Driver alignment on the A7

[Scottf]

I've read numerous articles discussing physical alignment of drivers on the a7. Generally it's stated that you want your voice coils aligned on the same plane. This seems puzzling to me, I would think what you would actually want is that your sound emitting diaphragms (speaker cone and tweeter diaphragm) be aligned on the same plane. Since electricity essentially moves at the speed of light and sound is considerably slower it seems reasonable that aligning the sound radiating diaphragms would be far more critical than the electro mechanical assembly that animates those diaphragms. With the a7 aligning the voice coils on the same physical plane has the tweeter diaphragm approximately 4 or 5 inches recessed relative to the woofer cone since the tweeter diaphragm is behind the magnet coil assembly and the woofer is mounted in front of the magnet. Seems like the woofer signal would lead the tweeter signal approximately 90 degrees at the 500 Hz crossover point. Any comments?

[marcuswilson]

Most crossovers will cause the HF to lead the LF electrically, before you even get to the drivers. This makes physical placement crossover dependant. The only way to get it right is measure it with a microphone and appropriate gear.

[Fitts]

I too have considered the placement of horn driver to woofer driver and moved my horns forward and backward trying to find the best location.

I think I have some Altec literature (on paper) telling how to align the units but if I recall correctly it requres a rather expensive and specialized audio spectrum analyzer.

I did find in one of my speaker enclosure texts that around 1935 a study was done on phasing between a very long (13') woofer horn and a shorter (2') high frequency horn with various crossovers in the 350 to 800 hertz range. The finding was that the human ear could not detect delays of less than 3 milliseconds. At the speed of sound 3ms translates into about 3 and a half feet.

I think that even thought the ear can't necessarily hear less than a 3ms delay, it would probably be desirable to get the horn and woofer aligned exactly. However, as Marcus pointed out in his post, delays caused by the crossover are going to affect where the horn goes so physically lining up the horn and woofer is probably not going to give you the best alignment.

Long and the short of it, I suggest lining up the horn and woofer and then move the horn back and forth an inch at a time until the sound is to your liking.

If you want me to try to find the audio spectrum analyzer article, I would be glad to look for it.

[marcuswilson]

David,

It is true that the human ear cannot detect short delays, it can detect the frequency anomalies caused by delays between speaker units producing the same sound. The critical problem area is where the two units (woofer and horn) are both radiating the same sound. This occurs at the crossover. If one unit is delayed with respect to the other, then there will be a phase difference between the two, the angle of this phase varying with frequency. The sound from the two speakers will not add properly with this phase difference and the amount of addition will change with frequency.

Professional crossovers have delay adjustments to fractions of a millisecond for driver alignment and delays less than a millisecond are quite noticeable in PA situations. In a domestic situation they may be even more audible.

[Fitts]

Marcus raises some excellent points so it would seem that we should try to find some way to align units either by physical means or through the crossover.

In this case, we can move the HF horn on the A7 cabinet so physical placement is what I will try.

As an experiment, I am going to set up a microphone where I usually sit in my listening room and attach it to an oscilloscope. Then, using an audio generator I will feed a sine wave matching the frequency of the crossover point to my speakers. Doing this, I should be able to "see" on the oscilloscope where the HF needs to be as any phase difference should be apparent in the oscilloscope trace. Once I think I have it, I will reverse the HF polarity and should "see" and hear a reduction in the signal intensity.

I will let you know the results of this test once I am finished. If any one has any additional suggestions, please write.

[David Deratany]

The Altec paper on the subject can be found here:

http://www.soundpractices.com/altec.html

under "Application Notes on Loudspeaker Phasing".

It describes the method to which you are referring.

[stpower]

dderat,
I tried this Altec method in my basement and ended up just leaving the horn mechanically aligned. It seems room effects in a smaller home environment have more of an effect than moving the horn. What seemed right in one measurement location was completely different just moving the spl meter a few inches. The null point seemed to be a moving target.
Regards,
stpower

[marcuswilson]

Dave,

You will find the single tone method of alignment insensitive due to standing waves in your room. Do it outdoors or use the Altec method, again outdoors if you have a small room.

[Fitts]

Well, Marcus was right, a small room is no place to do an alignment.

We have a foot of snow on the ground and right now Mother Nature is in the process of adding another foot. So, I guess I will have to wait until spring to drag my A7's out doors for a test. If it works, I'll post a note for all later in the spring.

[stpower]

Dave,
Apparently Altec was aware of the problem in smaller environments. I notice with the A-7 system , they recommend fixed positions for the 811 and 511 horns. It was with the larger format multicells that they used the movable mounting sleds for the horns. My 210 enclosure instructions mention the aligment as part of the instructions. They knew these systems would be used in huge spaces and would show a benefit from aligment electronically.
Regards,
stpower