# Time alignment

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.

#### zedda

Hy
I have a pa system with:
Dbx driverack pa+
Handbuilt sat with 15ndl76 and hf10at on 270*270mm horn 60*40°

Active three way

The center of woofer is 1,70meters high from floor,
And the horn center is 33,6cm higher
The horn depht from baffle to driver is 23cm...
The woofer depth is only 4cm from baffle to dome

I can delay only the woofer, is correct?

But the max delay is only 10ms...

How I can calculate the optimal delay?
If is bigger than 10ms, is better set 10ms or turn off the delay?

#### JonSnell Electronic

Have you any idea what delay means? The speed of sound in a dry atmosphere and at sea level is 1126 feet per second. So, 1/100th of a second delay is 1.126 feet.
23cm is 9 inches, so the delay is negligible and will not be noticed.

#### phase_accurate

10 ms is 11... feet BTW.

Zedda, are you talking of the delay between woofer and tweeter ? This would be around 20 cm (it is actually frequency dependant). Once you get your acoustical slopes sorted you can just invert one driver and vary the delay until the deepest possible suckout at the crossover frequency is reached. The delay with the deepest suckout is the actual delay. You then keep this setting and switch the inverted driver back to normal.

Regards

Charles

#### speakerdoctor

Didn't the OP say his speakers were for P.A. (i.e. public address)? Um, I have to wonder how critical t.a. is for that application.

#### phase_accurate

I'd say it is. Without proper time-alignment you can get non optimal driver integration. I.e. humps, dips and asymmetrical lobing.

Regards

Charles

I'd use 0.55 ms delay on the woofer channel as a starting point. Calculation:

Speed of sound in air at 22 degrees C is 345 m/s (34501 cm/s). The net depth of the horns vs. the woofer that you described is 19 cm, I believe.

19(cm)/34501 (cm/s) = 0.00055 s

You didn't mention the crossover frequency between the woofer and horn. This will tell you the relative wavelength mismatch at center crossover frequency, uncorrected. For instance, if it is 800 Hz, then:

relative phase angle = 158 degrees

Chris

Last edited:

#### speakerdoctor

I'd say it is. Without proper time-alignment you can get non optimal driver integration. I.e. humps, dips and asymmetrical lobing.

Regards

Charles
You cite all good reasons for optimal t.a. in a pair of audiophile speakers, but public address speakers????

#### zedda

I know what means delay, but i can't explain correctly it in english...
I had use a simulator on internet, for 19cm this simulator recommends 12ms...

Ok I try 0.56ms

Thanks

#### zedda

a pair of professional speaker, not public address, sorry

#### speakerdoctor

a pair of professional speaker, not public address, sorry

No problem, my bad. In the USA PA usually stands for public address.
Press on!

#### zedda

I know what means delay, but i can't explain correctly it in english...
I had use a simulator on internet, for 19cm this simulator recommends 12ms...

Ok I try 0.56ms

Edit. Sorry the dbx concede 0.55ms... perfect

Thanks

#### chris661

You cite all good reasons for optimal t.a. in a pair of audiophile speakers, but public address speakers????

So?
They're good components, and ought to be treated as such.

#### Mr. dB

You cite all good reasons for optimal t.a. in a pair of audiophile speakers, but public address speakers????

Time alignment is much more common in PA systems than it is in hifi these days. Above a certain size and price level, you'll find it in every system. All the digital crossover/processor units have the facility, and factory settings for speakers all use it.

#### Mr. dB

How I can calculate the optimal delay?
If is bigger than 10ms, is better set 10ms or turn off the delay?

Most people don't calculate it, they measure it. Measurement tools such as SmaartLive include the ability to measure time. http://www.rationalacoustics.com/

#### zedda

Most people don't calculate it, they measure it. Measurement tools such as SmaartLive include the ability to measure time. http://www.rationalacoustics.com/

The dbx have a rta input for microphone...
If I use the automatic calibration with a undefinied microphone (I have a set of three mic thomann) can I delete the equalization parameters and use only the auto time alignment?

Or can I use the dayton imm-6 with an adaptor?

Thanks

#### zedda

I have a big problem.... in the middle of the dance floor (open space) the medium frequency is cancelled...
The distance from left and right speaker is the same...
Same thing when I turn the speaker looking the center of the floor...
And when I walk on the floor the sound is very innatural, and change with my position...

When I tested the speaker in my house there isn't this problem...

For camouflage the problem I turn the right speaker look right and the left speaker look left...

The change of sound when I walk on the dancefloor was resolved... but in the middle there isn't sound...

What can I do?

Are all the polarities on each channel of your active crossover set the same, i.e., normal polarity, and not inverted? Did you recheck the polarity of the midrange connections?

A picture of your dance floor with speakers would help. Any acoustic reflectors in the area or a low ceiling?

#### zedda

Really open space...
But in my home there isn't any problem...

Ok I want to chek the polarity...
In the driverack pa+ i don't remember the polarity inversion...

#### zedda

Thank you thomann.de
3 hours of inferno...
An externally hosted image should be here but it was not working when we last tested it.

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.