Do you time align your drivers ?

[ginetto61]

Hello to Everyone :)
First of all thank you very much for this extremely interesting source of audio informations.
Second, I am very beginner but eager to learn as well.:bouncy:

I understand that when the centers of emission of drivers are not equally distant from the listening point they must be time aligned. :blink:
1) Is there a tolerance on this distance difference ? an acceptable gap ?
2) How do you perform this alignment ? actively with an active crossover or passively ?
3) Is this issue overrated for home applications ?

Thank you sincerely and kind regards,:D
Gino

[BMWCCA]

Sounds like a question intended for the D.I.Y. forum. There are several examples of time-aligned consumer systems from JBL such as the L250, 250ti, and the L-series such as the L7 which simply use slanted baffles to "correct" driver alignment passively if that's what you're asking. Since you seem to be specifically asking about consumer systems, and it's early here and I've nothing better to consider ;) :

Personally, from an untrained and uneducated audio dabbler, I just don't get the physics of the concern. Knowing the speed of sound and that even the worst case of miss-alignment in a normal home system is usually not much more than ten-inches, if we call it twelve-inches (on English foot) that would make the differential in time for the sound to hit your ears of even the most-poorly aligned drivers at something like 0.0009 seconds, regardless of your distance from the speaker. Without compression drivers and horns (L7, 250ti) that disparity is even a fraction of that tiny figure.

I realize the human brain and hearing's ability to discriminate such differences is truly amazing, and important to the species' survival much as is binocular vision, and I have no idea what the threshold of that ability to discriminate the difference, or correct for it, is. I'm sure there's more to the issue that I'm missing, like perhaps harmonics, but the name "time alignment" seems to imply a simple application of physics and the known speed of sound. If I'm missundertanding it, I'm sure the engineers here will correct my assumptions. Even those from the land down-under where they have their own laws of physics. ;)

As a competitive swimmer (in a distant past), we were always taught to launch on the smoke or spark from the starter's gun rather than the sound, so I realize we're a remarkable species in how we process subtle difference in sight and sound even over short distances.

My assumption has always been that true time-alignment is important only in live-sound reinforcement for large installations or where repeating systems are used for PA or SR over a large area. I'm happy to be properly educated concerning the problems in my assumptions and how the phenomenon really applies. :D

[ginetto61]

Thank you very much for your kind and helpful reply.
Moreover I am really sorry to have posted in the wrong forum.
I would be grateful if anyone could redirect this 3D in the correct forum.

Kind regards,
gino

[Robh3606]

I try too when I build my DIY stuff. I use the step response to see how well integrated they look. You have to remember that the alingment is only good in a specific window, point in space, so you shoot for alignment in your listening window.


1) Is there a tolerance on this distance difference ? an acceptable gap ?


Good question obviously you don't want extreme path lengths it also skews the null lobes which can be an issue. You don't want a null on axis.


2) How do you perform this alignment ? actively with an active crossover or passively ?


I used physical driver alignments with a passive crossover. My active system I have not really addressed it and never measured to see what the step response looks like. It sounds fine so don't fix what's not broken.


3) Is this issue overrated for home applications ?

Who knows? Looking at the step response of a 4344 the integration is not all that good at all but it doesn't stop it from being a fine sounding speaker. That said I feel better when the step resonse looks good

Rob:)

[rdgrimes]

Having the L250 with it's time-aligned baffle I have assumed that at least some of what I really like about this speaker is just that. It has a "cohesiveness" and clarity that really shouldn't be due to the drivers or XO themselves. (Having similar drivers and XO in other systems where I don't hear the same level of clarity). In short, I'm sold on the alignment for lack of anything better to point to for the origens of this clarity. It just sounds "right", and listening fatigue is non-existent. It's also true that in the case of the 250s, there are other design features that may well be contributing.

[ginetto61]

I try too when I build my DIY stuff.
I use the step response to see how well integrated they look.
You have to remember that the alingment is only good in a specific window, point in space, so you shoot for alignment in your listening window.

Thank you very much for your kind reply.
I realize that the all procedure involves sophisticated equipment I cannot access to, unfortunately.
I will try to get points of emission alignment in a vertical plane, caring to keep the middle point between the drivers on ear axis (please excuse my very poor english. I have not been able to improved it even if I am trying)

Good question obviously you don't want extreme path lengths it also skews the null lobes which can be an issue.
You don't want a null on axis.
I used physical driver alignments with a passive crossover

This is much more difficult for me to understand. Too technical I am afraid.

My active system I have not really addressed it and never measured to see what the step response looks like.
It sounds fine so don't fix what's not broken. Who knows?
Looking at the step response of a 4344 the integration is not all that good at all but it doesn't stop it from being a fine sounding speaker.
That said I feel better when the step resonse looks good
Rob:)

But this is very important I think.
If I understand well you are listening to a 4344 system without any kind of time alignment and nevertheless get an accurate response ?
How is the soundstage rendition of the system ?
Do you even get a good image depth ?
I know that also the listening room plays an important role
I am looking for that so much amazing effect when it seems that sounds come from points beyond the front wall.
I live in a small flat ... this would give a nice sensation of living in a much bigger space ...:D

Thank you very much indeed
Kind regards,
gino

[BMWCCA]

If I understand well you are listening to a 4344 system without any kind of time alignment and nevertheless get an accurate response ?
How is the soundstage rendition of the system ?
Do you even get a good image depth ?
I know that also the listening room plays an important role
I am looking for that so much amazing effect when it seems that sounds come from points beyond the front wall.
I live in a small flat ... this would give a nice sensation of living in a much bigger space ...:DI have the 4345s in a tiny living room. Nothing time-aligned about those big boxes! And yet when I sit on the couch and close my eyes I'm immersed in the performance. Granted some recordings are much better than others, but my impression of the soundstage is that it's darn good. But then I'm the guy who thinks L7s aren't that finicky when it comes to placement and I enjoy walking around them in the same room and enjoying the feeling of being in the middle of the performance. So, once again...:dont-know

[Mr. Widget]

This is an interesting question and one that I hope to someday do some serious studying of.

First of all, audio is full of pseudo scientific BS and worse. Bearing that in mind, if a speaker is advertised as being time aligned, I wouldn't assume it necessarily to be true.

So what is time alignment? As far as I know there are three types.

A full frequency driver such as a single full range cone or an electrostatic panel. These will inherently be time aligned.

A properly designed coaxial driver can be time aligned.

Lastly we have the vast majority of speakers claiming to be time aligned. These typically have a sloped baffle or a series of baffles that are stepped back. This design can only be really time aligned on axis and at one specific listening height. I am not so sure these designs are truly ever time and phase coherent in absolute terms in the real world.

So how important is time alignment?

That is a huge question. as Mr. BMW said, his 4345 clones are far from time aligned, a horn based speaker with a baffle mounted 2405 will have a step (impulse) response that shows a distinct separate arrival time from the tweeter and horn by several milliseconds. Does this make them unlistenable? No. Can everyone hear this? No. Can anyone hear this? I don't know. I'd like to find out.

That said, when you take a system like the 4345 and correct it's time alignment, it will sound different... better? I am not sure. This is an area that I find most interesting.


Widget

[Robh3606]

Hello Gino

Just to see what we are talking about here are a couple of Step Response measurements. First one is a DIY biamped 3 way and the second a biamped 4344. The step response from a 4345 is going to be very similar. You can see all 4 drivers with the 2405 first the 2122 a close second then the woofer out of phase with the 2425 right in the middle of the woofers response. It's the physical offset of the 2425 mounted on the 2307 that really skews the time alignment. When you start talking about soundstage and so on I would hazard to quess that "Time Alignment" may contibute but it certainly is not going to be the deciding factor. To many other variables to consider.

Rob:)

[jcrobso]

Passive X-overs can add phase shifts, different distance of the voice coil from the mounting baffle can cause time shifts. How can we survive all of this ??:banghead:
The speed of sound for most of us is close to 1 foot per mill-second, this is a general rule of thumb that I have used.

