A Question About Time Alignment


I was reading a review of the Wilson Alex V on Stereophile recently. (Published just in time. I’m thinking about picking up a pair. Maybe a couple for the bedroom, too.) And it raised a long-standing question of mine, one that I hope the wiser minds on this site can answer. 
 

Wilson’s big selling point is aligning the different frequencies so they all reach your ear simultaneously. As I understand it, that’s why they have minute adjustments among the various drivers. The woofers put out bass notes that move slowly thanks to their long sound waves while the tweeters are playing faster moving, high frequency notes with short waves. Wilson lets you make adjustments so that they all arrive at the ear at once. 
 

It seems to me, however, that live music isn’t time aligned. Suppose I’m playing the piano and you’re sitting across the room. When I stretch out my left hand to hit the low notes, those notes travel along the same long, slow wavelengths as the notes from Wilson’s woofers. Similarly, the treble notes I play with my right hand move quickly through the short wavelengths. The notes from the piano are naturally out of alignment. If Wilson’s goal is to achieve a lifelike sound, aligning the frequencies doesn’t seem like the way to do it. 
 

Wilson has been selling lots of zillion dollar speakers for lots of years and people continue to gobble ‘em up. Something must be wrong with my line of reasoning. Would someone please point out where I’ve gone wrong? Nicely?

paul6001

Time alignment: a simpler illustration. The acoustic centers of a cone midrange and a dome tweeter when both are mounted in the same plane (e.g., on a flat front of a speaker) will not be in the same plane. The tweeter would typically be ahead of the cone speaker. Thus even if the two speakers are in electrical phase the harmonics from the tweeter will arrive slightly ahead of the lower frequencies of a cone speaker whose acoustic center is near the apex of the cone which could be a couple inches behind the plane if both components are mounted on the same plane. 

@soundscapemd

The acoustic centers of a cone midrange and a dome tweeter when both are mounted in the same plane (e.g., on a flat front of a speaker) will not be in the same plane. The tweeter would typically be ahead of the cone speaker.

 

Monopulse employ this mechanical way of aligning the drive units.

Others claim to be able to do it electrically via the crossover.

Some prefer to horn load their tweeters so that the tweeter is behind the physical plane of the woofer(s).

 

Time alignment might not be the be all end all of speaker design but I think it does help to reduce the so called ’listener fatigue’.

I remember reading that the more unnatural the sound is, the more tiring it is to listen to. The reason usually given is that your brain will have to engage in considerably more processing with those recordings that don’t sound authentic or genuine.

I would bet that this is particularly true when listening to a recording of the human voice.

 

Hold on for one moment. Consider the piano. It's roughly 5 feet wide. So sound from a bass string is going to arrive at the mic way after sound from the treble side. Worse, that bass string is several feet long, so sound will arrive at different times from different parts of the string. Worse yet there are the reflections off the lid arriving all out of phase and time  alignment. So explain how moving a tweeter a fraction of an inch is going to make any audible difference.

Hi Paul,

    I have a time / phase aligned, and room corrected system at the listening position by way of digital crossovers processed through my Mini DSP using a mic that has a 30Hz-30KHz sensitivity range. Let me first speak to the basic principle of time alignment. You are correct, all frequencies being played travel at different speeds to your ears. Assume that the microphone is your ears. It has captured that moment in time of the recording when all those different frequencies met. You want to reproduce that at the listening position so that you hear what the microphone heard. If you time align the speakers, all the frequencies hit your ears at the same moment in time so you then hear the same “time distortion” that the microphone picked up. You are hearing what you would have experienced live. If the system isn’t time aligned you are compounding distortion by the variation in wavelength speeds. The highest sound we can hear is a sine wave about an inch long and the deepest sound we can hear is a wave 25ft long, they travel at very different speed so the higher sounds need to be delayed and the lowest played first. If one wants to truly create a “realistic” experience the system needs to be time aligned. This is not quackery as some have suggested but one of many important pieces of the puzzle in music reproduction. The speakers and their interaction with the room combined with their time / phase alignment are the factors with the most impact on creating a 3-D Holographic image “like being there.” I think physically aligning the system by ear would be next to impossible and the manufacturer can’t do it unless they tell you exactly the distance you would need to sit for them to be aligned as will as how far apart the speakers should be placed. I can turn off time alignment and room correction with a click of the button. Without time alignment the system is flat, with it I would put my listening experience against the most expensive system you can find. I hope this explanation helps. 

