Auto-align for dummies

Auto-align for Dummies

SPECIAL THANKS TO

Clem Ong

from the Philippines for his translation of the original french page into english !

You may also want to see the DCX2496 Frequently Asked Questions about the software

1 - What is the Auto-align function for ?

Purpose of this section is not to give a lengthy "know everything" lecture entitled "Phase, it's life, it's works". It is just to learn how to properly use the function of self-align(ment) of our DCX (Uri thank you) to obtain the results that are more probable than winning a lottery draw. For the facetious clever (Behringer) engineers have a few hidden traps that can give spice to the life of an otherwise monotonous way of the audiophile.

Basically, this system should make it possible for the sounds emitted by various drivers to arrive simultaneously at your ear, at your usual listening point. To achieve this, a delay will be applied to the HF (high frequency) whose emission zone (known as acoustic center) is the closest so they are aligned with those whose area of emission is farthest away (midrange/bass driver). Is this important? Yes, because at a minimum, this time lag may introduce irregularities in the response curve.

2 - How shall I use it ?

First, read very carefully pages 10 & 11 (Chapter 4.2.2) of the documentation provided by Behringer, then forget it!

Then proceed with the installation. The microphone, connected to the Input C of the DCX via a balanced cable - remember to select the analog input in the SETUP menu - must be placed at the height of your ears, in your usual listening position and pointed toward the center of the space between your beloved baby (in all senses of the term). On the DCX select "YES" to all available parameters, namely: LONG DELAY, DELAY SHORT, and POLARITIES. You will see hidden farther a few small trappings of our ingenious engineers. Now, using the "muting" of each track, we can choose what we want to have:

If you're configuration 2 x 3 active channels, I strongly recommend you carry out the action on one side at a time. As the microphone is theoretically atop an isosceles triangle, you should get the same results on the right and left, which is a way to check about the reliability of the measure. Provided of course that two speakers are identical (do not laugh, I saw so many "things").

If you're configuration 2 x 2 + channels sub., do not do direct measurement with the subwoofer. In this case the self-align function is indeed likely to create more problems than it solves. I treat this configuration separately. You are lucky, huh!

Now you should be able to get a measurement. You control the output volume knob with the central DCX. In most cases the results are reliable and DCX effectively fulfills its role, but we must keep in mind that things are sometimes, uhâ€¦, slightly more complicated than it seems, let alone Murphy's law!

1st trap: the microphone

Based on your reading, you get out of its wooden box your precious golden electrostatic Bruelheiser & Schoepsmann studio microphone that was entirely built by hand - and you install it as it should on its base. You've turned the volume up, and yet nicht, nada, that slab's display tells you:

Expected, as Uri has not the generosity to provide a real 48V phantom power on Input C. It does produce 15V. "That does it, you say, I'll use a dynamic mic." The result, despite a noise level so high that you do not hear the police who ring your door, you also get this:

Expected, because he also thought to prevent you from using any simple microphone on Input C, with such poor sensitivity! What do you need? Just a microphone with a sufficient sensitivity (-50 to -60 dB). i.e. this: 60.6482-6 or even that: 60.6482-5 . You can also adopt measurement microphone Behringer ECM8000, 60.4164 And voila, some Euros more in the piggy Uri! Well, I have nothing against this microphone - it is sufficiently linear to serve a real-time analyzer. It's up to you.

2nd trap: configuration

That's it, you made a measurement. The DCX played his small pulse train (tchiiit, tchiiit you've heard) and time values have been obtained. Everything seems perfect. Champagne? A little premature perhaps, because the device does not support these values... I know it happened to me! I used my FFT measurement system to verify among other things performance of the self-align function of the DCX. Although my work matched almost perfectly those provided by the equipment, I saw right at the moment that the tweeter was not properly aligned. Until I realized that I had not checked the little box that is necessary... beast, isn't it? On the DCX itself, you must check the "delay / on" on each output (documentation p 17, Chapter 4.5.5):

If you use the DCX software adjustment by computer, do not forget to check the appropriate box, which is at the Long Delay page. That's the same (page) where one never looks because it's not under a domestic system. Remember that Behringer is originally intended for Pro sound.