If your speakers have all cone drivers, most of the VCs will be within a few inches of each other, so don't throw your speaker out!
Now if you have horns, the difference could be a whole foot! In the case of my K-horns the woofer is about 6' in back of the mid-range driver. This is because if the horn were straight it would be about 6' long. :blink:
When I wired everything up I did swap the woofer polarity back and forth to get the best sound, I don't remember which way it's set. Phase/time delays are more critical at higher frequencies than at low.
I remember years ago I went over to a friends house to listen to his new speakers, he said he like them but was disappointed with the bass. I listened then went over to one of the speakers and reversed the leads, yes he had wire one backwards. All of a sudden there was plenty of bass. I knew what to listen for, so I knew what the problem was.
As others have said everything is a comprise, To get sound of a full range horn system I have accept a small amount of time misalignment.:o:
Could I fix this? Yes I could, I could bi-amp and put a 6 mill-second delay in the high frequency feed to that amp. So far I not felt the need to do this, everything you do cost you something.;)

[grumpy]

R. Heyser's excellent Klipschorn review from Audio, 1986:

http://forums.klipsch.com/forums/storage/3/1037187/Richard%20C%20Heyser%20KHorn%20Review.pdf

[Robh3606]

This can be a can worms!


Yes but I certainly wouldn't loose any sleep over it. Damn near every commercial speaker system out there is not time aligned and any vertical driver array is only aligned in a small window. A coax driver is the way to go and they are few and far between.

Rob:)

[ginetto61]

Having the L250 with it's time-aligned baffle I have assumed that at least some of what I really like about this speaker is just that.
It has a "cohesiveness" and clarity that really shouldn't be due to the drivers or XO themselves. (Having similar drivers and XO in other systems where I don't hear the same level of clarity). In short, I'm sold on the alignment for lack of anything better to point to for the origens of this clarity. It just sounds "right", and listening fatigue is non-existent. It's also true that in the case of the 250s, there are other design features that may well be contributing.

I have never seen directly a L250, unfortunately
I read that is one of if not the most prestigious JBL home speaker.
I understand that the front panel being not vertical but sloped aligns in the vertical plane the centers of emission of the various drivers.
Is this the real situation ?
Thanks a lot and regards,
gino

[grumpy]

http://www.lansingheritage.org/images/jbl/specs/home-speakers/1982-l250-b460/page02.jpg

[ginetto61]

I have the 4345s in a tiny living room.
Nothing time-aligned about those big boxes!
And yet when I sit on the couch and close my eyes I'm immersed in the performance.
Granted some recordings are much better than others, but my impression of the soundstage is that it's darn good. But then I'm the guy who thinks L7s aren't that finicky when it comes to placement and I enjoy walking around them in the same room and enjoying the feeling of being in the middle of the performance. So, once again...:dont-know

Very clear to me, thank you
Maybe this issue is overvalued after all.
As I understand from another 3D the role of the listening environment is much more important and decisive to get an accurate reproduction
Thanks again and regards,
gino

[jcrobso]

R. Heyser's excellent Klipschorn review from Audio, 1986:

http://forums.klipsch.com/forums/storage/3/1037187/Richard%20C%20Heyser%20KHorn%20Review.pdf

This came out soon after I had built my K-Horns, I forgot the the length is 8'.:biting:
It would be interesting to see how my JBL loaded K-horns would test. I know because of the mid horn I'm using the horizontal dispersion would be better.
The Beyma CP21 tweeter I use is much better than the EV-T35 that K uses.
I love my JBL K-horns and the is what matters,,to me.:)

As others have stated DON'T LOOSE SLEEP OVER THIS!

[ginetto61]

This is an interesting question and one that I hope to someday do some serious studying of.
First of all, audio is full of pseudo scientific BS and worse.
Bearing that in mind, if a speaker is advertised as being time aligned, I wouldn't assume it necessarily to be true.
So what is time alignment?
As far as I know there are three types....
So how important is time alignment?
That is a huge question. as Mr. BMW said, .....
That said, when you take a system like the 4345 and correct it's time alignment, it will sound different... better? I am not sure.
This is an area that I find most interesting.
Widget

Thank you very much Mr Widget
I am fascinated by this issue, but also confused
Nevertheless I am gathering extremely interesting information from this discussion
Someone think that some time disalignment can be tolerate in the real world.
Thank you very much again.
Kind regards,
gino

[rdgrimes]

I think it's fair to say that there are a lot of OTHER things you can be looking at first, to improve audio in your listening room. Such as room treatments, careful layout of the components and furniture and of course the hardware itself. The room can introduce massive time and phase issues very easily which would render moot any speaker adjustments in that realm.

All that being equal, I think time and phase aligned speakers are worth it based on what my ears tell me.

[toddalin]

I believe that my 4430s are time aligned and they do image better than my quasi-L300s, though I do prefer the overall sound of the quasi-L300s. Maybe that imaging is partially a benefit of the time alignment.

[1audiohack]

First of all, audio is full of pseudo scientific BS and worse.


That's for sure!


A full frequency driver such as a single full range cone. These will inherently be time aligned.


Well,,, kinda,,, it will have frequency dependant delay products caused by that long wire coiled in the magnetic gap. Why this seems less offensive I don't know but will speculate that since it is a single driver you don't have another reference creating a step. In effect its acoustic center will not be stable, it will move forward and back in space.



If your speakers have all cone drivers, most of the VCs will be within a few inches of each other.

This gaurantees you nothing, a fact overlooked by many "speaker enfineers." It's back to that inductor in the gap, a woofer or subwoofer is not "slow" due to it's moving mass, that makes it inefficient however, they are slow due to the huge inductor that is the voice coil. In more than a few projects I have done and measured the woofer will still be "later" than the compression driver who's VC is a foot behind the woofers.

This subject is still pretty hotly debated as to how audible it is, here is my experience with time aligning multiway systems.

An example,when the woofers sound slow on say kick drums, it's likely because you heard the tweeter or midrange first. A time aligned system sounds tight and fast.

As some have touched on, the interaction with the room is a large part of what you hear, a time aligned powerflat speaker system sounds better in a much larger sweet spot, sounds much more natural even in another room and seems much more imune to the sonic abberations imposed by the room.

We are currently prepping for the Baje 1000, it is the busiest time of the year for us. I have a lot of measurments I can share when I get my life back in a couple of weeks if there is still interest. Sorry to be short and not well worded.

My opinon on time alignment is it is important.

All the best, Barry.

[MikeBrewster77]

I have a lot of measurments I can share when I get my life back in a couple of weeks if there is still interest.

I'm pretty sure there absolutely is, at least on my part. This is really interesting stuff for a layman like me. :thmbsup:

[Oldmics]

An example,when the woofers sound slow on say kick drums, it's likely because you heard the tweeter or midrange first. A time aligned system sounds tight and fast.
__________________________________________________ ____________

Usually time alignment consists of applying delay to the componets in the box that have there acoustical centers closest to the leading edge (front) of the cabinet.

This allows the rear most componets producing sound to pass thru and catch up to the delayed componets.

The term "time alignment" is not an absolute term.

It should be more along the description of "All devices producing sound will arrive at a specified distance simutainiously".

I have found by manipulating the arrivals of different frequency producing band devices that a sound system can be "taylored" for specific applications.

E.G.-Delaying the low freq section give a system a loose (almost to a point of sloppy if so choosen) rolling low end that is excellent for reggie music.

Thats one of the reasons that a horn loaded system sounds the way they do.

Delaying all but the vocal region gives a "forward" sounding system-great for speech-bad for music.

Delaying all but the tweets gives a pretty bad sounding shrill system.

I dont recommend applying these techniques.Its just some observations we have come to over the years.

So the only real way to understand what a "time aligned" system is is to measure one and see if all bands of freqs are arriving simutainiously-or not.

Oldmics

[Mr. Widget]

Delaying all but the vocal region gives a "forward" sounding system-great for speech-bad for music.I'm not so sure...

I have noticed that my TAD 4003 midrange drivers on their TAD horns tend to sound deeper than a typical JBL system... deeper in stage depth not tone. Well, this combo is physically deeper or more delayed than say a 2440 or 2441 on a short fog horn... OK so that kinda backs up what you are saying, however when Bo and I swapped out his 2421s on his 4345s with a pair of TAD 2002s, the geometry didn't change but the sound stage grew deeper. We didn't measure the impulse response, but I doubt there was a significant change.

Now, one might argue that the JBL sound has a peak in the mids that make it sound forward and the TAD is more neutral so perhaps this sends the stage depth back into the room... well, maybe, but in both cases Bo very carefully EQed each configuration so a simple tonal shift is unlikely the cause.

