Scottmoose said:
yes, there is such a risk but it would never be as bad as in case of multi-way loudspeaker
see image attached - also a real life example - straight non-tapered array of 32 drivers
the step is jagged but still it is SINGLE
nevertheless my personal choice is a single 8 inches driver
best compromise for my omni
unfortunately in practice one has to choose - an array or an omni
I choose omni as it gives the best room-loudspeaker interface
though an omnidirectional line array is possible but a lot of money (for suitable Bandor drivers) and of complicated woodworking would be needed
best regards!
graaf
Attachments
"Scottmoose I was refering to the FR rather than the step et al response. "
tell us your experience with such constructions,
fotos, measurements etc.
may be you don´t know and only theoritically guess,
i´ve testeted it over years in my double horns.
specially the invers indirect driver is not listenable at the
listening place but makes soundstage, bass,
i tested last month the TUBA against the KORNETT
compair the constructions.
please read the first wavefront on WIKI:
http://de.wikipedia.org/wiki/Gesetz_der_ersten_Wellenfront
look my Step response of the SAXOPHON,
it is better than can be, better than a sigle driver solution,
and that by a double system with 2 different driver.
tested over years and proofed by measurement.
tell us your experience with such constructions,
fotos, measurements etc.
may be you don´t know and only theoritically guess,
i´ve testeted it over years in my double horns.
specially the invers indirect driver is not listenable at the
listening place but makes soundstage, bass,
i tested last month the TUBA against the KORNETT
compair the constructions.
please read the first wavefront on WIKI:
http://de.wikipedia.org/wiki/Gesetz_der_ersten_Wellenfront
look my Step response of the SAXOPHON,
it is better than can be, better than a sigle driver solution,
and that by a double system with 2 different driver.
tested over years and proofed by measurement.
Attachments
Whom are you asking? Graaf or me?
If me, you will have noticed that I in fact mentioned several instances when multiple FR units can be used: if they are "...focused upon a specific point in space, or run bipole, or quasi-omni." And so on.
OTOH, a straight array / multiple FR drivers on the same baffle = a bad idea due to the lobing of the outputs with increasing frequency; a problem that only gets worse with larger drivers. The general trend can be Eq'd ~flat to compensate for the HF losses entailed, but it can't do anything for the loss of refinement. Worse, a straight array will loose time coherence due to the ends being a greater distance from the listening position, & will benefit from some delay time built in, or ~faking it via power tapering. Step response might be OK, but the rest, especially the frequency response, is not.
If me, you will have noticed that I in fact mentioned several instances when multiple FR units can be used: if they are "...focused upon a specific point in space, or run bipole, or quasi-omni." And so on.
OTOH, a straight array / multiple FR drivers on the same baffle = a bad idea due to the lobing of the outputs with increasing frequency; a problem that only gets worse with larger drivers. The general trend can be Eq'd ~flat to compensate for the HF losses entailed, but it can't do anything for the loss of refinement. Worse, a straight array will loose time coherence due to the ends being a greater distance from the listening position, & will benefit from some delay time built in, or ~faking it via power tapering. Step response might be OK, but the rest, especially the frequency response, is not.
Scottmoose said:
Time to post the result of a bad idea ... can't resist
.Since my "Dipol 08" design is not a prototype anymore we had
an extended listening. A neighbour of mine brought his
TANNOY System 800 for comparison, which is a coaxial
"Point Source" Design.
There were about 6 people listening from experienced listeners
to a 11 Year old boy, whose opinion counted double for me,
because young children are not that prejudiced and have
normally better ears. The boy gave 100 points to the
tapered line array and 50 points to the point source.
Concerning the Tannoy he said
"It is not good, it is stopping the music."
I love childrens statements on loudspeakers.
By myself i would not have been so radical in my judgement.
The Tannoy was recognized as lacking detail and imaging,
thereby sounding somewhat harsh.
Even when changing listening places in the room, there was
no advantage ascertained by the listeners.
Although, the comparison of concepts was very interesting.
Kind regards
Attachments
Hi,
yes to be honest it was NOT a comparison of concepts,
but the comparison of two realizations of concepts.
