Home Theater Forum and Systems banner

1 - 20 of 29 Posts

·
Premium Member
Joined
·
82 Posts
Discussion Starter #1 (Edited)
I have recently updated my family room A/V system with several objectives (which I met). In the process I have gone from driving my Infinity Kappa 7.1 from Bi-wired from an Adcom GFA-5800 to Bi-amped from a Marantz MM8003. While I am quite happy with the update in most all respects, it seems to lack the tight, powerful bass from the previous setup.

Unfortunately, the AV8003 does not provide an active cross-over when used in this mode, so the passive crossovers remain in the speakers. I am considering adding a active crossover between the AV8003 and MM8003 and modifying the passive crossovers in the Kappa 7.1 to remove the low pass filter for the woofers and the high pass filter to the mid range (while leaving the crossover for mid range and tweeters).

I have been searching the web for active crossovers and am considering between digital and analog solutions. Looking to the forum here for comments, thoughts and recommendations.
 

·
Registered
Joined
·
2,625 Posts
I have recently updated my family room A/V system several objectives (which I met). In the process I have gone from driving my Infinity Kappa 7.1 from Bi-Wired from an Adcom GFA-5800 to Bi-amped from a Marantz MM8003. While I am quite happy with the update in most all respects, it seems to lack the tight, powerful bass from the previous setup.

Unfortunately, the AV8003 does not provide an active cross-over when used in this mode, so the passive crossovers remain in the speakers. I am considering adding a active crossover between the AV8003 and MM8003 and modifying the passive crossovers in the Kappa 7.1 to removed the low pass filter for the woofers and the high pass filter to the mid range (while leaving the crossover for mid range and tweeters).

I have been searching the web for active crossovers and am considering between digital and analog solutions. Looking to the forum here for comments, thoughts and recommendations.
These speakers already have crossovers. If you want to improve performance your best bet will be to make internal modifications. Bi-wiring makes no difference in measurable performance assuming adequate speaker wire is used to feed the signal to your entire speaker.
 

·
Registered
Joined
·
1,911 Posts
I have been searching the web for active crossovers and am considering between digital and analog solutions. Looking to the forum here for comments, thoughts and recommendations.
There are no off-the-shelf active crossovers that will supplant and improve on what is already in the speakers. To do this, you must be able to design custom networks to accommodate the crossover as well as other correctable anomalies in the speakers. Most easily done with a digital device but the required skills are not insignificant.

Best leave as is unless you qualify (but then you wouldn't be asking). :rofl2:
 

·
Registered
Joined
·
2,625 Posts
There are no off-the-shelf active crossovers that will supplant and improve on what is already in the speakers. To do this, you must be able to design custom networks to accommodate the crossover as well as other correctable anomalies in the speakers. Most easily done with a digital device but the required skills are not insignificant.

Best leave as is unless you qualify (but then you wouldn't be asking). :rofl2:
Well these speakers are capable of significant improvements with a step up in the build quality. But even then I suggest a DCX2496 be used to correct the speaker further and not internal crossover mods. The drivers are among the best in the world. To go fully active would require 3 amps per speaker and a crossover per speaker.
 

·
Registered
Joined
·
1,911 Posts
Well these speakers are capable of significant improvements with a step up in the build quality. But even then I suggest a DCX2496 be used to correct the speaker further and not internal crossover mods. The drivers are among the best in the world. To go fully active would require 3 amps per speaker and a crossover per speaker.
Yeah, it is a pretty flexible tool but I still wonder how the less than skilled can construct a proper system with it. Do you think that a complete auto-calibration is sufficient?
 

·
Premium Member
Joined
·
91 Posts
Making use of an active crossover is possible and will yield higher performance from your existing drivers. Apparently, you already thought of this. The main considerations are the crossover points and slopes. Sensitivity of each driver unit can be addressed within the crossover itself. You can also make corrections to the driver responses very easily. This is not that hard to do. The main thing it will require is that you have an understanding of the bandwidth your drivers can handle. And obviously, as Isiberian mentioned you will need 6 amplifier channels. What are your crossover points now? Do you know the slopes - 1st order 3rd order, etc?

Try this link.

You can make measurements using room EQ wizard and correct your driver response as needed.
 

