• 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!

Measuring Time Delay

TimW

Major Contributor
Joined
Aug 15, 2018
Messages
1,065
Likes
1,406
Location
Seattle, WA
Sorry if this has been covered in a tutorial already but I have not been able to find much information on it. I'm trying to integrate a subwoofer into a 2.1 system. I can apply delay to the main speakers to compensate for the delay of the subwoofer caused by DSP and added distance. To find the necessary delay I attempted to follow this guide using REW and a umik-1. I followed the directions and got the first measurement of the left speaker as my acoustic timing reference. Then I set the output to the right speaker and attempted to measure my subwoofer. The problem is that the chirp REW plays before the frequency response measurement is well... a high pitch chirp. My subwoofer, even with the low pass filter disabled, does not play the chirp at sufficient volume. REW gives me a message that says something like system delay could not be measured. And of course it can't if it's asking the subwoofer to play high frequency chirps. I set the sub and preamp to max volume and still could not get REW to recognize the quiet chirp. What am I doing wrong? How do other people do this? Is there a way to change the acoustic timing reference to a lower frequency that the subwoofer can actually play?

For reference my sub is powered by the Dayton Audio SPA250DSP amp which is performing all of the DSP. Source is an SMSL Sanskrit 10th taking an optical signal from a TV.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,202
Likes
16,982
Location
Riverview FL
That might be a good question for @mitchco...

I have a similar situation, and haven't quite figured it out myself.
 

amirm

Founder/Admin
Staff Member
CFO (Chief Fun Officer)
Joined
Feb 13, 2016
Messages
44,368
Likes
234,389
Location
Seattle Area
You can set the start and end of the chirp frequencies in the measurement dialog. Is this what you are asking about?

Also, you can use a tape measure and use 1 millisecond per foot.

Actual delay may not be the most optimal by the way.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,202
Likes
16,982
Location
Riverview FL
You can set the start and end of the chirp frequencies in the measurement dialog. Is this what you are asking about?

The test sweep, yes, the acoustic timing reference "beep/sweep" , I don't think so.

The timing reference (1) is a quick high frequency sweep that comes before the actual test sweep (2).

1558764594986.png
 
Last edited:

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
Sorry if this has been covered in a tutorial already but I have not been able to find much information on it. I'm trying to integrate a subwoofer into a 2.1 system. I can apply delay to the main speakers to compensate for the delay of the subwoofer caused by DSP and added distance. To find the necessary delay I attempted to follow this guide using REW and a umik-1. I followed the directions and got the first measurement of the left speaker as my acoustic timing reference. Then I set the output to the right speaker and attempted to measure my subwoofer. The problem is that the chirp REW plays before the frequency response measurement is well... a high pitch chirp. My subwoofer, even with the low pass filter disabled, does not play the chirp at sufficient volume. REW gives me a message that says something like system delay could not be measured. And of course it can't if it's asking the subwoofer to play high frequency chirps. I set the sub and preamp to max volume and still could not get REW to recognize the quiet chirp. What am I doing wrong? How do other people do this? Is there a way to change the acoustic timing reference to a lower frequency that the subwoofer can actually play?

For reference my sub is powered by the Dayton Audio SPA250DSP amp which is performing all of the DSP. Source is an SMSL Sanskrit 10th taking an optical signal from a TV.

The point of measuring time delay is that "Timing Reference Output" signal is always played by the same speaker. In the article you linked it is the left main speaker. When you measure right main speaker the output is sent to the right speaker but timing reference signal is again played by the left speaker, as shown here:

Capture.JPG


To measure the sub you first need to route front right input channel to the sub output channel via the channel switching matrix (as shown in the article you linked) but you again leave front left speaker to play the timing reference signal. Having the left front speaker ALWAYS playing the timing reference sets the time reference point so the timing delay of all other speakers is measured relative to it.

Doing so you won't have a problem as sub will never have to play the timing reference signal anyway. ;)
 
Last edited:

MZKM

Major Contributor
Forum Donor
Joined
Dec 1, 2018
Messages
4,240
Likes
11,463
Location
Land O’ Lakes, FL
Another way, as described in this video.

1) Do a normal measurement of L+R channels (assuming you have a crossover in play). If just caring about the subwoofer integration, limit to say 20Hz-200Hz.
2) Go to Spectrogram.
3) Go to Spectrogram settings and change Mode from Fourier to Wavelet (the video has the axes flipped as well).
4) Check to see if the subwoofer is in-line with the speakers at time 0. Adjust phase/delay if not.