Thanks,

Steve

Post removed 

You are correct, all frequencies being played travel at different speeds to your ears

BS

they travel at very different speed so the higher sounds need to be delayed and the lowest played first

BS

This is like high school science stuff.

If it it did work like you say, then the lightening bolt would a crack and woofer notes arriving later. But roll of thunder is more likely stuff we cannot see happening above the clouds.

Please don’t be rude and call science BS because you don’t believe in it. These are measurable results that I can produce which you can hear and isn’t BS in any way.  Your example with thunder and lightening is a great practical high school example. Thunder is caused by the lightening strike by expanding air around the bolt. Light travels at the speed of 186,291 miles per second and sound travels at 1,088 feet per second (depending on the air temp) so if you count the number of seconds from when you see the lightening bolt to the sound you hear and divide by 5 that is roughly how many miles the strike is from you. If there is no time between the strike and the sound find cover! Science my save your life in a thunderstorm and it can help reproduce the best recording reproduction you will ever hear. I learned a lot from this site and feel it is important to respect all views on the forum. If you have measurable scientific data that I’m full of BS I’d love to research it but I’m afraid the entire study of Physics and my ears disagree with you. 
 

Thanks,

Steve

@hifidream Steve, your post was great except for the velocity being a function of the frequency.

I can see how one might think that the velocity is a function of frquency as t links like this show v = freg * wavelength

 

let’s just round off the speed to 1 foot/millisecond.
And at a 1000 feet away the 1Hz takes a second to travel it.
what amount of time does 20Hz take to arrive?
And what amount of time does 20kHz take to arrive?

I have no doubt you MiniDSP sound good, and I suspect that the tweeter to woofer delays are on the order or inches or fractions of a millisecond.

 

if the speed was a function of frequency the the lightening crack would sound like a descending chirp from 20KHz down to 20Hz.
But it doesn’t, we see the lightening and we hear the crack a bit later. When it is so close that it is almost simultaneous our we also tend to crap our britches.

I tried to read every post here but too many have confused light waves with sound waves, light travels as a electromagnetic wave composed of photons, all electromagnetic waves travel at the same speed which is the speed of light.

Sound waves are a vibration, does not contain any photons, the vibration alternately compresses and decompresses the air particles next to it, creating a wave that is composed of compressions and refractions. Sound cannot therefore travel through space where there are no particles to create a sound wave.

Everything affects sound when it is in the air there is no free ride it is not a electromagnetic wave...

I copied and pasted the following for those of you in need of quick refresher:

You can calculate the wavelengths of audible sound in air. Audible sounds in air have frequencies that range from roughly 20 Hz to 20 kHz. Not surprisingly, the wavelengths of audible sounds also vary widely. Assuming a speed of sound of 340 m/s,

For 20 Hz sound in air: λ=vf=340m/s20Hz=17m

 

For 20 kHz sound in air: λ=vf=340m/s20,000Hz=0.017m=1.7cm

 

This calculation shows that wavelengths of sounds in air are distinctly human sized. The wavelengths range from roughly the diameter of a dime (for the highest frequencies) to roughly the length of a city bus (for the lowest frequencies). For comparison, the wavelengths of visible light are all far smaller than the thickness of a single human hair and have a very narrow range (from roughly 400 to 700 nm).

 

  1. The shorter wavelength sound has the higher frequency. Both sounds travel at the same speed.
  2. When the sound goes from cooler to warmer air, it's speed increases (because sound travels faster in warmer air). The frequency doesn’t change (unless the source changes). Since speed increases and frequency is unchanged, the wavelength must increase. Increasing the number for wave speed in the equation λ=v/f
  1. without changing the number for frequency will lead to a bigger value for wavelength.