·
Registered
Joined
·
2,625 Posts
Making use of an active crossover is possible and will yield higher performance from your existing drivers. Apparently, you already thought of this. The main considerations are the crossover points and slopes. Sensitivity of each driver unit can be addressed within the crossover itself. You can also make corrections to the driver responses very easily. This is not that hard to do. The main thing it will require is that you have an understanding of the bandwidth your drivers can handle. And obviously, as Isiberian mentioned you will need 6 amplifier channels. What are your crossover points now? Do you know the slopes - 1st order 3rd order, etc?

Try this link.

You can make measurements using room EQ wizard and correct your driver response as needed.
Not necessarily. For an experienced speaker builder sure, but for a first timer it's a lot of work with little benefit if cabinet mods aren't made. Replacing the cheap lining with r11 would yield a much better improvement in overall system performance.
 

·
Premium Member
Joined
·
91 Posts
Not necessarily. For an experienced speaker builder sure, but for a first timer it's a lot of work with little benefit if cabinet mods aren't made. Replacing the cheap lining with r11 would yield a much better improvement in overall system performance.
Well, if the cabinet damping is poor then that could help, but in general most people assume that implementing an active or digital crossover is much more difficult than it actually is.

It does take some reading to understand the extra parts, extra channels, and individual drivers in question but you posted the manual above with the crossover points. So you already have that as a starting point - the rest can be done with measurements and final tweaking done by ear.

Plus, you already know that the drivers were designed to work with one another. I think it would be worth a shot. Save your passive crossover just in case you don't like the results.

I would be comfortable recommending some beginning settings if you chose to attempt the digital solution. Or you could just use an active crossover with identical crossover points and slopes as are present in the passive crossover. But at the very least you could use steeper slopes. This stuff really isn't rocket science. It just seems complicated at first glance...

You could then take some measurements with Room EQwizard and make final adjustments. It would take some work but I think work that would be well rewarded. Just take your time and be patient!

The only real drawback here is that - as Kal pointed out -you will need more amp channels etc.

And to make sure you don't wire the wrong drivers to the output frying the tweeters accidentally.

Just my opinion... don't mean to argue but I think you can do it!
 

·
Registered
Joined
·
887 Posts
The Behringer DCX2496 seems to be most peoples choice as the best out there because it has adjustable slopes plus parametric eq and delay on all channels. One unit can do stereo 3-way.

In your link to the Kappa 7.1 the crossover network is specified. One could extract x-over frequencies, etc from this document for initial active settings...
 

·
Premium Member
Joined
·
82 Posts
Discussion Starter #10
Thanks for all the comments guys!

I am glad to see the DCX2496 keep being brought up as that is what I had in my Amazon cart when I started this thread (have not pulled the trigger just yet). I downloaded the software and it was pretty easy to use. (and yes, I was using the cross over freq in the spec as baseline).

I do have some physical limitations to deal with as I have to live with the wiring that was put as I had the wiring put in place when the house was built so unfortunately I don't have an option of tri-amping.

Some suggestions above were that I would need 6 channels, but I assume it is still worthwhile with 4, just more complicated to have to still have to modify the passive crossovers for the mid/high (as outlined above, reversible of course).
 

·
Premium Member
Joined
·
82 Posts
Discussion Starter #12
Just wanted to report back and say that I finally got some time to spend on this so I ordered the unit and installed it several weeks back.

The unit is really easy to use, both from remote from the PC as well as through the front panel interface. So I am really happy with the purchase and the level of control it provides me.

Although I don't have any hard data, the bass/power I felt was lost in the update the Marantz separates is seems addressed.

For now I have only removed the low-pass filters for the woofers, but have not not removed the passive high pass for the midrange and am using the following settings:

Low Freq:
- 20Hz -24db Butterworth
- Playing both with and without this. I noticed this in the
sample filters, and could see the logic in removing the low freq
if the speakers are not able to reproduce them anyway.

- 500Hz -24db Butterworth
- This is -6db more per octive than that passive filters in the
speakers, but

High Freq:
- 375Hz -12db Butterworth
- I offset this from the Low Freq cut-off by 1/2 an octive to
try and match the -3db and mirror the slope when combined with
(what looked to me to be) -12db per octive for the passive
filter still in place, for a combined slope of -24db per
octive.

I still have more work to do and I want to play with the filtering quite a bit more which sent me off in on a quest to find some tools, which led me to REW and excuses to go buy more stuff like a netbook to run it on (as well as remotely control the DCX2496 in my family room).