So what causes the TAD drivers to sound like they are playing music from a spot deeper in the sound stage and the stock JBL to sound more forward? I am not sure.


Widget

[Oldmics]

So what causes the TAD drivers to sound like they are playing music from a spot deeper in the sound stage and the stock JBL to sound more forward? I am not sure.


Widget[/quote]

Its gotta be materials difference.

AAHHHHH,The quest for audio :p

Oldmics

[Robh3606]

Its gotta be materials difference.


I was wondering about that as well Be vs Ti or Al. The Be is better damped and should have a cleaner ETC/CSD plot. Could be you are better able to hear whatever ambiance clues there are so you get a better sense of space and depth??

Face it any JBL speaker using a 2307 or 2312 is going to have issues in the time domain looking at the Step Response. According to the Selecting a Monitor technote on the site they are all below the audible threshhold for Group Delay. Even a 4333/L300 which has the longest midrange horn and the most offset is also below the threshhold.

http://www.audioheritage.org/vbulletin/showthread.php?t=4408

Rob:)

[ginetto61]

Hello Gino
Just to see what we are talking about here are a couple of Step Response measurements.
First one is a DIY biamped 3 way and the second a biamped 4344.
The step response from a 4345 is going to be very similar. You can see all 4 drivers with the 2405 first the 2122 a close second then the woofer out of phase with the 2425 right in the middle of the woofers response. It's the physical offset of the 2425 mounted on the 2307 that really skews the time alignment. When you start talking about soundstage and so on I would hazard to quess that "Time Alignment" may contibute but it certainly is not going to be the deciding factor.
To many other variables to consider.
Rob:)

Hello Rob
Thank you very much for the kind and interesting reply
I am trying to get more information on the subject
Maybe two experimet could be
1) put some increasing delay on one of the two L and R channel and listen for differences in soundstage rendition
2) listen to one speaker with a mono signal and delaying one driver in respect of the other and again listen for variation in sound

I understand it is a very raw approach... not that scientific I am afraid
But often is not so easy to correlate measurements and audio listening impressions

With the occasion I would like to thank you and all the other Members for this very kind disclosing your remarkable expertise with me
Have a nice day :D
gino

[ginetto61]

quote=1audiohack;269123
.... In more than a few projects I have done and measured the woofer will still be "later" than the compression driver who's VC is a foot behind the woofers....
An example,when the woofers sound slow on say kick drums, it's likely because you heard the tweeter or midrange first.
A time aligned system sounds tight and fast

Good Morning !
Very interesting points
Lately my interest in coaxial drivers has increased a lot
In these drivers the tweeter is placed behind the woofer
Even if there the center of emission of the tweeter is some cm behind that of the woofer someone argue if any delay of the woofer is necessary
It would be extremely interesting to know something more about the procedure you follow for time aligning your drivers

As some have touched on, the interaction with the room is a large part of what you hear, a time aligned powerflat speaker system sounds better in a much larger sweet spot, sounds much more natural even in another room and seems much more imune to the sonic abberations imposed by the room.

Another important point. Time alignment seems really a good thing, even if the listening environment is not that good.
Very very interesting.


We are currently prepping for the Baje 1000, it is the busiest time of the year for us.
I have a lot of measurments I can share when I get my life back in a couple of weeks if there is still interest.
Sorry to be short and not well worded.
My opinon on time alignment is it is important.
All the best, Barry.

Thank you very much Barry :D
You have convinced me that time alignment is worth the effort
I will wait for your further information.
And I wish you a great Baje 1000 !
Kind regards,
gino

[ginetto61]

Hello Gino
Just to see what we are talking about here are a couple of Step Response measurements. ... Rob:)

Please excuse me Rob could you tell me something more about the kind of test signal you use to get these very interesting plots ??
Thank you very much indeed
Kind regards,
gino

[tomee]

I don't think it can be easily generalized - a speaker with a passive crossover at 800Hz is different than one at 3000Hz, and 'time alignment' is probably more important for the design with the higher crossover if just to prevent comb filtering effects. Inductance in the crossover and voice coils alters everything with frequency, so at one specific frequency a design can be 'aligned' but then an octave above and below it's not, and this even applies for coaxial designs (eg. the Mastering lab crossover corrected the inherant mis-alignment of the stock Altec 604 using a time delay circuit for the woofer). 'Full range' or single cone speakers often use cone resonance or wizzer cones to extend the frequency range, and when a phase plot is shown it's easily seen that even these are not always 'aligned' at all frequencies. IMHO I think where the 'mis-alignment' occurs is as important as how it's dealt with - so the sound of 'mis-alignment' at 800Hz is probably not the same to our ears as one at 3000Hz (and I don't know which is better. :-))

[4313B]

According to the Selecting a Monitor technote on the site they are all below the audible threshhold for Group Delay. Even a 4333/L300 which has the longest midrange horn and the most offset is also below the threshhold.

http://www.audioheritage.org/vbulletin/showthread.php?t=4408

Rob:)What is really interesting is the paper that came before that wherein JBL discussed that only the 4301, 4311, 4313 and 4315 met the threshold criteria and that all the larger format studio monitors were above the threshold, the 4333 being the worst offender of all. Not that any of us really care though. They all sound groovy, group delay and all. :rotfl:

(If I remember correctly there were two versions of that paper, one of them evidently a first draft)


Face it any JBL speaker using a 2307 or 2312 is going to have issues in the time domain looking at the Step Response.Yeah, alot of horns aren't going to "work right" but we'll muddle through anyway.

Which reminds me... how is your Array project coming? :)

[1audiohack]

Real quick a couple of easy experiments to try with suprising results.

The issue at crossover with two drivers emitting the same frequency can be demonstated with the help of a friend, two small speakers like a pair of pro II's and a pink noise generator.

Drive the speakers at a comfortable level with pink noise, have your friend hold both speakers, one in each hand right next to each other with the sides touching and the front faces even with each other while you listen 10 feet or so away. Have him (or her) slide one speaker forward a few inches then back and observe the changes in what you hear, and let us know. Some say on the alignment issue, if your going to miss, miss big. You will see why.

The issue with the effects of the room, or some of them.

While listening in your usual spot, simply cup your hands behind your ears (giving your self Micky Mouse or Dumbo ears) and note the difference when your not hearing as much of the back of the room. When you do that out doors (like if you have a patio system) the sound does not change that much, in most "living rooms" the change is , well,,, try it. You might end up with a thick fiberglass pillow for you listening pleasure.

I have a pair of Altc 604 8-G's that I haven't listened to in some time and do have the desire to set them back up and get active with them. My real job and life just keep getting in the way.

Best,
Barry.

[ginetto61]

Real quick a couple of easy experiments to try with suprising results....
Best, Barry.

Really interesting Barry. Thank you very much.
Moreover I would be extremely interested to know anything about your time alignment procedure as well.
If now, as I understand, you are very busy please feel free in the future to mail me.
Thank you sincerely again.:D
Kind regards,
gino

P.S. when you talked about BAJE 1000 you mean BAJA 1000 ?? if not what's BAJE 1000 event ? I am quite curious :blink:

[robertbartsch]

Some of this "advanced" theory makes for interesting reading but little else.

About 30 years ago, I pulled some VOTT drivers from a non-time-aligned cab and built time-aligned cabs and installed the Altec divers. At the time, this theory was being advocated in Stereo and High Fidelity magazines as a very important development and space age theory and concluded it is absolute necessary for pure fidelity enjoyment.

Anyway, for me, the results were unimpressive and I would assume any blind tests using similar drivers in similar rooms, etc would result in no significant differences for listeners. After a few months the drivers returned to the factory cabs.

I think this theory is similar to the wonders of magic speaker cables that costs hundreds of buxs but produce no measurable improvment.

[4313B]

I think this theory is similar to the wonders of magic speaker cables that costs hundreds of buxs but produce no measurable improvment.Actually the improvements are measurable. The breakdown occurs when people are asked whether or not they care. Yeah, they can hear a difference... So what!? Then they go back to listening contentedly to their iPods...

[Mr. Widget]

Actually the improvements are measurable. The breakdown occurs when people are asked whether or not they care. Yeah, they can hear a difference... So what!? Then they go back to listening contentedly to their iPods...:yes:

The measured differences are readily seen... huge in fact. Not like some BS cable theory or bringing RF design parameters into an audio discussion.

