This cute little tweeter will work fine with a 2226:
http://www.diyaudio.com/forums/showthread.php?postid=1303475#post1303475
Regards
Charles
http://www.diyaudio.com/forums/showthread.php?postid=1303475#post1303475
Regards
Charles
BlueWizard said:
Hi,
Still different things do different people, or sometimes the same person.
To be accurate this should mean acoustic, see here for 2nd order L/R acoustic :
http://www.zaphaudio.com/ZD5.html
But often this is taken to mean 2nd order electrical, which in most
cases ends up meaning 4th order L/R acoustic to Zaph, see here :
http://www.zaphaudio.com/audio-speaker19.html
To go back to your original post about 1st order filters, practical 1st
order acoustic filters are near impossible with most standard driver
combinations, but see here for 1st order electrical = 2nd order L/R :
http://www.zaphaudio.com/Waveguidetmm.html
The point being your acoustic targets are set by the characteristics
of the drivers - what works well rather than theoretically ideal.
You are correct in what you stated earlier. you should not buy the
drivers and then get a crossover to work with them, the ease in
which the crossover can be developed or a driver should affect its
initial choice.
As suggested before Zaph's site is a very good place to peruse
driver choice criteria, where being able to manipulate the driver
into the wanted acoustic response is more important than its
flat response range.
/sreten.
""I'm not a fan of the buttcheeks.""
In my best Beavis voice...."huh huh you said buttcheeks..."
I was being a bit cheeky suggesting the 2226 driver as I knew it had a 'decent' response for Bluewizard but would give him nightmares if he tried to build something with it....
Personally for me it's a proper driver to be looking at.
I'd probably mix a pair of 2226's with something like an altec 511b as a basic starter system.
Nice seeing some alternatives to the 6" + tweet crowd anyhoo.
Cheers,
Rob.
In my best Beavis voice...."huh huh you said buttcheeks..."
I was being a bit cheeky suggesting the 2226 driver as I knew it had a 'decent' response for Bluewizard but would give him nightmares if he tried to build something with it....
Personally for me it's a proper driver to be looking at.
I'd probably mix a pair of 2226's with something like an altec 511b as a basic starter system.
Nice seeing some alternatives to the 6" + tweet crowd anyhoo.
Cheers,
Rob.
I cant believe no one has mentioned linkwitz's 3way SB article!ww
Unfortunately his site seems to have used up its bandwidth!
http://www.linkwitzlab.com/x-sb80-3wy.htm
Unfortunately his site seems to have used up its bandwidth!
http://www.linkwitzlab.com/x-sb80-3wy.htm
BlueWizard said:
I may have the wrong end of the stick here, but are you looking at how to select drivers for a speaker without first deciding what speaker design you would like? It sounds like it. You may have success using this method, but it's probably doubtful. A better way might be to decide what sort of speaker you wish for (or at least what you want it to do). Then you can pick the drivers you wish to use for such a job.
You can't put the cart before the horse (in most cases).
P.S. If I'm wrong then ignore this post...
bigwill said:
Thats another way to go,but has its own disadvantages like irregular polar response.
http://www.harman.com/about_harman/technology_leadership.aspx
Keep in mind that my original question is in the preliminary stage of evaluation of potential speakers. There is no need to look at polar response if you don't have a sufficient frequency overlap. In fact, there is no need to look at any other factor, if the overlap isn't there.
People keep pointing out that you can't use frequency response alone, a point I've readily acknowledged several times, because within that frequency response there are likely to be unpleasant aberrations. But again if the response overlap isn't there to begin with, why investigate further to see if there are any aberrations?
And, for the record, I'm not JUST interest in response overlap. I'm saying it is a starting point in evaluating potential speakers to determine if further, more detailed, investigation is warranted.
As to the statement -
"I'm not looking at speaker FOR a design, I'm looking at speaker to CONSIDER for a design. "
I think it should be obvious to anyone who has read the entire thread, that I am not looking at a speaker for a design, because I can't do that detail look until I have determine whether a give speaker is even a possibility, and that is the stage I am at. I'm trying to separate the 'possible' from the 'impossible'. Is a given speaker a possibility that warrants deeper investigation? And what criteria do I use to make that preliminary determination? One early-on parameter is frequency response, because, as I have said, if it isn't there, there is no point in investigating further.