The harshness i heared with the Tannoy 800 was not
necessarily due to the point source concept ...
Opinions on this speaker are very split among users, i tend
to share your impression ...
Cheers
yes to be honest it was NOT a comparison of concepts,
but the comparison of two realizations of concepts.
The harshness i heared with the Tannoy 800 was not
necessarily due to the point source concept ...
Opinions on this speaker are very split among users, i tend
to share your impression ...
Cheers
Dipole arrays I like.
An array with a tapered C-to-C spacing is an interesting notion. If it could even still be called an array in the usual sense. In a sense, I suspect that you'll be seeing some variance in the transition from near to farfield due to the varying distance between the drivers, although of course the fact that they're all producing the same signal, if I read what you're doing correctly, throws that logic out a bit. Whatever it is in practice, it's certainly a bit different, which is always interesting to see.
An array with a tapered C-to-C spacing is an interesting notion. If it could even still be called an array in the usual sense. In a sense, I suspect that you'll be seeing some variance in the transition from near to farfield due to the varying distance between the drivers, although of course the fact that they're all producing the same signal, if I read what you're doing correctly, throws that logic out a bit. Whatever it is in practice, it's certainly a bit different, which is always interesting to see.
Not all 6 drivers are producing the same signal ...
The array is split in an uppper 3 and lower 3 group.
The lower group is rolled off
v e e r y s m m o o o o t h l y (<6db)
beginning around 2Khz.
Both groups are wired parallel internally, the 2 groups
are in series. Briefly, the rolloff in upper highs expected from
a usual line array is compensated in bypassing the lower
group by an RC network.
This approach keeps transition from nearfield to far field more
independent from frequency since tapering is combined
by power and by distribution of drivers on the baffle.
In the beginning i thought an active network would be good,
using separated amps for the groups at least. But the passive
network does the job flawlessly IMO.
The presence and brillance region is produced predominantly
from the upper 3 drivers. The interference patterns in FR
caused by the uneven distribution of drivers are very smooth.
Listeners say, there is no noticeable change when moving
your head vertically from sitting to standing position.
Of cause listening distance should be around 1,5 meters or more.
From there you can go 7 meters away in my room
(then the room ends), without the system changing its
character subjectively. It simply sounds nearly the same.
Everywhere. Room contribution increases smoothly with
distance, thats all. You can choose where to listen.
Of cause by measurement there is noticeable change. But
distribution and character of the humps and dips over differing
position varies so smoothly and random - no serious overall
tendency detectable - that it is not an issue.
I modelled this behaviour using software and several
prototypes before. Believe me - i cannot afford wasting
kilos of marble and a dozen of fairly expensive drivers. Not
to say the working hours.
I was going for a speaker to sound homogenous. I was
prepared to some compromise in the brillance region and
i was prepared to accept it. I was a "more datail please !"
listener when younger you know, since i shifted my priorities
towards coherence. It came as a gift to me that the current
design is brilliant and detailed whithout harshness.
You don't miss a tweeter at all. In fact it sounds as detailed
and without harshness in the highs, that i had only achieved
that using a ribbon tweeter before.
Maybe ageing of my 42 year old ears helps ...
But i do not think it is only due to that.
Cheers
The array is split in an uppper 3 and lower 3 group.
The lower group is rolled off
v e e r y s m m o o o o t h l y (<6db)
beginning around 2Khz.
Both groups are wired parallel internally, the 2 groups
are in series. Briefly, the rolloff in upper highs expected from
a usual line array is compensated in bypassing the lower
group by an RC network.
This approach keeps transition from nearfield to far field more
independent from frequency since tapering is combined
by power and by distribution of drivers on the baffle.
In the beginning i thought an active network would be good,
using separated amps for the groups at least. But the passive
network does the job flawlessly IMO.
The presence and brillance region is produced predominantly
from the upper 3 drivers. The interference patterns in FR
caused by the uneven distribution of drivers are very smooth.
Listeners say, there is no noticeable change when moving
your head vertically from sitting to standing position.
Of cause listening distance should be around 1,5 meters or more.
From there you can go 7 meters away in my room
(then the room ends), without the system changing its
character subjectively. It simply sounds nearly the same.