...and the changes are readily heard, if you set up an A/B comparison.

All that said, most speaker system designs do not lend themselves to a true time aligned design and most of us, even those of us who do not use MP3 files as a source can still be happy with them. ;)


Widget

[ginetto61]

Please excuse me if I insist in on a topic so debatable
But I swear this is the last question I have :o:
If I understand correctly a time aligned speaker usually defines a speaker which have the centers of emission of its drivers on the same vertical plane
Am I correct ?
Thank you very much to everyone
Kind regards,
gino

[Mr. Widget]

If I understand correctly a time aligned speaker usually defines a speaker which have the centers of emission of its drivers on the same vertical planeThat would be some people's definition.

I am not so sure that design will truly yield a perfect impulse or step response. If this response isn't perfect, then the design isn't truly time aligned.


Widget

[jcrobso]

Time or phase alignment sounds very good if you can do it correctly, but alas it is not easy to do. Aligning the voice coils may not be the only problem, don't forget about phase delays caused by the crossover. There are many things to consider, without the necessary test gear you will just be using the SWAG* method.


*Scientific Wild Ass Guess.:blink:

[ginetto61]

That would be some people's definition.
I am not so sure that design will truly yield a perfect impulse or step response.
If this response isn't perfect, then the design isn't truly time aligned.
Widget

Thank you very much Mr Widget :D
Very clear now. The impulse response is the tool.
Kind regards,
gino

[robertbartsch]

I beleive time aligned systems refer to having all the drivers placed so the voice coils of each are all on the same plane (e.g. verticle).

For systems imploying compression drivers and horns, this is no small task since the VC is usually deep away from the baffle/woofer VC.

Yeh, crossovers also cause time delay but trying to cancel this is beyond most hobists.

[1audiohack]

Originally Posted by ginetto61
I swear this is the last question I have :o:


I hope not! That would end all our fun.

I can't type in a hurry, yes I meant Baja 1000.

The Klipsch review is, in typical Richard Heyser style, very informative, Nyquist plots are so cool. Thanks Grumpy, I didn't have that one.

As easy as it is to do, I'll bet no one does the two small speaker experiment. It is hard to believe how much a couple of inches, or micro seconds of time changes the sound when two drivers are radiating the same sound in close proximity to each other. It colors what you hear to a degree that will surprise you. Try it,,, I double dog dare you!

Have a good weekend all,
Barry.

[4343]

I hope not! That would end all our fun.

I can't type in a hurry, yes I meant Baja 1000.
...

Have a good weekend all,
Barry.

Say Hi to my bud Vince at VMAR Racing if you see him. He used to run a Bug with the surfboard on top... Not sure if he's still racing, last time I saw him was a month or so back, we were both driving north past the Bay Bridge and happened to both take Ashby Ave exit and came to a stop side by side at the first stoplight... He was in some sort of Smart Car sized vehicle with his VMAR Racing decal on the back, I yelled out "Hey Man, where's the surfboard", then asked him if that's what he was taking to Baja?. Just got a huge grin back...

[Ian Mackenzie]

Perhaps the point to appreciate is that you can only time align in one point is space.

Therefore, if the listener were 2.5 meters from the loudspeaker the designer can only setup the time alignment of two or more drivers at that distance.

The reason is that if an arch or radius was drawn vertically at 2.5 metres with the horn on axis and the woofer centre placed at a distance of 46 cm below the horn centre the true plane of focus as far a s time alignment is concerned would be 4cm closure to the listener as viewed from the woofer. This is simple geometry.

What is means is that the system will not pass a perfect square way (phase response) at the crossover point unless the woofer and the horn at aligned at that point. At progressively higher frequencies the amplitude response at the crossover point is is more sensitive the time displacement factor "D" because the wavelengths are shorter.

So a 50% displacement error at 4000 hertz (in terms of distance) would only be a 12.5 % error at 1000 hertz (for the same distance)

The human ear is apparently most sensitive in the 1 - 4 hertz region while significant delay known as group delay is readily dectectable at low frequencies where the delay could be 10's of milliseconds behind the upper fequencies.

Provided the crossover slopes are reasonably steep (18 -24 db)the response variations caused by poor time alignment at not a major factor.

Unfortunately other factors come into play that cause the sound stage to collapase other than time alignment such as "how" the power response of the woofer and horn sum at the crossover point.

The paper on the 4430-4435 discusses that stuff in great detail.

There is no question the 4435 images better than the 4345 in the right environment but that does not mean the 4345 or the 4343 are not more fun the listen too!

[Robh3606]

Which reminds me... how is your Array project coming? :)

Hello 4313B

I am about where I thought I would be. I am doing some renovations got most of my Sheetrock up yesterday so I get to tape and Spackle today:barf:Not my favorite part of the job. Target is still Thanksgiving.



Perhaps the point to appreciate is that you can only time align in one point is space.


Agreed but if you try some measurements you see the most change in the vertical. The horizontal seems to hang together so as long as you have the "window" height right I doubt normal distance changes are going to be that significant where you would have audible consequences as long as the driver spacing is reasonable. You are still going to get a tight step response package. It would also seem that you would want to avoid a 3K crossover point like the plague.

Rob:)

[Allanvh5150]

That would be some people's definition.

I am not so sure that design will truly yield a perfect impulse or step response. If this response isn't perfect, then the design isn't truly time aligned.


Widget

To be truely time aligned I guess the front baffle would have to be curved to focus the sound on one point.......

Allan.

[ginetto61]

I hope not! That would end all our fun.
I can't type in a hurry, yes I meant Baja 1000.
The Klipsch review is, in typical Richard Heyser style, very informative, Nyquist plots are so cool. Thanks Grumpy, I didn't have that one.
As easy as it is to do, I'll bet no one does the two small speaker experiment.
It is hard to believe how much a couple of inches, or micro seconds of time changes the sound when two drivers are radiating the same sound in close proximity to each other.
It colors what you hear to a degree that will surprise you.
Try it,,, I double dog dare you!
Have a good weekend all,
Barry.

Thank you very much Barry for your very kind and interesting advice
I think that the all issue is very fascinating
Kind regards,
gino

[1audiohack]

Say Hi to my bud Vince at VMAR Racing if you see him.


I have not seen him in some time. Those old cars were cool for sure. The grommet in the top of the Parker Pumper with the label that said "For Fatties" still makes me laugh every time I think about it.




Perhaps the point to appreciate is that you can only time align in one point is space. Ian


Well,


Critics of synchronization often say that full synchronization can only be achieved at a single point in space. What's overlooked is that signal synchronization, when performed at the overlaps between devices, results in the patterns retaining their directional integrity whereas just inches of missynchronization cause the polar responses of both devices to be corrupted. The above from Don Davis and Eugene Patronis Jr. Sound System Engineering, Third Edition 2006, page 336.



Unfortunately other factors come into play that cause the sound stage to collapase other than time alignment such as "how" the power response of the woofer and horn sum at the crossover point. Ian

Absolutely.



To be truely time aligned I guess the front baffle would have to be curved to focus the sound on one point......Allan.


Pass, then you would also be locked in to a fixed listening distance.

I find some of the most interesting work on the subject is in the array studies by Ralph Heinz of Renkus-Heinz, the Quadratic Throat Waveguide work of Charlie Hughes of Peavey, and the way out front work currently under way by Tom Danley with the Synergy Horn. Those I can't wait to hear!

[ginetto61]

Good Morning from Italy.

All in all I think it is a pity that the coaxial scheme (I am referring to Altec 604 and similar driver) has been abandoned lately.
Nevertheless I understand that this old driver keeps a wide group of admirers even today, with technology evolving
There must be a reson after all ... and I think it could be correlated to its geometry
By the way I do not know if the time alignment issue was considered in design its crossover
In a 2 ways coaxial driver with fixed displacement between the the center of emission of the W and the TW, the W can be easily delayed (I suppose) just designing correctly the inductor in series to it ( not sure about this), passively I mean.
After starting this 3D I look at time response in some speakers review (at Sterophile usually they perform this test)
A correctly time aligned speaker is usually an exception more than the rule
Nevertheless I have become a believer in time alignment worthiness
As usually I do not even try to explain why ... it's a sensation :o:
Have a nice day :D
gino

[Mr. Widget]

All in all I think it is a pity that the coaxial scheme (I am referring to Altec 604 and similar driver) has been abandoned lately.The Altecs and Tannoys are not really time aligned since the compression driver is so far behind the woofer. Of course with a simple delay they can be brought into alignment... group delay and all notwithstanding.