 

 

 

Time delay is something that your brain uses to locate sounds in space. We are extremely sensitive to it. We can perceive fractions of a second. The reason we know that a sound comes from right center is because our ears and brain are sensitive enough to detect the difference in the timing of the sound hitting our right ear and then traveling across our face to our left ear. If time alignment didn’t matter than we couldn’t locate sound in space. The reason we aren’t completely fooled that we are listening to a live performance is because our left ears are picking up what the right speaker is playing and the right ear is picking up what the left speaker is playing. That is why highly directional speakers can often make a spooky 3-D image if set up and corrected properly but that sweet spot is extremely small (like move your head fractions of an inch small). That is why time alignment and room correction are both important to mitigate room nulls to get the best image possible. There is a ton of disinformation here and I spent years sifting through the chaff to create a system based on factual measurable scientific results. If anyone is struggling to figure out why things just don’t quite sound right the answer is probably in this realm and it is often overlooked because time alignment and room correction are perceived as difficult to do and frankly not many people have heard a system set up correctly employing these techniques. This forum is about sharing ideas. People put great weight into small physical changes in their systems. It would behoove them to respect the power time alignment and room correction can truly make as well. Perhaps a user will find this insightful, easy to understand, and seek to better their system because of it. That’s why I’m still posting from time to time even if I get trolled by someone who doesn’t believe in science. 
 

Thanks,

Steve

Steve,

 

@hifidream You should read up on Head Related Transfer Function research.  It's a lot more than timing.

Having said that, I've heard speakers with great, near perfect impulse responses and they did not by themselves move me.

Hi Eric,

    Yes I’ve based my system on the principles of Edgar Choueiri. The missing piece to my system is his BACCH 3D sound implementation. I didn’t get into all that for the sake of simplicity. Just because a system is set up correctly doesn’t mean you’ll enjoy the sound of the speakers, you are at least listening to them at their best. His work in this field is profound. I am wrapping up paying for my daughter’s wedding and then I’ll be saving for the BACCH 3D package to run on my Mac Mini, I can’t wait!
 

Thanks,

Steve

 

 

 

My creation, the "Mermans" use a JBL 18" (2241H), a JBL ~10" (2251J) and Heil AMT.  Each is separate and I time align the drivers to my seating position by physically moving the three cabinets relative to each other.  I use a Behringer 61-band RTA to set the drivers for the smoothest transition in the crossover region.

 

The result is that, in addition to the sound being its "smoothest," the imaging is greatly improved.  BTW, in this pic, the 10" cabinets are a bit too far forward.

  

 

Hello Everyone,  Since Kenjit mentioned me, I thought that I'd clarify his comments:

I often get a pm from others asking for help and EVERYTIME until Kenji I have helped others.  Kenji contacted me asking for help with his speaker design.  While asking for my help,  He quoted designers, telling me "how it is and how I must design his speaker"...He was convinced that only Linkwitz Riley design could achieve a proper summed response.  I explained that in his case,  I would not be using Linkwitz Riley base design, but in the end would end up with Linkwitz Riley Alignment.  He argued until, I said forget it, do your own.... I had designed a near perfect anachoic flat response and told him that we needed a starting point and a flat response is it.... He accused me of not listening to him and didn't understand that a flat response is the starting point and we could tune from there. He had not at that point did he reveal what type of customization that he wanted .. I asked and found that he wanted a very specific frequency curve.. I did some basic work and again sent him info of how to achieve the frequency curve that he was trying to produce. He then proceeded to tell me that he has been burnt before from someone else and really had no trust for anything that I would do, but wanted me to dive in deep to prove to him that I could achieve a very specific frequency curve that he wanted to achieve. .... What he wanted was not crazy difficult.  I actually did the basic design, but still needed FRD and IMP to give him completely accurate info. He told me again that because of his history and mistrust for me that he may not ever even build what I sent him because crossovers can be expensive. I told him that he had no trust for me,  no respect for my time and no trust.  I would not move forward..... I told him to provide me frd and imp files and I would consider knocking this out for him... He did not send them and I was not going to waste any more time on Kenji.  

Now you see his response to me in this thread.   

Have fun with Kenjit.....

Phase really is time, if drivers are out of phase with each other, frequencies are leaving the drivers at different times, thus effecting time alignment.  I hope this all makes sense and helps in some way.  Tim

No you are just confusing the issue even more. Phase and time are two different things. time cohesion implies phase cohesion but not vice versa. So not equivalent at all. Time coherence between two drivers is about aligning the start of the waves coming out of each driver. You can have phase aligned yet time non aligned. 

