|
My spectrum analyser indicates that the high end of most recorded music rolls off around 12-14 KHz most of the time. Sometimes (not often) there is significant signal up to 20 KHz. The low end varies greatly. Usually nothing below about 40 Hz, but again there are times when the 20 Hz third octave is as high as any other. |
I know most acoustic instruments are between 50 hz and 3800 hz according to a chart in an old TAS magazine. |
What you guys are missing here are the overtone series that these insturments and voices produce. These are the sonic signatures that make the 440 from a piano sound different from that of a violin or other insturment. These freqencies can extend well beyond the human spectrum of hearing and are vital in listening to music. |
Close! Middle C is 261.63Hz. Nice links Pabelson, but where is the triangle? Oh the horror ... the horror! Liszt they have forsaken thee..... ;-)
|
I should mention I was referring to the fundamental frequencies of the instruments. |
My observations include the total music waveform, harmonics and all. We could debate whether a harmonic that is 40 dB down is important, but it is 40 dB down. |
Nice links, Thank you.
I look for a speaker that will output down to 40Hz. Thanks for the responses. |
Picking 40Hz means you will loose much of the reproduction of the bottom 7 keys on a piano and organ.
Remember that in general, a speaker that goes to 30Hz at 3db down will reproduce 40Hz notes clearer and more realistically than a speaker that goes to 40Hz at 3 db down. |
Zargon...My point is that those seven lowest piano keys rarely get played. A speaker that doesn't try for the lowest frequencies may reproduce higher ones better (not worse as you suggest) because it can use a smaller lighter cone driver, which is not being driven in and out to the ends of its excursion range. If one must compromise, skipping low end extension is a reasonable choice. It depends also on the quality of the bass that is present...is the response smooth or lumpy. The Magnepanar MG1.6 speaker (which I use) quits at 40 Hz, but for most music the bass sounds OK because the response is unusually smooth.
I am not saying that very low frequency capability is worthless: rather I am suggesting that a separate LF driver (called a subwoofer) is better. (That's what I use). |
Could not agree more with Eldartford. The same can be said of tuned ports to extend frequency range...often the results are less perfect than can be achieved with a separate dedicated driver (usually from a subwoofer) just for the low end. In the end it is often about compromise. Getting transducers to cover a broader frequency range helps in some sense by eliminating the challenges of cross overs but the downside is that the optimum freq range for that particular driver (size/material) may be exceeded and this usually results in more THD/IMD distortion and significantly less dynamic range.
Discrete transducers appropriately sized for their frequency range and precisely integrated using separate active electronics for each transducer is a solution that some speaker manufacturer's have adopted. A system of this type with large dynamic range and low distortion requires a four way system: subwoofer for 20 to 60 Hz, Woofer for 60 to 400 Hz, Mid range from 400 to 3000 Hz and tweeter from 3K to 20K Hz. The drawback of a system using more transducers and discrete active electronics is higher cost. The advantage is relatively low distortion and high dyanmic range (for a speaker). Nevertheless, harmonic distortion is in the 0.3% range, which is still hundreds of times more distortion than typical digital source and SS circuits. |
My comments did not assume a sub, and it didn't sound like Odnok was considering one as well.
Yes, I agree with you there are compromises. One more you didn't mention is that the speaker with the higher cutoff will also probably be more efficient and require less power.
I am using a 4 way speaker (sub built in), where the sub goes to 18Hz. Granted, not a lot going on that low, but it's great to know that if it is present, it will get attended to. |
I like this chart from the PSB website. I believe that all frequencies are important. There are orchestra pieces that go down to 16 Hz and some that are only going down to 30 Hz. I believe that there are harmonics above and below the principle tone that are important and anything less is a significant compromise. Add a sub to get that low (integration difficulties assumed).
