Xover design help - Page 2 - diyAudio
Go Back   Home > Forums > Loudspeakers > Multi-Way

Multi-Way Conventional loudspeakers with crossovers

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 16th December 2002, 05:07 PM   #11
Wizard of Kelts
diyAudio Moderator
 
Join Date: Sep 2001
Location: Connecticut, The Nutmeg State
Coming back in five minutes with a longer post. Hang on. Wanted to grab you while you were still here.
  Reply With Quote
Old 16th December 2002, 05:18 PM   #12
Wizard of Kelts
diyAudio Moderator
 
Join Date: Sep 2001
Location: Connecticut, The Nutmeg State
I would disagree with Francois, although I understand his philosophy. There is a school of thought that says that two drivers crossing over should have relatively flat response for an octave past the crossover point. Then the electrical components you insert should be responsible for the various slopes of the drivers around the crossover frequency.

The other school says that is alright to use the natural slopes of the drivers in your crossover. You will notice the original AR woofer you originally had in there had no inductor or capacitor attached to it. AR simply used the woofer's natural rolloff. That is what we are trying to do once we knock off that response surge around 2800 Hz. Knock that off, and you have a woofer rolling off at 2000 Hz.

The other thing Francois was probably referring to was the fact that it is a rule of thumb that you should cross over a tweeter at least an octave above it's resonance frequency. The chart appears to show the tweeter's resonance at 1300 Hz, but the resonance is listed at 1090 Hz. Other Audax tweeters have a 1090 Hz resonance as well. An octave above 1090 Hz is 2080 Hz, and 2000 Hz is certainly close that.

I have crossed tweeters over less than an octave above the tweeter resonance, and it came out alright.

I will not disagree completely with Francois, because if the question were: if you were an engineer and had to select speakers to make a run of 1,000 pairs, would you select these two units? The answer there would be: probably not. But the issue here was, given a certain amount of money, can we buy two units that will give us decent, hifi sound? And the answer there is, I believe, that these units will deliver that. These were about the only two units you could afford, and I believe that can be made to work together without stretching things too much.

I've stretched things a lot worse and had it work out, LOL.

More to post in a moment.
  Reply With Quote
Old 16th December 2002, 05:47 PM   #13
Wizard of Kelts
diyAudio Moderator
 
Join Date: Sep 2001
Location: Connecticut, The Nutmeg State
Quote:
Originally posted by Kram

A notch filter you say, what is one of those?
A passive crossover is a set of inductors, cpapacitors and resistors connected to a loudspeaker driver that suppresses the response of the loudspeaker above or below a certain point. The lower any frequency is in a circuit with a high pass filter, the more it is suppressed. Similarly, the higher any frequency is in a circuit with a low-pass filter, the more that frequency is suppressed.

A notch filter is a circuit connected to a loudspeaker, composed of similar components to a crossover, only it suppresses certain frequencies only in a certain band. Any frequencies above or below the notch filter are unaffected.

As you can see with our Audax AP210ZO woofer, it begins to roll off sharply at 2000 Hz, then surges up around 2800 Hz, then rolls off naturally again. A notch filter will knock off that surge, and we can have a woofer that acts like it rolls off naturally at 2000 Hz.

The picure below illustrates. The red line i the response of the input signal going through the notch filter.

I am trying to find how the notch filter at 2800 affects the phase of the AP210ZO at 2000 Hz. There are programs that will tell you, but I can't seem to get them set up.

We have such a program, written by a member here, that tells you, but I have to figure it out. Let me figure this out, and get back to you , hopefully in a day or two.
  Reply With Quote
Old 16th December 2002, 05:57 PM   #14
Wizard of Kelts
diyAudio Moderator
 
Join Date: Sep 2001
Location: Connecticut, The Nutmeg State
Below, superimposed over the AP210ZO response curve, is the action of the proposed notch filter, illustrated in red. They are simple to build, I just want to get some phase information.

The output of the speaker will be midway between the response curve in black, and the filter in red.

This should give us a woofer with a resonably workable, though not perfect, rolloff around 2000 Hz.
Attached Images
File Type: gif ap210zo notch illustration 1.gif (8.0 KB, 618 views)
  Reply With Quote
Old 16th December 2002, 06:13 PM   #15
Wizard of Kelts
diyAudio Moderator
 
Join Date: Sep 2001
Location: Connecticut, The Nutmeg State
As you recall, we had some difficulty measuring the tweeter's response. Now that we have a tweeter, the Audax TM025f1, with known response, you can measure the final response of the speaker, with the crossover, by comparing the response with the Audax tweeter without a crossover..

Your mic does not have to be linear. We know the Audax tweeter is linear. We simply send specific tones through the Audax tweeter alone, low level, (less than 1 volt) and take a reading for those tones. Then, wehen we have the crossover together, we can compare the readings to those of the Audax tweeter alone, and tell if the crossover is giving us acceptable response.

Incidentally, there is a test for your sound card that I think is worthwhile. I checked my sound card, with which I had some difficulty installing, and found that I had terrible frequency response. I re-installed,and found out I had relatively falt response down to 35 Hz.

The test requires an inexpensive cable, with two Walkman like plugs at each end. One end goes into the Line In of the sound card, the other to the Line Out. Then you can download and run the sound card test program, the link I am about to give.

By the way, Mpalin's has the cable. I believe it is the CK27E. It should have a stereo Walkman type plug-three sections-at each end. Check to make sure.

The link for the program is givien in the following thread, page 2:
cabinet design is giving me issues!