Everywhere. Room contribution increases smoothly with
distance, thats all. You can choose where to listen.
Of cause by measurement there is noticeable change. But
distribution and character of the humps and dips over differing
position varies so smoothly and random - no serious overall
tendency detectable - that it is not an issue.
I modelled this behaviour using software and several
prototypes before. Believe me - i cannot afford wasting
kilos of marble and a dozen of fairly expensive drivers. Not
to say the working hours.
I was going for a speaker to sound homogenous. I was
prepared to some compromise in the brillance region and
i was prepared to accept it. I was a "more datail please !"
listener when younger you know, since i shifted my priorities
towards coherence. It came as a gift to me that the current
design is brilliant and detailed whithout harshness.
You don't miss a tweeter at all. In fact it sounds as detailed
and without harshness in the highs, that i had only achieved
that using a ribbon tweeter before.
Maybe ageing of my 42 year old ears helps ...
But i do not think it is only due to that.
Cheers
Ah, that makes much more sense, with the lower units rolled off around that frequency, which will put the transition for the upper 3 units from near to farfield, if I remember my Jim Griffin & my SWAG at the length, at roughly 1.5 - 2m so I'd expect them to be pretty stable.
The drivers look like CSS FR125s, right? I can imagine that working far better than some other combinations, what with them being pretty smooth, & having decent detail (but not overly so). Love the marble too -nice touch.
Yeah, believe me, I know what you mean about not being able to afford to waste money / time / drivers etc. I'm a member of the same club.
We all have our different priorities, right? Being 29 my HF is still pretty good & makes 20KHz (dropped off a bit recently -2 years ago I could still get up to 21KHz 
) so I doubt I could cope with the combing through the upper midrange & treble, even though that's probably the best I've seen from this configuration -seems on a quick glance to be better than, say, the IDS or Kuzma arrays. Now, if they were focused, then my ears would really start to prick up...
Best
Scott
The drivers look like CSS FR125s, right? I can imagine that working far better than some other combinations, what with them being pretty smooth, & having decent detail (but not overly so). Love the marble too -nice touch.
Yeah, believe me, I know what you mean about not being able to afford to waste money / time / drivers etc. I'm a member of the same club.
We all have our different priorities, right? Being 29 my HF is still pretty good & makes 20KHz (dropped off a bit recently -2 years ago I could still get up to 21KHz ) so I doubt I could cope with the combing through the upper midrange & treble, even though that's probably the best I've seen from this configuration -seems on a quick glance to be better than, say, the IDS or Kuzma arrays. Now, if they were focused, then my ears would really start to prick up... Best
Scott
Hi,
yes they are FR 125 S.
I select the drivers which play in the
"upper group" for the best definition in the highs.
Like this:
Driver sounds dull -> lower group.
Driver has good high definition
but a bit nasty in presence/brillance -> upper group, lowest pos.
Driver has unobtrusive highs, bit dull -> upper group, top
Best driver out of 6 -> upper group mid.
Drivers are balanced same way for both speakers and
numbered before mounting, to get left and right system
similar. Two speakers are NEVER equal, are they ?
If there will be the opportunity to build a "MK II" version
someday, i'd rather experiment with driving them actively,
one amp for each of the upper drivers at least and reduce the
array to one single driver in the upper brillance region.
like
6->3->2->1
or
6->3->1
With separated amps there are lots of options to tweak that
concept further. I hope there will be an opportunity to try that
as long as my ears work fairly well, since i cannot build such a
speaker every year due to time and effort spent ...
Using digital signal processing there will be lots of options
including delaying and focussing ...
But for now i am happy with the result.
Kind regards
yes they are FR 125 S.
I select the drivers which play in the
"upper group" for the best definition in the highs.
Like this:
Driver sounds dull -> lower group.
Driver has good high definition
but a bit nasty in presence/brillance -> upper group, lowest pos.
Driver has unobtrusive highs, bit dull -> upper group, top
Best driver out of 6 -> upper group mid.
Drivers are balanced same way for both speakers and
numbered before mounting, to get left and right system
similar. Two speakers are NEVER equal, are they ?
