How about these? 1K Hz, 4th order crossover:
Measures at 1 M. The small insert to the upper right show the reproduction with a standard 1k Hz LR4 crossocer. The white trace is the electrical input the red the acoust output on the design axis. 300, 500, 1k and 2 kHz.
The original figure can be viewed at http://www.musicanddesign.com/pubimages/sqwavediy.gif
An externally hosted image should be here but it was not working when we last tested it.
Measures at 1 M. The small insert to the upper right show the reproduction with a standard 1k Hz LR4 crossocer. The white trace is the electrical input the red the acoust output on the design axis. 300, 500, 1k and 2 kHz.
The original figure can be viewed at http://www.musicanddesign.com/pubimages/sqwavediy.gif
John, that definitely looks different from your icta design - is this a teaser or is it documented somewhere? Are those results measured or simulated?
what happened to this thread? it suddenly stopped...
I followed it with great interest.
Yesterday I wanted to actually do something, so I constructed simple 2 transistor circuit, found the schematics on the web, it did not want to oscillate at first, but once I raised the voltage from 1.5 to 5 or 9, it produced nice square wave 100 Hz, 2.6 Vpp.
I have only simple hand held velleman oscilloscope, but it is good enough for experimentation.
So I fed it to my second system and looked at trace on the mike. It looks just like John k... small white insert on the bottom.
So...where do we go now? What do I do next? I want to experiment, improve...
My system is sounding freaking good, but measures poorly. The square wave out is nothing like square wave in. Am I disappointed? Not really. System sounds good. But I would like to improve it.
System is biamplified, active crossover, solid state for 15" woofer in closed box, tube amp for mid/hights on open baffle. The way I like it.
Can you comment? Lets keep this thread rolling.
I followed it with great interest.
Yesterday I wanted to actually do something, so I constructed simple 2 transistor circuit, found the schematics on the web, it did not want to oscillate at first, but once I raised the voltage from 1.5 to 5 or 9, it produced nice square wave 100 Hz, 2.6 Vpp.
I have only simple hand held velleman oscilloscope, but it is good enough for experimentation.
So I fed it to my second system and looked at trace on the mike. It looks just like John k... small white insert on the bottom.
So...where do we go now? What do I do next? I want to experiment, improve...
My system is sounding freaking good, but measures poorly. The square wave out is nothing like square wave in. Am I disappointed? Not really. System sounds good. But I would like to improve it.
System is biamplified, active crossover, solid state for 15" woofer in closed box, tube amp for mid/hights on open baffle. The way I like it.
Can you comment? Lets keep this thread rolling.
Those are measured results using SoundEasy's new Ultimate Equalizer active dsp crossover.
Quick question John...is getting the perfect square wave you showed audible? Is there big improvement in sound when compared to normal crossover? (between the red line and white insert) Can you hear the difference? Can you discribe the difference?
The reason I am asking is because over this weekend I fed the square wave signal to various amps. Just for fun, what the heck. Since I made the circuit, why not. So I fed the signal and looked at the output on the scope. Well, I did that with about dozen amplifiers and got different results. Not quite sure how to interpret it. Two solid state Onkyo amps gave perfect square wave on the output, yet one was sounding great, the other quite poorly. One solid state SONY measured poorly and sounded poorly. One Magnavox (SE) tube amp measure poorly but sounded great. Other Magnavox (PP) measured better but sounded worse than SE. One OTL tube amp measured great and sounded great. Anyway, do not want to make short story long, I thing I got all the combinations. Perfect square wave on the output of the amp did not guarantie good sound. I am asking if there is a possibility that the same happens with speakers. Do not want to open another can of worms, but how much is single measurement relevant to the sound?
The reason I am asking is because over this weekend I fed the square wave signal to various amps. Just for fun, what the heck. Since I made the circuit, why not. So I fed the signal and looked at the output on the scope. Well, I did that with about dozen amplifiers and got different results. Not quite sure how to interpret it. Two solid state Onkyo amps gave perfect square wave on the output, yet one was sounding great, the other quite poorly. One solid state SONY measured poorly and sounded poorly. One Magnavox (SE) tube amp measure poorly but sounded great. Other Magnavox (PP) measured better but sounded worse than SE. One OTL tube amp measured great and sounded great. Anyway, do not want to make short story long, I thing I got all the combinations. Perfect square wave on the output of the amp did not guarantie good sound. I am asking if there is a possibility that the same happens with speakers. Do not want to open another can of worms, but how much is single measurement relevant to the sound?
