This is true lieven. If you re-read the first post, you'll see that I'm not disagreeing with you. I know what happens when you apply DC to a voice coil so stop shouting and get a grip.
What I'd like my 4-way system to do is reproduce the initial impact of a sound (like the leading edge of a square wave) with out the frequency dependent phase shift associated with analog x-overs, active or passive.
Rodd Yamas***a
What I'd like my 4-way system to do is reproduce the initial impact of a sound (like the leading edge of a square wave) with out the frequency dependent phase shift associated with analog x-overs, active or passive.
Rodd Yamas***a
lieven,
You've said the magic word (or letters), JBL. A man after my own heart. Not to hijack my own thread, take a look at this thread.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=4840
I've aways like the 4343 concept.
Rodd Yamas***a
You've said the magic word (or letters), JBL. A man after my own heart. Not to hijack my own thread, take a look at this thread.
http://www.diyaudio.com/forums/showthread.php?s=&threadid=4840
I've aways like the 4343 concept.
Rodd Yamas***a
rodd, saw your system on the other treath.
Beautiful, but aren't your overdoing a bit the midbass-section.
I would go for only one mid-bass in a much smaller volume so the you can use the natural fall-off-frequency aided perhaps by only a 6db x-over even below fb if this enhance phase between bass and mid-bass (better WAF and less phase correction).
!!!! Phase-correction could also be done by moving mid-bass (and horns) around till satisfacion (passive way).
2 woofers at those frequencies in a (relativly) little room is asking for trouble.
1 woofer in smaller enclosure would increase efficiency as much as 3 dB or better, causing it to match more with the TAD's and the horn above it. Also transients would improve (better cello and spanish guitar).
Make a 6-10liter closed baffle for only one le10a and try it first without any filter.
You like it fast, so don't switch to 12 inch, that won't make it upto 1200 Hz in a decent way.
Buy a Behringer Ultracurve to level all, you'll be suprised what this thingy could do for only 300 $ or so.
Beautiful, but aren't your overdoing a bit the midbass-section.
I would go for only one mid-bass in a much smaller volume so the you can use the natural fall-off-frequency aided perhaps by only a 6db x-over even below fb if this enhance phase between bass and mid-bass (better WAF and less phase correction).
!!!! Phase-correction could also be done by moving mid-bass (and horns) around till satisfacion (passive way).
2 woofers at those frequencies in a (relativly) little room is asking for trouble.
1 woofer in smaller enclosure would increase efficiency as much as 3 dB or better, causing it to match more with the TAD's and the horn above it. Also transients would improve (better cello and spanish guitar).
Make a 6-10liter closed baffle for only one le10a and try it first without any filter.
You like it fast, so don't switch to 12 inch, that won't make it upto 1200 Hz in a decent way.
Buy a Behringer Ultracurve to level all, you'll be suprised what this thingy could do for only 300 $ or so.
lieven,
I had 1 - LE10A in a critically damped closed box for years. It was always strained at high output and lacked the dynamics. The dual arrangement sounds cleaner at volume and a touch more dynamic. Look at this thread, and you'll see why I went with the stereo 16" Tads and dual 10".
http://www.diyaudio.com/forums/showthread.php?s=&postid=56468#post56468
I would still like the 10"s to be more dynamic, but it might be just a phase alignment problem, which brings us back to this thread.
I would eventually like to implement some time delay to get better alignment of the drivers acoustic centers. I know that would only be a partial solution, but since I've been working on them for over 20 years, it's a good step in the evolution. Ultimately, I would go quad-amp, and implement an active 4-way digital xover with time delay. I don't think that will happen very soon.
SAF, not a problem, unless she wants my speakers.
Then I'm in trouble.
What are you running for your main system?
Rodd Yamas***a
I had 1 - LE10A in a critically damped closed box for years. It was always strained at high output and lacked the dynamics. The dual arrangement sounds cleaner at volume and a touch more dynamic. Look at this thread, and you'll see why I went with the stereo 16" Tads and dual 10".
http://www.diyaudio.com/forums/showthread.php?s=&postid=56468#post56468
I would still like the 10"s to be more dynamic, but it might be just a phase alignment problem, which brings us back to this thread.
I would eventually like to implement some time delay to get better alignment of the drivers acoustic centers. I know that would only be a partial solution, but since I've been working on them for over 20 years, it's a good step in the evolution. Ultimately, I would go quad-amp, and implement an active 4-way digital xover with time delay.
I don't think that will happen very soon.SAF, not a problem, unless she wants my speakers.
Then I'm in trouble.What are you running for your main system?
Rodd Yamas***a
does a speaker have to reproduce a square wave to be musical.
Note I said musical not accurate. some of teh speakers I love are not accurate but quite musical.
lest we forget we are listening to music. no speaker will reproduce a live concert or even a live solo BAD piano player on a malfunctioning piano.