I think it would be a good idea to have a means to check the final response. I hear Speaker Workshop is very good for this, but I seem to be unable to to get the program going.
  Reply With Quote
Old 16th December 2002, 09:31 PM   #16
Bose(o) is offline Bose(o)  Canada
diyAudio Member
 
Join Date: Dec 2001
Location: Ottawa, Ontario, Canada
If I'm not mistaken, you (kelticizard), want to smooth out that bump in red. But, aren't notches used for bringing down the bump in frequency caused by driver resonances? Also, Crossover around 2000Hz, or 2800Hz, you'll be fine. I crossed my TW010E1 (Told you, it's a 'budget' system), at 4000Hz. The fs of this driver is 3000Hz, making the desired point 6000Hz! (Not sure if the rule is above..or just an octave away) or, 1500Hz. Well, at 6000Hz, sounds were becoming harsh and piano frequency response ends around 3600-4000Hz, so, 4000Hz was the decision, but enough about MY system. Honestly IMO, point would be somewhere before 2000Hz, or maybe 2000Hz. However, I'd also smooth out any bumps in the response from this.
  Reply With Quote
Old 16th December 2002, 10:14 PM   #17
diyAudio Member
 
FrankDIY's Avatar
 
Join Date: Mar 2001
Location: Quebec City
Yeah, I agree with kelticwizard but I am afraid that the normal rolloff curve of the driver introduce distortion and affect the sound. I have never tried this so who knows if you can get good results.
__________________
Imagination is more important than knowledge. knowledge is limited, imagination encircles the world. Albert Einstein
FrankDIY's Audio Corner
  Reply With Quote
Old 17th December 2002, 05:47 AM   #18
Wizard of Kelts
diyAudio Moderator
 
Join Date: Sep 2001
Location: Connecticut, The Nutmeg State
Kram:

Of course, to test your sound card you don't have to buy that cable. As long as you have a multimenter, you can test the output of an online tone generator playing through the sound card-all the online tone generators do.

You can get a good online tone generator from:
http://duncan.rutgers.edu/physicsfreewares.htm

You can either measure the output of the sound card directly, run it into an amp and test the voltage at the loudspeaker outputs. Either way, you can see if the voltage frequency is equal for all frequencies, and at which frequencies it starts rolling off.
  Reply With Quote
Old 17th December 2002, 05:59 AM   #19
Wizard of Kelts
diyAudio Moderator
 
Join Date: Sep 2001
Location: Connecticut, The Nutmeg State
Quote:
Originally posted by Bose(o)
[B]If I'm not mistaken, you (kelticizard), want to smooth out that bump in red. But, aren't notches used for bringing down the bump in frequency caused by driver resonances? .... I crossed my TW010E1 (Told you, it's a 'budget' system), at 4000Hz. The fs of this driver is 3000Hz, making the desired point 6000Hz! (Not sure if the rule is above..or just an octave away) or, 1500Hz. Well, at 6000Hz, sounds were becoming harsh and piano frequency response ends around 3600-4000Hz, so, 4000Hz was the decision, but enough about MY system.
A) Are notch filters used to smooth out the bump in driver resonances? Yes. And I would say there is a rather big bump at 2800 Hz. Hence, the notch.

B) "but enough about MY system." There is nothing wrong with talking about your system if it is relevant. In this case, it is. You said that you got good results crossing over near the Fs of your tweeter-not an octave above-and that was mentioned as an issue in this thread.

C) The TW010E1 might have been selected by you for it's cost. However, it's .4 inch, (10 mm) diaphragm would make it a good choice for anyone who wants good dispersion above 12,000 Hz, regardless of cost. What was the size of the driver that crossed into your tweeter? Generally, "beaming" is what causes harshness, and too large a driver for the frequency causes "beaming". I would think that for a 6,000 Hz crossover to be used, the midrange should be five inches or less-preferably less.
  Reply With Quote
Old 17th December 2002, 06:40 AM   #20
Wizard of Kelts
diyAudio Moderator
 
Join Date: Sep 2001
Location: Connecticut, The Nutmeg State
Quote:
Originally posted by François
Yeah, I agree with kelticwizard but I am afraid that the normal rolloff curve of the driver introduce distortion and affect the sound. I have never tried this so who knows if you can get good results.
I have heard of several respectable speakers that use the natural rolloff of the woofer, without an inductor, as the crossover. The AR 18 BX, which is the box that these components are being mounted in, used that system-no electrical crossover components at all in the woofer circuit. And AR is a good brand, especially considered so in the early seventies, which is when these were manufactured.

Of course, these brands that used the natural rolloff did not use notch filters to suppress peaks in the rolloff area. That is what we are going to do, and yes, that makes it unique.

I think the idea of using the natural rolloff was used more in the sixties and seventies than it is now. But with gradual slope crossovers coming back from the seventies, why not this, LOL?

The notch filter is resistive, therefore in phase, at the notch frequency. It is the phase of of the frequencies around the notch filter's center frequency that I am concerned with.

Audax has a home theater system, designed by Joe D'Appolito, which features this woofer's little brother, the AP210, with a notch filter to smooth out some peak in the response. However, I don't think the peak is in the crossover region. Ours will be, unless someone has a better crossover.

Frankly, with the notch filter beng resistive in the notch frequency, I wonder if doing this way might not be an improvement over the normal crossover. It is a possiblity, I would think.
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
dual atc mid xover design mclsound Multi-Way 12 29th July 2009 12:56 AM
Proprietary Xover design? auriches Multi-Way 4 12th January 2009 09:10 PM
My 2 way design... need xover help AudioGeek Multi-Way 4 2nd April 2007 01:28 PM
Series xover design ronjg Multi-Way 26 4th September 2002 07:25 AM


New To Site? Need Help?

All times are GMT. The time now is 08:10 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright ©1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2