Next step for me now that I have some tools to capture some data and tune the crossover points and equalize the systems. Between the Audisy in the Marantz and capabilities in the DCX2496, there is a lot to play with. :dumbcrazy:
 

·
Premium Member
Joined
·
91 Posts
First, I must say I am really glad you decided to give this a shot. As you found out - it is not that difficult to achieve good or even great results. Bettering the stock passive crossover is very easy to do in almost all situations.

I'm a little confused by two lines in your post. "Although I don't have any hard data, the bass/power I felt was lost in the update the Marantz separates is seems addressed."

And "High Freq:
- 375Hz -12db Butterworth"

do you have a crossover point at both 375 and 500Hz?

Peter



Read more: Active crossover recommendations for bi-amping Infinity Kappa 7.1 with a Marantz AV8003/MM8003 - Page 2 - Home Theater Forum and Systems - HomeTheaterShack.com
 

·
Premium Member
Joined
·
82 Posts
Discussion Starter #14
First, I must say I am really glad you decided to give this a shot. As you found out - it is not that difficult to achieve good or even great results. Bettering the stock passive crossover is very easy to do in almost all situations.
No doubt to my ears that this was an improvement.

I'm a little confused by two lines in your post. "Although I don't have any hard data, the bass/power I felt was lost in the update the Marantz separates is seems addressed."
If you are referring my "hard data" comment, I meant that it was subjective and I did not have any data to back it up. I did use a consistent play list, settings, etc, but still very subjective none the less.

And "High Freq:
- 375Hz -12db Butterworth"

do you have a crossover point at both 375 and 500Hz?
Yes (at the moment, may change after my experiments this weekend).

I have the the HP/LP skewed by 1/2 octave and set to -12db per octave in attempt to achieve an overall -24db between both the digital and the analog filter (not sure it right mind you :dontknow:).

This is what led me to go find and way to measure my results somehow.

Below is the filter in the DCX2496 (at the moment):

5-20-2010 9-30-55 PM.jpg

And the changes to the passive crossover (at the moment):

5-22-2010 10-47-52 AM.jpg

I am going to make some measurement playing with the filters in this configuration.

The next step I am considering is to remove the low pass on the midrange, but I am not quite sure of my changes since I am struggling a bit to reverse engineering the filter design.

Any comments here would be greatly appreciated.

5-22-2010 11-00-50 AM.jpg

5-22-2010 10-58-29 AM.jpg
 

·
Premium Member
Joined
·
91 Posts
Looks really good so far!

I went back and looked at your speaker crossover slopes etc from the manufacturers data again. I can see your method behind using two slope points but I think I would put the xover point at 500Hz and us 24 db slopes for both drivers.

You could do some final tweaking of this crossover point and you should be able to flex it a bit both ways because you are using 24 db slopes instead of the 12 db that were on the passive filter. Try anything from 400 to 600 Hz and make notes from what you hear. I would guess that you will end up pretty close to 500 though.

If you have not done so already get a copy of REW available on this site and get it working. Then make some measurements. This will tell you a lot about your driver integration.

You may see straight away that your levels are incorrect. Maybe bass too strong, etc. Peaks that need to be tamed etc.

I have found that you can really make the speaker 'sound the right way for you' by balancing the levels of the different crossover legs and taming peaks in the response. Especially, between the midrange driver and tweeter. This is the relationship that gets the presence right on the loudspeaker in the room. This is one of the major advantages to active crossovers. All rooms are different and getting the balance right cannot be fixed on passive crossovers. Not easily, anyway...

Right now you are only able to adjust the amount of bass that is correct for your room and tastes. Definitively better, but you're not getting the full advantage yet.

You didn't mention it but were you able to align the drivers with delays too? I think you can time align with that software as well.


Peter

An interesting page on speaker box design you might be interested in.
 

·
Premium Member
Joined
·
91 Posts
If you get your extra two amp channels for the tweeters put your crossover at 4500 Hz @ 24 db and experiment from there. With that small 3 in midrange driver you could probably crossover anywhere from 3500 to 6000 Hz without experiencing any negative lobeing effects. In fact, if that 3 inch driver will play clean out to 5000Hz or 5500Hz, I would see what that sounds like just for fun.

A x-over point that high would really release the tweeter and increase its ability to play at higher SPL if needed. But there may be a reason the manufacturer put the point at 4500Hz, could be a breakup for the 3 in driver up in the response somewhere...

You can also measure this and find it by taking some REW measurements after you go to all active and removing the passive networks. The peaks should be pretty obvious in the measured response.
 

·
Premium Member
Joined
·
82 Posts
Discussion Starter #17
Looks really good so far!
Thanks!

I went back and looked at your speaker crossover slopes etc from the manufacturers data again. I can see your method behind using two slope points but I think I would put the xover point at 500Hz and us 24 db slopes for both drivers.

...

If you have not done so already get a copy of REW available on this site and get it working. Then make some measurements. This will tell you a lot about your driver integration.

...

You may see straight away that your levels are incorrect. Maybe bass too strong, etc. Peaks that need to be tamed etc.
I had found REW a few weeks back and have been working to get all the pieces together (netbook, usb audio, cables, etc.) to do exactly this. Yesterday was my first attempt to be able sit down and spend some time with this.

I am currently using my RS SPL meeting which shoudl be good enough for the ~500Hz crossover point. I understand I will need to upgrade to mic/pre-amp setup to be able make make a meaningful full range scan, but I am capturing the full range anyway right now, ignoring the high end results (>5Khz).

Below are my results of measuring the difference between the two settings (red = 375 @ -12db and green = 499 @ -24db (mirrors low-pass), but still with passive crossover in place). The results show a pretty subtle differences between the two.

But I am also seeing some anomalies that are very suspicious so I am taking all the results with a grain of salt until I can resolve these (one being that the results tend to mirror the SPL calibration/C weighted curve, and the other is the levels I need to be at to capture seem to be high (+90db?), I am going to replace the battery in the SPL meter and try again).

With that said, here are the results of that change (w/1/1 octave smoothing):
374at12vs499at24.jpg

You could do some final tweaking of this crossover point and you should be able to flex it a bit both ways because you are using 24 db slopes instead of the 12 db that were on the passive filter. Try anything from 400 to 600 Hz and make notes from what you hear. I would guess that you will end up pretty close to 500 though.
Good suggestion. I will need to wait until I make the changes to remove the 500Hz crossover point from the passive crossover to play with this.

You didn't mention it but were you able to align the drivers with delays too? I think you can time align with that software as well.
No. I don't have a mic that will work with the DCX2496 to try the auto align capabilities.

I am not sure how (or if I can) go about it with REW. Is there a graph from REW that I can use to evaluate the alignment?

I used the phase graph (again, w/1/1 octave smoothing) to check for phase inversion at the crossover point (which looks fine):

374at12vs499at24-phase.jpg

An interesting page on speaker box design you might be interested in.
Thanks!
 

·
Premium Member
Joined
·
82 Posts
Discussion Starter #18
If you get your extra two amp channels for the tweeters put your crossover at 4500 Hz @ 24 db and experiment from there. With that small 3 in midrange driver you could probably crossover anywhere from 3500 to 6000 Hz without experiencing any negative lobeing effects. In fact, if that 3 inch driver will play clean out to 5000Hz or 5500Hz, I would see what that sounds like just for fun.

A x-over point that high would really release the tweeter and increase its ability to play at higher SPL if needed. But there may be a reason the manufacturer put the point at 4500Hz, could be a breakup for the 3 in driver up in the response somewhere...

You can also measure this and find it by taking some REW measurements after you go to all active and removing the passive networks. The peaks should be pretty obvious in the measured response.
Unfortunately based on my setup, I am limited to 5 channels (only using 4) to the front speakers with my current house wiring.
 

·
Premium Member
Joined
·
82 Posts
Discussion Starter #20 (Edited)
Current house wiring????
The wiring from my amps to me speakers is built into the wall of the house.

Unfortunately, the builder had a policy of not letting customers do work on there own homes, so I had to let the builder put in the wiring, and while I was able to specify the quality of the wiring used, I failed to make sure they used PVC piping so I could run new/updating wiring, so I am stuck with what I have unless I want to tear our the cabinet and the sheet rock and re-do it, which is difficult justify the cost to rectify.

FWIW, the pictures below show the cabinet and the room I am talking about. The wires run from behind the built in cabinet, up the wall, over the ceiling and come down in the corner behind the media shelves in the right corner as that is the only place the structure would allow the wires to be exposed on that wall (~50' in all, despite being just a few feet away :foottap:)
 
1 - 20 of 29 Posts
Top