|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
63.142.209.123
I would look this up online, but I don't know where to start...
A typical speaker crossover divides up frequency reproduction duties based on hz. Lets say a driver is busy with a certain sound (or frequency) and along comes another sound in its "range of duty". How do those multiple "sounds" affect each other? What if one of those sounds is exactly twice or 1/2 the frequency of the other? My friend once told me that almost any speaker is better off with a subwoofer. Not because they go lower, but because they take some of the work away from the mains so that they can do what they are best at a little easier.
Follow Ups:
Ok. Let's see if this works...
Take your hand, extend it palm down fingers together.
Now move your hand quickly up and down in short movements. Like you're shaking from a bad night of drinking.
Now... move your hand... *while continuing the shaking movement*... slowly up and down about six inches up and six inches down.
Congratulations. Your hand is moving at two frequencies at once!
Interesting fact: no matter how many "frequencies" a membrane or object is moving at once, if it's a theoretically rigid membrane then it can only move in ONE direction at any one time, even when producing ten different 'frequencies' at once. The trick is that moving things don't reproduce 'frequencies' - they move in complex ways. A single tone (frequency) is just the most simple way a membrane can move. :)
Cheers,
Presto
I get what you are saying, but (perhaps because I am human) those motions sometimes affect each other. And what about the 2nd part of my question...what about if one frequency is exactly twice that of another?
Just for example, Richard Modaferri (formerly of McIntosh Labs) developed an "infinite slope" crossover design. This is just one of scores of crossover designs from "crossover-less" to very complicated ones. It is the implementation of crossovers that makes the difference between an ordinary speaker and an extraordinary one.
...basic answer - do a lot of listening.
If you can hear the effect of the crossover or the different drivers playing music, then it is not very well designed.
Yes a subwoofer will take the burden off the the main speaker's woofer, but blending the sub's sound in seamlessly can be more challenging.
If you can hear individual drivers then you do have a poor design.
Part of the design is to seamlessly crossover between drivers. Like a whisper in the wind.
Not an easy task is it!
A driver doesn't play any given frequency when playing music. It plays a single complex wave form that is the algebraic sum of all the frequencies of the sound at any given moment.
This'll start you off better than any "on-line" quickie article.
nt
Dman
Analog Junkie
Here's another way to look at things. Unless you are playing pure sine waves, your speaker already has a set of complex sound waves to deal with. For example, middle C on a piano is 261 Hz. Lots of other instruments play the same note at the same frequency, yet don't sound the same as the piano's note. This is due to their differing harmonics, attack & decay spectrum and other characteristics that make the note sound unique to that instrument.
In order to accurately reproduce the piano note, your speaker is already playing multiple frequencies at once within its active frequency range (keep in mind that most piano notes have two or three separate strings that are struck at once). Adding the notes of additional instruments within the driver's frequency range is just more of the same.
Yes, some speakers are better than others in terms of accurate reproduction, but this also depends on what sound characteristic you're focusing on at the moment.
The harmonics on the continuous tone idea is irrelevant to a piano - or any other percussion instrument - as its notes contain just an attack and a decay. But their characteristic timbre is easily identified right away.
Even when listening to instruments playing a note with a continuous tone as well we pay far less attention to that bit than to the attack and decay. The continuous bit comes last in importance.
This science predates 2ch stereo.
It goes right to the heart of musical expression, interplay, nuance and expression.
When I found this out - in my late 20s nearly 40 years ago - it changed the way i thought about audio in the home. The room and gear and you, and where you sit, are a system trying to decode music performance.
Viz. Q, rise and decay times are orthogonal to frequency.
One of the most ignored feature of electronic audio circuits is matching rise and decay times.
Clear thinking about audio will remain difficult until 'the harmonics on the continuous tone = timbre' shibboleth is dead and buried.
Warmest
Tim Bailey
Skeptical Measurer & Audio Scrounger
Apparently you are not familiar with books. Go to your local library -- you can get directions on your computer -- and ask for books on loudspeakers. There are several good ones. At your beginning level, most any one will suffice.
Check one out and read it.
Read on how a crossover works. How to choose drivers. What are the different circuits in a speaker.
After you read the article you should be more confused. As well as appreciate the complexity in designing a good speaker.
Also you may understand how today with the aid of computer modeling a lot of the guess work has been taken out of the design and building phase. As opposed to the old days when they just used some basic math and winged it.
But still you can use all the computer modeling you want. Sometimes it works and sometimes it still sounds like crap. It's still a mystery at times.
Read it's the only way you'll understand the basic ideas of how a speaker works. I don't claim to know but I do tell a lot of tall tails.
.
Lets say a driver is busy with a certain sound (or frequency) and along comes another sound in its "range of duty". How do those multiple "sounds" affect each other?
So long as the driver exhibits consistent directivity with both sets of frequencies and is driven in its linear range, you simply hear the added content. Other times, however, an instrument generates overtones or another instrument's output is added outside the driver's optimum directivity pattern. Then the image's width clamps down like a fun house mirror effect. Listen to most any 70's JBL monitor like a 4310/4311 to hear what I'm referring to.
My friend once told me that almost any speaker is better off with a subwoofer.
Only those that are inherently bass challenged at the intended output level. It does reduce overdriving a driver that cannot handle the higher excursion requirements of the bottom octaves and reduces doubling distortion.
wasn't the 4310f and 4311 designed as actual studio monitors. Predating our idea of "monitor" speaker. But to be using in a studio and not called a studio monitor.
Floyd Toole would be appalled at the excess of bandwidth attempted by the 5" midrange. Fortunately, you don't find that overt fkup with current designs like the LSR6332 . It cuts off the same size driver a full octave lower.
The 4310/4311 series imaging is comic book - wide at the bottom and top and decidedly pinched in the upper midrange where the piston diameter is significantly larger than the wavelength it attempts to reproduce.
The kind of distortion you are describing is the kind Bob Carver specifically designed to diminish with his new Amazing Loudspeakers. Read the review in TAS by Robert E. Green to an excellent description.
Which is one reason why I've favored and used full range electrostats for nearly forty years. And confess that I've never been fan of hybrid monopole woofer / dipolar planar or electrostatic based systems to avoid discontinuity experienced elsewhere.
One pebble in the pond. :)
nt
The driver's piston diameter (typically about 1" less than driver size) should never be used for frequencies smaller than the wavelength or else the imaging will be compromised.
Huh? Start with a wavelength calculator as found here .
First example: New Advent with 9" piston crossed over at 1500 hz.
Wavelength is 9.04 inches. Ok.
Second example: Polk LSi9 with 4" piston crossed over at 2500 hz.
Wavelength is 5.4 inches. Ok.
JBL 4311 with 4" piston crossed over at 6000 hz.
Wavelength is 2.25 inches. Fail!
So that would indicate that pretty much all full range drivers cannot image properly?
I don't have any experience using FR speakers but a lot of experienced audiophiles seem to like them.
Directivity varies greatly by frequency. What I find less disturbing is those designs start exhibiting a gradual narrowing of dispersion as frequency goes up. There's no drastic contrast.
With the JBL's I mentioned, however, the 5" midrange is driven beyond its ideal range - so far ok - but then is transitioned immediately to a wide dispersion dome tweeter sitting right in the middle of its sweet spot.
Huge contrast!
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: