delay line settings to time-align DTS-10 tapped horn to mids & high drivers

[cyclecamper]

I have a Rane active crossover with adjustable time dalay for the mid-bass towers and treble ribbon towers. So normally for a folded horn I measure the distance of the acoustical path of a folded horn and use speed of sound to convert that to time. Of course I also adjust that distance and time to accomodate the distance between the mid & treble drivers and front of the horn, etc.So my question is how to determine the correction delay for a tapped horn, like the Danley DTS-10 kit I'm about to purchase. Can't quite measure the acoustcal path length from driver to mouth...when the drivers are so far apart. I intend to cross around 100 Hz.

 

[GM]

Looks like searching 'ms' in this humongous thread will get you there: http://www.avsforum.com/avs-vb/search.php?searchid=17628218

GM

 

[weltersys]

I have a Rane active crossover with adjustable time dalay for the mid-bass towers and treble ribbon towers. So normally for a folded horn I measure the distance of the acoustical path of a folded horn and use speed of sound to convert that to time. Of course I also adjust that distance and time to accomodate the distance between the mid & treble drivers and front of the horn, etc.So my question is how to determine the correction delay for a tapped horn, like the Danley DTS-10 kit I'm about to purchase. Can't quite measure the acoustcal path length from driver to mouth...when the drivers are so far apart. I intend to cross around 100 Hz.

Set an approximate delay for the top cabinet to get close to the path length of the low horn. Using pink noise and an RTA, check that there is not a large dip at the crossover point, if there is the polarity will need to be reversed.

After determining proper polarity, reverse the polarity of the LF. Now there should be a dip at the acoustic crossover point. Change the delay time in short increments longer and shorter until the dip is at maximum.
Then reverse polarity back to normal for use.

Using Smaart I found the dip to be a maximum of about 18 dB when the delay was dialed in properly with a polarity reversal.

A test tone at the crossover frequency and a dB meter can also be used, but since the acoustic crossover frequency is not necessarily the same as the electronic crossover frequency, results won’t be as precise.

One mistake frequently made is swapping polarity, then adding one wavelength of delay time to make up for it.

The other mistake made with path lengths of around 8-12 feet and crossovers around 100 Hz is using no delay at all on the top cabinets. This can still result in a flat frequency and phase response, but the LF will lag by around 8-13 ms , or the time of one wavelength at the crossover frequency.

What we hear first sounds louder, if the LF lags, we tend to turn it up more to compensate for the apparently louder low mids.

 

[cyclecamper]

I sure wish I had the equipment to try some waterfall plots that might show time problems as distinct from phase issues.

 

[paulspencer]

It isn't expensive to get a basic measurement setup. You have more to think about in terms of phase alignement. I have used as much as 23ms delay to the mains just for a sealed sub. With DCX you can do "auto align," the lazy cheat way!