As for a lack of modern coaxial drivers? KEF, Thiel and Pioneer/TAD are all making very good coaxial drivers that are intended to replicate a point source type of sound. All of these are fare more accurate than any vintage design though admittedly they don't have the dynamics or the vintage coloration that some prefer.


Widget

[4313B]

The Altecs and Tannoys are not really time aligned since the compression driver is so far behind the woofer.Depends on where the acoustic center of the horn/c.d. is. Aligning the top plates of the woofer and horn/c.d. doesn't necessarily count.

[Mr. Widget]

Depends on where the acoustic center of the horn/c.d. is. Aligning the top plates of the woofer and horn/c.d. doesn't necessarily count.Agreed, however as I recall they all require a delay on the woofer. They all place the compression driver several inches behind the woofer.


Widget

[speakerdave]

Agreed, however as I recall they all require a delay on the woofer. They all place the compression driver several inches behind the woofer.


Widget

The classic Tannoys, yes. Not the 2558R used in the SRM and SGM 10's and Manley/MasteringLab 10b.

[ginetto61]

The Altecs and Tannoys are not really time aligned since the compression driver is so far behind the woofer.
Of course with a simple delay they can be brought into alignment... group delay and all notwithstanding.
As for a lack of modern coaxial drivers?
KEF, Thiel and Pioneer/TAD are all making very good coaxial drivers that are intended to replicate a point source type of sound.
All of these are fare more accurate than any vintage design though admittedly they don't have the dynamics or the vintage coloration that some prefer.
Widget

Thank you very much for your reply
I said this because in this 3D someone states that a woofer first situation in the time response test is less critical than a tweeter first, that usually is the common situation with bookshelf where the tweeter is mounted on the same baffle of the woofer
Pulling back the tweeter or use a horned one could be an interesting solution, one that I am studying with a great interest
Basically my point is that this kind of Altec drivers, but I could be wrong, keep a very high fame even today ... there must be something of intrinsically very good in their geometry, at least in some situation, like near field listening
This near field situation maybe is quite common in European cities where people live in small flat one over the other ... more or less :D
Warmest regards,
gino

[Ian Mackenzie]

Depends on where the acoustic center of the horn/c.d. is. Aligning the top plates of the woofer and horn/c.d. doesn't necessarily count.

Agreed,

A woofer can be seen as alow pass filter and a horn/driver as a high pass filter. If the drivers have a flat phase response beyond the desired crossover point then the top plate might be near enough.

The reactive aspect of the driver motor will effect the phase response.
A horn will also effect the phase response at the low end due to the varying acoustic impediance. It pays therefore not to run your woofer too high or the driver and horn too low but diy peole do it all the same.

With suitable equipment a phase shift of a driver can be measured however its easier to measure the phase integrity by measuring max cancellation at the crossover point with the drivers out of phase.

So its reasonable to conclude that aside from physical alignment any phase shift of the drivers in the crossover overlap region is going to upset any attempt at time / phase alignment and the response regularity. Add to this the group delay of certain crossover types and it makes the whole topic far from simple.

[JBL 4645]

Real quick a couple of easy experiments to try with suprising results.

The issue at crossover with two drivers emitting the same frequency can be demonstated with the help of a friend, two small speakers like a pair of pro II's and a pink noise generator.

Drive the speakers at a comfortable level with pink noise, have your friend hold both speakers, one in each hand right next to each other with the sides touching and the front faces even with each other while you listen 10 feet or so away. Have him (or her) slide one speaker forward a few inches then back and observe the changes in what you hear, and let us know. Some say on the alignment issue, if your going to miss, miss big. You will see why.

The issue with the effects of the room, or some of them.

While listening in your usual spot, simply cup your hands behind your ears (giving your self Micky Mouse or Dumbo ears) and note the difference when your not hearing as much of the back of the room. When you do that out doors (like if you have a patio system) the sound does not change that much, in most "living rooms" the change is , well,,, try it. You might end up with a thick fiberglass pillow for you listening pleasure.

I have a pair of Altc 604 8-G's that I haven't listened to in some time and do have the desire to set them back up and get active with them. My real job and life just keep getting in the way.

Best,
Barry.

Now that seems make sense. :applaud:

I have time alignment delay on the Behringer DCX2496 for my modified JBL control 5 that have had the passive crossover disabled and the DCX2496 takes over for the rest of their function.

I have tinkered around with the delay to get the best possible sound right down to crazy wild dialogue panning shifts over the matching LCR.

Also if, one of the speakers (left and right has mismatched frequency too much or too less of the other) then its going to sound misbalanced. The centre phantom will be off my so much db and you’ll find yourself sitting in an awkward listening position.



Time or phase alignment sounds very good if you can do it correctly, but alas it is not easy to do. Aligning the voice coils may not be the only problem, don't forget about phase delays caused by the crossover. There are many things to consider, without the necessary test gear you will just be using the SWAG* method.


*Scientific Wild Ass Guess.:blink:

:rotfl:
Now that was funny SWAG!

What about real life then? How is that aligned in terms of time delay I’m just curious?

[jblsound]

While listening in your usual spot, simply cup your hands behind your ears (giving your self Micky Mouse or Dumbo ears) and note the difference when your not hearing as much of the back of the room. When you do that out doors (like if you have a patio system) the sound does not change that much, in most "living rooms" the change is , well,,, try it. You might end up with a thick fiberglass pillow for you listening pleasure.



I don't know that's right. I've held my hands behind my ears and the difference is I hear more of the high end, as those frequencies are more directional.
I don't see that the back wall is the difference, as I sit in fairly high back seats with no full walls directly behind the seat (a large 8'-8" x 7'-0" archway centered on the room's centerline). And with another archway 12 ft behind the first. So any full walls, directly behind the seats are about 35 ft behind the seat.

[JBL 4645]

Even holding your hands next to ears and moving them in and out from distance of 4” makes for phasy like sound as the pressure changes on the ear.

[duaneage]

I've built systems where the drivers ended up in close time alignment and others where they were seriously off. Most alignment errors can be compensated and corrected in the crossover but it's a tedious task. The closer the drivers are to each other on the baffle, and the nearer the voice coils are in the baffle plane the easier it will be.

Plenty of great systems had poor time alignment to begin with. Tannoy concentric drivers have near perfect time alignment but that doesn't make them perfect speakers for some people. A single full range driver, while limited in range , offers natural time alignment as well.

Headphones have perfect TA and can be a good reference. They eliminate the room and outside influences, although the coloration is a matter of preference.

[ginetto61]

I've built systems where the drivers ended up in close time alignment and others where they were seriously off.
Most alignment errors can be compensated and corrected in the crossover but it's a tedious task.
The closer the drivers are to each other on the baffle, and the nearer the voice coils are in the baffle plane the easier it will be.
Plenty of great systems had poor time alignment to begin with.
Tannoy concentric drivers have near perfect time alignment but that doesn't make them perfect speakers for some people.
A single full range driver, while limited in range , offers natural time alignment as well.
Headphones have perfect TA and can be a good reference. They eliminate the room and outside influences, although the coloration is a matter of preference.

Thank you very much for your reply, very interesting
Your point that plenty of great systems had poor TA put a serious question mark on the importance of this property
At Stereophile they perform the TA test and TA speakers are the exception more than the rule.
Nevertheless, as you state, this speakers get very good reviews-
On principle TA seems a nice and "right" property
I have a pair of old but nice Tannoy DC100 sporting a coaxial of 8"
I have also a Behringer cx3400 electronic xover that allows a time delay of up to 2ms on the low output
My idea would be to do some test, when I will find the time
Maybe even with only a mono signal on one channel only, changin the delay on the woofer.
It would be very interesting to hear from coaxial speakers Owners like Altec or Tannoy if they have any experience in this regard.

Thanks again and have a nice day !
gino

[JBL 4645]

Headphones have perfect TA and can be a good reference. They eliminate the room and outside influences, although the coloration is a matter of preference.

