Need help with modification of DIY center speaker

[kjgarrison]

Hello all,

I'm new to this, just so you know.

I built a DIY 3-way "built-in" center speaker for my HT. WT/MW, I belive is the proper description. It was more of a "paint by numbers" project where I followed the instructions of somebody else. The TV and the center are built-in a so-called entertainment center, so it's close to being "in-wall". I am not happy with the sound quality of this center speaker. I think the midrange is the weak link, and am looking at two things to start with: first change to tri-amp with (DIY) active xo, then experiment with the current midrange just to see if I can avoid upgrading it.

Here is what I have installed already.
Woofers, 2 in parallel, ported: Seas Prestige here.
Midrange Tang Band here, crossover with woofers is 750Hz.
Tweeter SB Acoustics here, crossover with midrange is 2700Hz.

Both the tweeter and the Tang Band midrange have enclosures, and that is why I want to at least experiment with active crossover and triamping to try to keep the current Tang Band midrange.

The baffle can not be removed, and therefore if I change the midrange I need to find a good alternative that fits the routed opening for the existing one. I have found one that fits perfectly and has an excellent "full range" which will let me give the whole "intelligence band" to it: the Jordan JX92s.. The "real" T/S parameters for the Jordan are here.

If I change to the Jordan I have to build or come up with an enclosure for it. The enclosure can not be the nominal size of ~17 liters that would result in Qtc of 0.7 and F3 of 67Hz. Because of the space between the woofers and the midrange, the presence of the tweeter above, and the need not to take too much volume away from the woofers, I think I can only give around 4 liters to the Jordan, maybe a little more, maybe less. This results, according to Unibox, of a Qtc of around 1.0 and raises F3 to around 82Hz.

So, I don't really care about the low end for this midrange since I plan to cross it with the woofers at ~300Hz. I'm thinking that the gap between F3 (82Hz) and the crossover of 300Hz being a little shy of two octaves will not be a problem for the 24dB/octave active crossover. And I'm thinking the higher Qtc of ~1 will not be an issue either for this midrange, but there are things about the sound quality of different Qtc values that I do not understand.

So before I go and order all the components that let me do this crossover with both the existing Tang Band and the Jordan, I am asking you folks what mistakes I have made in my thinking, or what other considerations I am missing.

Secondary question: Does anybody know of a good way to build the enclosure for the Jordan? Maybe something like a can of some sort?

Thanks for any help.

 

[bjorno]

Hi kjgarrison,

Would you mind posting a picture showing your current filter design and the corresponding component values in use?

Have you thought about baffle impact on the center FR and do you have any FR measurements done at 1 m on axis?...See the picture for hints

b

 

[sreten]

Hi,

TBH in your shoes I'd be lloking at the crossover design and your original
choice of drivers, I would be looking at replacing the two bass drivers
with lower efficiency drivers (they will have more bass extension) and
using the Seas in another project for a pair of speakers.

The two seas in parallel are too efficient for your mid and will be for
the jordan as well. Active level controls would help but its a fairly
expensive way of fixing a fundamental problem.

FRD Consortium tools guide

And looking at Zaphs centre version of the ZDT3.5 should help.

I'd say look at a pair of these
Dayton DA175-8 7" Aluminum Cone Woofer | Parts-Express.com
The reference RS180's might be still a little too efficient,
but if not an obviously better choice.

If you simulate correctly and use Zaphs centre response alignment
you should get better results than what you currently have, and
using the Seas bass units data confirm what your problem was.

rgds, sreten.

Paul Carmody's DIY Audio Projects - undefinition
Zaph|Audio
RJB Audio Projects
Speaker Design Works
HTGuide Forum - A Guide to HTguide.com Completed Speaker Designs.
Humble Homemade Hifi
Click below to go to
Quarter Wavelength Loudspeaker Design
The Frugal-Horns Site -- High Performance, Low Cost DIY Horn Designs
Linkwitz Lab - Loudspeaker Design
Music and Design

 

[kjgarrison]

bjorno, thank you for the reply.
I will look at the hints you provided to see if I can figure out what you are asking.