EDIT: After getting close to being aligned with the wavelet, do what @Kvalsvoll suggested and do an RTA, play a sine wave at the crossover point and adjust phase/delay until output is at its max. I would do the wavelet first though, as there is a possibility of being phase-aligned but offset by a cycle.
 
Last edited:

Kvalsvoll

Addicted to Fun and Learning
Audio Company
Joined
Apr 25, 2019
Messages
878
Likes
1,643
Location
Norway
Fortunately this is much easier to do, just measure the whole system with subwoofers, and analyze this measurement. No need to mess with timing and try to measure the delay, you will never get it right that way.

The correct delay depends on several factors, such as crossover frequency and slopes (acoustic), the subwoofer, location of sub and speakers, room acoustics. But don't worry about that.

Adjust delay to get close to flat/zero group delay, fine tune in .5ms/20cm steps so that freq resp sum correctly, use RTA. It is possible to get flat phase (overall) and flat group delay, depends on the crossover slopes you ended up with, but some deviation from a flat phase/zero GD curve is not important.

There are a couple of articles in the blog-section on my web page with some information on bass systems and calibration, there you can see examples of measurements and what to look for (link is in signature below).

I don't know if this breif descripiton makes sense. If there is interest for it, I may put it up in a future article. I already have it in my user manuals.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,202
Likes
16,982
Location
Riverview FL
3) Go to Spectrogram settings and change Mode from Fourier to Wavelet (the video has the axes flipped as well).
4) Check to see if the subwoofer is in-line with the speakers at time 0. Adjust phase/delay if not.

Hmmm....

My "normal" settings

1558811015829.png


And with 80ms delay on the Cheezewoofers just to assist in figuring out what I'm looking at:

1558811083287.png
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,202
Likes
16,982
Location
Riverview FL
Adjust delay to get close to flat/zero group delay, fine tune in .5ms/20cm steps so that freq resp sum correctly, use RTA. It is possible to get flat phase (overall) and flat group delay, depends on the crossover slopes you ended up with, but some deviation from a flat phase/zero GD curve is not important.

Ok...

Normal (1/24th octave smoothing):

1558811313602.png


And with 80ms Cheezewoofer delay just for grins:

1558811335938.png


Uh-huh...
 
OP
TimW

TimW

Major Contributor
Joined
Aug 15, 2018
Messages
1,065
Likes
1,406
Location
Seattle, WA
Thanks for clearing that up @Krunok, it makes sense now and I was able to get delay measurements for my sub with this method. I was surprised to find the delay was about 30 ms. At first I took a measurement of the sub from 10 to 200 Hz and got an even longer delay. Then I tried measurements from 30 to 200 Hz and 45 to 200 Hz where results were more consistent at about 29.7 ms. My DSP only allows up to 10 ms of delay on the mains so I have that maxed. Honestly the difference between no delay and 10 ms is only mildly noticeable while listening to music. What do you plan to do about delay @RayDunzl? I'm not sure what's going on in that last graph but it doesn't look pretty.
 

Attachments

  • 10 - 200 Hz Sub Delay.PNG
    10 - 200 Hz Sub Delay.PNG
    51.4 KB · Views: 233
  • 30 - 200 Hz Sub Delay.PNG
    30 - 200 Hz Sub Delay.PNG
    51.2 KB · Views: 244
  • 45 - 200 Hz Sub Delay.PNG
    45 - 200 Hz Sub Delay.PNG
    51 KB · Views: 258

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,202
Likes
16,982
Location
Riverview FL
What do you plan to do about delay @RayDunzl? I'm not sure what's going on in that last graph but it doesn't look pretty.

The second Wavelet shown is with a delay of 80ms on the Cheezewoofers, just to see what it looks like, to get a better idea of what I'm looking at with a normal (first Wavelet) measurement.

Time is in the vertical direction, later is higher in the picture, so with 80ms delay there is a second image of low frequencies above the image of the mains (which run full bandwidth - the subs overlap the mains in my setup)

---

Oh, you mean the Group Delay image.

I'd call it related to comb filtering somehow - 80ms experimental delay on the subs vs normal on the overlapping mains.
 

mitchco

Addicted to Fun and Learning
Audio Company
Joined
May 24, 2016
Messages
640
Likes
2,397
That might be a good question for @mitchco...

I have a similar situation, and haven't quite figured it out myself.