Whether ot not the ability to reproduce a square wave aont he desing axis makes an audible difference is, has been, and I suspect will continure to be a highly debated topic. The least that can be said is that is will be dependent on source material.
Amplifier quality is dependent on a lot more parametersa that the ability to treproduce square wave. It's not an "all other things equal" kind of test.
The thing is that with the SE UE you can design a speaker and play it digitally with and without phase linearization. So, when done correctly is it an all other things equal test. Same crossover, same amplitude response, same polar response.... The only change is normal vs linear phase.
Amplifier quality is dependent on a lot more parametersa that the ability to treproduce square wave. It's not an "all other things equal" kind of test.
The thing is that with the SE UE you can design a speaker and play it digitally with and without phase linearization. So, when done correctly is it an all other things equal test. Same crossover, same amplitude response, same polar response.... The only change is normal vs linear phase.
IOW - it was irrelavent (not that this study was controlled well enough to draw any conclusions from.)
You are right Gedlee, it was not controlled enough. It was not even study. I was just having fun. And that is allowed, right?
Back to the square wave and speakers then. Is there an audible correlation between the square wave reproduction of a given speaker and the sound quality?
Can you, Gedlee, post square wave measurement of your speakers? Don't be shy.
Back to the square wave and speakers then. Is there an audible correlation between the square wave reproduction of a given speaker and the sound quality?
Can you, Gedlee, post square wave measurement of your speakers? Don't be shy.
No! I've never measured a square wave.
Is there a correlation between X and Y? Well as a scientist I would would have to assume no, unless there was some evidence to say otherwise. Do you know of any such evidence?
Is there a correlation between X and Y? Well as a scientist I would would have to assume no, unless there was some evidence to say otherwise. Do you know of any such evidence?
Its never too late!
I believe that is the heart of the whole thread. Lets ask the same question again slightly differently. Lets take some "highly acclaimed" loudspeakers, which some say, sound great, and measure how they reproduce square wave.
If they can not reproduce square wave properly, then maybe its not really that important.
As a scientist I would proceed by measuring large number of commercial speakers, clustered into groups by the sound quality, evaluated in blind tests, following by the measurement of the square wave reproduction. If, by the end of large testing, one could see the correlation, then reproduction of square wave is important. I am not aware of such study, I would not be asking, would I?
I remember reading somewhere that Dahlquist DQ-10 crossover was supposedly optimized to reproduce the best square wave possible...so maybe there is something to it.
I believe that is the heart of the whole thread. Lets ask the same question again slightly differently. Lets take some "highly acclaimed" loudspeakers, which some say, sound great, and measure how they reproduce square wave.
If they can not reproduce square wave properly, then maybe its not really that important.
As a scientist I would proceed by measuring large number of commercial speakers, clustered into groups by the sound quality, evaluated in blind tests, following by the measurement of the square wave reproduction. If, by the end of large testing, one could see the correlation, then reproduction of square wave is important. I am not aware of such study, I would not be asking, would I?
I remember reading somewhere that Dahlquist DQ-10 crossover was supposedly optimized to reproduce the best square wave possible...so maybe there is something to it.
markus76, thanks for the link, but I do not see square wave there...
I believe we are talking about reproducing the square wave in this thread.
I believe we are talking about reproducing the square wave in this thread.
In my previous post I was asking how relevant to the sound is the ability, or un-ability, of the speakers to reproduce of square wave. You cut out small part of my sentence and provide me with the link to the article about how flat the frequency response correlates with perceived sound. I do not dispute that.
As with so many topics in audio square wave reproduction is just a time consuming sidestep. Olive's blog (and the references) would open the eyes of many.
Best, Markus
Let's try and not substitute the discussion of reproducing the square wave with discussion of whether or not it is audible. Audibility is a science in itself, where one can easily prove for instance, that all CD players, or (gah!) loudspeakers sound the same anyway...
Now, in my opinion, ability to reproduce the square wave is not a virtue that we are trying to achieve for the sake of reproducing the square wave, but just a sign that your speaker has minimal phase and time distortion. That's all...
Now, in my opinion, ability to reproduce the square wave is not a virtue that we are trying to achieve for the sake of reproducing the square wave, but just a sign that your speaker has minimal phase and time distortion. That's all...
Last edited:
- Status
- Not open for further replies.