avoiding lobing/comb filtering in center channel

[bjames8720]

I am currently starting a project. A dreaded center channel.....I have been doing tons of research on lobbing/ comb filtering. The best way that I have found to fight this issue is running a 3 way set-up and having x-over points low as possible. I have done a lot of car audio work including competing in MECA (152.4 in M3), recently started SQ in my car and some home audio for myself. So I don’t feel I have advanced skills by any means but not a complete idiot either. I am hoping with this post to kind of bridge the gap between the two or at least help some people if I can. I would greatly appreciate any comments or discussion here. I am going to post some pictures get things started. These aren't the exact drivers I will be using. Instead of the Dayton Alum 5.25 I will be using a Dayton 2" Dome Midrange. I will post links to the drivers I am using. The woofers I just had lying around so I decided to use them but I am sure to save space you wouldn’t have to use a 7" driver. I understand that this will be a somewhat large center channel but the way I see it is if you are that concerned with lobbing and comb filtering it’s a sacrifice that more than likely you'd be will to make. Hopefully we have some positive and educational discussions!

Thanks,
Brock

[mdocod]

Check out the ZDC3 here: Zaph|Audio - ZDT3.5

Looks very similar to what you are doing, might help you with some ideas.

-----

General rule of thumb, if you can get your crossover point less than an octave below your driver center to center spacing wavelength size, then lobing/combing should be inaudible. The 3-way center is a great way to get the x-over from the woofers low enough to accomplish this, another option is a 2 way with a 2-3" full range used for the center driver, facilitating a very low x-over (only useful if you listen to your center very on-axis), or a 2.5 way with 4 smaller woofers and a very robust dome that will allow for a really low x-over point.

The dome mid and dome tweeter in a vertical arrangement will generally favor a listening position that is on or slightly above axis, ideal for center channel configurations that are commonly placed below the TV. The Zaph designed center linked above takes advantage of this characteristic.

Eric

[bjames8720]

I guess I should have posted my x-over information and points as well. 3-way, 2nd-order (12 dB/octave) at 500/4k. I wanted to go lower but the midrange more then likely wouldnt like it very much I dont think. I am already a little concerned with the power handling here because I have not ever used anything like this.

I did look at the fact that it isnt very often that I would be more then 30 degrees off axis when listening to it so to be honest I'm not sure where to draw the line as to when enough is enough but I am enjoying this project so I'm probably going overboard considering.

[bjames8720]

Dayton 7" Woofers - Dayton Audio RS180-4 7" Reference Woofer 4 Ohm 295-374

Dayton 2" Dome Midrange - Dayton Audio RS52AN-8 2" Dome Midrange 285-020

Dayton 1 1/8 Silk Dome Tweeter - Dayton Audio DC28FT-8 1-1/8" Silk Dome Truncated Tweeter 275-076

Dayton X-Over 500/4k (12db per octave) - Dayton Audio XO3W-500/4K 3-Way Crossover 500/4,000 Hz 260-152

I am currently in the process of tweaking the enclosure design and looking at different options to decrease reflections in the enclosure.

[mdocod]

You're from the world of car audio where those prefabricated crossovers like the one you have in mind there are commonplace. The dirty little secret is that in the car and home and pro audio segments, most if not all pre-made passive crossovers produce results that are very far from flat with the vast majority of drivers they could be paired up with. The odds of getting a desirable result (from a "hifi" standpoint) are pretty bad. Also, I would argue that for $50, those pre-made jobs are sort of a rip, you can build your own for about the same or less with often better quality and more importantly, properly matched components. The sad thing is the number of "component" systems out there for car audio, that are radically far from flat, and yet, boast about their "dedicated high flutin "separates" crossovers" and all that crap. The car audio industry creates probably more misconceptions about audio reproduction than every other source of bad information combined.

You've already hashed out 2 out of 3 drivers used in that Zaph design I linked to... Considering how close your build is to that, why not just copy it to the "t" and not loose any sleep over a design decision?

[bjames8720]

Originally Posted by mdocod

You're from the world of car audio where those prefabricated crossovers like the one you have in mind there are commonplace. The dirty little secret is that in the car and home and pro audio segments, most if not all pre-made passive crossovers produce results that are very far from flat with the vast majority of drivers they could be paired up with. The odds of getting a desirable result (from a "hifi" standpoint) are pretty bad. Also, I would argue that for $50, those pre-made jobs are sort of a rip, you can build your own for about the same or less with often better quality and more importantly, properly matched components. The sad thing is the number of "component" systems out there for car audio, that are radically far from flat, and yet, boast about their "dedicated high flutin "separates" crossovers" and all that crap. The car audio industry creates probably more misconceptions about audio reproduction than every other source of bad information combined.

You've already hashed out 2 out of 3 drivers used in that Zaph design I linked to... Considering how close your build is to that, why not just copy it to the "t" and not loose any sleep over a design decision?

I do agree. I've never used passive x-overs everything I have done in the past is active so this is kind of new to me and I really don't even know where to begin except that I know how to figure out what componets i need to use.

[mdocod]

Passive crossover design can be taken to different degrees of perfection. There is a great sticky in the multi-way section here by AllenB that covers some nice basics for designing a crossover without measurements. Ideally speaking, it's best to at least run a simulation of the intended drivers in order to make crossover component value decisions.

Eric