I recently built a set of 30w's which ,if they all worked out right, have a frequency resp. of 32hz- 20khz.
Is having such a wide frequncy responce from a single speaker (only 2 ways) bad for sound quality?
I find when playing bassy music that the lows tend to interfere with the low-mids... could this be that the woofers over exerting itself?
Woofer is a 6.5' poly cone Redback (sold by altronics in perth).
Is having such a wide frequncy responce from a single speaker (only 2 ways) bad for sound quality?
I find when playing bassy music that the lows tend to interfere with the low-mids... could this be that the woofers over exerting itself?
Woofer is a 6.5' poly cone Redback (sold by altronics in perth).
Yes, there is intermodulation distortion, and it is terrible news.
Take 2 tones. F1 = 50 Hz, F2 = 300 Hz. When a speaker produces both tones, frequencies are generated that are neither 50 Hz or 300 Hz. They rarely fall on the musical scale, and are very easily detected. These are called intermodulation distortion.
The more the woofer moves to create the lower tones, the greater the intermodualtion distortion when higher frequencies are played.
Any speaker trying to reproduce a 40 Hz tone at 101 dB must move 12 cu inches of air. A 6 1/2 " speaker has about 24 sq in of cone area. Therefore, in a sealed box it must move 1/2" to produce 40 Hz at 101 dB. That will definitely interfere with midrange and higher frequencies.
I will give you the formula to calculate this later. All this is taken from an article by late and very much missed Paul Klipsch in the Journal of the Audio Engineering Society.
Right now, I will just give you a graph of an 8 inch speaker, playing two tones at 90 dB. The first tone is 50 Hz, and is so low in frequency that it is not on the chart. The second tone is 300 Hz, and is clearly marked with F2. The vertical divisions are 50 Hz, the horizontal ones 10 dB. Look at the interference. Remember, this is from an 8 inch speaker-which needs to move less than a 6 1/2" speaker-played at a comparatively soft level. In a perfect world, the only tone that should show on the graph is F2 at 300 Hz. The others are all distortion tones.
The louder the level, the worse the situation becomes. Ported boxes offers an improvement over sealed box in this regard. Nevertheless, the phenomenon exists in the ported box as well.
Take 2 tones. F1 = 50 Hz, F2 = 300 Hz. When a speaker produces both tones, frequencies are generated that are neither 50 Hz or 300 Hz. They rarely fall on the musical scale, and are very easily detected. These are called intermodulation distortion.
The more the woofer moves to create the lower tones, the greater the intermodualtion distortion when higher frequencies are played.
Any speaker trying to reproduce a 40 Hz tone at 101 dB must move 12 cu inches of air. A 6 1/2 " speaker has about 24 sq in of cone area. Therefore, in a sealed box it must move 1/2" to produce 40 Hz at 101 dB. That will definitely interfere with midrange and higher frequencies.
I will give you the formula to calculate this later. All this is taken from an article by late and very much missed Paul Klipsch in the Journal of the Audio Engineering Society.
Right now, I will just give you a graph of an 8 inch speaker, playing two tones at 90 dB. The first tone is 50 Hz, and is so low in frequency that it is not on the chart. The second tone is 300 Hz, and is clearly marked with F2. The vertical divisions are 50 Hz, the horizontal ones 10 dB. Look at the interference. Remember, this is from an 8 inch speaker-which needs to move less than a 6 1/2" speaker-played at a comparatively soft level. In a perfect world, the only tone that should show on the graph is F2 at 300 Hz. The others are all distortion tones.
The louder the level, the worse the situation becomes. Ported boxes offers an improvement over sealed box in this regard. Nevertheless, the phenomenon exists in the ported box as well.
Attachments
Here is the formula and examples. I will probably be devoting a whole thread to this soon.
All of this information is taken from Paul Klipsch’s two articles in the Journal of the Audio Engineering Society. The articles are: Modulation Distortion In Loudspeakers, Part I in the JAES, April 1969, and Part II in February 1970.
All of this information is taken from Paul Klipsch’s two articles in the Journal of the Audio Engineering Society. The articles are: Modulation Distortion In Loudspeakers, Part I in the JAES, April 1969, and Part II in February 1970.
