poles (which also make the fringing field more symmetrical) are used in more expensive drivers. It is important to saturate the iron near the gap to minimize this type of distortion. Even the conductivity of the iron near the gap can cause distortions. Skaaning had an interesting article about this in the JAES sometime in the past few years.
For those who want access to the JAES papers for free - you can get access to all but the preprints for free at a library in many cities and/or universities. ;-)
If you look at a DUMAX spec sheet you will see that the Compliance (Cms - stiffness of the suspension) and the Bl factor vary with excursion, which means that the T/S parameters (these are small signal linear parameters) vary with drive level.
One of the effects of these nonlinearities is power compression - you don't get 3dB for a 3dB increase in signal. This is manifested as a gradual rounding of the signal tops and not as "clipping" as stated previously. Clipping is better used to refer to amplifiers, which remain linear, up to the voltage rail limits, when overdriven and then cut off anything above that.
Even if you had a perfectly linear suspension and Bl product, the magnetic field created by the coil can modulate the magnetic field created by the magnet system - causing distortion. That is why things like shorting rings and T-shaped poles (which also make the fringing field more symmetrical) are used in more expensive drivers. It is important to saturate the iron near the gap to minimize this type of distortion. Even the conductivity of the iron near the gap can cause distortions. Skaaning had an interesting article about this in the JAES sometime in the past few years.
For those who want access to the JAES papers for free - you can get access to all but the preprints for free at a library in many cities and/or universities. ;-)
First of all: The JAES IS available online since this spring (But only for this years issues). But you have to subscribe in order to get access.
2nd: I have the article by P.W. Klipsch, but I'd have to undig it first. There are two statements that I clearly remember: Despite being famous for his corner bass-horn, PWK's main attention regarding reproduction quality was the midrange. PWK original statement: "The midrange is where we live!". As Jan already mentioned PWK talked about reduced doppler distortion (i.e. FM distortion) due to the very small excursion in compression drivers.
3rd: I don't know what the xmax of the Bandor drivers is (I once tried to ask Doreen Bance by e-mail but never got an answer at all!) but I think that these drivers "run out of steam" before the excursion is large enough to generate relevant FM distortion. Even the most linear drivers will generate IMD as well. I assume that very small drivers with small xmax will generate more IMD than FM distortion (and therefore the former might mask the latter).
The problem might be largest for two-way systems that have to go really loud, like small P.A. speakers.
4th: There is nothing like "the best speaker" (at least to me, but maybe some scanting sales-people know better) so every principle will be better in one or more discipline than all the other ones. And I am also very confident that the perfect speaker will never be built.
Depending upon source material and mood I can appreaciate many different types of them and I don't claim to get sick when having to listen to another principle than my favourite one.
Though PWK was indeed a genious he also had to SELL his stuff and so his views are as biased as anyone else's.
To mikee: While intended for P.A. usage this family of JBL drivers ARE definitely serious high-quality items (like many other professional drivers as well!). They were amongst the first comercially available drivers with paper-glassfibre composite cones BTW.
P.A. drivers in general are intended to generate high SPLs with reasonable nonlinear distortion figures. If driven at the low levels usually needed for domestic usage they can really excel in this regard.
This behaviour will get even more extreme with good horn speakers, which you don't even have to "tickle" to get domestic SPLs.
Regards
Charles
2nd: I have the article by P.W. Klipsch, but I'd have to undig it first. There are two statements that I clearly remember: Despite being famous for his corner bass-horn, PWK's main attention regarding reproduction quality was the midrange. PWK original statement: "The midrange is where we live!". As Jan already mentioned PWK talked about reduced doppler distortion (i.e. FM distortion) due to the very small excursion in compression drivers.
3rd: I don't know what the xmax of the Bandor drivers is (I once tried to ask Doreen Bance by e-mail but never got an answer at all!) but I think that these drivers "run out of steam" before the excursion is large enough to generate relevant FM distortion. Even the most linear drivers will generate IMD as well. I assume that very small drivers with small xmax will generate more IMD than FM distortion (and therefore the former might mask the latter).
The problem might be largest for two-way systems that have to go really loud, like small P.A. speakers.
4th: There is nothing like "the best speaker" (at least to me, but maybe some scanting sales-people know better) so every principle will be better in one or more discipline than all the other ones. And I am also very confident that the perfect speaker will never be built.
Depending upon source material and mood I can appreaciate many different types of them and I don't claim to get sick when having to listen to another principle than my favourite one.
Though PWK was indeed a genious he also had to SELL his stuff and so his views are as biased as anyone else's.
To mikee: While intended for P.A. usage this family of JBL drivers ARE definitely serious high-quality items (like many other professional drivers as well!). They were amongst the first comercially available drivers with paper-glassfibre composite cones BTW.
P.A. drivers in general are intended to generate high SPLs with reasonable nonlinear distortion figures. If driven at the low levels usually needed for domestic usage they can really excel in this regard.
This behaviour will get even more extreme with good horn speakers, which you don't even have to "tickle" to get domestic SPLs.
Regards
Charles
http://www.dse.co.nz/cgi-bin/dse.storefront/3f3072f809346e2c273fc0a87f9906e6/Product/View/K2806
Can be configured for either stereo amplification of up to 1 watt output per channel or bridge (mono) amplification of up to 2W output, without heatsinks
LOL
http://www.dse.co.nz/cgi-bin/dse.st...2c273fc0a87f9906e6/Export/catalogs/DTS0000104
the problem with the jbl is that it wants 104litres rear box,so a shiva is much more worth it.
someday...
Can be configured for either stereo amplification of up to 1 watt output per channel or bridge (mono) amplification of up to 2W output, without heatsinks
LOL
http://www.dse.co.nz/cgi-bin/dse.st...2c273fc0a87f9906e6/Export/catalogs/DTS0000104
the problem with the jbl is that it wants 104litres rear box,so a shiva is much more worth it.
someday...
- Status
- Not open for further replies.