I agree with headphones but I don’t care to use them but I do agree since the audio isn’t going to be disrupted by objects that stand in its path. Its good sense to listen on headphones then go back to the loudspeakers and see what is missing in terms of sound imaging detail that makes up for the sound stage.

[1audiohack]

I don't know that's right. jblsound


Your right! In my haste I didn't type half of it. Sorry.

To continue turn around and listen to the back half of the room with your hands cupped behind your ears. Your room by the description of it is probably pretty quiet back there. My living room however is almost as bright off the back wall.

No more half conscious posts from me 'till after Baja. Sorry again.

Barry.

[JBL 4645]

Your right! In my haste I didn't type half of it. Sorry.

To continue turn around and listen to the back half of the room with your hands cupped behind your ears. Your room by the description of it is probably pretty quiet back there. My living room however is almost as bright off the back wall.

No more half conscious posts from me 'till after Baja. Sorry again.

Barry.

A tiny amount of high frequency absorbent placed around the back half sidewalls and along the back wall might reduce the brightness of shrilling highs.;)

How does such a manufacture like JBL time align their LSR models which are active powered?

Do they use narrow range pink noise or full spectrum pink noise or sine wave tone centred near the crossover points?

Where is the microphone placed and are there any pictures as pictures says a lot more.

[ginetto61]

... Its good sense to listen on headphones then go back to the loudspeakers and see what is missing in terms of sound imaging detail that makes up for the sound stage.

Maybe with this type of recording
http://en.wikipedia.org/wiki/Dummy_head_recording
is it possible to get spatial information even with headphone
I have not tried for myself yet, but I still think it could be interesting
In general I like very much that the recording transducers (mics) have a very "similar" geometry with that of the reproducing ones (headphones).
It makes me a lot of sense.
Kind regards,
gino

[Ian Mackenzie]

I will post a pic of a time aligned system (diy) I am building next week.

The problem is not that manufacturers don't want to time align.

Its how!

Often physical constraints make it practically impossible to time align unless done so digitally.

This is often the case with horn/ woofer two way systems.

In a practical example the first take in my diy project I used an Emilar sand cast EH500 horn with a Tad 4001 driver. The total depth to the VC is about 25 cm.

The woofer depth to the VC 13.5 cm.

The difference is obviously 11.5cm

By measuring the path lengths at a fixed point in space of 2.5 metres on axis with the horn we must compensate for about 4 cm between the horn and the woofer.

The difference therefore becomes 7.5 cm.

I arranged this with the dressed baffle using layers of 2.5cm structural plywood above and below the woofer is built up by 5 cm. A 2.5 cm flange around the horn mouth makes up the final amount of alignment. On a baffle of 56 cm width and 120 cm height the profile is visually acceptable.

This makes the front baffle 3 inches thick in some places.

The Emilar EH 500 is an exponential horn so the overall depth is not that great.

If it were a hyperbolic horn the depth would be over 40 cm for the same cut off making time alignment more difficult. Tad solved this challenge by using very steep 6th order crossover slopes on the woofer @650 hertz which by design have enough group delay at the crossover point to align the woofer with the horn.

Is time alignment audible?

I think it depends on the application.

[Allanvh5150]

The difference is obviously 11.5cm

Try to physicaly time align a 2366 and a 2404.....in a cabinet.

[JBL 4645]

Maybe with this type of recording
http://en.wikipedia.org/wiki/Dummy_head_recording
is it possible to get spatial information even with headphone
I have not tried for myself yet, but I still think it could be interesting
In general I like very much that the recording transducers (mics) have a very "similar" geometry with that of the reproducing ones (headphones).
It makes me a lot of sense.
Kind regards,
gino

Yes, yes the dummy head, of course. I read in an article that I found in building a Hi-Fi magazine which I still have, found it around 1989.

Would one of those polystyrene heads work as dummy head the ones that they use for displaying wigs? Drill a few holes in the sides and tuck small microphone inside it.

http://shop.heavenlybody-uk.com/ekmps/shops/heavenlybody/images/ph_f150r(1).jpg


Edit::wtf::blink:
The prices here might make you faint!
I mean look at the face looks more like an android than human face.
http://www.dv247.com/microphones/neumann-ku-100-dummy-head-binaural-stereo-microphone--21004 (http://www.dv247.com/microphones/neumann-ku-100-dummy-head-binaural-stereo-microphone--21004)

Build Your Own Binaural Dummy-Head!
http://www.binaural.com/serendipity/index.php?/archives/109-Dummy-Head-DIY-Binaural-Recording-Head-Instructions.html
http://digdagga.com/dummy/index.html

There are several videos here on this link page for binaural dummy heads
http://www.youtube.com/results?search_query=+binaural+dummy+head&search_type=&aq=f (http://www.youtube.com/results?search_query=+binaural+dummy+head&search_type=&aq=f)

I liked this video for its ambient sound of the garden while it rains, sounds neat with the headphones on at soft level, so that you can hear yourself breathing, if you know what I mean because then it would seem even more real.

Binaural Example (you must wear headphones)
http://www.youtube.com/watch?v=9zQn-Ae9t1g (http://www.youtube.com/watch?v=9zQn-Ae9t1g)

[ginetto61]

Good evening and thank you very much for the very interesting link
looking at those price I am thinking to propose myself as a rentable dummy head :o:
My reasoning is maybe a nonsense but I wonder is there should be some similarities between the geometrical shape of the diaphragm of the microphone and that of the replaying speaker, of course with different dimension
If for example the diaphragm of the microphone is cilindrical than also that of the replaying speaker must be cilindrical, and in this case an omnidirectional speaker should be the best way to reproduce the recorded sound.
I could not find a detailed drawing of a microphone capsule
A scaled-up microphone capsule could result in a very nice speaker
I said .. this is more a nonsense than a reasoning
But I believe is the reason why binaural recording sound so realistic and impressive when listening with an headphone
Thanks again and kind regards,
gino

[lgvenable]

the sound engineer I'm working with has suggested time alignment as I move to a bi-amped or tri-amp system. His suggestion was to use a dBX driverack260 to time align any horns which might be slightly out of phase with the other speakers in a channel which is being bi or tri-amped.

He sets up systems locally, then uses a computer and an RTA to determine the offset to digitally correct the time alignment. The dBX is Harmon product, and contains some JBL speaker parameters in its firmware, price each; from a shop I found on Amazon 550, otherwise 1000 each.

[JBL 4645]

Also I find the time-align with the sub bass extension for the LCR fronts a benefit when running a few frequency sweeps I can see a few gaps between one of each LCR and then run the same sweep over again only with the sub on and the LCR fronts muted on the DCX2496 I can see where the gaps are and when applying one of the LCR one at time with sub bass extension and few adjustments of the time-align it strengths up the gaps within the listening area.

I’ll run a few sweeps and show some examples, may take an hour or two.

[JBL 4645]

Okay knocked a quick example of the difference when adding time delay on one the LCR fronts in this case I used the centre front with sub bass extension ON and OFF and one with time-delay ON and OFF.

You can see the gaps in the frequency response due to the dreaded null room issue that is common in every room. a few peaks and dips. I prefer spectrumlab as it shows the issue quicker over REW and I can run the test repeatedly on (loop) while making some minor adjustments.

The hot colours need to be softened a bit and I’m fresh out of EQ filters on the DCX2496 now.

I used my SPL db metre connected up to the pc as the preamp for my ECM8000 has buggered up on me. :banghead: Sigh at least I have back-up!;)

Note the gab between 78Hz and the time delay seems to fix that tiny little gab.

The gabs between 40Hz and 70Hz and about 80Hz to 105Hz is inexcusable and really needs to be sorted ether with a bit more power and trim the EQ down on the hot spots.

I’ll try the SPL db placed near to the LF and HF and see if it will show up on spectrumlab. Only thing is setting the spectrumlab up for higher frequency response is bit of dogs dinner as I have to change the settings.

[JBL 4645]

As you can see the spectrumlab shows up the gap in the short frequency sweep that I used for 900Hz to 2KHz. I added some delay which seems to fix the slight gap. I’m not going to muck around with the EQ settings as it took me hours to do a few weeks ago.

I had to lower the level of the LF as the colour was too strong and the HF was somewhat less but I can hear the top range.

Not the gaps between delay ON and OFF.

The SPL db was placed vertical between the control 5 at few inches away.

There is that other gap I’m, not that blind yet! That can be fixed with a bit more tinkering around.