If there will be the opportunity to build a "MK II" version
someday, i'd rather experiment with driving them actively,
one amp for each of the upper drivers at least and reduce the
array to one single driver in the upper brillance region.
like
6->3->2->1
or
6->3->1
With separated amps there are lots of options to tweak that
concept further. I hope there will be an opportunity to try that
as long as my ears work fairly well, since i cannot build such a
speaker every year due to time and effort spent ...
Using digital signal processing there will be lots of options
including delaying and focussing ...
But for now i am happy with the result.
Kind regards
The child represents and confirms there are such human realities.
As for point source or not - I take no position on approaches or technology either. It's just nice to hear about something that does not
Perhaps this demonstrates the human mind understands music in 3D so completely that anything that can successfully recreate a 1D entry point (line or point) for sound can slip into our world less noticed.
Hmm... I wonder if B&W still keeps a few dozen handy orphans in a dungeon somewhere...?
-jerry
As for point source or not - I take no position on approaches or technology either. It's just nice to hear about something that does not
Perhaps this demonstrates the human mind understands music in 3D so completely that anything that can successfully recreate a 1D entry point (line or point) for sound can slip into our world less noticed.
Hmm... I wonder if B&W still keeps a few dozen handy orphans in a dungeon somewhere...?
-jerry
have you considered an esl???
something along the line of an Acoustat 0ne plus 0ne. 175 mm by1100 mm (active diaphragm) with a mm or so stator to diaphragm spacer. Provides the most for the least. Lowest distortion widest range single driver you can find or make. You can find Acoustats in NA for as little as 3-4 hundred dollars if you look hard and 5-6 hundred just about anywhere. The drive is two way the driver is fullrange but you can biamp an Acoustat and run it as a two way. Lots of info on toroid transformers 9-15 volts to 240 volts will work well and not cost a lot run in parallel you can enhance the bandwidth. YOu can purchase 3 micron polyester film which is more than 5 times thinner than the stock Acoustat film which was 65 gage Mylar HS.
Drive your room as a line source and forget about floor and ceiling bounce. Smooth clean detailed stage and image to make you think the performers are there. Not really all that hard to build. Only thing that is as dynamic are ribbons and horns. Have played with single drivers and modified the heck out of them and they don't do the trick. A used set of Acoustats will provide you with all the parts you need to get going and you can build up your own new panels at your lesiure while you listen to the factory panels. No dynamic driver has the range nor can they compete from distortion point of view. Worth considering.
something along the line of an Acoustat 0ne plus 0ne. 175 mm by1100 mm (active diaphragm) with a mm or so stator to diaphragm spacer. Provides the most for the least. Lowest distortion widest range single driver you can find or make. You can find Acoustats in NA for as little as 3-4 hundred dollars if you look hard and 5-6 hundred just about anywhere. The drive is two way the driver is fullrange but you can biamp an Acoustat and run it as a two way. Lots of info on toroid transformers 9-15 volts to 240 volts will work well and not cost a lot run in parallel you can enhance the bandwidth. YOu can purchase 3 micron polyester film which is more than 5 times thinner than the stock Acoustat film which was 65 gage Mylar HS.
Drive your room as a line source and forget about floor and ceiling bounce. Smooth clean detailed stage and image to make you think the performers are there. Not really all that hard to build. Only thing that is as dynamic are ribbons and horns. Have played with single drivers and modified the heck out of them and they don't do the trick. A used set of Acoustats will provide you with all the parts you need to get going and you can build up your own new panels at your lesiure while you listen to the factory panels. No dynamic driver has the range nor can they compete from distortion point of view. Worth considering.
Thank you moray,
Myself I would prefer as much as possible not to use any crossover on my first project for the reasons I've outlined. But also so that it may serve as something of a point of common reference when I communicate with others; and as a standard for comparison with any future projects that may be more in the direction that you propose.
But, your suggestion moray, at least to me, is a particularly seductive example, I would think, of my third category (that which will result in obvious or perhaps subtle but very real trade-offs leaving the builder perhaps happier but bypassing the single driver experience) and so on behalf of other newbies unfamiliar with the menu I appreciate your including it here.