[EDIT: bjorno I looked at your hint, and I have no idea what software that comes from, and of course I don't have images such as that to provide for you.]

Meantime ... I will also attempt to answer your questions about a picture of my current filter design and component values. First, however, I need to understand what you mean by "current filter design and the corresponding component values in use" Do you mean the current passive crossover? Or do you mean the yet to be built upcoming active crossover?

sreten, thank you as well. I will also look at the information and suggestions you have supplied. Perhaps you have already verified that the Dayton woofers are the same size as the Seas. If so thank you. [EDIT: it is only 1 mm smaller, so that should fit.] If they require a different diameter of routing for the opening, then it is not possible to use them. The baffle can not be removed, and the holes cannot be changed except perhaps for a few mm larger by hand.


If I can salvage the situation simply with different parts that can be exchanged and with modification of my passive crossover, then great. I have no doubt that different components would have worked better, but that ship has sailed.

And I have the bug to build an active 3-way mono crossover and triamp this center, and it is my understanding that all these incompatibilities of the drivers are fixed by this. I'm not really trying to do it on the cheap, but if it is truly as simple as using different parts (that fit) then it would be wasteful not to. I still have acquired the electronics building bug and am feeling like this might become my "hobby"

 

[sreten]

Hi,

Simply tri-amping and using standard active filter functions
will not fix all of your issues, not if you want to do it properly.
It wiil fix some of your issues, but leave some of the others.

rgds, sreten.

 

[kjgarrison]

Hi,

Simply tri-amping and using standard active filter functions
will not fix all of your issues, not if you want to do it properly.
It wiil fix some of your issues, but leave some of the others.

rgds, sreten.

Thanks for reading my post.

I certainly do want to do it properly, so could you be specific about what will and what will not be fixed with active xo and triamping?

Here are the limitations we have to live with :
Baffle cannot be removed or replaced
Enclosure volume can only be reduced
Ports are in (2x2")
XO is external but I'm thinking will be gone with active xo
EQ is not an issue since my AVR has Audyssey,and this is a center speaker which is only in use with 5.1

 

[PeteMcK]

Sounds like a sensitivity mis-match issue.
get rid of the Tang band and convert the box to a 2-way (but those Seas drivers are too good to waste on a centre channel...)
Alternatively, wire your woofers in series and recalculate their crossover component values.

 

[kjgarrison]

Sounds like a sensitivity mis-match issue.
get rid of the Tang band and convert the box to a 2-way (but those Seas drivers are too good to waste on a centre channel...)
Alternatively, wire your woofers in series and recalculate their crossover component values.

Yes, I am thinking the biggest problem is the sensitivity mismatch, which, as I understand it, is perfectly corrected with active crossover/triamping.

I'm beginning to think I have posted this question in the wrong place, or more likely asked it in a vague way. Seems like all the advice I get has to do with re-configuring drivers and/or the passive crossover.

My question is about the box. I'm not saying active crossovers are better than passive crossovers. It was a mistake to even put all that extraneous information in the first post. I apologize for that.

If I could only find out one thing it would be:

Is a higher Qtc than 0.707, such as 0.9 or 1.0, for the midrange enclosure going to be a problem given the frequencies the midrange will be used at?

The midrange will be crossing with the woofers at 300hz, so I don't care that the higher Qtc will take away some of the low end. I'm not going to be using the midrange anywhere close to its F3 of ~80hz. I also know that the step-response will be a little less ideal, but I don't know enough about step-responses to know if it matters less, the same, or more at higher frequencies.

If I get two questions, then the next one would be:
Does the step-response matter at higher frequencies as much as at lower frequencies?

