Baffle Step Correction Question

[mgrabow]

I understand the basic concept of baffle step correction but arent you basically padding down the high pass to copemsate for the loss in bass when you pull the speakers away from the wall? Thus on a single full range driver system doing so would be doing nothing but padding down the entire spectrum?


I ask because I saw an artical and a build of s single driver system and saw some comments on a BSC circuit and was wondering why...

[maynardg]

Re: Baffle Step Correction Question

Simple solution for full range baffle step.


http://www.quarter-wave.com/General/BSC_Sizing.pdf

http://diyaudioprojects.com/Technica...it-Calculator/

[mgrabow]

Re: Baffle Step Correction Question

Simple solution for full range baffle step.


http://www.quarter-wave.com/General/BSC_Sizing.pdf

http://diyaudioprojects.com/Technica...it-Calculator/

Both fine sites but I am not sure they adress the reasoning for doing it on a single driver system...

[Pete Schumacher]

Re: Baffle Step Correction Question

Both fine sites but I am not sure they adress the reasoning for doing it on a single driver system...

A contouring network, not a "crossover" is employed with a full range driver, unless it's natural response is weighted toward the low end.

With a two-way, baffle step compensation is accomplished in two pieces. Most of it takes place in the woofer XO, where baffle step takes place, usually between 300 and 1000Hz. The tweeter is then padded down across its range to meet up with the woofer at the crossover point. In many designs, full baffle step compensation is employed. If the woofer has a flat response on an infinite baffle, then the transfer function at the high end of the range (2000Hz for example) would be 6dB lower than at 200Hz, not including the crossover slope.

[daryl]

Re: Baffle Step Correction Question

One driver or a hundered drivers you want a flat response right?

You'll likely need to make adjustments for more than just diffraction loss.

So you probably will design a filter to align the response throughout the entire spectrum.

You cannot truly use formulae to design diffraction loss circuits or conjugate circuits as shown in the link above.

If you are not taking acoustic measurements of your project then that's all you can do, hence the 'tweaking' described in the article.

It does seem a shame to lose efficiency but that's how it's done.

One day I hope we all will be using computers as our signal sources and adjusting response will be as easy as calculating a new filter kernel and installing it in your audio software.

That way it's all handled in the digital domain before the signal is even converted to analog without detriment to the signal and without cost or logistics issues of building filters.

[Deward Hastings]

Re: Baffle Step Correction Question

on a single full range driver system doing so would be doing nothing but padding down the entire spectrum?

Baffle step is an equalization issue, not a crossover issue (although it is often implemented as part of a crossover). It compensates for the 2pi/4pi transition inherent in any “box” loudspeaker. The inflection point of the compensation is determined by the transition frequency, which is itself determined by the baffle dimensions, and is independent of any crossover frequencies. The amount of compensation needed is determined by the room, and the speaker’s placement in the room . . . the only way to know “how much” is to measure, and listen, in your particular application (although “guesstimates” tend to work reasonably well in general, as the range of adjustment is only 6dB, and most small rooms have similar properties at low frequencies). Any competently designed small speaker, unless purpose built for a single mounting position, will include compensation adjustment to accommodate various rooms and placements. It is simple to implement (Behringer does it in the 2030A with a 4 position slide switch (0,2,4,6 dB compensation)) if done at line level, more difficult when done in a passive crossover (where a significant re-design may be necessary to make the “adjustment”).

[mgrabow]

Re: Baffle Step Correction Question

I am still missing something regarding its application on a single driver full range system. Are you all saying it can be used to cut back a portion of the spectum on a single driver system? if so then I get it...

[Jeff B.]

Re: Baffle Step Correction Question

I am still missing something regarding its application on a single driver full range system. Are you all saying it can be used to cut back a portion of the spectum on a single driver system? if so then I get it...

Yes. That is what they are saying.

[Mark65]

Re: Baffle Step Correction Question

If the response of your driver is flat on an infinite baffle, SPL in the range below the baffle step will be somewhere between 2 and 6 dB below the rest of the response curve. Baffle step compensation shapes the response ABOVE the step frequency DOWN to match the 4 pi response, thus bringing the the whole curve back to "flat". It's necessary for any driver that is producing wavelengths greater than the width of the baffle, if the speaker in question is going to produce "full range" response, whether that is a single driver, or a two-, three- (or more) way speaker.


Mark

[Æ]

Re: Baffle Step Correction Question

No, you are padding everything down above the bass, to make up for the apparent loss of bass.
With a full range driver you would be padding down the entire spectrum above the bass.

I understand the basic concept of baffle step correction but arent you basically padding down the high pass to copemsate for the loss in bass when you pull the speakers away from the wall? Thus on a single full range driver system doing so would be doing nothing but padding down the entire spectrum?


I ask because I saw an artical and a build of s single driver system and saw some comments on a BSC circuit and was wondering why...

[mgrabow]

Re: Baffle Step Correction Question

It all makes sense now. I was thinking of a traditional 3 way with the circuit in the HP filter. I wasn't thinking about the BSC being tuned to a specific band width...