-jerry
Myself I would prefer as much as possible not to use any crossover on my first project for the reasons I've outlined. But also so that it may serve as something of a point of common reference when I communicate with others; and as a standard for comparison with any future projects that may be more in the direction that you propose.
But, your suggestion moray, at least to me, is a particularly seductive example, I would think, of my third category (that which will result in obvious or perhaps subtle but very real trade-offs leaving the builder perhaps happier but bypassing the single driver experience) and so on behalf of other newbies unfamiliar with the menu I appreciate your including it here.
-jerry
keep in mind...
that any dynamic driver has resistance inductance and capacitance built into it so you don't get a "get out of jail" pass just because you choose to use a single wide range dynamic driver. With the true to life transient response of an esl you can hear things that you will never hear with a wide range driver even a 2 inch unit. I lived a lot of years with esl's and more than half a dozen with all sorts of Fostex drivers and having gone back to esl's I cannot ever see going back to cones. OB mounted drivers don't have the resonances of boxes which you can consider almost impossible to eliminate no matter how hard you try. If you have ever listened to an Acoustat don't be too hasty to write it off as kind of edgy and a little dull before you take a look at the very poor quality of the (very old now) input cap. Try a top quality Blackgate or some quality film cap there and you will be in for a surprise. There is a lot of extra performance that can be milked out of a seemingly old Acoustat if you want it. Good luck with your projects. I wll be interested to see how they go for you. All the best Moray James.
that any dynamic driver has resistance inductance and capacitance built into it so you don't get a "get out of jail" pass just because you choose to use a single wide range dynamic driver. With the true to life transient response of an esl you can hear things that you will never hear with a wide range driver even a 2 inch unit. I lived a lot of years with esl's and more than half a dozen with all sorts of Fostex drivers and having gone back to esl's I cannot ever see going back to cones. OB mounted drivers don't have the resonances of boxes which you can consider almost impossible to eliminate no matter how hard you try. If you have ever listened to an Acoustat don't be too hasty to write it off as kind of edgy and a little dull before you take a look at the very poor quality of the (very old now) input cap. Try a top quality Blackgate or some quality film cap there and you will be in for a surprise. There is a lot of extra performance that can be milked out of a seemingly old Acoustat if you want it. Good luck with your projects. I wll be interested to see how they go for you. All the best Moray James.
You make an excellent point and why I described the suggestion as being so attractive. Perhaps seductive has overtones, but I meant it in a good way.
Anyone who is looking for a solution that as much as possible "gets out of the way of the sound" must seriously consider the option you propose.
It's all about making sometimes easy and sometimes very difficult decisions about trade offs. And what you are saying is that going this route the trade offs are minimal - far less than one would think not knowing the facts. And, most importantly, you move beyond opinion and specifically describe how to achieve this. Which dramatically assists in the difficulty for anyone who is going to spend time researching this possibility. Because it would be best if people could avoid making choices out of exhaustion. And in this field, and with the richness of this forum, I believe that's a real possibility.
I form preconceptions I try hard to make objective - usually after the fact - usually after much work and trepidation. And I have already been a casualty of bias myself. In my case it has caused me to hang on to a couple of ideas that were simply in the end going to take me even further from my claimed goals - and I've barely begun! The lesson to me is the earlier someone new to all of this can have good information about all the options, including yours, the fewer and less influential - or even frustrating - these preconceptions will be.
Hence this thread.
-jerry
Anyone who is looking for a solution that as much as possible "gets out of the way of the sound" must seriously consider the option you propose.
It's all about making sometimes easy and sometimes very difficult decisions about trade offs. And what you are saying is that going this route the trade offs are minimal - far less than one would think not knowing the facts. And, most importantly, you move beyond opinion and specifically describe how to achieve this. Which dramatically assists in the difficulty for anyone who is going to spend time researching this possibility. Because it would be best if people could avoid making choices out of exhaustion. And in this field, and with the richness of this forum, I believe that's a real possibility.