website:
http://www.psbspeakers.com/audioTopics.php?fpId=8&page_num=1&start=0 |
I agree Tgrisham! If you look at the orchestra and the various pieces in the orchestra rep there are many that go very low. There are extensions for the double bass that allow it to play the lowest "c" on the standard keyboard, which is about 33 htz. Also if you really listen to things like bass drums you would realize that the decay can produce noises that are well below 30 htz. Additionally, there are a select group of pianos, the Boesendorfer concert grand, that have freqency extensions down to 16htz...Now the only written piece that intends to use these extra pitches were written specifically for this piano, but don't you want to have the capibility to hear every note as it was intended to be heard in the concert hall??? I actually had the opportunity to give a recital with a Bosendorfer concert grand, and I can honestly say that I enjoy its sound much more than your standard Steinway...Without the ability for your system to reproduce even the sympathetic low vibrations of any instrument, you will never be able to acheive the actual sound and hear any instrument for what it really is! |
Beware of the Black & White viewpoint that more LF extension is necessarily better....really it is more of a "gray" area. Extreme LF can be better if DONE CORRECTLY AND CAREFULLY using SPL measurements and careful Room design/treatments and a smidgen of EQ, however, in many cases extreme LF may actually be detrimental to the overall sound at LF and may even cloud some of the mid range detail. See this interesting demonstration about hearing perception at Southhampton University, which should be considered when looking at the chart suggested by Tgrisham. http://www.isvr.soton.ac.uk/SPCG/Tutorial/Tutorial/Tutorial_files/Web-hearing.htmIt shows how the pitch of an instrument can still be heard even if the loudspeaker lacks the ability to play the fundamental frequency! The higher harmonics (or partials) allow one to hear the pitch of the played note. The timbre of the note will, however, be affected by the relative magnitudes of the fundamental AND all the partials and how they decay. Since room modes (also covered on this website) become severe at very low frequencies (easily up to 15 db boosts and nulls in most cases) then a flat speaker frequency response down to extreme LF may not necessarily be desirable. The room may alter the timbre of low notes more adversely than the loudspeaker response curve....no matter the cost/quality of the loudspeaker! Therefore, with typical modest room sizes, a well designed speaker with less impressive LF extension can often outperform in LF timbre another speaker with greater LF extension. There is actually quite a lot on this website that may interest audiogoners. |
I have heard a number of pipe organ recitals in large churches, e.g. Stanford Memorial, and small, our little local 100+ year old Episcopal church has a fine pipe organ. The Velodyne HGS-15 sub gives that nearly inaudible but clearly felt sensation you get when a big pipe is opened. Though my KEF 104/2s can go much lower, I find their sound more open and airy when LF below 80 Hz is handed off to the sub.
I think what some of the earlier posters were alluding to is combination tones that can arise from the beating among harmonics, generating energy at lower (audible) frequencies. But Eldartford makes a good point: The sound pressure levels of higher harmonics are comparatively low, and combination tones from such even lower.
db |
The recording that I have with the most extensive very low frequency signal is a Wurlitzer theatre organ CD from Organ Stop Pizza (in Mesa Arizona...great place to visit). There is a continuous background at uniform level down to the 20 Hz 1/3 octave band of my spectrum analyser. Is this good? Well maybe yes, and maybe no. It is the real sound of the wind generation machinery of the organ, and if it were not there the reproduced sound would not match a live performance. But maybe this sound is a distraction, and some folk might prefer to have it gone. |
Eldartford,
One of the most popular organ works for recital is the Bach Tocatta & Fugue in D minor. My recording is Deutche Gramaphone 427 668-2. There seem to be no extraneous noises, the small pipes have the immediacy of attack you expect, and there is that glorious subtle vibration when one of the big pipes is opened. It's very close to what I have experienced in live performances -- hey, maybe the pipe organs and organists were just for show and they were actually playing a recording.
For symphonic music Mahler invokes the pedal notes to add substance, and Saint-Seans has a field day in his racous symphony. My preference has moved more to Bach, Mozart, baroque, and jazz, but I enjoy the pipe organ.
db |
Donbellphd...I too have many classical organ recordings. Theatre organs are a whole different animal.
The Wurlitzer in the Organ Stop Pizza is,they say, the largest in the world. Others have claimed this, so I don't know, but it is a big one, beautifully restored. The building was specially designed and constructed to hold the instrument. The wind generation machinery and all the air ducting is on display behind a glass wall.
And the Pizza and wine is good too. |