This came up in another discussion, where someone had an old set of speakers that they wanted to rebuilt into something functional. Though I suspect they trusted the uninformed advice of another person.
The mated a Dayton DC28F-8 1-1/8" Silk Dome Tweeter with a Dayton DC250-8 10" Classic Woofer and a 2khz 2-way 12db crossover.
Dayton DC28F-8 1-1/8" Silk Dome Tweeter
Frequency response (± 2 dB): 2,000-20,000 Hz
Dayton DC250-8 10" Classic Woofer
Frequency response: 25-2,500 Hz
If this person has some basic Rule of Thumb for frequency overlap, he would have never considered these speakers. One glance and he would have moved on.
If he was really tied to that design concept, this speaker might have been a better choice -
Pioneer W25GR31-51F 10" Butyl Surround Woofer
Frequency response: 26-6,000 Hz
or perhaps this-
Goldwood GW-10PC-40-8 10" Heavy Duty Woofer
Frequency response: 36-5,000 Hz
Now, the Pioneer could be completely flakey in the upper end of that 6000hz, but crossing over at 2khz at12db/octave, should keep the flakeyness out of the working range.
Once you have this preliminary determination that two speakers MIGHT work together, then you start to consider the other factors to see if they really will work together. Frequency response is not the End Game, it is the preliminary game. It is where you start, not where you finish.
Steve/bluewizard
People keep pointing out that you can't use frequency response alone, a point I've readily acknowledged several times, because within that frequency response there are likely to be unpleasant aberrations. But again if the response overlap isn't there to begin with, why investigate further to see if there are any aberrations?
And, for the record, I'm not JUST interest in response overlap. I'm saying it is a starting point in evaluating potential speakers to determine if further, more detailed, investigation is warranted.
As to the statement -
"I'm not looking at speaker FOR a design, I'm looking at speaker to CONSIDER for a design. "
I think it should be obvious to anyone who has read the entire thread, that I am not looking at a speaker for a design, because I can't do that detail look until I have determine whether a give speaker is even a possibility, and that is the stage I am at. I'm trying to separate the 'possible' from the 'impossible'. Is a given speaker a possibility that warrants deeper investigation? And what criteria do I use to make that preliminary determination? One early-on parameter is frequency response, because, as I have said, if it isn't there, there is no point in investigating further.
This came up in another discussion, where someone had an old set of speakers that they wanted to rebuilt into something functional. Though I suspect they trusted the uninformed advice of another person.
The mated a Dayton DC28F-8 1-1/8" Silk Dome Tweeter with a Dayton DC250-8 10" Classic Woofer and a 2khz 2-way 12db crossover.
Dayton DC28F-8 1-1/8" Silk Dome Tweeter
Frequency response (± 2 dB): 2,000-20,000 Hz
Dayton DC250-8 10" Classic Woofer
Frequency response: 25-2,500 Hz
If this person has some basic Rule of Thumb for frequency overlap, he would have never considered these speakers. One glance and he would have moved on.
If he was really tied to that design concept, this speaker might have been a better choice -
Pioneer W25GR31-51F 10" Butyl Surround Woofer
Frequency response: 26-6,000 Hz
or perhaps this-
Goldwood GW-10PC-40-8 10" Heavy Duty Woofer
Frequency response: 36-5,000 Hz
Now, the Pioneer could be completely flakey in the upper end of that 6000hz, but crossing over at 2khz at12db/octave, should keep the flakeyness out of the working range.
Once you have this preliminary determination that two speakers MIGHT work together, then you start to consider the other factors to see if they really will work together. Frequency response is not the End Game, it is the preliminary game. It is where you start, not where you finish.
Steve/bluewizard
Hi Mike.e,
Just looking at the reply above you can see what I meant.