I form preconceptions I try hard to make objective - usually after the fact - usually after much work and trepidation. And I have already been a casualty of bias myself. In my case it has caused me to hang on to a couple of ideas that were simply in the end going to take me even further from my claimed goals - and I've barely begun! The lesson to me is the earlier someone new to all of this can have good information about all the options, including yours, the fewer and less influential - or even frustrating - these preconceptions will be.
Hence this thread.
-jerry
if you have not tried tis do...
with a meter set for continuity check to see if you have electrical contact between the speaker top plate and the rivit that holds the two male connector prongs onto the chassis. If so solder a short wire between the negative terminal connection and to the mounting rivit. This will mean that you are now connecting your amplifiers negative connection to the speakers top plate. Most folks never try this but it is an excellent modification tat can be removed at any time for what ever reason. Substantial improvement in sound. Takes a minute to do and costs nothing see what you think. Regards Moray James.
You might also want to try an after the fact shorted turn. Some speakers (the good ones) have copper sleves on the pole piece as well as copper rings at the top and bottom of the pole piece and top plate. You can do the next best thing by taking some copper wire (solid core) and wrapping it arounf the outside surface on the ceramic magnet slug. Wrap it in a tight coil and fix it in place with epoxy glue, when the glue is set you can solder all the windings together at one point. This will give you a very low DCR wrap or ring (your shorted turn) that will help to lower the speakers inductance a little. It is easy and inexpensive to do and it helps. Every little bit and it all adds up.
with a meter set for continuity check to see if you have electrical contact between the speaker top plate and the rivit that holds the two male connector prongs onto the chassis. If so solder a short wire between the negative terminal connection and to the mounting rivit. This will mean that you are now connecting your amplifiers negative connection to the speakers top plate. Most folks never try this but it is an excellent modification tat can be removed at any time for what ever reason. Substantial improvement in sound. Takes a minute to do and costs nothing see what you think. Regards Moray James.
You might also want to try an after the fact shorted turn. Some speakers (the good ones) have copper sleves on the pole piece as well as copper rings at the top and bottom of the pole piece and top plate. You can do the next best thing by taking some copper wire (solid core) and wrapping it arounf the outside surface on the ceramic magnet slug. Wrap it in a tight coil and fix it in place with epoxy glue, when the glue is set you can solder all the windings together at one point. This will give you a very low DCR wrap or ring (your shorted turn) that will help to lower the speakers inductance a little. It is easy and inexpensive to do and it helps. Every little bit and it all adds up.
This is getting more into after the fact mods which is a separate topic.
But seeing as how you mention it and it's on this thread, I feel it's important to provide a basis or key words so someone unfamiliar with what you are suggesting can google it somehow.
So, in the same way you gave a basis for the improvement in your second point, and for those who would like to understand it better, can you provide a basis or some subject keywords for the first one about connecting your amplifier's negative connection to the speaker's top plate? ie. what electrical property or acoustic phenomenon would this be effecting?
-jerry
But seeing as how you mention it and it's on this thread, I feel it's important to provide a basis or key words so someone unfamiliar with what you are suggesting can google it somehow.
So, in the same way you gave a basis for the improvement in your second point, and for those who would like to understand it better, can you provide a basis or some subject keywords for the first one about connecting your amplifier's negative connection to the speaker's top plate? ie. what electrical property or acoustic phenomenon would this be effecting?
-jerry
Quite honestly, fullrange drivers only perform at their best with tubeamps ... well, FirstWatt wasnt there the years I played with that stuff and neither was notches in common use
You win some and you loose some ... drag in one end of a rope and there will be less at the other end, it is so in life and especially in hifi
You win some and you loose some ... drag in one end of a rope and there will be less at the other end, it is so in life and especially in hifi
tinitus said:
Hi tinitus,
i try some provocation, no offence meant:
What about a high resolution software emulation of the
distortion characteristics of a certain tube amp
- including microphony and all that stuff -
and use it with a - good - SS amp chain ?
Could it work ?
Maybe this way the most beautiful tube sound in the world
could be designed ...
But when i have this idea it is most probable, that some guy
out there had this particular idea before.
---
Yes i know, there are certain interesting traits of a good
tube amp. But how far do you think one can get with a
"tube emulator" ?
Cheers
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.