• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Time Alignment in Speakers

FrantzM

Major Contributor
Forum Donor
Joined
Mar 12, 2016
Messages
4,337
Likes
7,720
Hi

I would like some clear explanation on what it takes for a speaker to be time-aligned. I always thought the crossover was where it had to start. With first order crossovers being the only ones effectively allowing time alignment. Is it possible through moving (or the different drivers? Could that (Time-alignment) be maintained though the passband by such mechanical means?
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,834
Likes
16,496
Location
Monument, CO
<Please bear in mind that I am not a speaker designer, have not built my own in years, and those grad acoustics classes were decades ago.>

Another simple question (albeit a very good one!) with a complex answer. The basic idea is for all frequencies to arrive at the listener at the same time, i.e. in alignment, assuming they all started together at the source. The crossover can be tuned and phase/delays adjusted electrically and/or mechanically to provide correct time alignment whatever crossover order is used. A first-order crossover does not ensure perfect time alignment; far away from the crossover, the highs and lows will be out of phase (sometimes you will see drivers wired "out of phase" in the box and, assuming nothing is wrong, that is one reason why). The drivers and box design (sealed vs. vented, choice of size, etc.) also affect time alignment. Some designs stagger physical positions of the drivers to improve time alignment, but that is also affected by the crossover, since the crossover (as well as drivers and boxes) influence phasing. There is only so much you can do with driver positioning since the phase varies over frequency. Broadband correction these days means DSP since filters and delays can be readily designed in the digital domain to correct crossover and mechanical influences.

How much time alignment matters, i.e. how audible it is, is debatable. I have leaned towards speakers with good time alignment but they have other attributes that I prefer so I really could not say if my preference, or how much of it, is due to time alignment.
 
Last edited:

watchnerd

Grand Contributor
Joined
Dec 8, 2016
Messages
12,449
Likes
10,408
Location
Seattle Area, USA
One of the claimed benefits of electrostatic speakers is time and phase coherence across an extremely broad range and physical space.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
I believe it is true that unless a single or coaxial driver the time alignment will only occur in one direction or plane. Also for large panels the sound will not be in complete time alignment except up close in nearfield because different parts of the panel are at different distances. Quad ESL 63s were quasi point sources using concentric rings. Only a point source would be fully time aligned. Those Quads came close to a point source because the center ring started moving first then the next ring, and the next and so on with delays between. The delay was close to how sound would spread from a real point source in air.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
One of the claimed benefits of electrostatic speakers is time and phase coherence across an extremely broad range and physical space.

for large panels the sound will not be in complete time alignment except up close in nearfield because different parts of the panel are at different distances.

I see measured flatness here - down to 200Hz or so (cross to cone at 180), and projected 10 feet to the listening position.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
I see measured flatness here - down to 200Hz or so (cross to cone at 180), and projected 10 feet to the listening position.
Just a thought. Can you show us a square wave or saw tooth at a few different frequencies? And if not a huge bother at listening position and up close to the speakers.
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,834
Likes
16,496
Location
Monument, CO
RayDunzl has previously shared his impulse responses, and they were excellent...

The wavefront from a large planar speaker should be coherent all across the driver, since the panel is driven identically across its surface, and so the part that hits your ears is coherent; the reason for alternative schemes has to do more with dispersion and panel modes than time alignment IME/IMO.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
Can you show us a square wave or saw tooth

It doesn't really make pretty squares:

Without and with DRC at 80, 250, 2681 Hz, both speakers playing, at listening position

upload_2017-1-8_14-3-9.png
 

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,834
Likes
16,496
Location
Monument, CO
I think the square waves show crossover and other issues between the panels and woofers. The sloping tilt indicates dropping LF response.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
Data point:

On thing I was surprised by in the past, was how closely the calculated Step response (from a log sweep tone) matches the recorded step response - here, of the 80Hz square...

From REW Sweep:

upload_2017-1-8_14-10-30.png


From audacity audio recording 8oHz square

upload_2017-1-8_14-12-52.png
 
Last edited:

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
I think the square waves show crossover and other issues between the panels and woofers. The sloping tilt indicates dropping LF response.

The speaker can't "pressurize" this room - how could it hold a level at low frequencies? So, it slopes. The pressure wave it can generate passes on by... The cone may be extended (flatly) but it isn't doing anything to the air when it is extended and motionless.

At 80Hz the cone would have to keep moving out for 6.25ms, then reverse direction for 6.25ms to keep making the compression wave... It doesn't. It moves out according to applied voltage and sits there. That's my thought.
 
Last edited:

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,834
Likes
16,496
Location
Monument, CO
Yes, that is a good thought, sometimes the obvious slips me by. I was thinking more of the shape of the square waves, without DRC, but the response is so good any comments would be taken as nits. BTW, what is "DRC"? The only audio definitions I recall off-hand are "digital/dynamic room correction" and "dynamic range compression".

The excellent matching between calculated and measured step response indicates the time and frequency domains are nearly ideally related. What usually throws them off are modes and/or later reflections that show up in one or the other and require careful gating of the sweeps to provide such close congruence. In the past I have had much better impulse responses than I do in my current system, something I would like to remedy, but just haven't had or taken the time to tweak.
 

fas42

Major Contributor
Joined
Mar 21, 2016
Messages
2,818
Likes
191
Location
Australia
As a data point, at one stage I used a cheapy boring box on one channel, and a Technics construction aligned on the other channel, this one:

a.jpg


Didn't cause any audible effects that I could pick - sound stage, etc, with electronics in tune was as per normal.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
BTW, what is "DRC"?

It's shorthand version of "Digital Room Correction" which I think of more as "DDSC" as in "Distant Digital Speaker Correction".

It does nothing to "correct" my room, though it does modify the output of the speakers to measure better at the measuring (synonymous here with listening) position, to give the sometimes audible impression of playing better in the room.
 

Cosmik

Major Contributor
Joined
Apr 24, 2016
Messages
3,075
Likes
2,180
Location
UK
It's shorthand version of "Digital Room Correction" which I think of more as "DDSC" as in "Distant Digital Speaker Correction".

It does nothing to "correct" my room, though it does modify the output of the speakers to measure better at the measuring (synonymous here with listening) position, to give the sometimes audible impression of playing better in the room.
If that is an unalloyed good, would you consider BAACH and other methods for reducing the crosstalk between speakers? Your measurements would be even better.
 

watchnerd

Grand Contributor
Joined
Dec 8, 2016
Messages
12,449
Likes
10,408
Location
Seattle Area, USA
As a data point, at one stage I used a cheapy boring box on one channel, and a Technics construction aligned on the other channel, this one:

a.jpg


Didn't cause any audible effects that I could pick - sound stage, etc, with electronics in tune was as per normal.

What are all those tiny holes in the front?

Can't be bass reflex ports...
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
If that is an unalloyed good, would you consider BAACH and other methods for reducing the crosstalk between speakers? Your measurements would be even better.

Not at the price I saw.

$54,000
 

fas42

Major Contributor
Joined
Mar 21, 2016
Messages
2,818
Likes
191
Location
Australia
What are all those tiny holes in the front?

Can't be bass reflex ports...
Apparently so. It's the total area of opening that matters, and Technics decided to try this approach on this one it seems - from the late 70's. This pair was used with my DIY chip amps - conventional sensitivity, power handling - still fully operational.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
Top Bottom