Time aligning subs is really difficult. The issue is that the peak of the sub or the portion one wants to time align spans many samples, so where does it start? There is good advice from @Krunok and others. I have tried all of the ways mentioned and several other approaches. The "only" approach that have found that consistently works is with a commercial software, Audiolense and verified with REW.

I have a three way system with dual subs:

F18 front.JPG


Even though the subs are lined up as best as I can using physical measurements, Audiolense still picks out the individual delays for each speaker in a stereo setup:

time offsets.JPG


Looking at the measured step response that correlates with the delays:

non time aligned 50ms.JPG


After Audiolense time aligns all drivers and removes the maximum phase reflection:

Audiolense step.JPG


And verified with REW:

JBL 4722 with Dual F18 subs Step.jpg


And for fun the REW wavelet:

JBL 4722 with Dual F18 subs wavelet.JPG


Finally, this is just not for the single mic position at listening position in my 9ft triangle, but holds true across my 6ft 3 seat couch. I can post a series of measurements showing that, but for now, probably enough.

Cheers,
Mitch
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,202
Likes
16,982
Location
Riverview FL
And for fun the REW wavelet:

Wow.

Having perfomed this display for the first time here, and before seeing your result.

Mine isn't so purty.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,202
Likes
16,982
Location
Riverview FL
Mine isn't so purty.

Interesting, though...

Right side - where the room is rectangular

1558845424462.png


Left side, where the left rear corner of the otherwise rectangular room is wide open to the kitchen area:

1558845537924.png
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
Thanks for clearing that up @Krunok, it makes sense now and I was able to get delay measurements for my sub with this method. I was surprised to find the delay was about 30 ms. At first I took a measurement of the sub from 10 to 200 Hz and got an even longer delay. Then I tried measurements from 30 to 200 Hz and 45 to 200 Hz where results were more consistent at about 29.7 ms. My DSP only allows up to 10 ms of delay on the mains so I have that maxed. Honestly the difference between no delay and 10 ms is only mildly noticeable while listening to music.

Huh, 29.7ms is app 10 meters. Is your sub really 10 meters further away from your front left speaker?

Can you post delays for left and right speaker?
 

MZKM

Major Contributor
Forum Donor
Joined
Dec 1, 2018
Messages
4,240
Likes
11,463
Location
Land O’ Lakes, FL
Huh, 29.7ms is app 10 meters. Is your sub really 10 meters further away from your front left speaker?

Set distance for subwoofers should always be further away than they actually are, as that accounts for the delay due to their internal DSP. If you ever run automatic room correction, the subwoofer is always set further, I have Audyssey and it does the same, and I checked with REW that it was pretty bang on. However, it usually sets it around 2-5ft further than its physical difference, so 10 meters is very odd.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
Set distance for subwoofers should always be further away than they actually are, as that accounts for the delay due to their internal DSP. If you ever run automatic room correction, the subwoofer is always set further, I have Audyssey and it does the same, and I checked with REW that it was pretty bang on. However, it usually sets it around 2-5ft further than its physical difference, so 10 meters is very odd.

Exactly - that's why I'm asking..
 
OP
TimW

TimW

Major Contributor
Joined
Aug 15, 2018
Messages
1,065
Likes
1,406
Location
Seattle, WA
The delay of the main speakers was less then 0.10 ms. I also found the delay of the sub confusingly long. It is not within site of the listening position but it is only a few feet back from the speakers at most. The thing about the delay of the subwoofer caused by DSP is that the main speakers are receiving signal from the same DSP. The signal for the main speakers is only being modified with a high pass filter and delay.
 

Krunok

Major Contributor
Joined
Mar 25, 2018
Messages
4,600
Likes
3,065
Location
Zg, Cro
The delay of the main speakers was less then 0.10 ms. I also found the delay of the sub confusingly long. It is not within site of the listening position but it is only a few feet back from the speakers at most. The thing about the delay of the subwoofer caused by DSP is that the main speakers are receiving signal from the same DSP. The signal for the main speakers is only being modified with a high pass filter and delay.

I still don't understand what is causing such huge delay in subwoofer. Which miniDSP device are you using? Maybe it would be more clear if you post screenshot of your config screen.

Btw, are you using high pass filter on your mains to roll off bass to make life easier for them? If that is so I think that is a good idea.

Which sub are you using? Are you setting low pass filtering on the sub or within miniDSP software?
 
Top Bottom