Phase and time in speakers is very related, that is why I said in a sense... If you are out of phase, normal time alignment procedures will not time align. 

On speaker design what we align is the portion of each driver where the sound is emitted, which is normally, aligning the front of each voice coil.  This allows all drivers sound to reach the ear at the same time. .... drivers can be staggered or sloped.   

This is also nonsense. 90%  of speakers are not sloped or staggered and nobody is complaining. Time alignement has never been proven to be beneficial let alone audible it is only for marketing. 

It isn't nonsense at all.... I've never said that you cannot have a musically satisfying speaker without absolute time alignment... I was only commenting on how to achieve the objective being discussed, by the way, it is beneficial and audible.  

I am very disappointed in you Tim. Was waiting for your custom tuned circuits and they never materialized.

Actually Kenji,  the disappointment is mine.  You have disrespected me at every turn while asking for my help.... It doesn't sit well to have a Master Speaker tuner and Master Audiophile that is so uncapable of doing his own work to come to me with no humbleness and then post this publicly.... 

~Master Kenjit, speaker tuner and master audiophile.

I haven’t read through this entire thread but for reference, the speed of sound is basically 343m/sec or 761.2 mph. So factor that into your listening distance of what, 8’, 10’, 12’ then, meh!

It doesn't sit well to have a Master Speaker tuner and Master Audiophile that is so uncapable of doing his own work to come to me with no humbleness and then post this publicly.... 

Tim it does not sit well with me that you tried to tune my speakers remotely without even knowing what cabinets I was using. Sorry but any competent speaker designer knows that this is a bad idea if not impossible. There were so many variables to play with yet you were so confident, arrogant even to suggest that you could tune my speakers to perfection first time round. You then insinuated that should they not be satisfactory, you would be very upset with me. This is completely unreasonable.

If you are so confident I dare you to submit your circuit to me and it will be built and then judged to see how good it really is.

I hereby challenge you Tim.

Wow, just wow. I never would have dreamed that this topic could be so 'diverse'. I read one post after another, saying one thing only to be roughly corrected by the next couple of posts. 

 I suppose that my take here is that you want the drivers to launch the audio signal 'in step' with each other. Also to be taken into account is the ear/brain acceptance of receiving of the audio. Let's say that 8 milliseconds are insignificant to the ear/brain acceptance (meaning that this draws a meaningful amount of attention). OTOH, 30 milliseconds are another matter, and will be noticed. I am not using actual millisecond references, but just rough ideas of time.

 If I were to design a loudspeaker, I certainly would want a certain amount of alignment between drivers, whatever that number in milliseconds needs to be. If mounting the drivers makes no discernable difference to ear/brain were golden.

 

  

@paul6001     Sorry that I derailed your thread with Kenjit,  but overall and I can tell you that a handful of people that gave you answers had a very good understanding and gave accurate answers. It looks to me that regardless of the many answers, most concur.  The speed of sound makes all frequencies travel at the same speed regardless of wavelength.  Phasing does matter to some degree, but is not the be all end all.  

I can tell you that there are differences.  I have taken a nice 2way out of a standard type of cabinet and switched it to a sloped front cabinet. To be fair, also changed the crossover a bit to compensate for listening angle at the seating position, but in the end.  I was able to add depth to the soundstage as well as a more pinpoint placement of instruments...  Overall,  in some speakers, it may not be dramatic, but in others, depending on design, it very well could be.   Good luck,  I hope this helps in some way,  Tim

@jerryg123 

It is an audiophile that has achieved a higher level of knowledge and understanding than other audiophiles about how to obtain perfection. We know what to listen for and we do not settle for anything less than perfection.

@kenjit LOL!!!!! How are you at fishing and baiting hooks? Makes you a what?

Go stroke that ego. 

Lets get back to the OP’s question.

 

It seems to me, however, that live music isn’t time aligned. Suppose I’m playing the piano and you’re sitting across the room. When I stretch out my left hand to hit the low notes, those notes travel along the same long, slow wavelengths as the notes from Wilson’s woofers. Similarly, the treble notes I play with my right hand move quickly through the short wavelengths. The notes from the piano are naturally out of alignment.