In my city we occasionally get a monkey man who cmoes by with a small hand drum. the drum is about 4" dia at each end and has 2 string (one at each end) each stringhas a small weight attached to it. when he flips his wrist he strings swing and the weight hits the drum skin and produces a tone.
The best of speakers I have heard have not been able to reproduce this simplelest of instruments accurately. I used to play guitar and piano and I know how easy it is to discern live from recorded. my mom who is quite deaf can do so easily.
my 2 cents.
Note I said musical not accurate. some of teh speakers I love are not accurate but quite musical.
lest we forget we are listening to music. no speaker will reproduce a live concert or even a live solo BAD piano player on a malfunctioning piano.
In my city we occasionally get a monkey man who cmoes by with a small hand drum. the drum is about 4" dia at each end and has 2 string (one at each end) each stringhas a small weight attached to it. when he flips his wrist he strings swing and the weight hits the drum skin and produces a tone.
The best of speakers I have heard have not been able to reproduce this simplelest of instruments accurately. I used to play guitar and piano and I know how easy it is to discern live from recorded. my mom who is quite deaf can do so easily.
my 2 cents.
roddyama said:Now this has me wondering.
If this is not true, what is a "real ribbon", and what is a "psuedo-ribbon"??
In a true ribbon the entire diaphram is the conductor - usually aluminum foil (ie Decca Ribbon, Ravens, the ribbon in a Magnepan). A psuedo-ribbon has a voice coil "printed" on top of a plastic diaphragm (ie BG, Newform, Monsoons, Philips, ApexJr Ribbon). The latter are more correctly called magneto-stats.
dave
I like them. My Brother In Law has them called CLS or something like that.
other speakers I have liked.
Klipsch Forte (in 1990 I heard them at an ex GF's place in Chicago)
Klipschons and Belle Klipsch combo in 1987
ALR Factor 2 and Factor 7 (heard at ALR factory in Germany)
Dynaudio Audience 50/52
Vandersteen 2Ci (slightly modified using a different Dynaudio tweeter and Vifa mid).
Magneplanar SMG/MMG great for the price
Snell Type K, Type C and Type E again all were great for the money.
I have not liked many big speakers...
Wilson WATT 7 / Puppy - too little speaker for the room I guess
B&W Nautilus - too much speaker for the room
Joesph Audio (looks like WATT/Puppy) forget name - though Joe Joesph seems to a nice guy. Nice speaker too but something was not right (other than the $20,000 price).
other speakers I have liked.
Klipsch Forte (in 1990 I heard them at an ex GF's place in Chicago)
Klipschons and Belle Klipsch combo in 1987
ALR Factor 2 and Factor 7 (heard at ALR factory in Germany)
Dynaudio Audience 50/52
Vandersteen 2Ci (slightly modified using a different Dynaudio tweeter and Vifa mid).
Magneplanar SMG/MMG great for the price
Snell Type K, Type C and Type E again all were great for the money.
I have not liked many big speakers...
Wilson WATT 7 / Puppy - too little speaker for the room I guess
B&W Nautilus - too much speaker for the room
Joesph Audio (looks like WATT/Puppy) forget name - though Joe Joesph seems to a nice guy. Nice speaker too but something was not right (other than the $20,000 price).
Navin,
If you read my first post, you'll see that I dont believe it's possible to reproduce a square wave with a multi-way speaker. The point is, that if a multi-way speaker could reproduce the leading edge of a square wave, then it will be able to more faithfully reproduce the impact in music that has been typically lost in the phase shifts in the xovers.
Alway trying to make it better.
Rodd Yamas***a
Point taken. I can assure you that my speakers are far from perfect, but I still like the way they sound. That doesn't mean that will stop trying to improve them. That is, in large part, why we are here.
If you read my first post, you'll see that I dont believe it's possible to reproduce a square wave with a multi-way speaker. The point is, that if a multi-way speaker could reproduce the leading edge of a square wave, then it will be able to more faithfully reproduce the impact in music that has been typically lost in the phase shifts in the xovers.
Alway trying to make it better.
Rodd Yamas***a
Lieven,
a (DC-on,DC-off)* sequence is not DC anymore. It is AC. And depending on the switching frequency it falls in the audible band and will be heard as sound. A bit like a trumpet
Perfect reproduction of that sound requires a minimum-phase system of perfectly straight frequency response from DC to God. Curtailing treble gives rise to rounded edges. Curtailing bass makes the level parts of the wave droop. Making the system non-MP (multi-way crossovered speaker system) turns it into an utter mess.
a (DC-on,DC-off)* sequence is not DC anymore. It is AC. And depending on the switching frequency it falls in the audible band and will be heard as sound. A bit like a trumpet
Perfect reproduction of that sound requires a minimum-phase system of perfectly straight frequency response from DC to God. Curtailing treble gives rise to rounded edges. Curtailing bass makes the level parts of the wave droop. Making the system non-MP (multi-way crossovered speaker system) turns it into an utter mess.