Attachments
Good idea. The less the cone must move to produce bass, the clearer will be the midrange and higher.
For some reason, modulation distortion does not get much discussion in audio circles, but I think it is a major determinant of sound quality. I think such things as crossover slope, etc, get much more discussion, but here you see foreign tones being introduced into the program material at amplitudes not far behind the music you want to hear, and nobody talks about it.
Good for you for noticing it yourself and investigating.
For some reason, modulation distortion does not get much discussion in audio circles, but I think it is a major determinant of sound quality. I think such things as crossover slope, etc, get much more discussion, but here you see foreign tones being introduced into the program material at amplitudes not far behind the music you want to hear, and nobody talks about it.
Good for you for noticing it yourself and investigating.
How did I miss this thread until now! Must have been busy...
Peter, I am looking forward to the new thread with great interest! (It should get you up to the 1K mark, for one reason)
One of the reasons I think IM distortion is sidelined is that most measurement methods, using either swept sine waves or pink noise will not show it up. It can only be seen on specific tests like the one you illustrated, and these are not commonly carried out.
IM distortion was one of the main reasons I went to three way speakers years ago, and am planning the next set to be four ways, ( maybe even 5 if I can afford the ATC dome and find a driver to fit in nicely between that and my fave Audax HD3P at about 7-8K).
Yes, crossing over that many ways is a pain, but if the mid ranges are good and wide range, which the ATCs are, then having two drivers doing low mid/bass and bass/ sub can work really well, and as each driver is covering less at the bottom end, IM problems are minimised, as well as maximising power handling and sensitivity.
I say Pah to full range speakers, Pah...
Peter, I am looking forward to the new thread with great interest! (It should get you up to the 1K mark, for one reason
)One of the reasons I think IM distortion is sidelined is that most measurement methods, using either swept sine waves or pink noise will not show it up. It can only be seen on specific tests like the one you illustrated, and these are not commonly carried out.
IM distortion was one of the main reasons I went to three way speakers years ago, and am planning the next set to be four ways, ( maybe even 5 if I can afford the ATC dome and find a driver to fit in nicely between that and my fave Audax HD3P at about 7-8K).
Yes, crossing over that many ways is a pain, but if the mid ranges are good and wide range, which the ATCs are, then having two drivers doing low mid/bass and bass/ sub can work really well, and as each driver is covering less at the bottom end, IM problems are minimised, as well as maximising power handling and sensitivity.
I say Pah to full range speakers, Pah...
If I don't start that thread within the next couple of days, either remind me on the board, or Email me. In fact, Klipsch's articles about this are one of the motivating factors for my finally breaking down and buying a scanner some months ago.
You think that the above graph is startling? Wait until you see some of his other graphs!!
I tend to agree, at least for anyone who plans to listen at volumes over 96 dB or so.
You realize, Pinkmouse, that we are going to get ourselves hammered by the full-range crowd!!
You think that the above graph is startling? Wait until you see some of his other graphs!!
pinkmouse said:I say Pah to full range speakers, Pah...
I tend to agree, at least for anyone who plans to listen at volumes over 96 dB or so.
You realize, Pinkmouse, that we are going to get ourselves hammered by the full-range crowd!!
fr0st said:
I have said for almost forever, the most important thing an active sub does is to improve the midrange.
When i 1st started saying this it was Roger's LS3/5As, KEF B139 coffins, and dB Systems bessel XOs...
An externally hosted image should be here but it was not working when we last tested it.
dave
Kelticwizard wrote
Crossover slopes have in fact much to do with IMD !!!
Reduced IMD is one of the main reasons for using high-order slopes.
OTOH high-order filters are detrimental to a speaker's temporal behaviour.
The next time you're holding some high-gloss brochure of a manufacturer in hands, saying that their speakers were "non compromise", you'll know that this is a blatant lie.
Regards
Charles
Crossover slopes have in fact much to do with IMD !!!
Reduced IMD is one of the main reasons for using high-order slopes.
OTOH high-order filters are detrimental to a speaker's temporal behaviour.
The next time you're holding some high-gloss brochure of a manufacturer in hands, saying that their speakers were "non compromise", you'll know that this is a blatant lie.
Regards
Charles
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.