There is not “slow and quickly” happening as the frequency changes.
The notes all travel at the same speed across the room.

If the woofer and the tweeter are both involved, with say a note in the middle of the cross over, then we want both of those to be launched so as to arrive at the ears in unison.
Hence this is fractions of an inch or a few inches at most.

One usually either steps the speakers spatially, or with a tilt, or uses a DSP to align them electrically. And each sample of a 44.1 kHz signal is roughly a 1/4” from the previous or next sample.

@4krowme  “Wow, just wow. I never would have dreamed that this topic could be so 'diverse'.”.

 

This is why I leave equipment design to the designers,  ignore the marketing, and just use sound as a judgement on the equipment I buy.

ghdprentice,

 

 I mostly agree, yet it can be at least entertaining to DIY for some things. I would NOT dare to make a decent TT complete with arm!

  When there is a proven design that can be counted on, (usually speakers in my case) I am encouraged to go the route and build it to my taste.

 

  Partially for thinking about better time alignment, I chose to use full range drivers in my builds.

Back in my old days at Marcof,  Ed Built a few Turntables that were very good, however,  He modified several arms, but I have never seen him attempt to build an arm.  

I think that if you have the machines and knowledge, it would be possible. I wouldn't be able to get there, lacking in both.

 

 I did see a local guy who built his own tonearm using wood. I think that he understood the concept, but the wood taught him a lesson by a slight warp. It was flat and thin, and not so well made in my opinion. I didn't even give it a second look as to what was the pivot. He was proud but should have been embarrassed.

You need to start someplace.

Bet these guys built a few duds.

I did see a local guy who built his own tonearm using wood. I think that he understood the concept, but the wood taught him a lesson by a slight warp. It was flat and thin, and not so well made in my opinion. I didn't even give it a second look as to what was the pivot. He was proud but should have been embarrassed.TRI ART AUDIO TA-2 Turntable

jerryg123,

 That tonearm is way above what I saw. Again, if the right machines are available, that could push me over the edge.

 

 Would it be correct to say that some form of time alignment starts at the source? 

Even with my Dual Concentric Tannoy Berkeley’s where the pepperpot tweeter is located back into the mouth of the horn loaded 15 inch drivers the bass does sometimes seem to be slightly lagging behind the rest of the sound.

However, this may be due to cabinet resonance issues rather that drive unit alignment time differences.

Thankfully the Berkeley’s are said to crossover at 1kHz and thus everything above that should be aligned.

When it comes to low bass (sub 40 Hz) very few systems can match a good subwoofer for either clarity or power.

It can also be problematic to describe bass as either fast or slow but there’s no doubt that some systems can display superior attack and decay of bass notes than others.

And then there are those that don’t bother at all. In fact I’d guess that the bass response of the majority of loudspeakers starts to rapidly fall off the frequency response cliff below 50Hz.

In those instances, time alignment should hardly be an issue.

 

Would it be correct to say that some form of time alignment starts at the source?

@4krowme no. Microphones and electronics generally close to perfect.

The speaker dominates time and phase alignment by orders of magnitude compared to everything else.

Don't move your head even one inch while listening or your whole time alignment will be off.

 

Don't move your head even one inch while listening or your whole time alignment will be off.

I’m more worried about my chakras being aligned… 😋

Lots of theories abound and some even result in better sounding gear! I'll say that the Vandersteens I've owned have definitley had some special coherence that was enjoyable for me throughout the 2 and 3 series and seemed to be less dependent on equipment.

And that the Daedalus Athena's I now have are ethereal and heavenly.

The ears have it!

The time it is taking to read all the comments on this topic does not align with my morning schedule.

I hope to pick it back up when I have more time this afternoon.

like kids arguing about HP and valvetrains when not much of that horsepower makes it to the ground…. excellent systems engineering rarely focuses on a variable or two… Time, Phase, pistonic motion, low diffraction, energy storage in the radiating surface and enclosure, ringing in the filters, doppler in a coaxial driver…and the list goes on…..

and tastes vary…not everyone loves a Stradavarius…. or a Telecaster…