Beyond that I find sreten's comment intriguing. It seems to hint at somehow combining my plan for active xo with other enhancements. That I'm definitely looking forward to hearing (whether or not I get the main answer... lol)

 

[kbgl]

Yes, I am thinking the biggest problem is the sensitivity mismatch, which, as I understand it, is perfectly corrected with active crossover/triamping.

I'm beginning to think I have posted this question in the wrong place, or more likely asked it in a vague way. Seems like all the advice I get has to do with re-configuring drivers and/or the passive crossover.

My question is about the box. I'm not saying active crossovers are better than passive crossovers. It was a mistake to even put all that extraneous information in the first post. I apologize for that.

If I could only find out one thing it would be:

Is a higher Qtc than 0.707, such as 0.9 or 1.0, for the midrange enclosure going to be a problem given the frequencies the midrange will be used at?

The midrange will be crossing with the woofers at 300hz, so I don't care that the higher Qtc will take away some of the low end. I'm not going to be using the midrange anywhere close to its F3 of ~80hz. I also know that the step-response will be a little less ideal, but I don't know enough about step-responses to know if it matters less, the same, or more at higher frequencies.

If I get two questions, then the next one would be:
Does the step-response matter at higher frequencies as much as at lower frequencies?

Beyond that I find sreten's comment intriguing. It seems to hint at somehow combining my plan for active xo with other enhancements. That I'm definitely looking forward to hearing (whether or not I get the main answer... lol)

Could you remove one woofer and fill the hole? That would get your sensitivities much closer to where they need to be. The x-over would need some alteration though.

If you have an extra stereo amp, you could connect the mid and tweeter on one channel, and the woofers on the other. Use the balance control to raise the mid and tweeter.

 

[sreten]

Beyond that I find sreten's comment intriguing. It seems to hint at somehow
combining my plan for active xo with other enhancements. That I'm definitely
looking forward to hearing (whether or not I get the main answer... lol)

Hi,

Its possible but I don't recommend. I recommend using the seas bass drivers
for a another stereo pair and using cheaper but good and less sensitive
drivers instead, keeping ther midrange and treble, replacing the mid and
treble and then going active ? Whats the point ? Acgtive would fix the
old fudamental issues to a degree. I severely recommend lower senstivity
bass units and designing the crossover properly, if you can trust TB curves .....

rgds, sreten.

Paul Carmody's DIY Audio Projects - undefinition
Zaph|Audio
FRD Consortium tools guide
RJB Audio Projects
Speaker Design Works
HTGuide Forum - A Guide to HTguide.com Completed Speaker Designs.
Humble Homemade Hifi
Click below to go to
Quarter Wavelength Loudspeaker Design
The Frugal-Horns Site -- High Performance, Low Cost DIY Horn Designs
Linkwitz Lab - Loudspeaker Design
Music and Design

 

[sreten]

Beyond that I find sreten's comment intriguing. It seems to hint at somehow
combining my plan for active xo with other enhancements. That I'm definitely
looking forward to hearing (whether or not I get the main answer... lol)

Hi,

Its possible but I don't recommend. I recommend using the seas bass drivers
for a another stereo pair and using cheaper but good and less sensitive
drivers instead, keeping the midrange and treble, replacing the mid and
treble and then going active ? Whats the point ? Active would fix the
old fundamental issues to a degree. I severely recommend lower senstivity
bass units and designing the crossover properly, if you can trust TB curves .....

rgds, sreten.

Paul Carmody's DIY Audio Projects - undefinition
Zaph|Audio
FRD Consortium tools guide
RJB Audio Projects
Speaker Design Works
HTGuide Forum - A Guide to HTguide.com Completed Speaker Designs.
Humble Homemade Hifi
Click below to go to
Quarter Wavelength Loudspeaker Design
The Frugal-Horns Site -- High Performance, Low Cost DIY Horn Designs
Linkwitz Lab - Loudspeaker Design
Music and Design