[JBL 4645]

That’s just as about far as I can push the spectrumlab I can’t seem to get it any higher than 5KHz. Apart from the short sine wave sweep busting my ears almost the response looks fairly even enough on the centre front.

The noise yes that is noise being picked-up in the lower part of the graph as I live in crappy neighbourhood with loud noisy traffic! (Hey you noisy buggers out there keep it down I’m working in here)!:biting::D

The graph may look better for highs on REW as the resolution gets a bit messy on spectrumlab when running it at higher frequencies.

SPL db was placed at the listening location on this frequency sweep run.

[Dr.db]

This topic is ancient but very interesting and still relevant!


I do believe time alignment cannot be harmful, but it`s simply impractical and visionary to time align a big horn and a woofer in a box!
I don`t like the looks of several drivers flying around the room just to be TA (time aligned). It is a pretty low WAF too, I guess...
So if you`re instructed to mount all drivers in an usual enclosure, I simply don`t see a chance to TA ahorn/woofer such as a E-145 & 2395..
A 6" thicker baffle at the horn`s location wouldn`t look any good...

I´m somehow stuck on my future enclosure design due to the issue of TA.
On the one hand, it seems to make sense. On the other hand there a plenty of fantastic sounding speakers out there with very poor TA...
So what`s the deal....!? :dont-know:

[hjames]

I've had a number of pairs of time-aligned speakers. I owned a pair of Dahlquist DQ10s, nice, but a bit quirky. Replaced them with multiple models of the DCM Timeframe transmission line speakers (Tf400/TF500/TF600/TF700/TF1000). Set up a whole surround system with the larger ones and it was impressive - but sold them all off except the TF600s (they make crazy nice rear surrounds...) After them I bought a couple pairs of Vandersteen 2 series speakers (2CEs and 2CIs) and they are my current keepers down in the TV/multichannel room. None of those were particularly ugly or had a bad SAF ... as a matter of fact, the Vandersteens are actually pretty charming. I also have a pair of KEF 105.4s with time align mid/tweeter top cabinets that are steerable - they have a great sound as well. I've taken a real liking to these various speakers that have a near magical imaging capability.

I also have a pair of customized L200 with 22345. Walnut Smith horn on top with a 2445J, and top mounted 2405 slot at the very top -
I have not taken the time to align them for best effect, but they are a pretty exciting way to listen to my music ... very dynamic.

I'd say if you have a good listening space, try some Time Aligned speakers and see how they work for you.



This topic is ancient but very interesting and still relevant!

I do believe time alignment cannot be harmful, but it`s simply impractical and visionary to time align a big horn and a woofer in a box!
I don`t like the looks of several drivers flying around the room just to be TA (time aligned). It is a pretty low WAF too, I guess...
So if you`re instructed to mount all drivers in an usual enclosure, I simply don`t see a chance to TA ahorn/woofer such as a E-145 & 2395..
A 6" thicker baffle at the horn`s location wouldn`t look any good...

I´m somehow stuck on my future enclosure design due to the issue of TA.
On the one hand, it seems to make sense. On the other hand there a plenty of fantastic sounding speakers out there with very poor TA...
So what`s the deal....!? :dont-know:

[Ian Mackenzie]

Building on this disucssion I have sort details of my current tower floor stander (Legend Acoustic - Kantu 6, Roger Crawford - ex Linn designer).

This designer not only time aligned drivers but made use of customised drivers that could be mounted much closer together than normal and then researched the best overall driver locations for polar response optimisation and matched the crossver Q to drivers selected for good transient performance.

If you look at Zaph Audio site there are numerous tests on all kinds of drivers and he is not afraid to call out the good from the bad.

I think transient performance is often interpreted subjectively as time alignment when poor transient performance is symptomatic of delays or laggs in the impulse response of the signal after ot enters the crossover network and passes though the driver output.

[David Ketley]

I run a 4 way active system using a Marchand active crossover. All units are moveable so I time align by putting a signal through at the crossover frequency and check the response on my lapp top software DB meter. This I find works well for a particular listening position. Just lining up the speaker fronts the system sounds horrible.
I notice some quality speakers also Time Align.http://www.wilsonaudio.com/product_mezzo_tech.shtml

Dave

[Fort Knox]

The Altecs and Tannoys are not really time aligned since the compression driver is so far behind the woofer. Of course with a simple delay they can be brought into alignment... group delay and all notwithstanding.



Widget
I'm assuming your referring to the Altec 511 being so "far behind the woofer"..
as I measure the 811's @ 3.5" further foreword and @ a mod. (corrected) alignment w/woof....
(the 2" and 15" diaphragm align. got factory juggled around (I'm assuming 'cause of the phasing plug
and 828 horn anomalies.....) (but my eq no's show only a negligible correction @ x/o

[speedysteve]

I time align my 5 way system by using Holmimpulse / calibrated mic to measure the 1st peak positive transients, comparing each driver in turn.
Since I went Najda 4 way DSP X/O I can time align even the bass tapped horns, these are some 9ms behind the rest, that's probably about 12 or more feet they'd have to be nearer than the others - not phyically possible in most rooms:)
I do the lower 4 channels in DSP delaying the 3 above the tapped horns suitably. The tweeter is physically time aligned to the upper mid horn by moving it.
At 11000Hz you only have to move the tweeter very small amounts as the wavelength is so short.
Aligning the mid bass to mid horn we are talking moving many inches by comparison. Getting things roughly aligned phyically apart form the tapped horns helped make the alignment in DSP easier.

I like this diagram to explain what needs to be achieved with physical time alignment

http://i727.photobucket.com/albums/ww272/speedysteve7/hifi/Conical horn project/Horn thread on Wam/page38/curve.jpg (http://s727.photobucket.com/user/speedysteve7/media/hifi/Conical horn project/Horn thread on Wam/page38/curve.jpg.html)

What does it sound like once aligned then? Well for me, it's just sound right. The upper frequencies kind of get things to snap into focus better.
With the lower frequencies it is timing and smoothness - I guess as things are aligned and not interfering / cancelling.
I realise there that many reflections going on in a normal room too of course, but I can dep hear the difference between aligned and not aligned.

I suppose aligning with 2nd or 3rd peak would be better than trough against peak. Perhaps some systems do this to be sort of more right?

[Lee in Montreal]

Aligned diaphragm for diaphragm, source for source. Fostex system pictured.

http://img.canuckaudiomart.com/uploads/large/49079-fostex_wood_horn_assembly_d252_rare__tweeter_needs _repair.jpg

[Mostlydiy]

I think you are on the right path there Steve. When me and More10 used his filter on my system we also used holmimpulse to measure all the driver to properly delay all the horns/drivers. The result was very good. I use to have a little dip between the midbashorn and the mids prior to this adjustment but afterwards it disappeared. If you are to go even further I would recommend trying to adjust the phase of the system with a FIR filter. More10 has made a FIR filter of his own using pos Rephase software, that we used and the result night and day. Especially in the bass region. Multiway horn systems benefit alot from this treatment since the horns turn the phase on their own.

/Mostly

[jblsound]

I just time aligned my custom212s. Being they only have 8" MB the recline tilt is only 1.25". I cut out tapered wood doughnuts for mounting the 052Ti tweeters to keep them vertical.
The result is a more coherent sound.

[hjames]

Share the pictures -everyone would like to see!


I just time aligned my custom212s.
Being they only have 8" MB the recline tilt is only 1.25". I cut out tapered wood doughnuts for mounting the 052Ti tweeters to keep them vertical.
The result is a more coherent sound.

[frank23]

I time aligned my 2123 / 2420+2344. I measured it using pulses. It took about 11cm that the 2344 needed to be put forward in comparison to the 2123 baffe to get the time alignment right. In the end, I liked my sound better with the 2344 mounted on the baffle so to have a smoother radiation surface. I read that widget also thinks that smooth radiation might be more important than time alignment.

[jblsound]

Share the pictures -everyone would like to see!


When they are completely done. Still rebuilding the wrap-around grilles.
The avatar is from the mid '90s as they looked with the 066s.
For 12 years I only had a face grille, being they have black ash veneer all around.
And I got the urge to go back to the original look, with black speaker cloth all the way around the boxes. I actually found speaker cloth down here that looks quite similar to the original L212 cloth. A textured weave.