Hi Werner
Completely agree with you !! If Lieven were correct then all signals we listen to were just something "wiggling DC" !!
Navin
I do of course agree with you that only because a system can reproduce a correct squarewave wouldn't guarantee that it would sound musical or natural. There are many factors more that have to be "right". OTOH only a system that can reproduce a squarewave to some accuracy is able to reproduce any non-deterministic waveshape with at least a minimum degree of accuracy.
The simulation I showed before is a PSPICE simulation of my Manger project using a subtractive crossover. The drivers are simulated as subsequent 2nd order highpass and lowpass filters. It is of course only accurate to some extent at the lower end but surely not so anymore where diffraction is starting to play a major role. So the "real thing" will of course be much worse, but I assume that it will still be better in this respect, than the majority of speakers available.
Apart from the sqarewave response the system will act as a good approximation of a point source within an important part of the audio range. The MSW's Neodymium magnet system guarantees for low IMD and noise.
But still I am aware that this system will not be perfect. But I think life would be boring if a perfect speaker could be achieved that easily, wouldn't it ?
Rodd
I still have a two-way speaker using a JBL 2344 bi-radial horn and I can still appreciate the sound of it, simply because it is fun to listen to, though it is not as accurate as the manger system!
I don't know what levels you are listening to, but if you stay at the lower end of your driver's SPL capabilities then you could maybe live with 2nd order crossovers and then there would be some possibilities for active crossovers with improved transient capabilities. If you want to exploit the low IMD/low THD capabilities of your horns at high levels then it would be better to use LR4 for instance (I assume it is what you are currently doing, aren't you?)
Regards
Charles
Completely agree with you !! If Lieven were correct then all signals we listen to were just something "wiggling DC" !!
Navin
I do of course agree with you that only because a system can reproduce a correct squarewave wouldn't guarantee that it would sound musical or natural. There are many factors more that have to be "right". OTOH only a system that can reproduce a squarewave to some accuracy is able to reproduce any non-deterministic waveshape with at least a minimum degree of accuracy.
The simulation I showed before is a PSPICE simulation of my Manger project using a subtractive crossover. The drivers are simulated as subsequent 2nd order highpass and lowpass filters. It is of course only accurate to some extent at the lower end but surely not so anymore where diffraction is starting to play a major role. So the "real thing" will of course be much worse, but I assume that it will still be better in this respect, than the majority of speakers available.
Apart from the sqarewave response the system will act as a good approximation of a point source within an important part of the audio range. The MSW's Neodymium magnet system guarantees for low IMD and noise.
But still I am aware that this system will not be perfect. But I think life would be boring if a perfect speaker could be achieved that easily, wouldn't it ?
Rodd
I still have a two-way speaker using a JBL 2344 bi-radial horn and I can still appreciate the sound of it, simply because it is fun to listen to, though it is not as accurate as the manger system!
I don't know what levels you are listening to, but if you stay at the lower end of your driver's SPL capabilities then you could maybe live with 2nd order crossovers and then there would be some possibilities for active crossovers with improved transient capabilities. If you want to exploit the low IMD/low THD capabilities of your horns at high levels then it would be better to use LR4 for instance (I assume it is what you are currently doing, aren't you?)
Regards
Charles
Hi Charles,
you seem to be getting into this single driver (or wide range drvier) bit too.
1. How do you find the dispersion of the Manger esp. at higher frequencies?
2. You say you are usng a 2und order Xo with the Manger. What are you crossing over to?
Maybe we should start a seperate thread for this???
regards
navin
you seem to be getting into this single driver (or wide range drvier) bit too.
1. How do you find the dispersion of the Manger esp. at higher frequencies?
2. You say you are usng a 2und order Xo with the Manger. What are you crossing over to?
Maybe we should start a seperate thread for this???
regards
navin
Hi Navin
The MSW is indeed quite directive at it's upper end. So it is something for selfish listeners . Apart from that there is a plastic thingie available called "Manger Holoprofil" that improves things a bit but at a prohibitively high price so I'd rather stay selfish!
In the current experimental setup (which is still passive at the moment) I am using it together with a Dynaudio 20W75 driver.
The final speaker will use a driver made by Audio Technology (also a Danish company, owned by the founder od Dynaudio).
I already made some post according to this in the active subtractive X-over thread. I have also prepared some new stuff regarding filler driver crossovers I will post there soon.
Regards
Charles
The MSW is indeed quite directive at it's upper end. So it is something for selfish listeners
. Apart from that there is a plastic thingie available called "Manger Holoprofil" that improves things a bit but at a prohibitively high price so I'd rather stay selfish!In the current experimental setup (which is still passive at the moment) I am using it together with a Dynaudio 20W75 driver.
