Frequency range/ crossover question

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
With some rare exceptions, musical instruments' (including human voice) fundamental vibrations in usual music passages are from 100 Hz to 1 kHz. A very low pitch can go below 100 Hz and a very high pitch can go above 1 kHz. But those cases are not very often. Harmonics, which are also important but weak in SPL, can go much higher in frequency. Also sibilants go higher, too.

So, it is not quite informative to simply say 100 to 6000 Hz, for example.

Simple 300 and 3 kHz xover point recommendation is not very informative, either. It may be said that putting a xover in this critical range will cause audible phase or time domain distortion. But we should note that there are also many other factors. Using two drivers with two very different harmonic distortion profiles can also cause an audible effect of bad driver integration. Which type of xover is used can also affect the result because they give different on-axis and off-axis behaviors around the xover point.

So I'd not say putting a xover in the critical range is always a bad idea. It all depends on HOw the speaker is designed, which has many cosiderations. The question of which design choice is more audible in a good or bad way is very difficult to answer in general. Most of the time, you need to make a compromise to have a workable solution, and within the given limitation you can build sensibly good sounding speakers.
Here is a link to another discussion on the subject of human voice frequency range. In that thread this information was posted, though I don't actually know where the person got the information.

The following vocal range classifications are typically used in classical music (from highest to lowest):

Soprano (240 - 1170 Hz)
Mezzo-soprano (220 - 900 Hz)
Contralto (130 - 700 Hz)
Tenor (130 - 440 Hz)
Baritone (110 - 350 Hz)
Bass (80 - 330 Hz),2462526,2467053

It would seem that for fundamental notes, the range is from 80hz to about 1.2khz.

From another books, I know that the center of acoustical power for an orchestra is in the 500 hz range.

For telephony, the 'voice range' is considered 300 Hz to 3400 Hz. Which explains why the phone typically sounds very tinny.

I think more than anything, your speakers will dictate your crossover frequency.

I think a better question is, how much margin do I need to leave to be sure I'm well within the working range of my speakers. For example, I have a 3x9 mid-horn with a rated range of 500 to 10khz, so is 800 hz adequately for low/mid a 12db crossover? How about 700hz? How about 600hz?

Also note that frequency from the perspective of music is not linear. Octaves represent a doubling of the frequency. So the gap between one octave and the next at low frequencies is very small, at high frequencies the jump between octaves is very big. How does that come into play when determining crossovers?

Starting with 20hz, here are the octaves up to 20khz. Maybe this isn't important, maybe it is. Inquiring minds want to know.

20hz, 40, 80, 160, 320, 640, 1280, 2560, 5120, 10240, 20480hz

Sorry, for distracting from the central subject. But I feel these aspect are just as important and unanswered as the original question.

Once again, inquiring minds want to know.

Hi Everybody,

Thanks everyone, especially to Bluewizard who provided me with exactly what I was looking for. I wasn't really looking for a crossover point per se, I was more curious about where the typical voices lay on the spectrum. Also thanks for the octave information, I was curious about that too just hadn't asked.

I listen to a lot of female singers (blues, folk, etc.) and due to that I was wondering if it would be wise to avoid a crossover in that range for best effect. As it turns out my two ways crossed over at 2200 (1st order) work quite well at that (out of the critical area anyway), I was just curious, especially if I was to go with a 3 way design.

To me it didn't seem like a good idea at first to have a crossover smack dab in the middle. But then I wondered if maybe it made no difference because if the drivers are integrated well ...?

Anyway thanks for satisfying my curiosity.
With a 3-way crossover, in this day and age, you will rarely to never find a crossover above 800hz. Though at one time 800hz was the most common in both 2 way and 3-way systems, though in the case of 2-way, the mid/high was usually quite a large horn.

In a 3-way system that is sub-bass, bass, and high, the crossover would typically be in the rage of 150 to 350.

A more standard 3-way, that is bass, mid, and high, the low/mid crossover is between about 350hz and 800hz. With mid/high in the 3000hz to 6000hz range. Though I think 3khz to 5khz is most common.

Here is a link to some commercially available crossovers

Part Express 3-way -
It's going to have to drop another full octave down to 500hz and up to 8khz to make a substantial difference in the perceived output level of either the woofer or the tweeter.

That means a pretty substantial overlap between the woofer and the tweeter. At 4khz, the tweeter is still putting out a substantial amount of sound and at 1khz, the tweeter is still very much in service. That's why I prefer 12 db crossovers, even if I have to buy them off the shelf.

Also, do you have a true TWO-way First order filter? That is, do you have a capactor in-line with the tweeter and a coil in-line with the woofer? Many cheaper speakers get by with just a capacitor for the tweeter and count on the natural roll off of the woofer to do the job. Definitely not the best method.

Just a lot of talking to very little end.


PS: I'm more that willing to have people dispute what I've said.
If you have an analog or digital active crossover you might want to experiment with steep slopes, like analog LR8 48db/o or digital FIR 96db/o, around 80Hz and 3,000Hz.

Despite all the discussion on transient perfect and 1st order crossovers, I found it is better to put 80Hz-3,000Hz on the midrange driver even with phase ugly steep slopes.
That classical range given above looks about right for such music. Since that individual voices are not amplified it becomes difficult to get enough volume at much below G or A. In other musical styles the range goes deeper into the 60s. I was Bass II in an octet and we spanned a four octave range give or take.

The D two 8va below middle C is easy for me and I can do C or occasionally B on a good day. Our Soprano I had a range that spanned the two 8va above middle C. I have personally hear her nail the D above that but I don't think that was her absolute limit.

Now the thing with the lowest bass vocal notes is that they carry a fair amount of energy but very little directionality. So in theory you want the midrange to cover from say 65Hz up to 6Khz to avoid xover in the range of fundamentals and critical overtones for voice and winds but the lower end can easily be relaxed and give you the added benefit of the capability of the bass driver to move the air more effortlessly so my practical target is 100Hz to 5Khz with up to near 200 being acceptable.

This of course is essentially a WR/FR with helpers on both ends. My current setup uses a FR from about 180 on up. I can't really hear much beyond 12Khz so I have not even thought about a helper tweeter. I have played a 20Khz test tone through the system and I didn't hear a thing but my dog sure did so I guess there is some really high stuff getting through.

Just my take.

This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.