3rd trap: polarity of the loudspeaker

Some will wonder why let the DCX adjust the polarity and phase of the HP. I have not forgotten my goal from the beginning and I'll try to explain simply what happens: a loudspeaker itself acts as a filter that will be added to filtering the "signal", the one you have set on the DCX, which modifies the phase relationship in the critical crossover area. It is clear, is not it? Okay, an example is simpler: a tweeter overlaps with a midrange at 2500 Hz with a filter of 12 dB / octave. The relative phase between the two is 180 degrees. So by reversing the polarity of the tweeter compared to the midrange, the phase catches up on this difference using the connection. This is what happens in theory.

tweeter with normal polarity (0°)

tweeter with inverted polarity (180°)

In bona fide reality and actually truth, things are more complicated. For the tweeter itself cuts frequencies below 1000 Hz at a rate of 12dB/octave. So below 1000Hz there will be 12 +12 dB / octave or 24dB. The midrange works similarly to 5000Hz but with a slope of 18dB/octave. Under these conditions phase on the connection between the two HP is not at all the same, since it is approximately 270 degrees. The DCX will try to recover it all plumb and square so that the speakers have a phase relationship to the same doctrine in this area and that therefore they overlap without creating an unmanageable valley or peak in the response curve.

Phase difference before alignment = 270°

Phase difference after alignment = 180°

4th trap: "single subwoofer" system

There are two cases :

1) Your system is composed of a single subwoofer placed exactly at the center between two speakers. Everything is simple, proceed as if it were a 3-way active system, the box is aligned with the two satellites.

2) Your subwoofer is placed, say elsewhere. Wherever it is installed, unless it is in the 4th dimension or another floor, places which I declare stupid, you will have to proceed in two stages :

First step: move the subwoofer to put it exactly in the middle, between the two speakers and on the same plane. Of course, I am aware that your box, with a volume of 5m3, consists of a concrete / lead / concrete sandwich 150 mm thick and includes 20, 38cm speakers filled with 10kg of iron. It is workable, just do some weight training beforehand. Once this is done, you're aligning as noted above. The settings completed, get a tape rule or meterstick, take a point of reference on the box and measure the distance between it and microphone measurement point. Second move: put the box in its place of origin. I swear that despite appearances, I do not get paid by the physiotherapists and osteopaths.
Second step : with your tape rule or meterstick, you measure again the distance between your point of reference on the box and microphone measurement point. Now, we must do a little calculation (and it does not copy on its neighbour). Example: placed at the centre of your speakers, the subwoofer was 2.50m distant from the microphone measurement. Located near secretary Louis XV, under the magnificent vase taken from Vallauris, the distance from the box to microphone is now 3.82m. So the sound must now go 3.82m-2.50m: 1.32m further to reach your ears. Knowing that sound travels 344m per second, it will take 1.32/344: 0.0038 seconds, or 3.8 ms. You just have to add this value of 3.8 ms or 1.32m to the respective loudspeakers "satellites" delays, so when the system is played, that the sound of them reaches you simultaneously to the one issued by the subwoofer. You can also save the two configurations (one with and without the additional delay). You will hear quite clearly that the joining of the subwoofer and speakers is really effective with the extra time. When cut (i.e. when there is no additional delay), we can seriously regard the separation of sound from the main speakers.

Warning: This process does not function properly when not at the reference point of listening. That is why, after many experiments, I ended up abandoning the principle of a single listening point. I consider that a basic dimension of enjoyment of music is sharing. We must therefore be able to listen more appropriately at the same time. As in a concert hall, it must be the same at home. Otherwise, use headphones...