"""Keep in mind that my original question is in the preliminary stage of evaluation of potential speakers. There is no need to look at polar response if you don't have a sufficient frequency overlap. In fact, there is no need to look at any other factor, if the overlap isn't there.""
Rob.
Just looking at the reply above you can see what I meant.
"""Keep in mind that my original question is in the preliminary stage of evaluation of potential speakers. There is no need to look at polar response if you don't have a sufficient frequency overlap. In fact, there is no need to look at any other factor, if the overlap isn't there.""
Rob.
Steve: No, a person who had studied the issue a bit wouldn't think of trying to match the polar pattern of a 10" woofer (ANY ten inch woofer) to a 1" tweeter (ANY 1" tweeter).
You still haven't gotten over the idea that "frequency response" is a vague term, and becomes even more useless when talking about 3-D radiation patterns. A speaker is NOT an amplifier. If I had a magic 10" woofer with on-axis response of 20Hz-20kHz, it still couldn't be satisfactorily mated to a 1" (or 1-1/8") dome tweeter despite a 100% overlap (to use your term).
You still haven't gotten over the idea that "frequency response" is a vague term, and becomes even more useless when talking about 3-D radiation patterns. A speaker is NOT an amplifier. If I had a magic 10" woofer with on-axis response of 20Hz-20kHz, it still couldn't be satisfactorily mated to a 1" (or 1-1/8") dome tweeter despite a 100% overlap (to use your term).
You ask a simplistic question, and here is my simplistic answer - you probably need at least 1 octave of overlap, at the level of an initial pass to see if a closer look is justified.
The problem is that the headline "frequency response" quoted by the manufacturer is essentially meaningless, and in many cases there are overwhelming reasons why you can't use the frequency extremes.
For woofers, the two problems are beaming and breakup. For frequencies where the woofer is more than half a wavelength across, the sound starts to become more and more directional. Depending on your target usage, most woofers should not be used at frequencies above the point where they are a wavelength across, even if the on-axis response carries on for another couple of octaves. As the wavelength at 1kHz is 12", this basically means fmax(in kHz) < 12/D, where D is diameter in inches. Many designs run woofers much higher than this, at the expense of poor off-axis response. If you want really good dispersion, fmax<6/D.
Some woofers have horrible break-up modes with massive peaks (common in metal cone designs, but others have the problem too). The woofer shows massive excess distortion at sub-multiples of the peak frequencies, so you have to restrict usage to some small fraction of the peak (a 1/4 or 1/5) even though the flat response carries on up for a couple of octaves.
The big problem for tweeters is low end power handling and distortion. Most tweeters have a flat(ish) response down to the resonant frequency, but they vary wildly in how much power they can accept with reasonable distortion in the couple of octaves above the resonance.
The problem is that the headline "frequency response" quoted by the manufacturer is essentially meaningless, and in many cases there are overwhelming reasons why you can't use the frequency extremes.
For woofers, the two problems are beaming and breakup. For frequencies where the woofer is more than half a wavelength across, the sound starts to become more and more directional. Depending on your target usage, most woofers should not be used at frequencies above the point where they are a wavelength across, even if the on-axis response carries on for another couple of octaves. As the wavelength at 1kHz is 12", this basically means fmax(in kHz) < 12/D, where D is diameter in inches. Many designs run woofers much higher than this, at the expense of poor off-axis response. If you want really good dispersion, fmax<6/D.
Some woofers have horrible break-up modes with massive peaks (common in metal cone designs, but others have the problem too). The woofer shows massive excess distortion at sub-multiples of the peak frequencies, so you have to restrict usage to some small fraction of the peak (a 1/4 or 1/5) even though the flat response carries on up for a couple of octaves.
The big problem for tweeters is low end power handling and distortion. Most tweeters have a flat(ish) response down to the resonant frequency, but they vary wildly in how much power they can accept with reasonable distortion in the couple of octaves above the resonance.
SY said:
http://wduk.worldomain.net/acatalog/wd25ahfw1a.html
An externally hosted image should be here but it was not working when we last tested it.
dave
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.