Two years ago I replaced the ageing 066s with a pair of 052Ti. Wish I had done that back when the 052Ti was first available. Far superior.
So anyway, this is the final upgrade.
The total package: mirror imaged, time aligned boxes, Biased XOs (with upgraded components), 052Ti, 2105 midrange.
Only original part, 112A MB driver.

I could have sold these many times over. But imo, I would then have to buy something along the lines of the Revel Salon2, JBL M2, JBL Arrays, maybe the Revel Studio2 to have something better. Those all carry huge price tags.

[Ian Mackenzie]

Have a look at Drew Daniels project for his comments on time alignment.

I think the term is a generalisation for a number of technical related problem in loudspeaker design.

For example people talk of group delay when crossing over a woofer below 100 hertz with certain filter types.

In practise the woofers need to be physically moved 10 feet forward unless a digital delay is incorporated.

What is interesting is that subtractive filters do not demonstrate the same group delay as typical filters.

At much higher frequencies from 1000-4000 the human ear is acutely sensitive to time domain artefacts.

Theses artefacts are often impulse related to diffraction of the baffle, driver transient performance and filter transient performance.

When these artefacts are superimposed on physical time alignment things start to get complicated.

Some designers like J Dunlavy used simple first orders crossovers and good Dynaudio drivers on symmetrical stepped baffles with acoustic foam on the baffles to minimise these artefacts and minimise time alignment

[jblsound]

Share the pictures -everyone would like to see!
OK, the last upgrade. Took longer to do the cosmetic changes than the time aligning. Rebuilt the grilles to original wrap around style, removed the ash veneer from the top/sides and painted them red.
And really went retro with blue cloth....much like my old '73 L55s.
636936369463695

[Doctor_Electron]

Actually the improvements are measurable. The breakdown occurs when people are asked whether or not they care. Yeah, they can hear a difference... So what!? Then they go back to listening contentedly to their iPods...

LOL !!! +Chuckle, guffaw etc. etc. Just TOO real!

Could there be a higly lucrative, esoterics-driven market for Time-Aligned Ear Buds?

Wait, wait, don't answer that!

And I wish the best of the holidays' cheer and joy to all Lansingers, "DE"

[SEAWOLF97]

I used to think that time alignment was a gimmick. Then I got my Walsh F's . The factory describes them as coherent drivers and almost everyone who hears them are astounded by the different experience.

here is the explanation:

The voice-coil and magnet are located where you'd expect them—at the apex of the cone (see fig.1). As the apex of the cone is compressed or stretched, the area around the apex radiates a soundwave into the surrounding air at the speed of sound or about 1127ft/s. At the same time, the vibrations ripple down the cone toward the rim with a propagation velocity characteristic of the cone material. For a typical paper cone the radial velocity through the midrange frequencies is in the range 400–800ft/s. Is it possible to vibrate the rim of the cone exactly when the airwave from the apex is directly over the rim? Such synchronism would guarantee the propagation of a time-aligned coherent wavefront toward the listener, but obviously requires the cone vibrations to travel faster than the speed of sound in air. Walsh showed that this was possible when the cone is engineered with the proper angle and materials.
http://www.stereophile.com/images/archivesart/999Ohmfig1.jpg


Forty years ago, the Ohm F speaker was a game-changer; it still is

The Ohm Acoustics F omnidirectional speaker was such a radical advance in 1972 few 21st century speakers can match its sound.

http://www.cnet.com/news/forty-years-ago-the-ohm-f-speaker-was-a-game-changer-it-still-is/

[Fort Knox]

those conical sp. sound ok 'till the vocals start
then everything gets vary weird:blink:

[SEAWOLF97]

those conical sp. sound ok 'till the vocals start then everything gets vary weird:blink:

IF your opinion was more valid regarding the design, none would have ever been sold.
In fact , many have ..by Infinity, Ohm & the uber expensive German Physiks
(maybe you should notify them that they don't work on vocals ?) :rotfl::rotfl:

Have you EVER heard a good working pair of Walsh drivers ? :confused:

It's really strange how NONE of the reviews agree with you

Ohm Walsh F Floorstanding Speakers



http://crev.vo.llnwd.net/o42/audioreview/images/stars/50star.gif
4.81/5
(27 Reviews)



http://www.audioreview.com/cat/speakers/floorstanding-speakers/ohm/walsh-f/prd_121023_1594crx.aspx

[Fort Knox]

IF your opinion was more valid regarding the design, none would have ever been sold. In fact , many have ..by Infinity, Ohm & the uber expensive German Physiks (maybe you should notify them that they don't work on vocals ?) :rotfl::rotfl: Have you EVER heard a good working pair of Walsh drivers ? :confused:It's really strange how NONE of the reviews agree with you Ohm Walsh F Floorstanding Speakers http://crev.vo.llnwd.net/o42/audioreview/images/stars/50star.gif 4.81/5 (27 Reviews)http://www.audioreview.com/cat/speakers/floorstanding-speakers/ohm/walsh-f/prd_121023_1594crx.aspxAn article, by Ohm, "absolute phase" or something like that ...already knows about the "voice thing.....

[audiomagnate]

IF your opinion was more valid regarding the design, none would have ever been sold.
In fact , many have ..by Infinity, Ohm & the uber expensive German Physiks
(maybe you should notify them that they don't work on vocals ?) :rotfl::rotfl:

Have you EVER heard a good working pair of Walsh drivers ? :confused:

It's really strange how NONE of the reviews agree with you

Ohm Walsh F Floorstanding Speakers



http://crev.vo.llnwd.net/o42/audioreview/images/stars/50star.gif
4.81/5
(27 Reviews)



http://www.audioreview.com/cat/speakers/floorstanding-speakers/ohm/walsh-f/prd_121023_1594crx.aspx


I agree with him. I use to sell OHM Fs when I was in college, and they always sounded odd to me, and particularly colored on vocals. I was into IMFs back then, which are pretty good on voices. I heard the German Physics at 2014 RMAF and was not impressed. I don't even look at reviews from that site BTW. As Fat Bastard says, everybody loves their own brand, and that's all you get there.

[SEAWOLF97]

I agree with him. I use to sell OHM Fs when I was in college, and they always sounded odd to me, and particularly colored on vocals. I was into IMFs back then, which are pretty good on voices. I heard the German Physics at 2014 RMAF and was not impressed. I don't even look at reviews from that site BTW. As Fat Bastard says, everybody loves their own brand, and that's all you get there.

OK ...all those personal opinions add up to .... :moon: as for agreeing with him , he's never heard any.

the same RMAF reviews are made for top end JBL's "they sounded like crap at those shows"

as for those loving their own brand , I have plenty of complaints with "my brands" . Your post doesn't impress me.

yeah, no speaker is perfect , the problems that I have with the OHM's are very minor

[Mr. Widget]

Hey 'wolf... don't take it personally, there are many opinions out there. Then there is the fact that an omnidirectional speaker is going to be affected by the room and location in that room even more than most speakers so different setups will yield very different listening experiences.


Widget

[SEAWOLF97]

I agree with him. I use to sell OHM Fs when I was in college, and they always sounded odd to me, and particularly colored on vocals. I was into IMFs back then, which are pretty good on voices. I heard the German Physics at 2014 RMAF and was not impressed. I don't even look at reviews from that site BTW. As Fat Bastard says, everybody loves their own brand, and that's all you get there.

I'm sure that the 35-40 y.o. recollection is what you believe.

"As Fat Bastard says, everybody loves their own brand, and that's all you get there" .

OK , then let's see what non-owners say ....(OBTW , JH was considered one of the top reviewers in his days at SR) ..pick up at "COMMENT"

[grumpy]

I always came away from large Ohm (with large amp) demos with a smile
(the smaller ones, not so much), but I've never lived with them
so room or recording may have highlighted their strengths.

I'd be proud of them too, especially considering the deal you got on those "F-ers". ;)

[jblsound]

Share the pictures -everyone would like to see! This is the final upgrade. I tore off the ash veneer, as it not survive the long ocean trip. Sick of looking at it anyway after 12 years.What you see now is what I call elephant skin vinyl, 1/16" thick.The vinyl is one continual piece, with the seam on the back.I also rounded the edges on the sides/tops.And new wood doughnuts (for the 052Ti) that fit into the baffle...no nails, no glue...just fiction of a tight fit.647466474764748