The final speaker will use a driver made by Audio Technology (also a Danish company, owned by the founder od Dynaudio).
I already made some post according to this in the active subtractive X-over thread. I have also prepared some new stuff regarding filler driver crossovers I will post there soon.
Regards
Charles
my next project will be a speaker without XO (I might even try without box or XO) so I am looking seriously at cheap "full rnage" drivers....
these speakers will be used for surround so far the best speaker I can find are some 2" cone drivers with limited excursion but I suppose if I use multiples of these I might maange.
more later on this. right now I am trying to meake a dead box :-(
BTW which part of Schwiez are you from. I lived in Schaffhaussen for a bit. Skied at Chur. Nice country. Lovely people.
cheers
these speakers will be used for surround so far the best speaker I can find are some 2" cone drivers with limited excursion but I suppose if I use multiples of these I might maange.
more later on this. right now I am trying to meake a dead box :-(
BTW which part of Schwiez are you from. I lived in Schaffhaussen for a bit. Skied at Chur. Nice country. Lovely people.
cheers
Hi Charles,
Your suggestion for the LR4 would help with both of those problems. It's not the square wave, but it would probably sound better then my current xover. This may be the best reasons I've received so far on why I should replace my Nakamichi xover.
Thanks.
Actually, I'm using a 2nd order xover from the dual 10" midrange to the mid-horn at 1200Hz. I'm assuming, since you mention higher outputs, that your suggestion for a LR4 is to limit the band width and therefore the excursion of the horns diaphragm. I am doing this to an extent by crossing over at a little higher frequency, but in general, I would have to agree with you. This may be, in part, what's behind my dissatifaction with the dynamics of my mid-range as I indicated to Lieven when I told him:
This phase mis-alignment may be another reason that the sound isn't as good I would like it to be. Certianly in crossing over from cones to a horn using analog filters, the phase will be a problem. The 12db slope may be too shallow, forcing the two mis-aligned drivers to share the output for double the range of a xover with a 24db slope.
Your suggestion for the LR4 would help with both of those problems. It's not the square wave, but it would probably sound better then my current xover. This may be the best reasons I've received so far on why I should replace my Nakamichi xover.
Thanks.
Werner said:Lieven,
a (DC-on,DC-off)* sequence is not DC anymore. It is AC. And depending on the switching frequency it falls in the audible band and will be heard as sound. A bit like a trumpet
Although I don't really feel comfortable with the way it was previously put, applying and disconnecting a DC source is DC, not AC.
When a DC source is connected (forgetting the risetime - slanting rise and rounded top edge) a current will flow through the loudspeaker that is a constant value, when the DC source is disconnected no current will flow - this is still a direct current although it is intemittent it is not alternating.
Although if you were to apply a positive DC then a negative DC this would be an AC (although nothing near sinusoidal) as the current will flow in both directions around the circuit.
Hope this helps!
<center>_____________________________________</center>
An interesting note to finish on though...
...speaker manufacturers do not merely test peak power outputs and transient handling of their speakers, they also use what are known as speaker killers - square, triangular and spike waveforms to physically kill the speakers (although this test is more for mechanical stability and build quality than sound reproduction).
Annex,
no, it would be AC, with a DC (offset)
component and a spread of harmonics from the switching frequency on. If it was a single on/off cycle, it still would contain the spectrum of the two convolved step functions that you implement by the mere act of switching.
If you make the time between on and off long enough, then indeed it approaches what we conveniently call 'DC', even though mathematically the only 'real' DC signal has to be infinite in duration.
But since we were talking audio and square waves, we really are far away from any notion of DC.
no, it would be AC, with a DC (offset)
component and a spread of harmonics from the switching frequency on. If it was a single on/off cycle, it still would contain the spectrum of the two convolved step functions that you implement by the mere act of switching.
If you make the time between on and off long enough, then indeed it approaches what we conveniently call 'DC', even though mathematically the only 'real' DC signal has to be infinite in duration.
But since we were talking audio and square waves, we really are far away from any notion of DC.
Hi Eric
Neodymium magnet drive systems (and Alnico as well) are less prone to barkhausen noise.
John Watkinson for instance claims that the resolution of a transducer with ceramic magnet can't even achieve that of 16 bit audio !
See the two lowest paragraphs of :
http://www.celticaudio.co.uk/technical2.htm
Regards
Charles
Neodymium magnet drive systems (and Alnico as well) are less prone to barkhausen noise.
John Watkinson for instance claims that the resolution of a transducer with ceramic magnet can't even achieve that of 16 bit audio !
See the two lowest paragraphs of :
http://www.celticaudio.co.uk/technical2.htm
Regards
Charles
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.